The Issue Whether the adult open heart surgery rule in effect at the time the certificate of need (CON) applications were filed, and until January 24, 2002, or the rule as amended on that date is applicable to this case. Which, if any, of the applications filed by Martin Memorial Medical Center, Inc. (Martin Memorial); Bethesda Healthcare System, Inc., d/b/a Bethesda Memorial Hospital (Bethesda); and Boca Raton Community Hospital, Inc. (BRCH) meet the requirements for a CON to establish an adult open heart surgery program in Agency for Health Care Administration (AHCA) Health Planning District 9, for Okeechobee, Indian River, St. Lucie, Martin, and Palm Beach Counties, Florida.
Findings Of Fact The Agency for Health Care Administration (AHCA) is the agency which administers the certificate of need (CON) program for health care facilities and programs in Florida. It is also the designated state health planning agency. See Subsection 408.034(1), Florida Statutes. For health planning purposes, AHCA District 9 includes Indian River, Okeechobee, St. Lucie, Martin, and Palm Beach Counties. See Subsection 408.032(5), Florida Statutes. AHCA published a fixed need pool of zero for additional open heart surgery programs in District 9, for the January 2002, planning horizon. The mathematical need formula in the rule, using the use rate for open heart surgery procedures in the district as applied to the projected population growth, indicated a gross numeric need for 7.9 programs in District 9. After rounding off the decimal and subtracting four, for the number of existing District 9 open heart surgery programs, the formula showed a numerical need for four additional ones. The need number defaulted to zero, however, because one of the existing programs, at Lawnwood Medical Center, Inc., d/b/a Lawnwood Regional Medical Center (Lawnwood), had not reached the required minimum of 350 surgeries a year, or 29 cases a month for 12 months prior to the quarter in which need was published. Having initiated services in March 1999, the Lawnwood program had not been operational for 12 months at the time the applications were filed in October 1999. The other existing providers of adult open heart services in District 9, in addition to Lawnwood, are Palm Beach Gardens Community Hospital, Inc., d/b/a Palm Beach Gardens Medical Center (PBGMC); Columbia/JFK Medical Center Limited Partnership, d/b/a JFK Medical Center (JFK); and Tenet Healthsystem Hospitals, Inc., d/b/a Delray Medical Center (Delray). All are intervening parties to this proceeding. In the Pre-Hearing Stipulation, the parties agreed that the Intervenors have standing to participate in this proceeding. Despite the publication of zero numeric need, five hospitals in District 9 applied for CONs to establish new adult open heart programs asserting need based on not normal circumstances. Three of those applications are at issue in this case: CON Number 9249 filed by Martin Memorial Medical Center, Inc. (Martin Memorial); CON Number 9250 by Bethesda Healthcare System, Inc., d/b/a Bethesda Memorial Hospital (Bethesda); and CON Number 9248 by Boca Raton Community Hospital, Inc. (BRCH). AHCA initially reviewed and denied all of the applications. After changing its position before the final hearing, AHCA supports the approval of the applications filed by Martin Memorial and BRCH. Martin Memorial Martin Memorial, the only hospital in Martin County, and the only party/applicant not located in Palm Beach County, operates two facilities, a total of 336 beds, on two separate campuses under a single license. The larger hospital, in Stuart, has 236 beds and is located approximately 20 miles south of Lawnwood and 30 miles north of PBGMC. Martin Memorial owns and maintains, at the hospital, its own ambulance service used exclusively for hospital-to-hospital transfers. The drive from Martin Memorial to Lawnwood averages 38 minutes. The drive time to PBGMC averages 48 minutes. By helicopter, it takes 11 or 12 minutes to get from Martin Memorial to PBGMC. The remaining 100 Martin Memorial Hospital beds are located on its southern campus, approximately six miles south of the Stuart facility. Martin Memorial is a private not-for-profit hospital, established in 1939. The parent corporation also operates an ambulatory care center, physician group, billing and collection company, and a foundation. Martin Memorial is applying to operate an open heart program at its Stuart location, where it currently offers cardiology, hematology, nephrology, pulmonary, infectious disease, pathology, blood bank, anesthesiology, diagnostic nuclear medicine, and intensive care services. Martin Memorial has a 25-bed telemetry unit, a 14-bed medical intensive care unit, a nine-bed surgical intensive care unit, and a 22-bed progressive care unit, with an identically equipped 16-bed overflow unit used only for high seasonal occupancy, from approximately December to April. If its CON is approved, Martin Memorial will dedicate four surgical intensive care unit beds and six progressive care beds for post-open heart surgery patients. Martin Memorial agreed to condition its CON on the provision of 2.4% of the project's gross revenues for charity care and 2% for Medicaid. The total estimated project cost is $6.5 million. Martin Memorial intends to affiliate with the University of Florida and its teaching facility, Shands Hospital, to assist in establishing the program and training staff. The cardiovascular surgeon is expected to be a full-time faculty member who will live and work in Martin County. Although initially opposed, AHCA now supports Martin Memorial’s application primarily because (1) it has the largest cardiac catheterization (cath) program at any hospital in this state which does not also provide open heart services; (2) it has a medium size and growing Medicare population, which constitutes the age group most likely to require open heart surgery and related services; (3) Martin County residents now must receive open heart and related services at hospitals outside Martin County, primarily in areas ranging from Palm Beach County south to Dade County; (4) emergency heart attack patients who present at Martin Memorial-Stuart could receive primary angioplasties without transfer; and (5) it is a not-for-profit hospital, while all of the existing open heart providers in the District are for- profit corporate subsidiaries. Of the applicants, Martin Memorial is also located the greatest distance from the existing providers. Bethesda Memorial Bethesda has 362 licensed beds located in Boynton Beach. JFK is nine miles north or an average drive of 18 minutes from Bethesda. Delray is nine miles south or an average drive of 17 minutes from Bethesda. Established in February 1959, Bethesda is a not-for- profit subsidiary of Bethesda Health Care Systems, Inc., which also operates some for-profit subsidiaries, including Bethesda Medical/Surgical Specialists, Bethesda Management Services, and Bethesda Comprehensive Cancer Institute. Bethesda is a disproportionate share provider of Medicaid and Medicare services. The services currently available at Bethesda include obstetrics, Level II and III neonatal intensive care, cardiology, orthopedics, pediatrics, neurological and stroke care, peripheral vascular surgery, wound care, pulmonary and infectious disease care. Bethesda recently eliminated a 20-bed unit for adult psychiatric services, and a 20-bed skilled nursing unit. Currently, at Bethesda, the sickest patients are placed in a 10-bed critical care unit. The hospital also operates a 12- bed surgical intensive care unit, an eight-bed medical intensive care unit, and 30 and 25-bed telemetry units. Bethesda was planning to open a 20-bed extension to the telemetry unit, all in private rooms, in January 2002. If an open heart surgery program is established, Bethesda, will add an eight-bed cardiovascular intensive care unit to care post-operatively for the patients. Bethesda offered to condition its CON on the provision of 3% of total open heart surgeries to Medicaid and 3% of total open heart surgeries to indigent patients. Bethesda's estimated total project cost is $4 million, $1.7 million for equipment, and $2.24 for construction. Bethesda will receive assistance from Orlando Regional Medical Center in training personnel and developing protocols for an open heart program. At Orlando Regional, a statutory teaching hospital, the number of open heart cases ranges from 1,300 to 1,600 a year. Bethesda has a contract with a physicians' group to provide a board-certified cardiovascular surgeon to serve as medical director for the open heart program. AHCA’s position is that the Bethesda application is "approvable" but, of the Palm Beach County applicants, less desirable than that of BRCH. By contrast, Bethesda's experts emphasized (1) the absence of any overlap with the Lawnwood market; (2) the greater need for a new program, based on the volume of cases, in Palm Beach County than elsewhere in the District; (3) the size, growth, and age of the population within Bethesda's market area, and (4) the ability of Bethesda to enhance access for underserved groups, particularly Medicaid patients. Boca Raton Community Hospital BRCH is licensed for 394 beds. Located in southern Palm Beach County, close to the Broward County line, BRCH is from eight to nine miles south of Delray and approximately 15 miles north of North Ridge Medical Center (North Ridge), in adjacent Broward County. On average, the drive from BRCH to Delray takes 20 minutes. The drive from BRCH to North Ridge takes about 25 minutes. Founded in the late 1960's, BRCH operates as a not-for- profit corporation. BRCH has a staff of 750 physicians and 1,600 employees. Services at BRCH include cardiology, a 10-bed Level II neonatal intensive care unit, hematology, nephrology, pulmonology, radiology, nuclear medicine, and neurology. If approved and issued a CON for adult open heart surgery, BRCH will build a new facility for the program, including two new cath labs, an electrophysiology lab and 12 intensive care beds. In the CON, the estimated construction cost was $16.5 million and the estimated equipment cost was $2.7 million of the $20 million estimated for the total project. BRCH agreed to having conditions on its CON (1) to provide 5% of open heart cases in year two to uninsured patients, (2) to establish an outreach program to increase the utilization of open heart services among the uninsured, and (3) to relinquish the CON if it fails to perform at least 350 open heart surgery procedures a year in any two consecutive years after the end of the second year of operations. AHCA determined that it should change its initial position opposing the approval of the BRCH application to one of approval because of (1) the large Medicare population in the service area; (2) the volume of emergency room heart attack patients; (3) the district out-migration for services primarily to North Ridge; (4) the large, well-developed interventional cardiology program; and (5) the not-for-profit organizational structure. When AHCA decided to support the approval of the BRCH application, it did so, in part, based on erroneous data. The cath lab volume was assumed to be approximately 1,800 caths a year, as compared to the actual volume of 667 caths for the year ending March 2001. Having considered the corrected data, AHCA’s expert described BRCH’s application as significantly less compelling, but still preferable to that of Bethesda. BRCH is the largest hospital in number of beds in Florida which does not have an open heart surgery program. AHCA also responded favorably to identified "cultural" access issues, described as underservice to demographic groups, based on race, gender, and class. BRCH presented a plan to equip a mobile unit to provide diagnostic screenings and primary care in underserved areas. Pre-Hearing Stipulations The parties stipulated that all of the applications met the statutory requirements concerning the application content and filing procedures of Sections 408.037 and 408.039, Florida Statutes (1999), and Rule 59C-1.033, Florida Administrative Code. Martin Memorial, Bethesda, and BRCH have a history of providing quality care. See Subsection 408.035(1)(c), Florida Statutes (1999). There are no existing outpatient, ambulatory or home care services which can be used as alternatives to inpatient adult open heart and angioplasty services. See Subsection 408.035(1)(d), Florida Statutes (1999). Martin Memorial and Bethesda have sufficient available funds for capital and operating expenses required for their proposed open heart surgery programs. See Subsection 408.035(1)(h), Florida Statutes (1999). Martin Memorial complied with the requirements related to costs and methods of construction, and equipment for the proposed project. Except for the contention that it omitted $1,687,180 in fixed equipment costs and that the proposed construction project is excessively large and expensive, the parties stipulated that BRCH reasonably estimated construction and equipment costs, including costs and methods of energy provision. See Subsection 408.035(1)(m), Florida Statutes (1999). The parties agreed that Subsections 408.035(1)(p), and 408.035(2)(e), Florida Statutes, related to nursing home beds, are not at issue at in this proceeding. If Bethesda, BRCH, and Martin Memorial can recruit the necessary, competent nursing and surgical staff, they will meet the requirements of Rule 59C-1.033(3), (4)(b), (4)(c), and (5)(c), Florida Administrative Code. Adult open heart surgery services are currently available to District 9 residents within the two-hour travel standard of Rule 59C-1.033(4)(a), Florida Administrative Code. Bethesda, BRCH, and Martin Memorial are accredited by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), assuring quality as required by Rule 59C-1.033(5)(a), Florida Administrative Code. The parties agreed that if Bethesda, BRCH, and Martin Memorial can recruit the necessary nursing and surgical personnel, their programs would meet the requirements of Rule 59C-1.033(4)(b), (4)(c), (5)(b), and (5)(c), Florida Administrative Code, except that JFK and Lawnwood did not agree that the applicants satisfied the requirements related to cardiovascular surgeons. Martin Memorial will be able to obtain perfusionist services, as required by Rule 59C-1.033(5)(b)5, Florida Administrative Code. Bethesda and BRCH projected reasonable staffing patterns, in their CON schedules 6A, given projected census levels, although the ability to recruit staff and adequacy of projected salaries are at issue. The rule criteria related to pediatric open heart surgery are not applicable to this proceeding. Disputed Statutory and Rule Criteria The following statutory criteria and applicable in this case: Subsections 408.035(1)(a), (b), (c) - for comparison; (e), (f), (g), (h) - related to funding for BRCH, and related to staff recruitment and salaries; (i), (j), (k), (l), (m) - for Bethesda, and related to the size, scope, and fixed equipment cost for BRCH, (n), and (o); and Subsections 408.035(2)(a), (b), (c), and (d), Florida Statutes. The criteria in Rules 59C-1.030, and of Rule 59C-1.033(5)(b) - related to staffing, except as stipulated - are at issue. The parties have also raised the issue of whether AHCA is consistent in applying its agency rules related to open heart cases. The District 9 health plan contains two preferences for open heart applicant hospitals, the first for hospitals with established cardiac cath programs, the second for applicants with a documented commitment to serve patients regardless of their ability to pay or county of residence. All of the applicants have established diagnostic cardiac cath programs and related cardiology services. During the cardiac cath procedure, a catheter is inserted into a cardiac chamber to diagnose heart disease. During a therapeutic cardiac cath procedure, or angioplasty, the catheter with a balloon-tip is inserted into a coronary artery and inflated to open blockages. The latter requires open heart surgery back-up in case a vessel is ruptured and thus, an open heart surgery certificate of need. Martin Memorial operates the largest cardiac cath program at a hospital in Florida which does not also offer open heart surgery. At Martin Memorial, 1,885 inpatient and outpatient caths were performed in 1999, 1,770 in 2000, and 1,286 in the first nine months of 2001. Cardiac caths are only performed at the Stuart facility. Non-invasive cardiology services began in the 1970's at Martin Memorial. A CON to establish the first cardiac cath lab was issued in 1989, and a second, CON-exempt cath lab opened in 1998. Martin Memorial also offers pacemaker implants and peripheral angioplasties to eliminate clots in other areas of the body, for example, in the legs, electrocardiography, echocardiography, stress tests, and cardiac rehabilitation. Neither electrophysiology studies nor defibrillator implants are performed at Martin Memorial. Martin Memorial has an open staff of cardiologists, meaning that its cath lab is available for use by any of the invasive cardiologists on staff. The facilities include two cardiac cath procedure rooms, a control room for the laboratory, a five-bed holding room and a two-bay inpatient recovery area. Bethesda also has an established cardiac cath program with an open staff. Seventeen cathing physicians were listed on the Bethesda roster for the month of March 2001. Of those, five were also the only cardiologists allowed to perform caths at the closed lab at JFK. Some of these cardiologists are permitted to perform emergency angioplasties at Bethesda. Bethesda has, at least, two cardiovascular surgeons on staff. From 1995 to 1998, the volume of cardiac caths at Bethesda increased over 60%, from 133 to 213. For the 12 months ending August 31, 2000, Bethesda cardiologists performed 428 caths. For the 12 months ending September 30, 2001, the cath volume was 506 cases. Currently, cath procedures at Bethesda are performed in one lab with recently upgraded digital equipment. As part of the planned expansion of the hospital, the existing lab will be relocated and a second one added. Permanent pacemakers are implanted at Bethesda, but internal cardioverter defibrillator procedures, electrophysiology, and table studies are not performed. Cardiac cath services, at BRCH, started in 1987. Two cath labs with state-of-the-art digital equipment are used. In the 12 months ending March 31, 2001, there were 667 inpatient and outpatient caths performed at BRCH. Currently, cardiac services at BRCH are the largest source of admissions, approximately 20% of total admissions. The available services include echocardiography, tilt table studies, electrocardiography, stress tests, cardiac wellness and rehabilitation programs, electrophysiology studies, and internal cardioverter defibrillator implants. Each year, one or two "rescue" or salvage angioplasties are performed in extreme, life- threatening circumstances at BRCH. Forty-nine cardiologists are on the closed "invitation-only" medical staff at BRCH, 47 are board-certified and approximately half are invasive cardiologists. The staff also includes seven electrophysiologists, five of whom are board-certified, and seven thoracic surgeons, five of whom perform open heart surgeries at other hospitals. For the first two years of operating an open heart program, BRCH intends to have a closed program, by virtue of an exclusive contract with a single group of cardiovascular surgeons. Subsection 408.035(1)(a) - district health plan preference for serving patients regardless of county of residence or ability to pay; and Subsection 408.035 (1)(n) - history of and proposed services to Medicaid and indigent patients Martin Memorial, Bethesda, and BRCH will serve patients regardless of residence and, they contend, will enhance access for Medicaid, indigent, charity and/or self-pay patients. Each applicant has offered to care for patients in some of these categories as a condition for CON approval. The proposed conditions, are, for Martin Memorial, 2.4% of total project revenues for charity and 2% of admissions for Medicaid patients. Martin Memorial provides a number of services without charge, including follow-up education to former inpatients to assist them in managing diseases such as asthma, diabetes, congestive heart failure and chronic obstructive pulmonary disease. Obstetric care includes one free home visit by a nurse/midwife to check the health of newborns and mothers. Office space is provided for a free clinic for the "working poor" of Martin County, which receives approximately 10,000 annual visits from a patient base of about 2,000 patients. Over $100,000 a year is provided for an indigent pharmacy program. Combining the outreach services with other charitable contributions, including charity care, Martin Memorial valued "community benefits" at $24 million in 1998, $30.5 million in 2000. When Martin Memorial received an inpatient cardiac cath CON, it agreed to provide a minimum of 2.5% of total cardiac caths to Medicaid patients and 3% to charity care. Due to changes in state regulation, Medicaid and charity care for cardiac caths no longer needs to be reported to the state. That data, representing as it does, the base of patients from which open heart cases will come, is useful in evaluating Martin Memorial's projections. In 1999, seven-tenths of one percent of the patients in Martin Memorial's cath lab were Medicaid and four-tenths of one percent were indigent. In 2000, seven-tenths of one percent were Medicaid and two-tenths of one percent were indigent. Martin's cath lab data indicates that its projected open heart levels of Medicaid and indigent care are not attainable. Bethesda offered a commitment to provide 3% of total open heart cases for Medicaid patients and 3% to indigent patients annually. Historically, Bethesda has cared for a relatively large number of Medicaid, minority, and indigent patients. It is recognized as a disproportionate share provider of Medicaid care under the Florida program and of Medicare under the Federal program. The Palm Beach County Health Department provides approximately $1 million a year to Bethesda for charity care. As a percentage of gross revenue, Bethesda provided 8.8% Medicaid and 3.46% charity care in 1999. Approximately 54% of the charity care is attributable to obstetrics and pediatric services. Bethesda's younger patient base and the number of adult open heart Medicaid cases from Bethesda's service area, 2.4% or 7 cases in the year ending September 2000, raise the issue of its ability to generate sufficient cases to meet the proposed commitment. In 1995, 20 of the 36 total resident Medicaid open heart surgeries were performed at the three providers in District 9, Delray, JFK, and PBGMC. In 1999, when Lawnwood began open heart care, the Medicaid volume at the District providers increased to 51 of the 64 total Medicaid resident cases. In 2000, the four programs treated a net number of 56 of 60 resident Medicaid cases. A program at Bethesda also could reasonably be expected to increase the number of Medicaid and charity cases performed in the District, in volume and by reversing outmigration, but the patients must come from a base of patients with cardiac diagnoses. For the year ending September 2000, in Bethesda's service area, 4.9% of cardiac patients were Medicaid and charity patients combined, 1.6% Medicaid and 3.3% charity. Assuming that the same proportions could be maintained for open heart surgeries, Bethesda cannot achieve 3% Medicaid and, although unlikely, has a chance of reaching 3% charity only in the best case scenario. If approved, BRCH commits to providing 5% of total OHS in the second year to uninsured patients and to establish an outreach program to increase utilization by uninsured patients. BRCH has, over the past three and a half years, established outreach programs, which include having nurses and social workers in schools, providing free physical examinations to children who do not have primary care doctors, and performing echocardiograms for high school athletes, equipping police and fire rescue units with portable defibrillators, and operating mobile units for mammography screenings and vans to transport patients to and from their homes for hospital care. A free dental screening program is operated in conjunction with Nova Southeastern University. BRCH also operates a family medical center approximately seven miles west of the hospital. Recently, the Foundation for BRCH purchased, for $1.8 million, a large bus to equip as a mobile clinic. The mobile diagnostic unit is intended to reach uninsured patients to provide primary care and ultimately open heart surgery care to those who might not otherwise be screened, diagnosed and referred. No information was available and no decisions had been made about the staff and equipment, or service areas for use of the van. Because of the lack of more specific plans, it is impossible to determine whether the outreach effort has any reasonable prospects for success in meeting any unmet need. For the years ending June 1996, 1997, and 1998, BRCH provided six-tenths of one percent, and five-tenths of one percent of gross revenues for charity care. In 2000, BRCH provided one-half of one percent for charity care and, in 2001, twenty-seventh hundreds of a percent. The historical levels do not support the proposed commitment of 5% of open heart surgeries for uninsured patients in the second year of the program. Although worded to apply only to the second year, BRCH's President and CEO testified concerning the condition without limiting it to the second year. In Boca Raton Community Hospital, Inc.'s Proposed Recommended Order (Reformatted), filed on July 5, 2002, the condition is described as follows: 49. As conditions of CON approval, Boca will, beginning in the second year of operation of the program and continuing thereafter, provide a minimum of five percent each year of OHS cases to uninsured patients, and establish an outreach program to locate and provide OHS and cardiology services to uninsured patients in Palm Beach County. (Boca Ex. 3 at Schedule C; Pierce, 1899). Boca reasonably decided to focus on the needs of the uninsured, rather than Medicaid patients, because of the low volume of Medicaid patients who require OHS services. (Pierce, 1902). At BRCH, Medicaid and Medicaid health maintenance organization (HMO) care as a percent of total ranged from 1.3% to 1.4% from 1996 through 1998. BRCH projected serving 1.2% to 1.3% open heart Medicaid cases, or four patients in the first year and 1.5% to 1.6%, or seven Medicaid patients in the second year. The projections are consistent with its history although BRCH offered no Medicaid condition. Bethesda and BRCH also claimed not normal circumstances exist in District 9 due to the disparity in open heart care for uninsured and Medicaid patients as compared to the insured. For uninsured residents of Palm Beach County during the twelve months ending June 30, 2000, the use rate was 4.7 per 1000, as compared to 21.8 per 1,000 for insured open heart patients. For angioplasty patients, the insured use rate was 38.2, but the uninsured rate was only 8.9. Assuming that the use rates should not be so different, the discrepancy in access for the uninsured is significant and unfortunate but was not shown to be a not normal circumstance in the health care delivery system. The applicants' proposals, unlikely as they are to meet even the proposed conditions, are inadequate to increase access materially for the uninsured. Comparisons of the level of Medicaid provided statewide to that provided in District 9 without consideration of other factors, including age and income levels, were not useful in analyzing access. Assertions that any discrepancy in care for potential Medicaid open heart patients constitutes a not normal circumstance are not substantiated by this evidence. Subsection 408.035(1)(b) and (2)(b) - availability, quality of care, efficiency, appropriateness, accessibility, extent of utilization and adequacy of like and existing facilities in District Nine In 2006, the population in District 9 is projected to reach 1.2 million people, of which approximately 992,378 will reside in Palm Beach County, 119,573 in Martin County, 181,406 in St. Lucie County, 106,790 in Indian River County, and 31,140 in Okeechobee County. In District 9, throughout Florida, and in the United States, heart disease is the leading cause of death. In 2000, heart disease was the cause in 522 of 1,560 total deaths in Martin County, and 4,337 of 12,795 total deaths in Palm Beach County. From 1995 to 2000, the number of Florida residents having open heart surgeries increased 15.1%. During the same period of time, the number of District 9 resident cases, regardless of where the surgeries were performed, increased from 3,119, to 3,938, an increase of 755 OHS cases, or 24%. Palm Beach County residents represented 427 of the 755 increase, and 2,633 of the total of 3,938 resident cases. The distribution of the remaining 1,305 District resident cases by county was as follows: 597 from St. Lucie, 339 from Martin, 269 from Indian River, and 100 from Okeechobee County. More recent data, however, indicates trends towards a leveling off or even decline in the number, but an increase in the complexity of open heart procedures. Some experts describe open heart volumes having reached a "plateau" in the United States, in Florida, and in District 9. Last year, the number of open heart surgeries in the United States declined 22%. The statewide volume of cases was 32,199 in 1996, 33,507 in 1997, 34,013 in 1998, and 32,097 in 1999. At District 9 hospitals, open heart volumes were 1,670 in 1994, 1,841 in 1995, 2,152 in 1996, 2,407 in 1997, 2,527 in 1998, 2,656 in 1999, and 2,650 in 2000. Cardiac Catheterizations and Angioplasties The major reason given for the stable and declining open heart volume is the increase in the utilization of angioplasty, or therapeutic cardiac cathing, an alternative which costs less and is less invasive. Angioplasty procedures increased from 1995-2000, by over 2,500 cases for District 9 residents, and over 2,600 cases in District 9 hospitals, from 2,104 cases in 1995, to 4,714 in 2000. Among the procedures generally referred to as angioplasties are percutaneous transluminal angioplasty (PTCA) or balloon angioplasty, percutaneous transluminal coronary rotational atherectomy (PTCRA), and the insertion of scaffolding- like devices, called stents, to prevent re-occlusion of coronary arteries. In Florida, diagnostic cardiac caths may be performed at facilities which do not have angioplasty and open heart surgery programs, but angioplasties must be performed, except in rare emergency circumstances, only at hospitals which are licensed to provide open heart services, in case back-up surgery is needed. Lawnwood Regional Lawnwood is located in Fort Pierce, in St. Lucie County, which is second to Palm Beach County in population and in District 9 resident open heart cases. Lawnwood is owned by a subsidiary of HCA, the Hospital Corporation of America, formerly known as Columbia. HCA is a for-profit, investor-owned corporation which owns and operates approximately 200 hospitals in the United States. A $17 million addition at Lawnwood, designed for the open heart program, includes two dedicated operating rooms and a 12-bed intensive care unit. The Lawnwood program has a full-time staff of two surgeons and one additional surgeon who divides his time between Lawnwood and PBGMC. Lawnwood, having opened its program early in 1999, is not considered a mature program. In addition, Lawnwood has had some difficulties with accreditation and disputes with cardiologists. Lawnwood reported one open heart case in the first quarter of 1999, and 143 or 144 for the year. In calendar year 2000, between 330 to 340 open heart surgeries were performed at Lawnwood. In calendar year 2001, the volume was between 333 and 336 cases. Depending on the source of the data, the volume at Lawnwood was reported to be as high as 364 for the twelve months ending September 30, 2000; in a range from 336 to 396 for the twelve months ending March 31, 2001; and up to 412 for the twelve months ending July 2001. The variances result from seasonal patient utilization, and from AHCA’s use, for the fixed need pool, of the most current available data which it receives from the various local health councils. That data is submitted on handwritten or typed forms which are not uniform across districts. Subsequently, the hospitals provide electronic data tapes directly to AHCA, which if properly decoded, should provide more accurate statistics. While there may be variances either way, in this case, the lower volumes for Lawnwood were derived from the more reliable electronic tapes. Based on that data and the testimony of the cardiac surgeon who is the director of the program at Lawnwood, the annual volume of open heart surgeries was approximately 330 in 2000, and 348 in 2001. The new rule, adopted on January 24, 2002, reduces the minimum number required for existing programs to 300 a year, or 25 adult operations a month. The number of angioplasties performed at Lawnwood increased from 465 in 1999, to 845 in 2000. Palm Beach Gardens Medical Center South of the four relatively small northern counties in District 9, PBGMC has 204 beds located in northern Palm Beach County. It is a subsidiary of Tenet Healthsystem Hospitals (Tenet). Adult open heart surgery has been available at PBGMC since 1983. The surgeries are typically performed in two or three of the 11 operating rooms, although five are equipped to handle open heart cases. PBGMC has 94 telemetry beds, and 32 intensive care beds, eight designated for cardiovascular intensive care patients. PBGMC has four cardiac cath labs and separate electrophysiology labs. The medical staff of approximately 400 physicians includes about 200 cardiologists, 24 invasive cardiologists and seven cardiac surgeons. The number of open heart cases at PBGMC was 700 in 1994, 801 in 1995, 913 in 1996, 1,028 in 1997, 1,045 in 1998, 1,124 in 1999, 940 in 2000, and 871 in 2001. The number of angioplasties increased from 552 in 1994, to 1,019 in 1997, to 1,431 in 2000. JFK JFK, which has 387 beds, is located roughly in the center of Palm Beach County, in the City of Lake Worth. Like Lawnwood, JFK is an HCA's subsidiary, having been purchased by that corporation in 1995. Open heart services and cardiac cath services began simultaneously at JFK in 1987. JFK has three open heart operating rooms. JFK, after a major expansion, has a separate entrance to its three cardiac cath laboratories, a dedicated electrophysiology suite, for treatment of arrhythmias, and 17- patient holding area. JFK provides all cardiac services, except heart transplants. The average age of patients at JFK is 74 years old. The medical staff of 504 board-certified or board- eligible physicians includes 25 cardiologists, five invasive cardiologists, two electrophysiologists, and three cardiac surgeons. JFK has recently accepted applications from but not yet extended privileges to three additional cardiovascular surgeons. Volumes of open heart cases at JFK were, with some variances depending on the data source, approximately 428 in 1994, 434 in 1995, 630 in 1996, 674 in 1997, 711 in 1998, 613 in 1999, 621 in 2000, and 610 in 2001. The number of angioplasties ranged from 709 in 1994, to 1,152 in 1997, to 1,281 in 2000. Delray Delray, with 343 beds, in Delray Beach, is the trauma center for southern Palm Beach County. Open heart care began at Delray in 1986. The surgeries are currently performed in three of ten, but soon to be a total of twelve operating rooms with shelled-in spaces set aside for two more. Patients recover in a 15-bed surgical intensive care unit. The Delray medical staff of over 600 physicians has close to 60 cardiologists, including 15 invasive cardiologists and six cardiovascular surgeons. Delray has three cath lab rooms and seven bays for holding patients pre- and post-procedure. For the years 1994 through 2001, open heart volumes at Delray were 542, 606, 609, 705, 771, 758, 759, and 738, respectively. During the same period of time, the annual number of angioplasty procedures increased from 591 in 1994, to 810 in 1997, to 929 in 2000. The existing CON-planned and approved programs in the District are well distributed geographically and allocated appropriately based on population. Considering the declining utilization, the like and existing open heart surgery programs are available and accessible. Subsection 408.035(1)(f) - services that are not reasonably and economically accessible in adjoining areas Over 30% of District 9 resident open heart cases are performed in other districts, the vast majority at North Ridge in District 10 (Broward County). The district outmigration for a service when excessive or difficult can indicate access or quality concerns and constitute a not normal circumstance for approval of a new program. In this case, with adequate available services in District 9 and its close proximity, the outmigration to North Ridge, which is 15 miles or 25 minutes from BRCH is not a not normal circumstance. There is also substantial overlap in the medical staff at both hospitals which allows continuity of care for patients despite transfers. The argument that families, particularly an older spouse, will necessarily have to drive farther to visit the patient is rejected, since that depends on where in the district the person resides not on the distances between hospitals. North Ridge has 391 licensed beds, with 260 to 270 acute care beds in use. At North Ridge, cardiovascular surgeons usually use three OHS operating rooms, although a fourth is also available. Open heart patients recover in a six-bed cardiovascular intensive care unit. The reported volumes of open hearts at North Ridge have been from 1994 through 2001, respectively, 864, 935, 893, 826, 882, 890, 905, and 795. The total number of open heart cases in District 10 has been declining since 1998. The volume of angioplasties at North Ridge increased from 793 in 1994, to 829 in 1997, to 1,155 in 2000, consistent with a rising District 10 use rate from 2.95 to 3.66 over the same period of time. The staff at North Ridge includes 107 cardiologists, 27 interventional cardiologists, and 17 cardiovascular surgeons, many of whom also regularly perform open heart surgeries at Holy Cross, which is approximately a mile south of North Ridge in Fort Lauderdale. At Holy Cross, which also has established referral networks from District 9, open heart volumes declined from a high of 753 in 1998 to 693 in 2000. All of the open heart services proposed by the applicants are reasonably available in adjoining areas, in Districts 10 and 11 to the south and in the other districts to the north. Subsection 408.035(1)(c) - comparisons of quality; and Subsection 408.035(1)(e) - joint, cooperative or shared resources; and Subsection 408.035(1)(g), (h), and (k) - need for research, educational and training programs or facilities for medical and health care professionals; and Subsection 408.035(1)(h) and Rule 59C-1.033 - recruitment, training and salaries for staff The parties stipulated that the applicants have a history of providing quality care. Martin Memorial was accredited with commendation by the JCAHO in 1997, which is now called accreditation without Type I Recommendations. That was followed, in July 2001, with a score of 93 on survey items with some follow-up improvements required related to patient assessment and nutrition. Martin Memorial offers internships, and residencies for training non-physician medical personnel from Barry University, Indian River Community College, and Florida Atlantic University. The cancer center at Martin Memorial is affiliated with the Moffitt Center. Despite the absence of an open heart program, Martin Memorial has participated in clinical trials of cardiac drugs. The Shands Healthcare System of nine affiliated hospitals, including two research and teaching hospitals, is the model for the relationship proposed with Martin Memorial. The partnerships are intended to upgrade the care available in community hospitals and to establish, for complex cases, referral networks for the Shands teaching hospitals. Shands has already satisfied itself that Martin Memorial meets its due diligence test for the quality of its existing program and philosophical compatibility. If Martin Memorial's CON is approved, Shands will assist in training staff for the program. Initially, the program will have one cardiovascular surgeon, a University of Florida medical school faculty member, in Martin County. When that surgeon is ill or on vacation, others from the University of Florida will be available. The logistics of the plan raises questions about the adequacy of coverage to meet the 24-hour requirements of Rule 59C-1.033, Florida Administrative Code. In the JCAHO survey process, Bethesda received a score of 97, as a result of its survey in June 2000, and was accredited for the maximum allowable time, three years. Personnel for a Bethesda program can be appropriately trained at Orlando Regional, a statutory teaching hospital with a high volume open heart program. In June 2000, BRCH received a JCAHO score of 96. BRCH maintains a scholarship program for new nurses making a two-year commitment, and an on-site educational department with a preceptorship for training operating room and emergency room nurses. Nursing students from Florida Atlantic University (FAU), which is located across Glades Road from BRCH, rotate at BRCH. FAU is in the process of establishing a medical school. There is a severe shortage of nurses in the United States, in Florida, and in District 9. All of the hospitals in District 9 have resorted to highly competitive and innovative recruitment and retention strategies, including international recruiting, signing bonuses, child care and, of course, rising salaries and benefits. The demand is greater and shortages more severe in highly specialized areas, such as critical care, telemetry and open heart surgery nursing. The average age of nurses has also increased to 46 or 47 years old, while enrollment in nursing schools and the number of nursing school professors have declined. All of the applicants concede that recruiting and retaining nurses for new open heart program will be a challenge. The likely results are a loss of experienced nurses from existing programs, an increase in total health care costs, an increase in vacancies, and, at least temporarily a decline in the quality of experienced nursing care in existing open heart programs. At this time, there is no evidence that declining open heart utilization will eventually alleviate the shortage of experienced nurses. It has, so far, only eased the need to resort as frequently to other extreme and expensive alternatives, including pay overtime, contracting with private agencies, and bringing in traveling nurses. Subsection 408.035(1)(m) - size, scope and fixed equipment cost at BRCH; Subsection 408.035(2)(c) - alternatives to new construction; and Subsection 408.035(1) (h) - funding for BRCH BRCH plans to construct a 74,000 square-foot cardiac care facility, which will include two open heart operating rooms and two cardiac cath labs, an electrophysiology lab, 12 cardiovascular intensive care beds, and 18 cardiac cath lab bays. Only 18,568 square feet are attributable to the open heart operating rooms and cardiovascular intensive care unit which compares favorably with Bethesda's estimate of 17,759 square feet for the same functions. It is not possible, therefore, to conclude that the size of the BRCH project is excessive as compared to that proposed by Bethesda. BRCH underestimated the cost for fixed equipment for the open heart project by approximately $1.6 million. That omission resulted in understated estimates of depreciation by approximately $275,000. The total project cost for BRCH is approximately $2.2 million when almost $2 million in omitted equipment costs is added to the original estimate of $20 million. All pending capital projects, as shown on Schedule 2 of the BRCH application, total $54 million. With combined cash and investments of $160 million, the BRCH foundation has sufficient funds for the hospital's projects. Although BRCH earned profits of $6.6 million and $7.3 million in 1998 and 1999, respectively, the hospital lost $30 million from operations due to billing and collection errors in 2000. BRCH has a donor who has stated a willingness to donate $20 million for the cardiac care center. BRCH has the funds necessary to build the facility. With Medicare capital cost reimbursement completely phased out, there is insufficient evidence of a direct impact on health care costs based on this proposed capital expenditure. Subsection 408.035(1)(i) - short and long term financial feasibility Martin Memorial initially projected that its program would perform 360 open heart surgeries in year one and 405 in year two. As a result of changes in the use rate, Martin Memorial lowered its second year projection to 375 surgeries while increasing staffing levels. Even if projected open heart surgery revenues of $264,000 in the second year decline in proportion to expected lower utilization, estimated angioplasty revenues of $468,000, are sufficient to make up the deficit and to keep the combined program financially feasible in the short and long term. Bethesda projected volumes of 165 open heart surgeries in the first year and 270 in the second year. Assuming Bethesda's revenues are 90% of the district average, the combined net profit for open heart and angioplasty services is reasonably expected to be approximately $750,000 in the second year operations. The project is profitable, therefore, financially feasible in the short and long term. BRCH's expert projected volumes of 308 open heart surgeries and 289 angioplasties in the first year, and 451 open heart surgeries and 422 angioplasties in the second year. If utilization projections are correct, then BRCH will receive incremental net income of $1.6 million from the open heart surgery program and $825,000 from the angioplasty services. Factoring in claims that the Medicare case weight was overstated and depreciation underestimated, the BRCH project is, nevertheless, financially feasible for the short and long term. Typically, any open heart surgery program that can reach volumes in the range of 200 to 250 cases, will be financially feasible. The establishment of an open heart program also has a "halo effect," for the hospital, attracting more patients to the cardiac cath labs and other related cardiology services. Open heart surgery and angioplasty tend to be profitable, generating revenue which hospitals use to offset losses from other services. Subsection 408.035(1)(j) - needs of HMOs All of the applicants will enter into contracts with, but none is a health maintenance organization. Subsection 408.035(1)(l) - probable impact of fostering competition to promote quality assurance and cost-effectiveness Hospitals with higher volumes of open heart surgeries and angioplasties usually have higher quality as measured by lower mortality rates and fewer complications. The open heart surgery rule, in effect at the time the applications were filed, established a minimum volume of 350 annual admissions for existing providers. In the rule as amended on January 24, 2002, the minimum volume for existing programs was reduced to 300. The divisor in the formula for determining need, which represents the average size of a program in the district, was 350 prior to amendment and 500 subsequently. The minimum and average volumes in the rule set, in effect, the protected range for existing programs, not the optimal size, or "cut point" at which outcomes are worse below and better above. According to the American College of Cardiology and American Heart Association (ACC/AHA) the evidence is clear that outcomes are better if an individual performs at least 75 procedures at a high volume center with more than 400 cases. The ACC/AHA guidelines indicate, although more controversial and less clearly established, that acceptable outcomes may be achieved if the individual operator performs at least 75 procedures in centers with volumes from 200 to 400 cases. Because the relationship between higher volumes and better outcomes is continuous and linear, and because research showing the benefits of primary angioplasty with or without open heart surgery back-up is preliminary and limited, the position of the ACC/AHA is, in summary, as follows: The proliferation of small angioplasty or small surgical programs to support such angioplasty programs is strongly discouraged. (Journal of the American College of Cardiology, Vol. 37, no. 8 June 15, 2001, pp. 2239xvii (Tenet Exhibit 5)) An open heart program at Martin Memorial will redirect cases that would otherwise have gone to Lawnwood, PBGMC, and JFK. The proposed Martin Memorial Service area overlaps that of Lawnwood in southern St. Lucie County, an area which generates one quarter of the open heart cases at Lawnwood. Lawnwood is reasonably expected to lose 56 open heart cases a year with total volume going down below 300, resulting in loss of $1.8 million, or 20% of its total revenues. Lawnwood would have unacceptably low volumes threatening the quality of the open heart program. PBGMC, as a result of a new program at Martin Memorial, will lose approximately 170 and 180 open heart cases annually and an equal number of angioplasties reducing its open heart volume to approximately 700 a year. The financial loss would range from $4 to $5 million a year, as compared to total net income which was between $20 and $30 million a year for past three years. PBGMC would not suffer an adverse impact sufficient to threaten either the quality or the financial feasibility of the open heart program or total hospital operations. JFK, which currently receives most of the angioplasty referrals from Martin Memorial, is expected to lose from 25 to 30 open heart cases, and 65 to 70 angioplasties each year during the first two years of a Martin Memorial program. The estimated financial loss to JFK is $1.7 million, a significant detriment when compared to $2.8 million in net income from operations in calendar year 2000. Approval of open heart program at Bethesda will adversely affect case volumes at JFK and Delray. Bethesda projected that, in its first year, 75% of its cases would have gone to Delray and 25% to JFK, and that by the third year, the split would be even at 50% from Delray and 50% from JFK. JFK, depending on the approach to the impact analysis, will lose from 40 to 60 open heart cases in the first year, from 90 to 110 in the second year, and from 115 to 170 in the third year of a program at Bethesda. The volumes of lost angioplasties is expected to be slightly higher. The resulting combined open heart and angioplasty financial loss is $6.6 million, far greater than the significant detriment expected from a Martin Memorial program alone. The annual volume of open heart cases at JFK would be approximately 400 to 500, assuming flat not continued declining utilization. If Bethesda offered the service, Delray's open heart volumes would decline by 124 cases in the first year and by 248 cases in the third year of operations, decreasing total volume to 500 or 600 annual surgeries. Delray had a net income from operations of approximately $24.7 million in 2000, which would indicate that neither quality nor financial stability would be significantly adversely affected. If an open heart program is approved for BRCH, the volumes of cases at Delray and North Ridge will decline. Delray would be expected to lose 163 open heart cases and 235 in years one and two, respectively, and equal numbers of caths and angioplasties, resulting in annual open heart cases reduced from the low 700s to approximately 500 cases. Delray's pre-tax revenue was $39 million in 2001. In terms of quality and financial stability, Delray can withstand the adverse impact of a new program at BRCH. North Ridge would lose approximately 124 open heart cases in year one and 178 in year two, and similar numbers of caths, reducing open heart volumes from the upper 700s to approximately 600 annual cases. North Ridge's pre-tax income was $21 million for the year ending May 31, 2001. It appears that North Ridge could, even with the adverse impact of BRCH, maintain a quality, financially viable open heart program. Subsection 408.035(l) - probable impact on costs The applicants, all not-for-profit corporations, contend that the fact that District 9 has only for-profit open heart hospitals affects charges and is a not normal circumstance for the approval of one or more not-for-profit. District 9 is the only district in Florida in which all open heart providers are for-profit corporations. Statewide, not-for-profit open heart hospitals charge 31% less than for-profit. Martin Memorial's CON proposal includes a charge structure below that at existing programs. Bethesda's planned charges are 10% less than the District 9 average for open heart and angioplasty services. BRCH is the applicant which is most likely to increase competition in District 9, based on the Herfindahl-Hirschman Index (HHI). The HHI's measurement of competition in a market used by economists frequently to analyze anti-trust issues. Charges are not a factor in up to 75% of open heart/angioplasty cases reimbursed by payors, such as Medicare, at set flat rates. In approximately 10% of cases, including complex "outlier" cases exceeding the range for flat rate reimbursement and for other payors on a percent-of-charges basis, charges are not irrelevant. But, the evidence to demonstrate lower charges were applicable to patients of the same severity was questionable. Subsection 408.035(1)(o) - continuum of care There is insufficient evidence the any applicant is preferable based on its ability to promote a continuum of care in a multilevel system. Subsection 408.035(2)(a) - alternatives to inpatient services There are no alternatives to inpatient services for open heart surgery and angioplasty patients. Subsection 408.035(2)(d) - patients who will experience serious problems in the absence of the proposed new service The applicants and AHCA determined that new open heart surgery programs are needed mainly to provide emergency or "primary" angioplasty to patients suffering heart attacks (acute myocardial infections). Primary angioplasty is an alternative to "clot busting" medications, or thrombolytics, and to open heart surgery. Performed on an emergency basis, the three different treatments are used to restore blood flow before heart muscle dies. Because "time is muscle," patients benefit only if treated within a relatively short time after the onset of symptoms. The goal is 90 minutes from door-to-balloon for angioplasty. The decision to treat a patient with a particular therapy is based on a number of factors assessed during triage. Paramedics in consultation with ER doctors at the receiving hospital frequently begin triage and administering medications and oxygen in ambulances equipped with sophisticated diagnostic equipment. As the statistical data demonstrates, angioplasty, whether scheduled or emergency, is increasingly becoming the preferred therapy. Some studies have shown improved outcomes, higher survival rates and fewer complications, from primary angioplasty as compared to thrombolytics. Comparisons have not been made over extended periods of time, and the apparent benefits of angioplasty have not been duplicated in community hospitals as compared to clinical trials in high volume research centers. Estimates of the number of people who could benefit from the availability of angioplasty services at the applicants vary based on the number of elderly in the service area, the number of non-traumatic chest pain ER visits, delays in transfers of emergency patients, and the number of patients being transferred to existing providers for angioplasties or open heart surgeries. Martin Memorial selected five patients as examples of those who could be served in an open heart program at Martin Memorial. The anecdotal evidence of transfer "delays" is insufficient to demonstrate bed unavailability or capacity constraints. Martin Memorial-Stuart and Martin Memorial South transferred 240 heart attack patients to open heart surgery hospitals. Only 18 of the emergency heart attack patients who presented at the Martin Memorial ER were transferred from the ER. Approximately ten patients a year are so unstable that an intra- aortic balloon pump is required during transfer. Martin Memorial presented evidence of delays of two hours or more in transfers of 84 patients from its cath labs to open heart surgery hospitals. The transfer records, created for subsequent certificate of need litigation, were of questionable probative value. The case studies were inadequate to establish whether "delays" were reasonable or not. Factors such as physician consultation time, time to stabilize a patient for transfer and the assumed travel time seem to have been included in the time periods. Bethesda transferred 270 patients for cardiac care from October 1999 through September 2000. Thirty patients were transferred, from November 2000 to July 2001, for angioplasties or open heart surgery after having cardiac caths at Bethesda. Bethesda failed to establish that transfers were delayed due to capacity problems at existing hospitals because emergency patients were not classified separately, and the causes of the time lapses were not identified. Of the applicants, BRCH has the busiest ER, with 50,000 to 52,000 annual visits compared to approximately 48,000 at the two Martin Memorial locations combined. BRCH admitted 439 heart attack patients through its ER during the year ending June 30, 2000. The majority of patients are treated with thrombolytics at BRCH. BRCH transfers approximately one emergency heart attack patient a week on average, or from 30 to 50 a year, for interventional cardiac procedures. BRCH's presentation of evidence of delays in transfers was flawed. The data was collected and used only for litigation, and was incomplete. Some patient records were lost and others were deleted due to inaccurate data. Of the applicants, BRCH is located in an area with the largest percentage of the population age 65 and older, approximately 35%, as compared to 24% in Martin Memorial's service area. Agency Consistency Martin Memorial, through expert witness testimony, compared its situation to that of Brandon, a hospital in AHCA District 6, which was issued an open heart surgery CON in 2001. The expert noted that Martin Memorial and Brandon are both in five county health planning districts, and that they are 19 and from 15 to 17 miles, respectively, from the nearest open heart provider. Three of the counties in District 6 have open heart programs, including Hillsborough County where Brandon is located, as compared to two District 9 counties, St. Lucie and Palm Beach, but not Martin. The Martin Memorial primary service area projected population is 238,861 for 2004, 24.1% aged 65 and older. The Brandon service area population projection is 309,000 for 2004, with 10.5% aged 65 and older. Brandon has 255 beds, Martin Memorial-Stuart has 236. Brandon had 53,000 emergency room visits, and Martin Memorial, at both locations, had 48,503 in 1999. Before defaulting to zero, the numerical formula yielded a need for 3.27 additional open heart programs in District 6 as compared to 3.9 in District 9. Other specific comparisons favorable to Martin Memorial included the number of heart attack patients presenting at its ER, cath lab volumes, patient transfers for open heart and angioplasty procedures. Among others, there are several significant distinguishing facts in Florida Health Sciences Center, Inc. v. Agency for Health Care Administration, Case No. 00-0481CON, (R.O. Mar. 30, 3001, F.O. Oct. 17, 2001) aff'd per curiam sub nom, University Community Hospital v. Agency for Health Care Administration, Case No. 1DO1-3592, et al. (Fla. 1st DCA Sept. 19, 2002), the Brandon case. In that case, the two existing providers performing fewer than 350 cases a year, Blake Medical Center, and Manatee Memorial Hospital, both in Manatee County, were mature programs located 40 miles from Brandon with no service area overlap. By contrast, Lawnwood which is not a mature program and, therefore, has not reached its potential volume, is 20 miles from Martin Memorial, and has an overlapping service area. Martin Memorial's ER volume and the number of transfers from its ERs are the combined experience from two locations. The more accurate comparison is 27,000 ER visits at Martin Memorial-Stuart to 53,000 at Brandon. Emergency heart attack patients presenting at Martin Memorial South would continue to require transfers for primary angioplasty. Finally, the decision in Brandon was based, in large part, on transportation difficulties, inadequate interfacility ambulances and traffic congestion, which are not factors in District 9. Factually, the case of Halifax Hospital Medical Center, d/b/a Halifax Medical Center v. Agency for Health Care Administration, et al., Case No. 95-0742 (AHCA Jan. 14, 1997) is also distinguishable. The applicant could have no effect on the low volume providers located 80 miles to the north. That was one not normal circumstance. Need existed because of another not normal circumstance, i.e., capacity constraints at the only other provider in the same primary service area. In Oak Hill Hospital v. AHCA, Case No. 00-3216CON (R.O. Oct. 4, 2001, F.O. Jan. 22, 2002), appeal dismissed sub nom Hernando HMA, Inc. v. HCA Services of Florida, Inc., Case No. 1DO2-854 (Fla. 1st DCA June 6, 2002), the two approved applicants were in separate counties which constituted entirely separate health care markets. Neither applicant would adversely affect the low volume providers. After the Administrative Law Judge recommended approval of the Citrus County applicant, AHCA, engaging in what appears to be a comparative review of the two remaining applicants from Hernando County, approved a second applicant from the same district at the same time. Some facts are similar to those in this case: The average drive time between hospitals was 30 minutes; transfers and admissions procedures required additional time; there was a recognition of increasing preferences for reperfusion of heart muscle using primary angioplasty; patients and families experience stress and anxiety as a result of transfers. Institution-specific issues included the transfer of 600 cardiac patients by ambulance from Oak Hill, the size of the cardiology and cardiac cath programs (1,641 caths in 1999), the larger elderly population in the service area, and the hospital's size.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Agency for Health Care Administration enter a final order denying Certificate of Need Application Number 9248 filed by BRCH, Certificate of Need Application Number 9249 filed by Martin Memorial, and Certificate of Need Application Number 9250 filed by Bethesda. DONE AND ENTERED this 11th day of November, 2002, in Tallahassee, Leon County, Florida. S ELEANOR M. HUNTER Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 11th day of November, 2002. COPIES FURNISHED: Lealand McCharen, Agency Clerk Agency for Health Care Administration Fort Knox Building 3, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308-5403 Valda Clark Christian, General Counsel Agency for Health Care Administration Fort Knox Building 3, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308-5403 Gerald L. Pickett, Esquire Agency for Health Care Administration 525 Mirror Lake Drive, North Sebring Building, Suite 310K St. Petersburg, Florida 33701 Lori C. Desnick, Esquire Agency for Health Care Administration Fort Knox Building 3, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308-5403 Stephen A. Ecenia, Esquire David Prescott, Esquire Rutledge, Ecenia, Purnell & Hoffman, P.A. 215 South Monroe Street, Suite 420 Post Office Box 551 Tallahassee, Florida 32302-0551 W. David Watkins, Esquire R. L. Caleen, Jr., Esquire Watkins & Caleen, P.A. 1725 Mahan Drive, Suite 201 Post Office Box 15828 Tallahassee, Florida 32317-5828 H. Darrell White, Esquire William B. Wiley, Esquire McFarlain & Cassedy, P.A. 305 South Gadsden Street Post Office Box 2174 Tallahassee, Florida 32316-2174 Paul H. Amundsen, Esquire Amundsen, Moore & Torpy, P.A. 502 East Park Avenue Post Office Box 1759 Tallahassee, Florida 32302 Robert D. Newell, Jr., Esquire Law Firm of Newell & Terry, P.A. 817 North Gadsden Street Tallahassee, Florida 32303-6313 C. Gary Williams, Esquire Michael J. Glazer, Esquire Ausley & McMullen 227 South Calhoun Street Post Office Box 391 Tallahassee, Florida 32302 Seann M. Frazier, Esquire Michael J. Cherniga, Esquire Greenberg Traurig, P.A. 101 East College Avenue Tallahassee, Florida 32302
The Issue Whether proposed rule amendments to Rule 59C- 1.033(7)(c) and (7)(d), Florida Administrative Code, published in the Notice of Change on June 15, 2001, constitute an invalid exercise of delegated legislative authority. Whether the proposed rule is invalid due to the absence of a provision specifying when the amendments will apply to the review of certificate of need applications to establish open heart surgery programs.
Findings Of Fact The Agency is responsible for administering the Health Facility and Services Development Act, Sections 408.031-408.045, Florida Statutes. The goals of the Act are containment of health care costs, improvement of access to health care, and improvement in the quality of health care delivered in Florida. AHCA initiated the rulemaking process by proposing amendments to existing Rule 59C-1.033, Florida Administrative Code, the rule for determining the need for adult open heart surgery (OHS)1 services, which currently provides, in part, that: Adult Open Heart Surgery Program Need Determination. a new adult open heart surgery program shall not normally be approved in the district if any of the following conditions exist: There is an approved adult open heart surgery program in the district. One or more of the operational adult open heart surgery programs in the district that were operational for at least 12 months as of 3 months prior to the beginning date of the quarter of the publication of the fixed need pool performed less than 350 adult open heart surgery operations during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool; or One or more of the adult open heart surgery programs in the district that were operational for less than 12 months during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool performed less than an average of 29 adult open heart surgery operations per month. Provided that the provisions of paragraphs (7)(a) and (7)(c) do not apply, the agency shall determine the net need for one additional adult open heart surgery program in the district based on the following formula: NN =((Uc x Px)/350)) -- OP>=0.5 Where: NN = The need for one additional adult open heart surgery program in the district projected for the applicable planning horizon. The additional adult open heart surgery program may be approved when NN is 0.5 or greater. Uc = Actual use rate, which is the number of adult open heart surgery operations performed in the district during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool, divided by the population age 15 years and over. For applications submitted between January 1 and June 30, the population estimate used in calculating Uc shall be for January of the preceding year; for applications submitted between July 1 and December 31, the population estimate used in calculating Uc shall be for July of the preceding year. The population estimates shall be the most recent population estimates of the Executive Office of the Governor that are available to the department 3 weeks prior to publication of the fixed need pool. Px = Projected population age 15 and over in the district for the applicable planning horizon. The population projections shall be the most recent population projections of the Executive Office of the Governor that are available to the department 3 weeks prior to publication of the fixed need pool. OP = the number of operational adult open heart surgery programs in the district. Regardless of whether need for a new adult open heart surgery program is shown in paragraph (b) above, a new adult open heart surgery program will not normally be approved for a district if the approval would reduce the 12 month total at an existing adult open heart surgery program in the district below 350 open heart surgery operations. In determining whether this condition applies, the agency will calculate (Uc x Px)/(OP+1). If the result is less than 350 no additional open heart surgery program shall normally be approved. Based on the issues raised by the Petitioner, Bethesda, and the factual evidence presented on these issues, AHCA must demonstrate that its proposed amendments to the existing OHS rule are valid exercises of delegated legislative authority or, more specifically, that it (a) followed the statutory requirements for rule-making, particularly for changing a proposed rule; (b) considered the statutory issues necessary for the development of uniform need methodologies; (c) acted reasonably to eliminate potential problems in earlier drafts of the proposed rule; (d) used appropriate proxy data to project the demand for the service proposed; (e) appropriately included county considerations for a tertiary service with a two-hour travel time standard; and (f) was not required to include a provision advising when CON applications would be subject to the new provisions. Rule challenges and rule development process The existing rule was challenged by IRMH on June 27, 2000, in DOAH Case No. 00-2692RX. Martin Memorial intervened in that case, also to challenge the rule. Like IRMH, Martin Memorial was an applicant for a certificate of need (CON), the state license required to establish certain health care services, including OHS programs, in Florida. Both are located in AHCA health planning District 9, as is the Petitioner in this case, Bethesda. AHCA entered into a settlement agreement with IRMH and Martin Memorial on September 11, 2000, which was presented when the final hearing commenced on September 12, 2000. Prior to the rule challenge settlement agreement, staff at AHCA had been discussing, over a period of time, possible amendments to the OHS rule to expand access and enhance competition. Issues raised by AHCA staff included the continued appropriateness of OHS as a designated tertiary service and the anti-competitive effect of the 350 minimum volume of OHS cases required of existing providers prior to approval of a new provider in the same district. The staff was considering whether the rule was too restrictive and outdated given the advancements in technology and the quality of OHS programs. The relationship of volume to outcomes was considered as various studies and CON applications were received and reviewed, as was the increasing use of angioplasty also known as percutaneous coronary angioplasty, referred to as PTCA or simply, angioplasty, as the preferred treatment for patients having heart attacks. Angioplasty can only be performed in hospitals with backup open heart services. During an angioplasty procedure, a catheter or tube is inserted to open a clogged artery using a balloon-like device, sometimes with a stent left in the artery to keep it open. Discussions of these issues took place at AHCA over a period of years, during the administrations of the two previous Agency heads, Douglas Cook and Reuben King-Shaw. In August 2000, AHCA published notice of a rule development workshop to consider possible changes to the OHS rule. Because it could not get the parties to settle DOAH Case No. 00-2692RX at the time, rather than proceed with the workshop while defending the existing rule, AHCA cancelled the workshop. As a result of the September 11, 2000, settlement agreement, on October 6, 2000, AHCA published a proposed rule amendment and notice of a workshop, scheduled for October 24, 2000. That version of a proposed rule would have changed Subsection (7)(a) of the OHS Rule to allow approval of "additional programs" rather than being limited to approval of one new program at a time in a district. The October proposal would have also eliminated OHS from the list of tertiary health services in Rule 59C-1.002(41). Tertiary health services are defined, in general, in Subsection 408.032(17), Florida Statutes, as follows: "Tertiary health service" means a health service which, due to its high level of intensity, complexity, specialized or limited applicability, and cost, should be limited to, and concentrated in, a limited number of hospitals to ensure the quality, availability, and cost-effectiveness of such service. Examples of such services include, but are not limited to, organ transplantation, specialty burn units, neonatal intensive care units, comprehensive rehabilitation, and medical or surgical services which are experimental or developmental in nature to the extent that the provision of such services is not yet contemplated within the commonly accepted course of diagnosis or treatment for the condition addressed by a given service. The agency shall establish by rule a list of all tertiary health services. With this statutory authority, AHCA adopted Rule 59C- 1.002(41), Florida Administrative Code, to provide a more specific and complete list of tertiary services: The types of tertiary services to be regulated under the Certificate of Need Program in addition to those listed in Florida Statutes include: Heart transplantation; Kidney transplantation; Liver transplantation; Bone marrow transplantation; Lung transplantation; Pancreas and islet cells transplantation; Heart/lung transplantation; Adult open heart surgery; Neonatal and pediatric cardiac and vascular surgery; and Pediatric oncology and hematology. As an additional assurance that tertiary services are subject to CON regulation, the tertiary category is specifically listed in the projects subject to review in Subsection 408.036, Florida Statutes. The October 2000 version included a proposal to increase the divisor from 350 to 500 in the formula in Subsection (7)(b), to represent the average size of existing OHS programs, but to decrease from 350 to 250, the minimum number required of an existing provider prior to approval of a new program in Subsection (7)(a)2. The definition of OHS would have been amended to add an additional diagnostic group, DRG 109, to delete DRG 110 and to eliminate the requirement for the use of the heart-lung by-pass machine during the surgery. Most controversial in the October version was a separate county- specific need methodology for counties which have hospitals but not OHS programs, in which residents are projected to have 1,200 annual discharges with a principal diagnosis of ischemic heart disease. On October 24, 2000, AHCA held a workshop on the proposed amendments. At the workshop, AHCA Consultant, John Davis, outlined the proposed changes. As a practical matter, eight Florida counties are not eligible to provide OHS because they have no hospitals. When Mr. Davis applied the county-specific need methodology, as if it were in effect for the planning horizon of January 2003, six Florida counties demonstrated a need for OHS: Hernando, Martin, Highlands, Okaloosa, Indian River, and St. Johns. Two of these, Martin and Indian River are in AHCA District 9. AHCA has already approved an OHS program for Martin County, at Martin Memorial. Mr. Davis also presented a simplified methodology for reaching the same result. In support of the proposed rule, AHCA received data, although not adjusted by the severity of cases, showing better outcomes in hospitals performing from 250 to 350 OHS, as compared to larger providers. Although the majority of heart attack patients are treated with medications, called thrombolytics, for some it is inappropriate and less effective than prompt, meaning within the so-called "golden hour," interventional therapies. In these instances, angioplasty is considered the most effective treatment in reducing the loss of heart muscle and lowering mortality. Opposing the proposed rule at the October workshop, Christopher Nuland, on behalf of the FSTCS, testified that OHS is still a highly complex procedure, that it requires scarce resources, equipment and personnel, and should, therefore, be available in only a limited number of facilities. In general, however, the opponents complained more about process rather than the substance of the proposal. Having petitioned on October 13, 2000, for a draw-out proceeding instead of the workshop, those Petitioners noted that AHCA had obligated itself to predetermined rule amendments based on the settlement agreement, regardless of information developed in the workshop. The draw- out Petitioners were the Florida Hospital Association, Association of Community Hospitals and Health Systems of Florida, Inc., Delray, Lakeland Regional Medical Center, Punta Gorda HMA, Charlotte Regional Medical Center, JFK, HCA Health Services of Florida, Inc., d/b/a Regional Medical Center Bayonet Point; Tampa General and the FSTCS. While agreeing that OHS is complex and costly, supporters of the proposed rule, particularly the declassification of OHS as a tertiary service, noted that many cardiologists are now trained to do invasive procedures. In support of fewer restrictions on the expansion of OHS programs in Florida, other witnesses at the October workshop discussed delays and difficulties in arranging transfers to OHS providers, possible complications from deregulated diagnostic cardiac catheterizations at non-OHS provider hospitals, and hardships of travel on patients and their families, especially older ones. On December 22, 2000, AHCA published another proposal, which retained most of the October provisions, continuing the elimination of OHS from the list of tertiary services, the addition of DRG 109, the deletion of DRG 110, the elimination of the requirement for the use of a heart-lung by-pass machine, and the authorization for approval of more than one additional OHS program at a time in the same district. The minimum number of OHS performed by existing providers prior to approval of a new one continued from the October 2000 version, to be decreased from 350 to 250, and the divisor in the numerical need formula continued to be increased from 350 to 500. As in the October version, the requirement that existing providers be able to maintain an annual volume of 350 OHS cases after approval of a new program was stricken. The separate need methodology for counties without an OHS program was simplified, as proposed by Mr. Davis, and was as follows: Regardless of whether need for additional a new adult open heart surgery programs is shown in paragraph (b) above, need for one a new adult open heart surgery program is demonstrated for a county that meets the following criteria: None of the hospitals in the county has an existing or approved open heart surgery program; Residents of the county are projected to generate at least 1200 annual hospital discharges with a principal diagnosis of ischemic heart disease, as defined by ICD-9- CM codes 410.0 through 414.9. The projected number of county residents who will be discharged with a principal diagnosis of ischemic heart disease will be determined as follows: PIHD = (CIHD/CoCPOP X CoPPOP) Where: PIHD = the projected 12-month total of discharges with a principal diagnosis of ischemic heart disease for residents of the county age 15 and over; CIHD = the most recent 12-month total of discharges with a principal diagnosis of ischemic heart disease for residents of the county age 15 and over, as available in the agency's hospital discharge data base; CoCPOP = the current estimated population age 15 and over for the county, included as a component of CPOP in subparagraph 7(b)2; CoPPOP = the planning horizon estimated population age 15 and over for the county, included as a component of PPOP in subparagraph 7(b)2; If the result is 1200 or more, need for one adult open heart surgery program is demonstrated for the county will not normally be approved for a district if the approval would reduce the 12 month total at an existing adult open heart surgery program in the district below 350 open heart surgery operations. In determining whether this condition applies, the agency will calculate (Uc X Px)/(OP + 1). If the result is less than 350 no additional open heart surgery program shall normally be approved. County-specific need identified under paragraph (c) is a need occurring because of the special circumstances in that county, and exists independent of, and in addition to, any district need identified under the provisions of paragraph (b). A program approved pursuant to need identified in paragraph (c) will be included in the subsequent identification of approved and operational programs in the district, as specified in paragraph (a). On January 17, 2001, a public hearing was held to consider the December amendments. Opponents complained that the proposals resulted from a private settlement agreement rather than a public rule development workshop as required by law. They noted that declassification of OHS as a tertiary service is contrary to the recommendations of AHCA's CON advisory study group and the report of the Florida Commission on Excellence in Health Care, co-chaired by AHCA Secretary Reuben King-Shaw, created by the Florida Legislature as a part of the Patient Protection Act of 2000. The risk of inadvertently allowing some OHS procedures to become outpatient services was also raised, because of the statute that specifically states that tertiary services are CON-regulated. The reduction from 350 to 250 in the annual volume required at existing programs prior to approval of new ones was criticized for potentially increasing costs due to shortages in qualified staff, including surgical nurses, perfusionists, recovery and intensive care unit nurses, who are needed to staff the programs. The potential for approval of more than one program at a time, under normal circumstances, was viewed as an effort to respond to the needs of two geographically large districts out of the total of eleven health planning districts in Florida. That, in itself, one witness argued demonstrated that more than one approval at a time should be, as it currently is, a not- normal circumstance. The combination of the district-wide and county- specific need methodologies was criticized as double counting. The district formula which relied on the projected number of OHS, overlapped with the county formula, which used projected ischemic heart disease discharges, to the extent that the same patient hospitalization could result in first, the diagnosis, and then the OHS procedure. Approximately, eighteen percent of diagnosed ischemic heart disease patients in Florida go on to have OHS. The county-specific methodology was also characterized as inappropriate health planning based on geo- political boundaries rather than any realistic access barriers. Although 500, the average size of existing programs was the proposed divisor in the formula, and 250 was the threshold number existing providers, the proposal included the deletion of any provision assuring that existing programs maintain some minimum annual volume, which is 350 in subsection 7(e) of the current rule. AHCA representatives testified that the proposal to delete a minimum adverse impact was inadvertent. The combined effect of a district-wide need methodology, an independent but overlapping county need methodology, and the absence of an adverse impact provision, created concern whether approvals based on county need determinations could reduce volumes at providers in adjacent counties to unsafe levels. Some health planners predicted that, as a consequence of adopting the December draft, like the October version, a number of new OHS programs could be coming into service at one time, seriously draining already scarce resources. One witness, citing an article in the Journal of the American Medical Association, testified that higher volume OHS providers, those over 500 cases, do have better outcomes, and that the relationship persists for angioplasties, including those performed on patients having heart attacks. Florida has 63 or 64 OHS programs. Of those, 25 to 30 percent have annual OHS volumes below 350 surgeries a year. The demand for OHS is increasing slowly and leveling off. AHCA was warned, at the January public hearing by, among others, Eric Peterson, Professor of Cardiology, Duke University Medical Center (by videotaped presentation); and Brian Hummel, M.D., a Cardiothoracic Surgeon in Fort Myers, President of the Florida Society of Thoracic and Cardiovascular Surgeons, that simultaneously easing too many provisions of the OHS rule was a risk to the quality of the programs and the safety of patients. Among other specific comments made at the January public hearing related to the December proposal were the following: This change would authorize a county- specific methodology to support approving a program on the theory that that county needs better access to open heart surgery program. Yet there is no inquiry under the proposed provision into how accessible adjacent programs are or, indeed, how low the volumes of adjacent programs are. Most blatantly, the county provision requires double counting and double need projections. (AHCA Ex. 7, p. 14, by Elizabeth McArthur). The proposed rule creates an exemption for counties that are currently without open heart surgery programs. One can only surmise that the purpose of this exemption is to improve access, and certainly improving access is an appropriate goal and it is possible that there are few situations around the state where access to open heart surgery is a concern, but the proposed rule is completely inadequate and a thoroughly inappropriate way to identify which situations those are . . . (AHCA Ex. 7, p. 26, by Carol Gormley). With the county exemption provision, the Agency has stumbled on an entirely new method for estimating need. In fact, the only good thing about this provision is that it demonstrates that the Agency actually can look at some alternative ways to estimate need, and the use of data about incidence of ischemic heart disease might be one of those. Certainly it should be explored if there is ever a valid planning process that addresses open heart surgery. However, the proposed rules cobble together the county- based epidemiology with the district-wide demand based formula, and I believe that this method is not applicable for evaluating access to care. It is not applicable because the provision only considers the population's rate of ischemic heart disease and does not even attempt to assess the extent to which county residents with ischemic disease are, in fact, already receiving open heart surgery. Therefore, a determination that county residents generate at least 1,200 ischemic heart disease discharges annually does nothing to indicate whether or not they experience any barriers to obtaining that needed service. * * * Another problem with county exemption permission [sic: provision] is that the addition of this assessment, quote "regardless of the results of the district need formula," end quote, constitute double counting of a need in districts where counties without programs are located. (AHCA Ex. 7, p. 27-30, by Carol Gormley). * * * As further evidence of the benefits of limiting open heart surgery to a few high volume programs, the Society would like to place into record the following articles. The first one you've heard on several occasions is the Dudley article, "Selective referral to high volume hospitals." The second, from Farley and Osminkowski, is, "Volume-outcome relationships and in- hospital mortality: Effective changes in volume over time," from Medicare in January of 1992. There's another article from Grumbach, et al., "Regionalization of cardiac surgery in the United States and Canada," again from JAMA. Another article from Hannon, et al., "Coronary artery bypass surgery: The relationship between in-hospital mortality rate and surgical volume after controlling for clinical risk factors," Medical Care. Hughes, et al., "The effects of surgeon volume and hospital volume on quality care in hospitals," again from Medical Care; finally, Riley and Nubriz, "Outcomes of surgeries among Medicare aged: Surgical volume and mortality." Each of these scholarly articles comes to the same inevitable conclusion: outcomes improve as the volume of cardiac surgeries in any given program and hospital increases, therefore increasing the number of hospitals in which these services are provided inevitably will lead to an increase in morbidity. (AHCA Ex. 7, p. 83-84, by Christopher Nuland). * * * On or before the January public hearing, AHCA also received the following written comments: Martin Memorial supports the exception provision for Counties that do not have an open heart surgery program and have a substantial number of residents experiencing cardiovascular disease. This provision ensures an even dispersion of programs, and that adequately sized communities are not denied open heart surgery. (Martin Memorial Ex. 6, Letter of 10/24/2000, from Richard M. Harman, Chief Executive Officer, Martin Memorial, to Elizabeth Dudek) * * * Adding new open heart surgery programs to counties that currently lack programs will increase geographic access to coronary angioplasty services as well as open heart surgery. Primary angioplasty is now the treatment of choice for a significant percentage of patients presenting in the emergency department with acute myocardial infarction (patients who would otherwise be treated with thrombolytic drugs to dissolve blood clots in occluded coronary arteries). Thus, the provision of the proposed regulations that addresses the need for open heart surgery at a county level will also increase access to life-saving invasive cardiology services. The effect of the proposed rule changes is to slightly broaden the circumstances in which the Agency would see presumed need for new programs. Initially, the increase in the number of programs presumed to be needed would be only five. These potential new approvals would be in counties which currently have no programs. This is consistent with the reasoning that supports removing open heart surgery from the list of tertiary procedures. All else equal, distributing new programs to counties where they already exist is reasonable in light of the goal of improving geographic accessibility of advanced cardiology services. As with the other draft proposed rule changes, there is no certainty that any programs will be approved on the basis of the county-specific need formula in (7)(c). These proposed programs would still have to meet the statutory and rule criteria. As discussed above, a number applications for programs have been ultimately denied even when presumed need was shown by the need formula. We recommend adoption of this additional formula for demonstrating need. (IRMH Ex. 1, p. 25, Comments of Ronald Luke, J.D., Ph.D., 10/24/2000) In what could be interpreted as an admission that the process resulting in the development of the earlier drafts was flawed, Jeff Gregg, Chief of the AHCA CON Bureau, concluded the January public hearing by saying, . . . in terms of the analysis that the Agency did about the proposed rule, I would simply have to tell you that CON staff was not involved in that analysis, and that's CON staff including myself. So I cannot elaborate on what went into it. But having said that, I do want to assure you that CON staff will be involved in further analysis and we will do our best to consider all the points that have been made and present them as clearly and concisely as we can in assisting the Agency to formulate its response to this hearing. (AHCA Ex. 7, p. 86). The December draft was also challenged by a number of Petitioners in DOAH Case No. 01-0372RP, filed on January 26, 2001, and ten other consolidated cases. In response to the criticism that the adverse impact provision should not have been deleted and because that omission was unintended, AHCA published another proposed amendment to the OHS rule, on May 4, 2001, reinstating a minimum adverse impact volume, this time set at 250 OHS operations, down from 350 in the existing rule. On May 31, 2001, AHCA and the other parties to DOAH Case No. 01-0372RP and the consolidated cases entered into another settlement agreement, which provided: that in an effort to avoid further administrative proceedings, without conceding the correctness of any position taken by any party, and in response to materials received in to the record on or before the public hearing, the Agency for Health Care Administration agrees to publish and support . . . The Notice of Change . . . (Bethesda Ex. 34, p. 2-3). In upholding that agreement, AHCA superseded or revised all prior drafts and published a notice of change on June 15, 2001. In this final version, AHCA limited normal approval of a new OHS program to one at a time, used 500 as the numeric need formula divisor, increased the required prior-to-approval OHS minimum volume at mature existing providers from 250 in the October version to 300 (down from 350 in the existing rule) and for non- mature programs from a monthly average of 21 in the October draft to 25 (down from 29 in the existing rule), retained the classification of OHS as a tertiary service, and altered the separate, independent county need methodology to make it a county preference. The June 15th version, containing Subsections 7(c) and 7(d), which are challenged in this case is as follows: Adult Open Heart Surgery Program Need Determination. An additional open heart surgery programs shall not normally be approved in the district if any of the following conditions exist: There is an approved adult open heart surgery program in the district; One or more of the operational adult open heart surgery programs in the district that were operational for at least 12 months as of 3 months prior to the beginning date of the quarter of the publication of the fixed need pool performed less than 300 adult open heart surgery operations during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool; One or more of the adult open heart surgery programs in the district that were operational for less than 12 months during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool performed less than an average of 25 adult open heart surgery operations per month. * * * Provided that the provisions of paragraphs (7)(a) do not apply, the agency shall determine the net need for an additional adult open heart surgery programs in the district based on the following formula: NN=[(POH/500)-OP]> 0.5 where: NN = the need for an additional adult open heart surgery programs in the district projected for the applicable planning horizon. The additional adult open heart surgery program may be approved when NN is 0.5 or greater. POH = the projected number of adult open heart surgery operations that will be performed in the district in the 12-month period beginning with the planning horizon. To determine POH, the agency will calculate COH/CPOP x PPOP, where: COH = the current number of adult open heart surgery operations, defined as the number of adult open heart surgery operations performed in the district during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool. CPOP = the current district population age 15 years and over. PPOP = the projected district population age 15 years and over. For applications submitted between January 1 and June 30, the population estimate used for CPOP shall be for January of the preceding year; for applications submitted between July 1 and December 31, the population estimate used for CPOP shall be for July of the preceding year. The population estimates used for COP and PPOP shall be the most recent population estimates of the Executive Office of the Governor that are available to the agency 3 weeks prior to publication of the fixed need pool. OP = the number of operational adult open heart surgery programs in the district. In the event there is a demonstrated numeric need for an additional adult open heart surgery program pursuant to paragraph (7)(b), preference shall be given to any applicant from a county that meets the following criteria: None of the hospitals in the county has an existing or approved open heart surgery program; and Residents of the county are projected to generate at least 1200 annual hospital discharges with a principal diagnosis of ischemic heart disease, as defined by ICD-9- CM codes 410.0 In the event no numeric need for an additional adult open heart surgery program is shown in paragraphs (7)(a) or (7)(b) above, the need for enhanced access to health care for the residents of a service district is demonstrated for an applicant in a county that meets the criteria of paragraph (7)(c)1. and 2. above. An additional adult open heart surgery program will not normally be approved for the district if the approval would reduce the 12 month total at an existing adult open heart surgery program in the district below 300 open heart surgery operations. Bethesda objects to Subsections 7(c) and 7(d) as invalid. It challenges the rule promulgation process as a sham, having resulted from settlement negotiations rather than from statutorily mandated considerations and processes. That charge was, in effect, conceded by AHCA, as related to the October draft. That version carried over into the December draft, essentially unchanged, but did gain support at the October workshop. The October and December versions are not at issue in this proceeding. The proposed rule amendments at issue in this proceeding must have been supported by information provided to AHCA before or during the January public hearing. The proposal at issue differs substantially from the terms of the September settlement agreement, but is precisely what was attached to the May 31, 2001, settlement agreement. For example, the settlement agreement of September 11, 2000, included a proposal to reduce the prior minimum volume of cases at existing OHS providers from 350 to 250, but in May and June, that number was set at 300. AHCA, in the September settlement agreement, was to eliminate any limitation on the number of additional programs approved at a time, but the May and June version retains the one-at-a-time provision of the existing rule. AHCA agreed to determine county numeric need independent of and in addition to district numeric need, in September, but that provision is, in the May 31st and June 15th version, a preference. In September 2000, AHCA agreed to delete adult OHS from the list of tertiary services in Rule 59C-1.002(41), but it is a tertiary service in the May and June version. Bethesda is correct that the records of the October workshop and January public hearing contained criticisms of the county need methodology but no specific proposal to modify it into a preference. The first draft of that concept is the May 31, 2001, settlement agreement. (See Findings of Fact 26 and 27). Statutory rule-making issues Subsection 408.034(3), Florida Statutes, provides that: The Agency shall establish, by rule uniform, need methodologies for health care services and health facilities. In developing uniform need methodologies, the agency shall, at a minimum, consider the demographic characteristics of the population, the health status of the population, service use patterns, standards and trends, geographic accessibility, and market economics. As required by statute, AHCA considered the demographics and health status of the population and examined, as a part of the rule adopting process, age-specific calculations of ischemic heart disease. AHCA relied on statistical evidence of the relationship of ischemic heart disease and OHS. In 1999, for example, there were 33,027 OHS in Florida, and 25,257 of those patients had a primary diagnosis of ischemic heart disease. Consideration of service use patterns, and standards and trends related to OHS led AHCA to increase the divisor in the numeric need formula to maintain the average size of 500 surgeries for existing providers. The availability of more reliable data than that collected when the existing rule was promulgated allowed AHCA to propose reliance on residential use rates. The trend towards the use of angioplasty, as a preferred treatment for heart attack patients, and the need for timely geographical access to care are major factors for AHCA's proposal to consider a county services within the normal need analysis or as a not normal indication of a need for enhanced access when a county has a critical mass of heart disease patients. Geographical accessibility is also addressed in the travel time standard in the existing rule, which the proposal would not change. AHCA received testimony on the issue of market economics and health status, related to care for indigent and minority patients in not-for-profit, county-funded hospitals, and related to reimbursement formulas. The record demonstrates that AHCA was provided with evidence on the effect of scare resources on the costs of operating OHS programs. County-specific need methodology in earlier drafts as compared to the county preference in 7(c) and the need for enhanced access in 7(d) Bethesda alleges that the county preference in the June version is essentially another need methodology, like the county-specific need methodology in the earlier versions of the proposed rule. Bethesda also contends that a preference for a hospital because it is in a county which does not have an open heart program over a reasonably accessible facility in an adjoining county in the same district is irrational health planning which could lead to a maldistribution of programs. The county-specific need methodology was first included in the September settlement agreement, and the preference in 7(c) and need for access in 7(d), originated after the January 17, 2001, public hearing. During the public hearing, counsel for the Florida Hospital Association complained that the county-specific need methodology precluded any inquiry into accessibility and volumes at adjoining programs. Another representative of the Florida Hospital Association surmised that the goal of the county exemption was improved access but explained that it was an inappropriate means to identify access concerns. For example, while Hernando County would qualify for need with the separate methodology, most of its residents, 97 percent receive OHS services at a hospital in another district which is only 13 miles from the population center. (See Finding of Fact 26). The preference under normal circumstances in Subsection 7(c) and finding of need for enhanced access in Subsection 7(d), must be supported by evidence that county boundaries, in general, do create valid access issues. On or before the January workshop, information provided to AHCA indicated that some special inquiry into access issues related to CON applications for programs in counties without OHS programs is warranted. See Finding of Fact 27). AHCA found correctly that counties matter for several reasons. First is the fact that emergency services are funded and organized by counties, in general, and operated by municipal and county agencies. Approximately 60 percent of heart attack patient discharges in Florida are admitted through emergency rooms. Emergency heart attack patients who live in counties with OHS programs are twice as likely to be taken to a hospital with OHS as those who live in counties without an OHS provider. Second, whether a patient is taken to an OHS provider affects the care received. The probability of having an angioplasty performed is almost 50 percent greater for residents of counties with OHS programs as compared to those in counties without an OHS program. Third, some health care reimbursement plans and health care districts are operated within counties, limiting financial access to out-of-county hospitals. AHCA has always considered whether or not a county has an OHS program as a part of access issues. The issue of greater access to OHS was the basis for AHCA's initial consideration of the possibility of easing the OHS rule. With the May and June draft, it has codified and specified when that policy will apply. AHCA's deputy secretary noted that geographic access in the absence of numeric need was the basis for approvals of OHS CONs for Marion County, and for hospitals located in Naples and Brandon. In each instance, the applicants argued a need for enhanced access. AHCA has experience in applying preferences as a part of balancing and weighing criteria from statutes, rules and local health plans, particularly to distinguish among multiple applicants. In the totality of the review process, other factors which Bethesda's expert testified should be considered, including financial, racial and other potential access barriers, are not precluded. Preferences related to specific locations within health planning areas are included in CON rules governing the need for nursing home beds and hospices. Bethesda noted that these are not tertiary services, suggesting that a county location preference is inappropriate for tertiary services, but similar preferences for OHS exist in some of the local health plans. In AHCA District 1, the CON allocation factors for OHS and cardiac catheterization services include a preference for applicants proposing to locate in a county which does not have an existing OHS program. In District 4, the preference favors an applicant located in a concentrated population area in which existing programs have the highest area use rates. District 5 is similar to District 4, supporting OHS projects in areas of concentrated population with the highest use rates. The District 8, like District 1, preference goes to the applicant located in a county without an OHS program. There is no evidence that the existing preferences have been difficult to apply within the context of other CON criteria for the review of OHS applications. In effect, the proposed amendments establish an uniform state-wide county preference which is more concrete in terms of the requirements for a potential patient base. Bethesda has questioned the rationale for standards which are, in effect, different in Subsection 7(c) as compared to Subsection 7(d). The lower requirement, according to Bethesda, 1200 ischemic heart diagnoses, in 7(d), applies when there is no numeric need. But, the 500 divisor and 300 minimum at existing providers, when combined with 1200 ischemic heart diagnoses is a heavier burden to meet in 7(c), although under normal circumstances. Bethesda did not adequately explain reasons for this objection to the proposed rule. In addition, it is not inconsistent logically for AHCA to require applicants to demonstrate lower numeric need in situations in which AHCA has determined that these will be, in general, a greater need for enhanced access. Bethesda also raised a concern for the eventual maldistribution of programs as a result of the county preference. In 1999, Palm Beach county residents received 2700 OHS, or an average of 900 cases for each of the three programs. The total for District 9 was 3800 cases in 1999. When 500 St. Lucie County resident cases, in which Lawnwood is an OHS provider, are combined with 2700 Palm Beach resident cases, that leaves only 650 resident cases from Okeechobee, Indian River and Martin Counties. If programs are approved in all three, then the total will be inadequate for each to reach 300 cases, while, presumably, the demand in Palm Beach could be increasing disproportionately and not be met adequately. Disproportionate need, the appropriate dispersion of programs, and the benefits of enhanced competition are among the factors which AHCA can consider along with county need when choosing among competing applicants. 1200 ischemic heart disease discharges The proposed amendments require a projection that residents will reach a threshold of 1200 cases of ischemic heart disease discharges as a condition for the entitlement to the numeric need preference or to demonstrate a not normal need for enhanced access. In general, ischemic heart disease, which is also known as coronary heart disease, is characterized by blocked arteries which, in turn, limit blood to heart muscles causing first the onset of angina from acute coronary syndrome, progressing on to acute myocardial infarction, or a heart attack. The use of heart disease as a proxy for OHS utilization is consistent with AHCA's use of live births in pediatric open heart surgery and pediatric cardiac catheterization rules, deaths in the hospice rule, and related diagnoses in organ transplantation rules rather than actual utilization. It was supported by information received during or before the January workshop (See Finding of Fact 26 and 27). Bethesda's criticism of the use of a proxy per se is also not well-founded because any single statistical approach could be misleading. For example, historic use rates can understate future use with a growing service or an artificially imposed access limit. Using heart disease data in a preference or a need for enhanced access as opposed to a need formula or conclusive finding allows more flexibility in determining need in conjunction with other significant factors. One of Bethesda's expert health planners was also critical of the use of 1200 ischemic heart disease diagnoses as inadequate for projecting OHS cases, and for not equating to approximately 300 annual OHS cases, the minimum required of existing providers in Subsection 7(a) and the minimum adverse impact allowed in Subsection 7(e). Based on actual historical Florida data, 1200 ischemic heart disease diagnoses on average resulted in 207 OHS in 1997, 203 in 1998, and 203 in 1999. Ischemic heart disease has approximately an 18 to 20 percent conversion rate to OHS, and results in a total of 76 to 80 percent of all OHS cases. OHS cases from other diagnoses added statistically another 54 OHS in 1997, 59 in 1998, and 61 in 1999, to those from ischemic heart disease, giving, in each year a total less than 300. Bethesda presented evidence of wide variations in the ischemic heart disease to OHS conversion ratios from county-to- county. For example, only 14 percent of Bradford County ischemic heart diseases converted to OHS, and only 11 percent of the 700 cases in Columbia County converted to OHS. In Columbia County, the average state conversion rate of 20 percent yields 140 cases but, in reality, there were only 78 OHS cases from Columbia County in 1999. Bethesda's expert concluded that conversion ratio discrepancies resulting in the approval of a program that cannot achieve 300 OHS, as required in Subsection 7(a)2. and 7 (e), of the proposed rule, could bar the approval of new programs when needed in the district and would not be of minimum required quality. Bethesda also proved that the accuracy of projected OHS cases can also be affected by patterns of patient migration for health care, particularly if in- and out-migration do not offset each other. In counties with OHS programs, the average out-migration for acute care is 10.7 percent, varying widely from 3.8 percent in Alachua County to 70 percent in Seminole County. In counties without an OHS provider, average out- migration for acute care is 44 percent, but ranges from 17.6 percent in Indian River County to 98 percent in Baker County. An average of 18 percent of the residents of Florida counties with OHS programs have their surgeries performed elsewhere. Like out-migration, in-migration for acute care, for ischemic heart disease care, and for OHS varies from county to county in Florida. Counties without OHS programs have acute care in-migration from lows of 5.3 percent for Flagler County up to highs of 40 percent for Columbia County. In counties with OHS, in-migration for acute care is as low as 8 percent for Brevard and Polk, and as high as 60 percent for Alachua County. Similarly, in-migration, as determined by ischemic heart disease discharges averages 19.4 percent in counties without OHS programs and approximately 25 percent in those with OHS. In-migration for OHS, averages 35.7 percent for the state, but that is derived from a range from 9.2 percent in Pinellas County to 74 percent in Alachua and Leon Counties. Bethesda demonstrated, patterns of migration for health care vary throughout Florida, but there are trends due to the presence of OHS programs. Average net in-migration to counties with OHS is 29 percent, and is positive in sixteen of the twenty-four counties with OHS programs. All of these differences can be considered within the regulatory scheme proposed by AHCA. The issue of whether 1200 residential ischemic heart disease diagnoses is, in fact, the critical mass of prospective OHS patients needed or is deceptive due to migration patterns, due to access to alternative providers or any other review criteria listed in rule or statutes can be considered on a case-by-case basis with the proposed amendments. Bethesda's specific concern is that Indian River with well over 1200 ischemic heart disease discharges could be approved even though that represented only 255 OHS cases, and that if Indian River is approved under the county preference provision, then Bethesda would not be approved under normal circumstances until Indian River achieved and was projected to maintain 300 OHS cases a year. That Bethesda may be delayed in meeting the requirements for normal need is likely, but that appears to be a function of its location as compared to existing providers as much as it is the result of the county preference. Bethesda is not precluded, however, under either the existing or proposed rules from demonstrating not normal circumstances in District 9 for the issuance of an OHS CON to Bethesda. Bethesda's assumption that 300 is the minimum volume required for adequate quality is not supported by studies from various professional societies. The American College of Cardiology, the American Heart Association, and the Society of Thoracic Surgeons set minimums of 200 to 250 annual hospital cases as the volumes necessary to maintain the skills of the staff. The American College of Surgeons, in 1996, published their opinion that 100 to 125 cases per hospital is sufficient for quality, while at least 200 cases a year are needed for the economic efficiency of a program. AHCA has never used the required and protected volumes as the volume which must also be projected for a new programs. In the current OHS rule, the volume required is 350 a year for existing programs but that has not been required of applicants. In the recent approval of an OHS CON for Brandon Regional Hospital, the applicant projected reaching 287 cases in the third year of operation. County preference, tertiary classification and travel time Bethesda argued that the tertiary classification, suggesting a regional approach, is inconsistent with having a county access provision. Bethesda correctly noted that the county provision first appeared in a draft which included the elimination of OHS from the list of tertiary services. But AHCA proposes to establish the county preference and to maintain OHS on the list of tertiary services under Rule 59C-1.002(41), and to maintain the two-hour drive time standard in Rule 59C- 1.033(4)(a). Substantial information, mostly from medical doctors and studies linking morbidity to low volume, supports the view that OHS continues to be a complex service. Obviously, those services in the tertiary classification range in complexity and availability from OHS at the lower level to organ transplantation at the upper level. The tertiary classification is justified to assure AHCA's continued closer scrutiny of OHS CON applications. It is also consistent with the increase in the need formula divisor to 500, which together serve as restrains on the approval of additional programs. AHCA reasonably concluded, based on case law and precedents with local health plan that it is not inconsistent to apply county preferences to OHS while it is classified a tertiary service. The two-hour travel time standard, is as follows: Adult open heart surgery shall be available within a maximum automobile travel time of 2 hours under average travel conditions for at least 90 percent of the district's population. The counties most likely qualify for the preference, based on meeting or exceeding 1200 residential ischemic heart disease diagnoses, are Citrus, Martin, Hernando, St. Johns, Highlands, Indian River, and Okaloosa. The population centers in each of these counties are well within two hours of an existing provider. Citrus County, in which there is an approved but not yet operational OHS program, is about an hour's drive from Marion County. Hernando is approximately 25 minutes from the Pasco County provider. The population center of St. Johns County is approximately 40 minutes away from Duval County OHS providers. Okaloosa County is approximately a one-hour drive away from Escambia County OHS providers. In District 9, Indian River is approximately a 30- minute drive from the Lawnwood OHS program. Martin Memorial, is an approved provider, is approximately 20 miles or 35 minutes from Lawnwood and 30 miles or 40 minutes from Palm Beach Gardens, another existing OHS provider. In the next three to five years, it is foreseeable that Okeechobee County in northwestern District 9 could qualify for the county preference. Adjacent to Okeechobee, Highlands County's population can drive either an hour and thirty minutes to a Charlotte County OHS program or an hour and twenty minutes to a Polk County facility. The evidence related to travel times, according to one of Bethesda's experts, demonstrates that the county preference is not needed to assure access which is already provided for each and every likely qualifying county. But the population centers in the entire state of Florida are all within the two- hour travel standard, and there has been no suggestion that Florida cease approval of new OHS programs. Bethesda's contention that no need exists for enhanced access if the travel time standard is met, and its claim that the rule is internally inconsistent with a county preference and two-hour drive time are rejected. Two hours is, as the rule clearly states, a "maximum" not a bar, and has never been interpreted by AHCA as a bar, to more proximate locations. Any other interpretation is an impossibility considering the numerous counties across the state with multiple programs, including Dade, Broward, Palm Beach, Hillsborough, Pinellas, Orange, Volusia, Duval, and Escambia, among others. AHCA can appropriately and consistently establish reasonable guidelines for choosing among applicants to enhance access within the maximum travel standard. There is no language in the proposed rule indicating when it will take effect. Although the issue was raised in Bethesda's petition, it failed to provide evidence or legal arguments at hearing or subsequently to support its objection to the omission. AHCA's deputy secretary testified that the agency reviews applications using need methodology rules in effect when the applications are filed. Before new rules are applied, applicants are given the opportunity to reapply to address new provisions in a rule.
The Issue Whether the Certificate of Need application (CON 9239) of Galencare, Inc., d/b/a Brandon Regional Hospital ("Brandon") to establish an open heart surgery program at its hospital facility in Hillsborough County should be granted?
Findings Of Fact District 6 District 6 is one of eleven health service planning districts in Florida set up by the "Health Facility and Services Development Act," Sections 408.031-408.045, Florida Statutes. See Section 408.031, Florida Statutes. The district is comprised of five counties: Hillsborough, Manatee, Polk, Hardee, and Highlands. Section 408.032(5), Florida Statutes. Of the five counties, three have providers of adult open heart surgery services: Hillsborough with three providers, Manatee with two, and Polk with one. There are in District 6 at present, therefore, a total of six existing providers. Existing Providers Hillsborough County The three providers of open heart surgery services ("OHS") in Hillsborough County are Florida Health Sciences Center, Inc., d/b/a Tampa General Hospital ("Tampa General"), St. Joseph's Hospital, Inc. ("St. Joseph's"), and University Community Hospital, Inc., d/b/a University Community Hospital ("UCH"). For the most part, Interstate 75 runs in a northerly and southerly direction dividing Hillsborough County roughly in half. If the interstate is considered to be a line dividing the eastern half of the county from the western, all three existing providers are in the western half of the county within the incorporated area of the county's major population center, the City of Tampa. Tampa General Opened approximately a century ago, Tampa General has been at its present location in the City of Tampa on Davis Island at the north end of Tampa Bay since 1927. The mission of Tampa General is three-fold. First, it provides a range of care (from simple to complex) for the west central region of the state. Second, it supports both the teaching and research activities of the University of South Florida College of Medicine. Finally and perhaps most importantly, it serves as the "health care safety net" for the people of Hillsborough County. Evidence of its status as the safety net for those its serves is its Case Mix Index for Medicare patients: 2.01. At such a level, "the case mix at Tampa General is one of the highest in the nation in Medicare population." (Tr. 2452). In keeping with its mission of being the county's health care safety net, Tampa General is a full-service acute care hospital. It also provides services unique to the county and the Tampa Bay area: a Level I trauma center, a regional burn center and adult solid organ transplant programs. Tampa General is licensed for 877 beds. Of these, 723 are for acute care, 31 are designated skilled nursing beds, 59 are comprehensive rehabilitation beds, 22 are psychiatry beds, and 42 are neonatal intensive care beds (18 Level II and 24 Level III). Of the 723 acute care beds, 160 are set aside for cardiac care, although they may be occupied from time-to-time by non-cardiac care patients. Tampa General is a statutory teaching hospital. It has an affiliation with the University of South Florida College of Medicine. It offers 13 residency programs, serving approximately 200 medical residents. Tampa General offers diagnostic and interventional cardiac catheterization services in four laboratories dedicated to such services. It has four operating rooms dedicated to open heart surgery. The range of open heart surgery services provided by Tampa General includes heart transplants. Care of the open heart patient immediately after surgery is in a dedicated cardiovascular intensive care unit of 18 beds. Following stay in the intensive care unit, the patient is cared for in either a 10-bed intermediate care unit or a 30- bed telemetry unit. Tampa General's full-service open heart surgery program provides high quality of care. St. Joseph's Founded by the Franciscan Sisters of Allegheny, New York, St. Joseph's is an acute care hospital located on Martin Luther King Boulevard in an "inner city kind of area" (Tr. 1586) of the City of Tampa near the geographic center of Hillsborough County. On the hospital campus sit three separate buildings: the main hospital, consisting of 559 beds; across the street, St. Joseph's Women's Hospital, a 197-bed facility dedicated to the care of women; and, opened in 1998, Tampa Children's Hospital, a 120-bed free-standing facility that offers pediatric services and Level II and Level III neonatal intensive care services. In addition to the women's and pediatric facilities, and consistent with the full-service nature of the hospital, St. Joseph's provides behavioral health and oncology services, and most pertinent to this proceeding, open heart surgery and related cardiovascular services. Designated as a Level 2 trauma center, St. Joseph's has a large and active emergency department. There were 90,211 visits to the Emergency Room in 1999, alone. Of the patients admitted annually, fifty-five percent are admitted through the Emergency Room. The formal mission of St. Joseph's organization is to take care of and improve the health of the community it serves. Another aspect of the mission passed down from its religious founders is to take care of the "marginalized, . . . the people that in many senses cannot take care of themselves, [those to whom] society has . . . closed [its] eyes . . .". (Tr. 1584). In keeping with its mission, it is St. Joseph's policy to provide care to anyone who seeks its hospital services without regard to ability to pay. In 1999, the hospital provided $33 million in charity care, as that term is defined by AHCA. In total, St. Joseph's provided $121 million in unfunded care during the same year. Not surprisingly, St. Joseph's is also a disproportionate Medicaid provider. The only hospital in the district that provides both adult and pediatric open heart surgery services, St. Joseph's has three dedicated OHS surgical suites, a 14-bed unit dedicated to cardiovascular intensive care for its adult OHS patients, a 12-bed coronary care unit and 86 progressive care beds, all with telemetry capability. St. Joseph's provides high quality of care in its OHS. UCH University Community Hospital, Inc., is a private, not-for-profit corporation. It operates two hospital facilities: the main hospital ("UCH") a 431-bed hospital on Fletcher Avenue in north Tampa, and a second 120-bed hospital in Carrollwood. UCH is accredited by the JCAHO "with commendation," the highest rating available. It provides patient care regardless of ability to pay. UCH's cardiac surgery program is called the "Pepin Heart & Vascular Institute," after Art Pepin, "a 14-year heart transplant recipient [and] . . . the oldest heart transplant recipient in the nation alive today." (Tr. 2841). A Temple Terrace resident, Mr. Pepin also helped to fund the start of the institute. Its service area for tertiary services, including OHS, includes all of Hillsborough County, and extends into south Pasco County and Polk County. The Pepin Institute has excellent facilities and equipment. It has three dedicated OHS operating suites, three fully-equipped "state-of-the-art" cardiac catheterization laboratories equipped with special PTCA or angioplasty devices, and several cardiology care units specifically for OHS/PTCA services. Immediately following surgery, OHS patients go to a dedicated 8-bed cardiovascular intensive care unit. From there patients proceed to a dedicated 20-bed progressive care unit ("PCU"), comprised of all private rooms. There is also a 24-bed PCU dedicated to PTCA patients. There is another 22-bed interventional unit that serves as an overflow unit for patients receiving PTCA or cardiac catheterization. UCH has a 22-bed medical cardiology unit for chest pain observation, congestive heart failure, and other cardiac disorders. Staffing these units requires about 110 experienced, full-time employees. UCH has a special "chest pain" Emergency Room with specially-trained cardiac nurses and defined protocols for the treatment of chest pain and heart attacks. UCH offers a free van service for its UCH patients and their families that operates around the clock. As in the case of the other two existing providers of OHS services in Hillsborough Counties, UCH provides a full range of cardiovascular services at high quality. Manatee County The two existing providers of adult open heart surgery services in Manatee County are Manatee Memorial Hospital, Inc., and Blake Medical Center, Inc. Neither are parties in this proceeding. Although Manatee Memorial filed a petition for formal administrative hearing seeking to overturn the preliminary decision of the Agency, the petition was withdrawn before the case reached hearing. Polk County The existing provider of adult open heart surgery services in Polk County is Lakeland Regional Medical Center, Inc. ("Lakeland"). Licensed for 851 beds, Lakeland is a large, not-for- profit, tertiary regional hospital. In 1999, Lakeland admitted approximately 30,000 patients. In fiscal 1999, there were about 105,000 visits to Lakeland's Emergency Room. Lakeland provides a wide range of acute care services, including OHS and diagnostic and therapeutic cardiac catheterization. It draws its OHS patients from the Lakeland urban area, the rest of Polk County, eastern Hillsborough County (particularly from Plant City), and some of the surrounding counties. Lakeland has a high quality OHS program that provides high quality of care to its patients. It has two dedicated OHS surgical suites and a third surgical suite equipped and ready for OHS procedures on an as-needed basis. Its volume for the last few years has been relatively flat. Lakeland offers interventional radiology services, a trauma center, a high-risk obstetrics service, oncology, neonatal intensive care, pediatric intensive care, radiation therapy, alcohol and chemical dependency, and behavioral sciences services. Lakeland treats all patients without regard to their ability to pay, and provides a substantial amount of charity care, amounting in fiscal year 1999 to $20 million. The Applicant Brandon Regional Hospital ("Brandon") is a 255-bed hospital located in Brandon, Florida, an unincorporated area of Hillsborough County east of Interstate 75. Included among Brandon's 255 beds are 218 acute care beds, 15 hospital-based skilled nursing unit beds, 14 tertiary Level II neonatal intensive care unit ("NICU") beds, and 8 tertiary Level III NICU beds. Brandon offers a wide array of medical specialties and services to its patients including cardiology; internal medicine; critical care medicine; family practice; nephrology; pulmonary medicine; oncology/hematology; infectious disease; neurology; psychiatry; endocrinology; gastroenterology; physical medicine; rehabilitation; radiation oncology; pathology; respiratory therapy; and anesthesiology. Brandon operates a mature cardiology program which includes inpatient diagnostic cardiac catheterization, outpatient diagnostic cardiac catheterization, electrocardiography, stress testing, and echocardiography. The Brandon medical staff includes 22 Board-certified cardiologists who practice both interventional and invasive cardiology. Board certification is a prerequisite to maintaining cardiology staff privileges at Brandon. Brandon's inpatient diagnostic cardiac catheterization program was initiated in 1989 and has performed in excess of 800 inpatient diagnostic cardiac catheterization procedures per year since 1996. Brandon's daily census has increased from 159 to 187 for the period 1997 to 1999 commensurate with the burgeoning population growth in Brandon's primary service area. Brandon's Emergency Room is the third busiest in Hillsborough County and has more visits than Tampa General's Emergency Room. From 1997- 1999, Brandon's Emergency Room visits increased from 43,000 to 53,000 per year and at the time of hearing were expected to increase an additional 5-6 percent during the year 2000. Brandon has also recently expanded many services to accommodate the growing health care needs of the Brandon community. For example, Brandon doubled the square footage of its Emergency Room and added 17 treatment rooms. It has also implemented an outpatient diagnostic and rehabilitation center, increased the number of labor, delivery and recovery suites, and created a high-risk ante-partum observation unit. Brandon was recently approved for 5 additional tertiary Level II NICU beds and 3 additional tertiary Level III NICU beds which increased Brandon's Level II/III NICU bed complement to 22 beds. Brandon is a Level 5 hospital within HCA's internal ranking system, which is the company's highest facility level in terms of service, revenue, and patient service area population. Brandon has been ranked as one of the Nation's top 100 hospitals by HCIA/Mercer, Inc., based on Brandon's clinical and financial performance. The Proposal On September 15, 1999, Brandon submitted to AHCA CON Application 9239, its third application for an open heart surgery program in the past few years. (CON 9085 and 9169, the two earlier applications, were both denied.) The second of the three, CON 9169, sought approval on the basis of the same two "not normal" circumstances alleged by Brandon to justify approval in this proceeding. CON 9239 addresses the Agency's January 2002 planning horizon. Brandon proposes to construct two dedicated cardiovascular operating rooms ("CV-OR"), a six-bed dedicated cardiovascular intensive care unit ("CVICU"), a pump room and sterile prep room all located in close proximity on Brandon's first floor. The costs, methods of construction, and design of Brandon's proposed CV-OR, CVICU, pump room, and sterile prep room are reasonable. As a condition of CON approval, Brandon will contribute $100,000 per year for five years to the Hillsborough County Health Care Program for use in providing health care to the homeless, indigent, and other needy residents of Hillsborough County. The administration at Brandon is committed to establishing an adult open heart surgery program. The proposal is supported by the medical and nursing staff. It is also supported by the Brandon community. The Brandon Community in East Hillsborough County Brandon, Florida, is a large unincorporated community in Hillsborough County, east of Interstate 75. The Brandon area is one of the fastest growing in the state. In the last ten years alone, the area's population has increased from approximately 90,000 to 160,000. An incorporated Brandon municipality (depending on the boundaries of the incorporation) has the potential to be the eighth largest city in Florida. The Brandon community's population is projected to further increase by at least 50,000 over the next five to ten years. Brandon Regional Hospital's primary service area not only encompasses the Brandon community, but further extends throughout Hillsborough County to a populous of nearly 285,000 persons. The population of Brandon's primary service area is projected to increase to 309,000 by the year 2004, of which approximately 32,000 are anticipated to be over the age of 65, making Brandon's population "young" relative to much of the rest of the State. The community of Brandon has attracted several new large housing developments which are likely to accelerate its projected growth. According to the Hillsborough County City- County Planning Commission, six of the eleven largest subdivisions of single-family homes permitted in 1998 are located nearby. For example, the infrastructure is in place for an 8,000-acre housing development east of Brandon which consists of 7,500 homes and is projected to bring in 30,000 people over the next 5-10 years. Two other large housing developments will bring an additional 5,000-10,000 persons to the Brandon area. The community of Brandon is also an attractive area for relocating businesses. Recent additions to the Brandon area include, among others, CitiGroup Corporation, Atlantic Lucent Technologies, Household Finance, Ford Motor Credit, and Progressive Insurance. CitiGroup Corporation alone supplemented the area's population with approximately 5,000 persons. The community of Brandon has experienced growth in the development of health care facilities with 5 new assisted living facilities and one additional assisted living facility under construction. The average age of the residents of these facilities is much higher than of the Brandon area as a whole. Existing Providers' Distance from Brandon's PSA Brandon's primary service area ("PSA") is comprised of 12 zip code areas "in and around Brandon, essentially eastern Hillsborough County." (Tr. 1071). Using the center of each zip code in Brandon's primary service area as the location for each resident of the zip code area, the residents of Brandon's PSA are an average of 15 miles from Tampa General, 16.4 miles from St. Joseph's, 17.3 miles from UCH and 24.6 miles from Lakeland Regional Medical Center. In contrast, they are only 7.7 miles from Brandon Regional Hospital. Using the same methodology, the residents of Brandon's PSA are an average of more than 40 miles from Blake Medical Center (44.9 miles) and Manatee Memorial (41 miles). Numeric Need Publication Rule 59C-1.033, Florida Administrative Code (the "Open Heart Surgery Program Rule" or the "Rule") specifies a methodology for determining numeric need for new open heart surgery programs in health planning districts. The methodology is set forth in section (7) of the Rule. Part of the methodology is a formula. See subsection (b) of Section (7) of the Rule. Using the formula, the Agency calculated numeric need in the District for the January 2002 Planning Horizon. The calculation yielded a result of 3.27 additional programs needed to serve the District by January 1, 2002. But calculation of numeric need under the formula is not all that is entailed in the complete methodology for determining numeric need. Numeric need is also determined by taking other factors into consideration. The Agency is to determine net need based on the formula "[p]rovided that the provisions of paragraphs (7)(a) and (7) (c) do not apply." Rule 59C-1.033(b), Florida Administrative Code. Paragraph (7)(a) states, "[a] new adult open heart surgery program shall not normally be approved in the district" if the following condition (among others) exists: 2. One or more of the operational adult open heart surgery programs in the district that were operational for at least 12 months as of 3 months prior to the beginning date of the quarter of the publication of the fixed need pool performed less than 350 adult open heart surgery operations during the 12 months ending 3 months prior to the beginning date of the quarter of the publication of the fixed need pool; . . . Rule 59C-1.033(7)(a), Florida Administrative Code. Both Blake Medical Center and Manatee Memorial Hospital in Manatee County were operational and performed less that 350 adult open heart surgery operations in the qualifying time periods described by subparagraph (7)(a)2., of the Rule. (Blake reported 221 open heart admissions for the 12-month period ending March 31, 1999; Manatee Memorial for the same period reported 319). Because of the sub-350 volume of the two providers, the Rule's methodology yielded a numeric need of "0" new open heart surgery programs in District 6 for the January 2002 Planning Horizon. In other words, the numeric need of 3.27 determined by calculation pursuant to the formula prior to consideration of the programs described in (7)(a)2.1, was "zeroed out" by operation of the Rule. Accordingly, a numeric need of zero for the district in the applicable planning horizon was published on behalf of the Agency in the January 29, 1999, issue of the Florida Administrative Weekly. No Impact on Manatee County Providers In 1998, only one resident of Brandon's PSA received an open heart surgery procedure in Manatee County. For the same period only two residents from Brandon's PSA received an angioplasty procedure in Manatee County. These three residents received the services at Manatee Memorial. Of the two Manatee County programs, Manatee Memorial consistently has a higher volume of open heart surgery cases and according to the latest data available at the time of hearing has "hit the mark" (Tr. 1546) of 350 procedures annually. Very few residents from other District 6 counties receive cardiac services in Manatee County. Similarly, very few Manatee county residents migrate from Manatee County to another District 6 hospital to receive cardiac services. In 1998, only 19 of a total 1,209 combined open heart and angioplasty procedures performed at either Blake or Manatee Memorial originated in the other District 6 counties and only two were from the Brandon area. Among the 6,739 Manatee County residents discharged from a Florida hospital in calendar year 1998 following any cardiovascular procedure (MDC-5), only 58(0.9 percent) utilized one of the other providers in District 6, and none were discharged from Brandon. Among the 643 open heart surgeries performed on Manatee County residents in 1998, only 17 cases were seen at one of the District 6 open heart programs outside of Manatee County. There is, therefore, practically no patient exchange between Manatee County and the remainder of the District. In sum, there is virtually no cardiac patient overlap between Manatee County and Brandon's primary service area. The development of an open heart surgery program at Brandon will have no appreciable or meaningful impact on the Manatee County providers. CON 9169 In CON 9169, Brandon applied for an open heart surgery program on the basis of special circumstances due to no impact on low volume providers in Manatee County. The application was denied by AHCA. The State Agency Action Report ("SAAR") on CON 9169, dated June 17, 1999, in a section of the SAAR denominated "Special Circumstances," found the application to demonstrate "that a program at Brandon would not impact the two Manatee hospitals . . .". (UCH Ex. No. 6, p. 5). The "Special Circumstances" section of the SAAR on CON 9169, however, does not conclude that the lack of impact constitutes special circumstances. In follow-up to the finding of the application's demonstration of no impact to the Manatee County, the SAAR turned to impact on the non-Manatee County providers in District The SAAR on CON 9169 states, "it is apparent that a new program in Brandon would impact existing providers [those in Hillsborough and Polk Counties] in the absence of significant open heart surgery growth." Id. In reference to Brandon's argument in support of special circumstances based on the lack of impact to the Manatee County providers, the CON 9169 SAAR states: [T]he applicant notes the open heart need formula should be applied to District 6 excluding Manatee County, which would result in the need for several programs. This argument ignores the provision of the rule that specifies that the need cannot exceed one. (UCH No. 6, p. 7). The Special Circumstances Section of the SAAR on CON 9169 does not deal directly with whether lack of impact to the Manatee County providers is a special circumstance justifying one additional program. Instead, the Agency disposes of Brandon's argument in the "Summary" section of the SAAR. There AHCA found Brandon's special circumstances argument to fail because "no impact on low volume providers" is not among those special circumstances traditionally or previously recognized in case law and by the Agency: To demonstrate need under special circumstances, the applicant should demonstrate one or more of the following reasons: access problems to open heart surgery; capacity limits of existing providers; denial of access based on payment source or lack thereof; patients are seeking care outside the district for service; improvement of care to underserved population groups; and/or cost savings to the consumer. The applicant did not provide any documentation in support of these reasons. (UCH No. 6, p. 29). Following reference to the Agency's publication of zero need in District 6, moreover, the SAAR reiterated that [t]he implementation of another program in Hillsborough County is expected to significantly [a]ffect existing programs, in particular Tampa General Hospital, an important indigent care provider. (Id.) Typical "not normal circumstances" that support approval of a new program were described at hearing by one health planner as consisting of a significant "gap" in the current health care delivery system of that service. Typical Not Normal Circumstances Just as in CON 9169, none of the typical "not normal" circumstances" recognized in case law and with which the Agency has previous experience are present in this case. The six existing OHS programs in District 6 have unused capacity, are available, and are adequate to meet the projected OHS demand in District 6, in Hillsborough County ("County"), and in Brandon's proposed primary service area ("PSA"). All three County OHS providers are less than 17 miles from Brandon. There are, therefore, no major service geographic gaps in the availability of OHS services. Existing providers in District 6 have unused capacity to meet OHS projected demand in January 2002. OHS volume for District 6 will increase by only 179 surgeries. This is modest growth, and can easily be absorbed by the existing providers. In fact, existing OHS providers have previously handled more volume than what is projected for 2002. In 1995, 3,313 OHS procedures were generated at the six OHS programs. Yet, only 3,245 procedures are projected for 2002. The demand in 1995 was greater than what is projected for 2002. Neither population growth nor demographic characteristics of Brandon's PSA demonstrate that existing programs cannot meet demand. The greatest users of OHS services are the elderly. In 1999, the percentage in District 6 was similar to the Florida average; 18.25 percent for District 6, 18.38 percent for the state. The elderly percentage in Hillsborough County was less: 13.21 percent. The elderly component in Brandon's PSA was less still: 10.44 percent. In 2004, about 18.5 percent of Florida and District 6 residents are projected to be elderly. In contrast, only 10.5 percent of PSA residents are expected to be elderly. Brandon's PSA is "one of the younger defined population segments that you could find in the State of Florida" (Tr. 2892) and likely to remain so. Brandon's PSA will experience limited growth in OHS volume. Between 1999 - 2002, OHS volume will grow by only 36. The annual growth thereafter is only 13 surgeries. This is "very modest" growth and is among the "lowest numbers" of incremental growth in the State. Existing OHS providers can easily absorb this minimal growth. Brandon's PSA, is not an underserved area . . . there is excellent access to existing providers and . . . the market in this service area is already quite competitive. There is not a single competitor that dominates. In fact, the four existing providers [in Hillsborough and Polk Counties] compete quite vigorously. (Tr. 2897). Existing OHS programs in District 6 provide very good quality of care. The surgeons at the programs are excellent. Dr. Gandhi, testifying in support of Brandon's application, testified that he was very comfortable in referring his patients for OHS services to St. Joseph and Tampa General, having, in fact, been comfortable with his father having had OHS at Tampa General. Likewise, Dr. Vijay and his group, also supporters of the Brandon application, split time between Bayonet Point and Tampa General. Dr. Vijay is very proud to be associated with the OHS program at Tampa General. Lakeland also operates a high quality OHS program. In its application, Brandon did not challenge the quality of care at the existing OHS programs in District 6. Nor did Brandon at hearing advance as reasons for supporting its application, capacity constraints, inability of existing providers to absorb incremental growth in OHS volume or failure of existing providers to meet the needs of the residents of Brandon's primary service area. The Agency, in its preliminary decision on the application, agreed that typical "not normal" circumstances in this case are not present. Included among these circumstances are those related to lack of "geographic access." The Agency's OHS Rule includes a geographic access standard of two hours. It is undisputed that all District 6 residents have access to OHS services at multiple OHS providers in the District and outside the District within two hours. The travel time from Brandon to UCH or Tampa General, moreover, is usually less than 30 minutes anytime during the day, including peak travel time. Travel time from Brandon to St. Joseph's is about 30 minutes. There are times, however, when travel time exceeds 30 minutes. There have been incidents when traffic congestion has prevented emergency transport of Brandon patients suffering myocardial infarcts from reaching nearby open heart surgery providers within the 30 minutes by ground ambulance. Delays in travel are not a problem in most OHS cases. In the great majority, procedures are elective and scheduled in advance. OHS procedures are routinely scheduled days, if not weeks, after determining that the procedure is necessary. This high percentage of elective procedures is attributed to better management of patients, better technology, and improved stabilizing medications. The advent of drugs such as thrombolytic therapy, calcium channel blockers, beta blockers, and anti-platelet medications have vastly improved stabilization of patients who present at Emergency Rooms with myocardial infarctions. In its application, Brandon did not raise outmigration as a not-normal circumstance to support its proposal and with good reason. Hillsborough County residents generally do not leave District 6 for OHS. In fact, over 96 percent of County residents receive OHS services at a District 6 provider. Lack of out-migration shows two significant facts: (a) existing OHS programs are perceived to be reasonably accessible; and (2) County residents are satisfied with the quality of OHS services they receive in the County. This 96 percent retention rate is even more impressive considering there are many OHS programs and options available to County residents within a two-hour travel time. In contrast, there are two low-volume OHS providers in Manatee County, one of them being Blake. Unlike Hillsborough County residents, only 78 percent of Manatee County residents remain in District 6 for OHS services. Such outmigration shows that these residents prefer to bypass closer programs, and travel further distances, to receive OHS services at high-volume facility in District 8, which they regard as offering a higher quality of service. In its Application, Brandon does not raise economic access as a "not normal" circumstance. In fact, Brandon concedes that the demand for OHS services by Medicaid and indigent patients is very limited because Brandon's PSA is an affluent area. Brandon does not "condition" its application on serving a specific number or percentage of Medicaid or indigent patients. There are no financial barriers to accessing OHS services in District 6. All OHS providers in Hillsborough County and LRMC provide services to Medicaid and indigent patients, as needed. Approving Brandon is not needed to improve service or care to Medicaid or indigent patient populations. Tampa General is the "safety net" provider for health care services to all County residents. Tampa General is an OHS provider geographically accessible to Brandon's PSA. Tampa General actively services the PSA now for OHS. Brandon did not demonstrate cost savings to the patient population of its PSA if it were approved. Approving Brandon is not needed to improve cost savings to the patient population. Brandon based its OHS and PTCA charges on the average charge for PSA residents who are serviced at the existing OHS providers. While that approach is acceptable, Brandon does not propose a charge structure which is uniquely advantageous for patients. Restated, patients would not financially benefit if Brandon were approved. Tertiary Service Open Heart Surgery is defined as a tertiary service by rule. A "tertiary health service" is defined in Section 408.032(17), Florida Statutes, as follows: health service, which, due to its high level of intensity, complexity, specialized or limited applicability, and cost, should be limited to, and concentrated in, a limited number of hospitals to ensure the quality, availability, and cost- effectiveness of such service. As a tertiary service, OHS is necessarily a referral service. Most hospitals, lacking OHS capability, transfer their patients to providers of the service. One might expect providers of open heart surgery in Florida in light of OHS' status as a tertiary service to be limited to regional centers of excellence. The reality of the six hospitals that provide open heart surgery services in District 6 defies this health-planning expectation. While each of the six provides OHS services of high quality, they are not "regional" centers since all are in the same health planning district. Rather than each being a regional center, the six together comprise more localized providers that are dispersed throughout a region, quite the opposite of a center for an entire region. Brandon's Allegations of Special Circumstances. Brandon presents two special circumstances for approval of its application. The first is that consideration of the low-volume Manatee County providers should not operate to "zero out" the numeric need calculated by the formula. The second relates to transfers and occasional problems with transfers for Brandon patients in need of emergency open heart services. "Time is Muscle" Lack of blood flow to the heart during a myocardial infarction ("MI") results in loss of myocardium (heart muscle). The longer the blood flow is disrupted or diminished, the more myocardium is lost. The more myocardium lost, the more likely the patient will die or, should the patient survive, suffer severe reduction in quality of life. The key to good patient outcome when a patient is experiencing an acute MI is prompt evaluation and rapid treatment upon presentation at the hospital. Restoration of blood flow to the heart (revascularization) is the goal of the treating physician once it is recognized that a patient is suffering an MI. If revascularization is not commenced within 2 hours of the onset of an acute MI, an MI patient's potential for recovery is greatly diminished. The need for prompt revascularization for a patient suffering an MI is summed up in the phrase "time is muscle," a phrase accepted as a maxim by cardiologists and cardiothoracic surgeons. Recent advances in modern medicine and technology have improved the ability to stabilize and treat patients with acute MIs and other cardiac traumas. The three primary treatment modalities available to a patient suffering from an MI are: 1) thrombolytics; 2) angioplasty and stent placement; and, 3) open heart surgery. Because of the advancement of the effectiveness of thrombolytics, thrombolytic therapy has become the standard of care for treating MIs. Thrombolytic therapy is the administration of medication to dissolve blood clots. Administered intravenously, thrombolytic medication begins working within minutes to dissolve the clot causing the acute MI and therefore halt the damage done by an MI to myocardium. The protocols to administer thrombolysis are similar among hospitals. If a patient presents with chest pain and the E.R. physician identifies evidence of an active heart attack, thrombolysis is normally administered. If the E.R. physician is uncertain, a cardiologist is quickly contacted to evaluate the patient. Achieving good outcomes in cases of myocardial infarctions requires prompt consultation with the patient, competent clinical assessment, and quick administration of appropriate treatment. The ability to timely evaluate patient conditions for MI, and timely administer thrombolytic therapy, is measured and evaluated nationally by the National Registry of Myocardial Infarction. The National Registry makes the measurement according to a standard known as "door-to-needle" time. This standard measures the time between the patient's presentation at the E.R. and the time the patient is initially administered thrombolytic medication by injection intravenously. Patients often begin to respond to thrombolysis within 10-15 minutes. Consistent with the maxim, "time is muscle," the shorter the door-to-needle time, the better the chance of the patient's successful recovery. The effectiveness of thrombolysis continues to increase. For example, the advent of a drug called Reapro blocks platelet activity, and has increased the efficacy rate of thrombolysis to at least 85 percent. As one would expect, then, thrombolytic therapy is the primary method of revascularization available to patients at Brandon. Due to the lack of open heart surgery backup, moreover, Brandon is precluded by Agency rule from offering angioplasty in all but the most extreme cases: those in which it is determined that a patient will not survive a transfer. While Brandon has protocols, authority, and equipment to perform angioplasty when a patient is not expected to survive a transfer, physicians are reluctant to perform angioplasty without open heart backup because of complications that can develop that require open heart surgery. Angioplasty, therefore, is not usually a treatment modality available to the MI patient at Brandon. Although the care of choice for MI treatment, thrombolytics are not always effective. To the knowledge of the cardiologists who testified in this proceeding, there is not published data on the percentage of patients for whom thrombolytics are not effective. But from the cardiologists who offered their opinions on the percentage in the proceeding, it can be safely found that the percentage is at least 10 percent. Thrombolytics are not ordered for these patients because they are inappropriate in the patients' individual cases. Among the contraindications for thrombolytics are bleeding disorders, recent surgery, high blood pressure, and gastrointestinal bleeding. Of the patients ineligible for thrombolytics, a subset, approximately half, are also ineligible for angioplasty. The other half are eligible for angioplasty. Under the most conservative projections, then at least 1 in 20 patients suffering an MI would benefit from timely angioplasty intervention for which open heart surgery back-up is required in all but the rarest of cases. In 1997, 351 people presented to Brandon's Emergency Room suffering from an acute MI. In 1998, the number of MIs increased to 427. In 1999, 428 patients presented to Brandon's Emergency Room suffering from an acute MI. At least 120 (10 percent) of the total 1206 MI patients presenting to Brandon's Emergency Room from 1997 to 1999 would have been ineligible for thrombolytics as a means of revascularization. Of these, half would have been ineligible for angioplasty while the other half would have been eligible. Sixty, therefore, is the minimum number of patients from 1997 to 1999 who would have benefited from angioplasty at Brandon using the most conservative estimate. Transfers of Emergency Patients Those patients who presented at Brandon's Emergency Room with acute MI and who could not be stabilized with thrombolytic therapy had to be transferred to one of the nearby providers of open heart surgery. In 1998, Brandon transferred an additional 190 patients who did not receive a diagnostic catheterization procedure at Brandon for either angioplasty or open heart surgery. For the first 9 months of 1999, 114 such transfers were made. Thus, in 1998 alone, Brandon transferred a total of 516 cardiac patients to existing providers for the provision of angioplasty or open heart surgery, more than any other provider in the District. In 1999, Brandon made 497 such transfers. Not all of these were emergency transfers, of course. But in the three years between 1997 and 1999 at least 60 patients were in need of emergency transfers who would benefit from angioplasty with open heart backup. Of those Brandon patients determined to be in need of urgent angioplasty or open heart surgery, all must be transferred to existing providers either by ambulance or by helicopter. Ambulance transfer is accomplished through ambulances maintained by the Hillsborough County Fire Department. Due to the cardiac patient's acuity level, ambulance transfer of such patients necessitates the use of ambulances equipped with Advanced Life Support Systems (ALS) in order to monitor the patient's heart functions and to treat the patient should the patient's condition deteriorate. Hillsborough County operates 18 ambulances. All have ALS capability. Patients with less serious medical problems are sometimes transported by private ambulances equipped with Basic Life Support Systems (BLS) that lack the equipment to appropriately care for the cardiac patient. But, private ambulances are not an option to transport critically ill cardiac patients because they are only equipped with BLS capability. Private ambulances, moreover, do not make interfacility transports of cardiac patients between Hillsborough County hospitals. There are many demands on the ambulance transfer system in Hillsborough County. Hillsborough County's 18 ALS ambulances cover in excess of 960 square miles. Of these 18 ambulances, only three routinely operate within the Brandon area. Hillsborough County ambulances respond to 911 calls before requests for interfacility transfers of cardiac patients and are extremely busy responding to automobile accidents, especially when it rains. As a result, Hillsborough County ambulances are not always available on a timely basis when needed to perform an interfacility transfer of a cardiac patient. At times, due to inordinate delay caused by traffic congestion, inter-facility ambulance transport, even if the ambulance is appropriately equipped, is not an option for cardiac patients urgently in need of angioplasty or open heart surgery. It has happened, for example, that an ambulance has appeared at the hospital 8 hours after a request for transport. Some cardiac surgeons will not utilize ground transport as a means of transporting urgent open heart and angioplasty cases. Expeditious helicopter transport in Hillsborough County is available as an alternative to ground transport. But, it too, from time-to-time, is problematic for patients in urgent need of angioplasty or open heart surgery. Tampa General operates two helicopters through AeroMed, only one of which is located in Hillsborough County. AeroMed's two helicopters are not exclusively devoted to cardiac patients. They are also utilized for the transfer of emergency medical and trauma patients, further taxing the availability of AeroMed helicopters to transfer patients in need of immediate open heart surgery or angioplasty. BayCare operates the only other helicopter transport service serving Hillsborough County. BayCare maintains several helicopters, only one of which is located in Hillsborough County at St. Joseph's. BayCare helicopters are not equipped with intra-aortic balloon pump capability, thereby limiting their use in transporting the more complicated cardiac patients. Helicopter transport is not only a traumatic experience for the patient, but time consuming. Once a request has been made by Brandon to transport a patient in need of urgent intervention, it routinely takes two and a half hours, with instances of up to four hours, to effectuate a helicopter transfer. At the patient's beside, AeroMed personnel must remove the patient's existing monitors, IVS, and drips, and refit the patient with AeroMed's equipment in preparation for flight. In more complicated cases requiring the use of an intra-aortic balloon pump, the patient's balloon pump placed at Brandon must be removed and substituted with the balloon pump utilized by AeroMed. Further delays may be experienced at the receiving facility. The national average of the time from presentation to commencement of the procedure is reported to be two hours. In most instances at UCH, it is probably 90 minutes although "[t]here are of course instances where it would be much faster . . .". (Tr. 3212). On the other hand, there are additional delays from time-to-time. "[P]erhaps the longest circumstance would be when all the labs are full . . . or . . . even worse . . . if all the staff has just left for the day and they are almost home, to then turn them around and bring them all back." (Id.) Specific Cases Involving Transfers Delays in the transfer process were detailed at hearing by Brandon cardiologists with regard to specific Brandon patients. In cases in which "time is muscle," delay is critical except for one subset of such cases: that in which, no matter what procedure is available and no matter how timely that procedure can be provided, the patient cannot be saved. Craig Randall Martin, M.D., Board-certified in Internal Medicine and Cardiovascular Disease, and an expert in cardiology, wrote to AHCA in support of the application by detailing two "examples of patients who were in an extreme situation that required emergent, immediate intervention . . . [intervention that could not be provided] at Brandon Hospital." (Tr. 408). One of these concerned a man in his early sixties who was a patient at Brandon the night and morning of October 13 and 14, 1998. It represents one of the rare cases in which an emergency angioplasty was performed at Brandon even though the hospital does not have open heart backup. The patient had presented to the Emergency Room at approximately 11:00 p.m., on October 13 with complaints of chest pain. Although the patient had a history of prior infarctions, PTCA procedures, and onset diabetes, was obese, a smoker and had suffered a stroke, initial evaluation, including EKG and blood tests, did not reveal an MI. The patient was observed and treated for what was probably angina. With the subsiding of the chest pain, he was appropriately admitted at 2:30 a.m. to a non- intensive cardiac telemetry bed in the hospital. At 3:00 a.m., he was observed to be stable. A few hours or so later, the patient developed severe chest pain. The telemetry unit indicated a very slow heart rate. Transferred to the intensive care unit, his blood pressure was observed to be very low. Aware of the seriousness of the patient's condition, hospital personnel called Dr. Martin. Dr. Martin arrived on the scene and determined the patient to be in cardiogenic shock, an extreme situation. In such a state, a patient has a survival rate of 15 to 20 percent, unless revascularization occurs promptly. If revascularization is timely, the survival rate doubles to 40 percent. Coincident with the cardiogenic shock, the patient was suffering a complete heart block with a number of blood clots in the right coronary artery. The patient's condition, to say the least, was grave. Dr. Martin described the action taken at Brandon: . . . I immediately called in the cardiac catheterization team and moved the patient to the catheterization laboratory. * * * Somewhere around 7:30 in the morning, I put a temporary pacemaker in, performed a diagnostic catheterization that showed that one of his arteries was completely clotted. He, even with the pacemaker giving him an adequate heart rate, and even with the use of intravenous medication for his blood pressure, . . . was still in cardiogenic shock. * * * And I placed an intra-aortic balloon pump . . ., a special pump that fits in the aorta and pumps in synchrony with the heart and supports the blood pressure and circulation of the muscle. That still did not alleviate the situation . . . an excellent indication to do a salvage angioplasty on this patient. I performed the angioplasty. It was not completely successful. The patient had a respiratory arrest. He required intubation, required to be put on a ventilator for support. And it became apparent to me that I did not have the means to save this patient at [Brandon]. I put a call to the . . . cardiac surgeon of choice . . . . [Because the surgeon was on vacation], [h]is associate [who happened to be in the operating room at UCH] called me back immediately . . . and said ["]Yes, I'll take your patient. Send him to me immediately, I will postpone my current case in order to take care of your patient.["] At that point, we called for helicopter transport, and there were great delays in obtaining [the] transport. The patient was finally transferred to University Community Hospital, had surgery, was unsuccessful and died later that afternoon. (Tr. 409-412). By great delays in the transport, Dr. Martin referred to inability to obtain prompt helicopter transport. University Community Hospital, the receiving hospital, was not able to find a helicopter. Dr. Martin, therefore, requested Tampa General (a third hospital uninvolved from the point of being either the transferring or the receiving hospital) to send one of its two helicopters to transfer the patient from Brandon to UCH. Dr. Martin described Tampa General's response: They balked. And I did not know they balked until an hour later. And I promptly called them back, got that person on the telephone, we had a heated discussion. And after that person checked with their supervisor, the helicopter was finally sent. There was at least an hour-and-a-half delay in obtaining a helicopter transport on this patient that particular morning that was unnecessary. And that is critical when you have a patient in this condition. (Tr. 413, emphasis supplied.) In the case of this patient, however, the delay in the transport from Brandon to the UCH cardiovascular surgery table, in all likelihood, was not critical to outcome. During the emergency angioplasty procedure at Brandon, some of the clot causing the infarction was dislodged. It moved so as to create a "no-flow state down the right coronary artery. In other words, . . ., it cut off[] the microcirculation . . . [so that] there is no place for the blood . . . to get out of the artery. And that's a devastating, deadly problem." (Tr. 2721). This "embolization, an unfortunate happenstance [at times] with angioplasty", id., probably sealed the patient's fate, that is, death. It is very likely that the patient with or without surgery, timely or not, would not have survived cardiogenic shock, complete heart block, and the circumstance of no circulation in the right coronary artery that occurred during the angioplasty procedure. Adithy Kumar Gandhi, M.D., is Board-certified in Internal Medicine and Cardiology. Employed by the Brandon Cardiology Group, a three-member group in Brandon, Dr. Gandhi was accepted as an expert in the field of cardiology in this proceeding. Dr. Gandhi testified about two patients in whose cases delays occurred in transferring them to St. Joseph’s. He also testified about a third case in which it took two hours to transfer the patient by helicopter to Tampa General. The first case involves an elderly woman. She had multiple-risk factors for coronary disease including a family history of cardiac disease and a personal history of “chest pain.” (Tr. 2299). The patient presented at Brandon’s Emergency Room on March 17, 1999 at around 2:30 p.m. Seen by the E.R. physician about 30 minutes later, she was placed in a monitored telemetry bed. She was determined to be stable. During the next two days, despite family and personal history pointing to a potentially serious situation, the patient refused to submit to cardiac catheterization at Brandon as recommended by Dr. Gandhi. She maintained her refusal despite results from a stress test that showed abnormal left ventricular systolic function. Finally, on March 20, after a meeting with family members and Dr. Gandhi, the patient consented to the cath procedure. The procedure was scheduled for March 22. During the procedure, it was discovered that a major artery of the heart was 80 percent blocked. This condition is known as the “widow-maker,” because the prognosis for the patient is so poor. Dr. Gandhi determined that “the patient needed open heart surgery and . . . to be transferred immediately to a tertiary hospital.” (Tr. 2305-6). He described that action he took to obtain an immediate transfer as follows: I talked to the surgeon up at St. Joseph’s and I informed him I have had difficulties transferring patients to St. Joseph’s the same day. [I asked him to] do me a favor and transfer the patient out of Brandon Hospital as soon as possible by helicopter. The surgeon promised me that he would take care of that. (Tr. 2261). The assurance, however, failed. The patient was not transferred that day. That night, while still at Brandon, complications developed for the patient. The complications demanded that an intra-aortic balloon pump be inserted in order to increase the blood flow to the heart. After Dr. Gandhi’s partner inserted the pump, he, too, contacted the surgeon at St. Joseph’s to arrange an immediate transfer for open heart surgery. But the patient was not transferred until early the next morning. Dr. Gandhi’s frustration at the delay for this critically ill patient in need of immediate open heart surgery is evident from the following testimony: So the patient had approximately 18 hours of delay of getting to the hospital with bypass capabilities even though the surgeon knew that she had a widow-maker, he had promised me that he would make those transfer arrangements, even though St. Joseph’s Hospital knew that the patient needed to be transferred, even though I was promised that the patient would be at a tertiary hospital for bypass capabilities. (Tr. 2262). Rod Randall, M.D., is a cardiologist whose practice is primarily at St. Joseph’s. He had active privileges at Brandon until 1998 when he “switched to courtesy privileges,” (Tr. 1735) at Brandon. He reviewed the medical records of the first patient about whom Dr. Gandhi testified. A review of the patient’s medical records disclosed no adverse outcome due to the patient’s transfer. To the contrary, the patient was reasonably stable at the time of transfer. Nonetheless, it would have been in the patient’s best interest to have been transferred prior to the catheterization procedure at Brandon. As Dr. Randall explained, [W]e typically cath people that we feel are going to have a probability of coronary artery disease. That is, you don’t tend to cath someone that [for whom] you don’t expect to find disease . . . . If you are going to cath this patient, [who] is in a higher risk category being an elderly female with . . . diminished injection fraction . . . why put the patient through two procedures. I would have to do a diagnostic catheterization at one center and do some type of intervention at another center. So, I would opt to transfer that patient to a tertiary care center and do the diagnostic catheterization there. (Tr. 1764, 1765). Furthermore, regardless of what procedure had been performed, the significant left main blockage that existed prior to the patient’s presentation at Brandon E.R. meant that the likely outcome would be death. The second of the patients Dr. Gandhi transferred to St. Joseph’s was a 74-year-old woman. Dr. Gandhi performed “a heart catheterization at 5:00 on Friday.” (Tr. 2267). The cath revealed a 90 percent blockage of the major artery of the heart, another widow-maker. Again, Dr. Gandhi recommended bypass surgery and contacted a surgeon at St. Joseph’s. The transfer, however, was not immediate. “Finally, at approximately 11:00 the patient went to St. Joseph’s Hospital. That night she was operated on . . . ”. (Tr. 2267). If Brandon had had open heart surgery capability, “[t]hat would have increased her chances of survival.” No competent evidence was admitted that showed the outcome, however, and as Dr. Randall pointed out, the medical records of the patient do not reveal the outcome. The patient who was transferred to Tampa General (the third of Dr. Ghandhi's patients) had presented at Brandon’s ER on February 15, 2000. Fifty-six years old and a heavy smoker with a family history of heart disease, she complained of severe chest pain. She received thrombolysis and was stabilized. She had presented with a myocardial infarction but it was complicated by congestive heart failure. After waiting three days for the myocardial infarction to subside, Dr. Gandhi performed cardiac catheterization. The patient “was surviving on only one blood vessel in the heart, the other two vessels were 100 percent blocked. She arrested on the table.” (Tr. 2271). After Dr. Gandhi revived her, he made arrangements for her transfer by helicopter. The transfer was done by helicopter for two reasons: traffic problems and because she had an intra-aortic balloon pump and there are a limited number of ambulances with intra- aortic balloon pump maintenance capability. If Brandon had had the ability to conduct open heart surgery, the patient would have had a better likelihood of successful outcome: “the surgeon would have taken the patient straight to the operating room. That patient would not have had a second arrest as she did at Tampa General.” (Tr. 2273). Marc Bloom, M.D., is a cardiothoracic surgeon. He performs open-heart surgery at UCH, where he is the chief of cardiac surgery. He reviewed the records of this 54-year-old woman. The records reflect that, in fact, upon presentation at Brandon’s E.R., the patient’s heart failure was very serious: She had an echocardiogram done that . . . showed a 20 percent ejection fraction . . . I mean when you talk severe, this would be classified as a severe cardiac compromise with this 20 percent ejection fraction. (Tr. 2712). Once stabilized, the patient should have been transferred for cardiac catheterization to a hospital with open- heart surgery instead of having cardiac cath at Brandon. It is true that delay in the transfer once arrangements were made was a problem. The greater problem for the patient, however, was in her management at Brandon. It was very likely that open heart surgery would be required in her case. She should have been transferred prior to the catheterization as soon as became known the degree to which her heart was compromised, that is, once the results of the echocardiogram were known. Adam J. Cohen, M.D., is a cardiologist with Diagnostic Consultative Cardiology, a group located in Brandon that provides cardiology services in Hillsborough County. Dr. Cohen provided evidence of five patients who presented at Brandon and whose treatments were delayed because of the need for a transfer. The first of these patients was a 76-year old male who presented to Brandon’s ER on April 6, 1999. Dr. Cohen considered him to be suffering “a complicated myocardial infarction.” (Brandon Ex. 45, p. 43) Cardiac catheterization conducted by Dr. Cohen showed “severe multi-vessel coronary disease, cardiogenic shock, severely impaired [left ventricular] function for which an intra-aortic balloon pump was placed . . .”. (Id.) During the placement of the pump, the patient stopped breathing and lost pulse. He was intubated and stabilized. A helicopter transfer was requested. There was only one helicopter equipped to conduct the transfer. Unfortunately, “the same day . . . there was a mass casualty event within the City of Tampa when the Gannet Power Plant blew up . . .”. (Brandon Ex. 45, p. 44). An appropriate helicopter could not be secured. Dr. Cohen did not learn of the unavailability of helicopter transport for an hour after the request was made. Eventually, the patient was transferred by ambulance to UCH. There, he received angioplasty and “stenting of the right coronary artery times two.” (Id., at p. 47.) After a slow recovery, he was discharged on April 19. In light of the patient’s complex cardiac condition, he received a good outcome. This patient is an example of another patient who should have been transferred sooner from Brandon since Brandon does not have open heart surgery capability. The second of Dr. Cohen’s patients presented at Brandon’s E.R. at 10:30 p.m. on June 14, 1999. He was 64 years old with no risk factors for coronary disease other than high blood pressure. He was evaluated and diagnosed with “a large and acute myocardial infarction” Two hours later, the therapy was considered a failure because there was no evidence that the area of the heart that was blocked had been reperfused. Dr. Cohen recommended transfer to UCH for a salvage angioplasty. The call for a helicopter was made at 12:58 a.m. (early the morning of June 15) and the helicopter arrived 40 minutes later. At UCH, the patient received angioplasty procedure and stenting of two coronary arteries. He suffered “[m]oderately impaired heart function, which is reflective of myocardial damage.” (Brandon Ex. 45, p. 58). If salvage angioplasty with open heart backup had been available at Brandon, the patient would have received it much more quickly and timely. Whether the damage done to the patient’s heart during the episode could have been avoided by prompt angioplasty at Brandon is something Dr. Cohen did not know. As he put it, “I will never know, nor will anyone else know.” (Brandon Ex. 45, p. 60). The patient later developed cardiogenic shock and repeated ventricular tachycardia, requiring numerous medical interventions. Because of the interventions and mechanical trauma, he required surgery for repair of his right femoral artery. The patient recently showed an injection fraction of 45 percent below the minimum for normal of 50 percent. The third patient was a 51-year-old male who had undergone bypass surgery 19 years earlier. After persistent recurrent anginal symptoms with shortness of breath and diaphoresis, he presented at Brandon’s E.R. at 1:00 p.m. complaining of heavy chest pain. Thrombolytic therapy was commenced. Dr. Cohen described what followed: [H]he had an episode of heart block, ventricular fibrillation, losing consciousness, for which he received ACLS efforts, being defibrillated, shocked, times three, numerous medications, to convert him to sinus rhythm. He was placed on IV anti- arrhythmics consisting of amiodarone. The repeat EKG showed a worsening of progression of his EKG changes one hour after the initiation of the TPA. Based on that information, his clinical scenario and his previous history, I advised him to be transferred to University Hospital for a salvage angioplasty. (Brandon Ex. 45, p. 62). Transfer was requested at 1:55 p.m. The patient departed Brandon by helicopter at 2:20 p.m. The patient received the angioplasty at UCH. Asked how the patient would have benefited from angioplasty at Brandon without having to have been transferred, Dr. Cohen answered: In a more timely fashion, he would have received an angioplasty to the culprit lesion involved. There would have been much less occlusive time of that artery and thereby, by inference, there would have been greater salvage of myocardium that had been at risk. (Brandon Ex. 45, p. 65). The patient, having had bypass surgery in his early thirties, had a reduced life expectancy and impaired heart function before his presentation at Brandon in June of 1999. The time taken for the transfer of the patient to UCH was not inordinate. The transfer was accomplished with relative and expected dispatch. Nonetheless, the delay between realization at Brandon of the need for a salvage angioplasty and actual receipt of the procedure after a transfer to UCH increased the potential for lost myocardium. The lack of open heart services at Brandon resulted in reduced life expectancy for a patient whose life expectancy already had been diminished by the early onset of heart disease. The fourth patient of Dr. Cohen’s presented to Brandon’s E.R. at 8:30, the morning of August 29, 1999. A fifty-four-year-old male, he had been having chest pain for a month and had ignored it. An EKG showed a complete heart block with atrial fibrillation and change consistent with acute myocardial infarction. Thrombolytic therapy was administered. He continued to have symptoms including increased episodes of ventricular arrhythmias. He required dopamine for blood pressure support due to his clinical instability and the lack of effectiveness of the thrombolytics. The patient refused a transfer and catheterization at first. Ultimately, he was convinced to undergo an angioplasty. The patient was transferred by helicopter to UCH. The patient was having a “giant ventricular infarct . . . a very difficult situation to take care of . . . and the majority of [such] patients succumb to [the] disease . . .”. (Tr. 2703). The cardiologist was unable to open the blockage via angioplasty. Dr. Bloom was called in but the patient refused surgical intervention. After interaction with his family the patient consented. Dr. Bloom conducted open heart surgery. The patient had a difficult post-operative course with arrythmias because “[h]e had so much dead heart in his right ventricle . . .”. (Id.) The patient received an excellent outcome in that he was seen in Dr. Bloom’s office with 40 percent injection fraction. Dr. Bloom “was just amazed to see him back in the office . . . and amazed that this man is alive.” (Tr. 2704). Most of the delay in receiving treatment was due to the patient’s reluctance to undergo angioplasty and then open heart surgery. The fifth patient of Dr. Cohen’s presented at Brandon’s E.R. on March 22, 2000. He was 44 years old with no prior cardiac history but with numerous risk factors. He had a sudden onset of chest discomfort. Lab values showed an elevation consistent with myocardial injury. He also had an abnormal EKG. Dr. Cohen performed a cardiac cath on March 23, 2000. The procedure showed a totally occluded left anterior descending artery, one of the three major arteries serving the heart. Had open heart capability been available at Brandon, he would have undergone angioplasty and stenting immediately. As it was, the patient had to be transferred to UCH. A transfer was requested at 10:25 that morning and the patient left Brandon’s cath lab at 11:53. Daniel D. Lorch, M.D., is a specialist in pulmonary medicine who was accepted as an expert in internal medicine, pulmonary medicine and critical care medicine, consistent with his practice in a “five-man pulmonary internal medicine critical care group.” (Brandon Ex. 42, p. 4). Dr. Lorch produced medical records for one patient that he testified about during his deposition. The patient had presented to Brandon’s E.R. with an MI. He was transferred to UCH by helicopter for care. Dr. Lorch supports Brandon’s application. As he put it during his deposition: [Brandon] is an extremely busy community hospital and we are in a very rapidly growing area. The hospital is quite busy and we have a large number of cardiac patients here and it is not infrequently that a situation comes up where there are acute cardiac events that need to be transferred out. (Brandon Ex. 42, p. 20). Transfers Following Diagnostic Cardiac Catheterization Brandon transfers a high number cardiac patients for the provision of angioplasty or open heart surgery in addition to those transferred under emergency conditions. In 1996, Brandon performed 828 diagnostic cardiac catheterization procedures. Of this number, 170 patients were transferred to existing providers for open heart surgery and 170 patients for angioplasty. In 1997, Brandon performed 863 diagnostic catheterizations of which 180 were transferred for open heart surgery and 159 for angioplasty. During 1998, 165 patients were transferred for open heart surgery and 161 for angioplasty out of 816 diagnostic catheterization procedures. For the first nine months of 1999, Brandon performed 639 diagnostic catheterizations of which 102 were transferred to existing providers for open heart surgery and 112 for angioplasty. A significant number of patients are transferred from Brandon for open heart surgery services. These transfers are consistent with the norm in Florida. After all, open heart surgery is a tertiary service. Patients are routinely transferred from most Florida hospitals to tertiary hospitals for OHS and PCTA. The large majority of Florida hospitals do not have OHS programs; yet, these hospitals receive patients who need OHS or PTCA. Transfers, although the norm, are not without consequence for some patients who are candidates for OHS or PCTA. If Brandon had open heart and angioplasty capability, many of the 1220 patients determined to be in need of angioplasty or open heart surgery following a diagnostic catheterization procedure at Brandon could have received these procedures at Brandon, thereby avoiding the inevitable delay and stress occasioned by transfer. Moreover, diagnostic catheterizations and angioplasties are often performed sequentially. Therefore, Brandon patients determined to be in need of angioplasty following a diagnostic catheterization would have had access to immediate angioplasty during the same procedure thus reducing the likelihood of a less than optimal outcome as the result of an additional delay for transfer. Adverse Impact on Existing Providers Competition There is active competition and available patient choices now in Brandon's PSA. As described, there are many OHS programs currently accessible to and substantially serving Brandon's PSA. There is substantial competition now among OHS providers so as to provide choices to PSA residents. There are no financial benefits or cost savings accruing to the patient population if Brandon is approved. Brandon does not propose lower charges than the existing OHS providers. Balanced Budget Act The Balanced Budget Act of 1997 has had a profound negative financial impact on hospitals throughout the country. The Act resulted in a significant reduction in the amount of Medicare payments made to hospitals for services rendered to Medicare recipients. During the first five years of the Act's implementation, Florida hospitals will experience a $3.6 billion reduction in Medicare revenues. Lakeland will receive $17 million less, St.Joseph's will receive $44 million less, and Tampa General will receive $53 million less. The impact of the Act has placed most hospitals in vulnerable financial positions. It has seriously affected the bottom line of all hospitals. Large urban teaching hospitals, such as TGH, have felt the greatest negative impact, due to the Act's impact on disproportionate share reimbursement and graduate medical education payment. The Act's impact upon Petitioners render them materially more vulnerable to the loss of OHS/PTCA revenues to Brandon than they would have been in the absence of the Act. Adverse Impact on Tampa General Tampa General is the "safety net provider" for Hillsborough County. Tampa General is a Medicaid disproportionate share provider. In fiscal year 1999, the hospital provided $58 million in charity care, as that term is defined by AHCA. Tampa General plays a unique, essential role in Hillsborough County and throughout West Central Florida in terms of provision of health care. Its regional role is of particular importance with respect to Level I trauma services, provision of burn care, specialized Level III neonatal and perinatal intensive care services, and adult organ transplant services. These services are not available elsewhere in western or central Florida. In fiscal year 1999, Tampa General experienced a net loss of $12.6 million in providing the services referenced above. It is obligated under contract with the State of Florida to continue to provide those services. Tampa General is a statutory teaching hospital. In fiscal year 1999, it provided unfunded graduate medical education in the amount of $19 million. Since 1998, Tampa General has consistently experienced losses resulting from its operations, as follows: FY 1998-$29 million, FY 1999-$27 million; FY 2000 (5 months)-$10 million. The hospital’s financial condition is not the result of material mismanagement. Rather, its financial condition is a function of its substantial provision of charity and Medicaid services, the impact of the Act, reduced managed care revenues, and significant increases in expense. Tampa General’s essential role in the community and its distressed financial condition have not gone unnoticed. The Greater Tampa Chamber of Commerce established in February of 2000 an Emergency Task Force to assess the hospital's role in the community, and the need for supplemental funding to enable it to maintain its financial viability. Tampa General requires supplemental funding on a continuing basis in order to begin to restore it to a position of financial stability, while continuing to provide essential community services, indigent care, and graduate medical education. It will require ongoing supplemental funding of $20- 25 million annually to avoid triggering the default provision under its bond covenants. As of the close of hearing, the 2000 session of the Florida Legislature had adjourned. The Legislature appropriated approximately $22.9 million for Tampa General. It is, of course, uncertain as to what funding, if any, the Legislature will appropriate to the hospital in future years, as the terms which constitute the appropriations must be revisited by the Legislature on an annual basis. Tampa General has prepared internal financial projections for its fiscal years 2000-2002. It projects annual operating losses, as follows: FY 2000-$20.1 million; FY 2001- $20.6 million; FY 2002-$31.9 million. While its projections anticipate certain "strategic initiatives" that will enhance its financial condition, including continued supplemental legislative funding, the success and/or availability of those initiatives are not "guaranteed" to be successful. If the Brandon program is approved, Tampa General will lose 93 OHS cases and 107 angioplasty cases during Brandon's second year of operation. That loss of cases will result in a $1.4 million annual reduction in TGH's net income, a material adverse impact given Tampa General’s financial condition. OHS services provide a positive contribution to Tampa General's financial operations. Those services constitute a core piece of Tampa General's business. The anticipated loss of income resulting from Brandon's program pose a threat to the hospital’s ability to provide essential community services. Adverse Impact on UCH UCH operated at a financial break-even in its fiscal year 1999. In the first five months of its fiscal year 2000, the hospital has experienced a small loss. This financial distress is primarily attributed to less Medicare reimbursement due to the Act and less reimbursement from managed care. UCH's reimbursement for OHS services provides a good example of the financial challenges facing hospitals. In 1999, UCH's net income per OHS case was reduced 33 percent from 1998. Also in 1999, UCH received OHS reimbursement of only 32 percent of its charges. UCH would be substantially and adversely impacted by approval of Brandon's proposal. As described, UCH currently is a substantial provider of OHS and angioplasty services to residents of Brandon's PSA. There are many cardiologists on staff at Brandon who also actively practice at UCH. UCH is very accessible from Brandon's PSA. UCH reasonably projects to lose the following volumes in the first three years of operation of the proposed program: a loss of 78-93 OHS procedures, a loss of 24-39 balloon angioplasties, and a loss of 97-115 stent angioplasties. Converting this volume loss to financial terms, UCH will suffer the following financial losses as a direct and immediate result of Brandon being approved: about $1.1 million in the first year, and about $1.2 million in the second year, and about $1.3 million in the third year. As stated, UCH is currently operating at about a financial break-even point. The impact of the Balanced Budget Act, reduced managed care reimbursement, and UCH's commitment to serve all patients regardless of ability to pay has a profound negative financial impact on UCH. A recurring loss of more than $1 million dollars per year due to Brandon's new program will cause substantial and adverse impact on UCH. Adverse Impact on St. Joseph’s If Brandon's application is approved, St. Joseph’s will lose 47 OHS cases and 105 PTCA cases during Brandon's second year. That loss of cases will result in a $732,000 annual reduction in SJH's net income. That loss represents a material impact to SJH. Between 1997 and 2000, St. Joseph’s has experienced a pattern of significant deterioration in its financial performance. Its net revenue per adjusted admission had been reduced by 12 percent, while its costs have increased significantly. St. Joseph's net income from operations has deteriorated as follows: FYE 6/30/97-$31 million; FYE 12/31/98- $24 million; FYE 12/31/99-$13.8 million. A net operating income of $13.8 million is not much money relative to St Joseph's size, the age of its physical plant, and its need for capital to maintain and improve its facilities in order to remain competitive. St. Joseph’s offers a number of health care services to the community for which it does not receive reimbursement. Unreimbursed services include providing hospital admissions and services to patients of a free clinic staffed by volunteer members of SJH's medical staff, free immunization programs to low-income children, and a parish nurse program, among others. St. Joseph’s evaluates such programs annually to determine whether it has the financial resources to continue to offer them. During the past two years, the hospital has been forced to eliminate two of its free community programs, due to its deteriorating financial condition. St. Joseph’s anticipates that it will have to eliminate additional unreimbursed community services if it experiences an annual reduction in net income of $730,000. Adverse Impact to LRMC The approval of Brandon will have an impact on Lakeland. Lakeland will suffer a financial loss of about $253,000 annually. This projection is based on calculated contribution margins of OHS and PTCA/stent procedures performed at the hospital. A loss of $253,000 per year is a material loss at Lakeland, particularly in light of its slim operating margin and the very substantial losses it has experienced and will continue to experience as a result of the Balanced Budget Act of 1997. In addition to the projected loss of OHS and other procedures based upon Brandon's application, Lakeland may experience additional lost cases from areas such as Bartow and Mulberry from which it draws patients to its open heart/cardiology program. Lakeland will also suffer material adverse impacts to its OHS program due to the negative effect of Brandon's program on its ability to recruit and retain nurses and other highly skilled employees needed to staff its program. The approval of Brandon will also result in higher costs at existing providers such as Lakeland as they seek to compete for a limited pool of experienced people by responding to sign-on bonuses and by reliance on extensive temporary nursing agencies and pools. Nursing Staff/Recruitment The staffing patterns and salaries for Brandon's projected 40.1 full-time equivalent employees to staff its open heart surgery program are reasonable and appropriate. Filling the positions will not be without some difficulty. There is a shortage for skilled nursing and other personnel needed for OHS programs nationally, in Florida and in District 6. The shortage has been felt in Hillsborough County. For example, it has become increasingly difficult to fill vacancies that occur in critical nursing positions in the coronary intensive care unit and in telemetry units at Tampa General. Tampa General's expenses for nursing positions have "increased tremendously." (Tr. 2622). To keep its program going, the hospital has hired "travelers . . . short-term employment, registered nurses that come from different agencies, . . . with [the hospital] a minimum of 12 weeks." (Tr. 2622). In fact, all hospitals in the Tampa Bay area utilize pool staff and contract staff to fill vacancies that appear from time-to- time. Use of contract staff has not diminished quality of care at the hospitals, although "they would not be assigned to the sickest patients." (Tr. 2176). Another technique for dealing with the shortage is to have existing full-time staff work overtime at overtime pay rates. St. Joseph's and Lakeland have done so. As a result, they have substantially exceeded their budgeted salary expenses in recent months. It will be difficult for Brandon to hire surgical RNs, other open heart surgery personnel and critical care nurses necessary to staff its OHS program. The difficulty, however, is not insurmountable. To meet the difficulty, Brandon will move members of its present staff with cardiac and open heart experience into its open heart program. It will also train some existing personnel in conjunction with the staff and personnel at Bayonet Point. In addition to drawing on the existing pool of nurses, Brandon can utilize HCA's internal nationwide staffing data base to transfer staff from other HCA facilities to staff Brandon's open heart program. Approximately 18 percent of the nurses hired at Brandon already come from other HCA facilities. The nursing shortage has been in existence for about a decade. During this time, other open heart programs have come on line and have been able to staff the programs adequately. Lakeland, in District 6, has demonstrated its ability to recruit and train open heart surgery personnel. Brandon, itself, has been successful, despite the on- going shortage, in appropriately staffing its recent additions of tertiary level NICU beds, an expanded Emergency Room, labor and delivery and recovery suites, and new high-risk, ante-partum observation unit. Brandon has begun to offer sign-on bonuses to compete for experienced nurses. Several employees who staff the Lakeland, UCH and Tampa General programs live in Brandon. These bonuses are temptations for them to leave the programs for Brandon. Other highly skilled, experienced individuals who already work at existing programs may be lost to Brandon's program as well simply as the natural result of the addition of a new program. In the end, Brandon will be able to staff its program, but it will make it more difficult for all of the programs in Hillsborough County and for Lakeland to meet their staffing needs as well as producing a financial impact on existing providers. Financial Feasibility Short-Term Brandon needs $4.2 million to fund implementation of the program. Its parent corporation, HCA will provide financing of up to $4.5 million for implementation. The $4.2 million in start-up costs projected by Brandon does not include the cost of a second cath lab or the costs to upgrade the equipment in the existing cath lab. Itemization of the funds necessary for improvement of the existing cath lab and the addition of the second cath lab were not included in Brandon's pro formas. It is the Agency's position that addition of a cath lab (and by inference, upgrade to an existing lab) requires only a letter of exemption as projects separate from an open heart surgery program even when proposed in support of the program. (See UCH No. 7, p. 83). The position is not inconsistent with cardiac catheterization programs as subject to requirements in law separate from those to which an open heart surgery program is subject. Brandon, through HCA, has the ability to fund the start-up costs of the project. It is financially feasible in the short-term. Long-Term Open heart surgery programs (inclusive of angioplasty and stent procedures, as well as other open heart surgery procedures) generally are very profitable. They are among the most profitable of programs conducted by hospitals. Brandon's projected charges for open heart, angioplasty, and stent procedures are based on the average charges to patients residing in Brandon's PSA inflated at 2 percent per year. The inflation rate is consistent with HCFA's August 1, 2000, Rule implementing a 2.3 percent Medicare reimbursement increase. Brandon's projected payor mix is reasonably based on the existing open heart, angioplasty, and stent patients within its PSA. Brandon also estimated conservatively that it would collect only 45 to 50 percent of its charges from third-party payors. To determine expenses, Brandon utilized Bayonet Point's accounting system. It provided a level of detail that could not be obtained otherwise. "For patients within Brandon's primary service area, . . . that information is not provided by existing providers in the area that's available for any public consumption." (Tr. 1002). While perhaps the most detailed data available, Bayonet Point data was far from an ideal model for Brandon. Bayonet Point performs about 1,500 OHS cases per year. It achieves economies of scale that will not be achievable at Brandon in the foreseeable future. There is a relationship between volume and cost efficiency. The higher the volume, the greater the cost efficiency. Brandon's volume is projected to be much lower than Bayonet Point's. To make up for the imperfection of use of Bayonet Point as an "expenses" proxy, Brandon's financial expert in opining that the project was feasible in the long-term, considered two factors with regard to expenses. First, it included its projected $1.8 million in salary expenses as a separate line item over and above the salary expenses contained in the Bayonet Point data. (This amounted to a "double" counting of salary expenses.) Second, it recognized HCA's ability to obtain competitive pricing with respect to equipment and services for its affiliated hospitals, Brandon being one of them. Brandon projected utilization of 249 and 279 cases in its second and third year of operations. These projections are reasonable. (See the testimony of Mr. Balsano on rebuttal and Brandon Ex. 74). Comparison of Agency Action in CONs 9169 and 9239 Brandon's application in this case, CON 9239, was filed within a six-month period of the filing of an earlier application, CON 9169. The Agency found the two applications to be similar. Indeed, the facts and circumstances at issue in the two applications other than the updating of the financial and volume numbers are similar. So is the argument made in favor of the applications. Yet, the first application was denied by the Agency while the second received preliminary approval. The difference in the Agency's action taken on the later application (the one with which this case is concerned), i.e., approval, versus the action taken on the earlier, denial, was explained by Scott Hopes, the Chief of the Bureau of Certificate of Need at the time the later application was considered: The [later] Brandon application . . ., which is what we're addressing here today, included more substantial information from providers, both cardiologists, internists, family practitioners and surgeons with specific case examples by patient age [and] other demographics, the diagnoses, outcomes, how delays impacted outcomes, what permanent impact those adverse outcomes left the patient in, where earlier . . . there weren't as many specifics. (Tr. 1536, 1537). A comparison of the application in CON 9169 and the record in this case bears out Mr. Hopes' assessment that there is a significant difference between the two applications. Comparison of the Agency Action with the District 9 Application During the same batching cycle in which CON 9239 was considered, five open heart surgery applications were considered from health care providers in District 9. Unlike Brandon's application, these were all denied. In the District 9 SAAR, the Agency found that transfers are an inherent part of OHS as a tertiary service. The Agency concluded that, "[O]pen heart surgery is a tertiary service and patients are routinely transferred between hospitals for this procedure." (UCH Ex. 7, pp. 51-54). In particular, the Agency recognized Boca Raton's claim that it had provided "extensive discussion of the quality implications of attempting to deal with cardiac emergencies through transfer to other facilities." (UCH Ex. 7, p. 52). Unlike the specific information referred to by Mr. Hopes in his testimony quoted, above, however, the foundation for Boca Raton's argument is a 1999 study published in the periodical Circulation, entitled "Relationship Between Delay in Performing Direct Coronary Angioplasty and Early Clinical Outcomes." (UCH Ex. 7, p. 21). This publication was cited by the Agency in its SAAR on the application in this case. Nonetheless, a fundamental difference remains between this case and the District 9 applications, including Boca Raton's. The application in this case is distinguished by the specific information to which Mr. Hopes alluded in his testimony, quoted above.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that a final order be entered granting the application of Galencare, Inc., d/b/a Brandon Regional Hospital for open heart surgery, CON 9239. DONE AND ENTERED this 30th day of March, 2001, in Tallahassee, Leon County, Florida. DAVID M. MALONEY Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 30th day of March, 2001. COPIES FURNISHED: Sam Power, Agency Clerk Agency for Health Care Administration 2727 Mahan Drive Building 3, Suite 3431 Tallahassee, Florida 32308-5403 Julie Gallagher, General Counsel Agency for Health Care Administration 2727 Mahan Drive Building 3, Suite 3431 Tallahassee, Florida 32308-5403 Richard A. Patterson, Esquire Agency for Health Care Administration 2727 Mahan Drive Building 3, Suite 3431 Tallahassee, Florida 32308-5403 Stephen A. Ecenia, Esquire R. David Prescott, Esquire Thomas W. Konrad, Esquire Rutledge, Ecenia, Purnell & Hoffman, P.A. 215 North Monroe Street, Suite 420 Post Office Box 551 Tallahassee, Florida 32302-0551 James C. Hauser, Esquire Metz, Hauser & Husband, P.A. 215 South Monroe Street, Suite 505 Post Office Box 10909 Tallahassee, Florida 32302 John H. Parker, Jr., Esquire Jonathan L. Rue, Esquire Sarah E. Evans, Esquire Parker, Hudson, Rainer & Dobbs 1500 Marquis Two Tower 285 Peachtree Center Avenue, Northeast Atlanta, Georgia 30303 Robert A. Weiss, Esquire Karen A. Putnal, Esquire Parker, Hudson, Rainer & Dobbs, LLP The Perkins House, Suite 200 118 North Monroe Street Tallahassee, Florida 32301
The Issue Whether there is need for a new Pediatric Heart Transplant and/or Pediatric Heart and Lung Transplant program in Organ Transplant Service Area (OTSA) 3; and, if so, whether Certificate of Need (CON) Application No. 10471 (heart) and/or 10472 (heart and lung), filed by The Nemours Foundation, d/b/a Nemours Children’s Hospital (Nemours or NCH), to establish a Pediatric Heart Transplant and/or Pediatric Heart and Lung Transplant program, satisfy the applicable statutory and rule review criteria for award of a CON.
Findings Of Fact Based upon the demeanor and credibility of the witnesses and other evidence presented at the final hearing and on the entire record of this proceeding, the following Findings of Fact are made: The Parties The Applicant, Nemours Nemours Children’s Hospital is a licensed Class II specialty children’s hospital located in Orange County, Health Planning District 7, Subdistrict 7-2, OTSA 3, which is owned and operated by The Nemours Foundation. Nemours is licensed for 100 beds, including 73 acute care, nine comprehensive medical rehabilitation, two Level II neonatal intensive care unit (NICU), and 16 Level III NICU beds, and is a licensed provider of pediatric inpatient cardiac catheterization and pediatric open-heart surgery. As the primary beneficiary of the Alfred I. duPont Testamentary Trust established in the will of Alfred duPont, the Foundation was incorporated in Florida in 1936. The Foundation set out to provide children and families medical care and services, its mission being “[t]o provide leadership, institutions, and services to restore and improve the health of children through care and programs not readily available, with one high standard of quality and distinction regardless of the recipient’s financial status.” Foundation assets reached $5.5 billion, by the end of 2015. The Foundation has funded $1.5 billion of care to Florida’s pediatric population through subspecialty pediatric services, research, education, and advocacy. Nemours has established a pediatric care presence throughout the State of Florida. Nemours operates over 40 outpatient clinics throughout Florida that offer primary care, specialty care, urgent care, and cardiac care services to pediatric patients in central Florida, Jacksonville, and the panhandle region. Nemours also provides hospital care to pediatric inpatients at Nemours Children’s Hospital in Orlando, as well as through affiliations with Wolfson’s Children’s Hospital in Jacksonville, West Florida Hospital in Pensacola, and numerous hospital partners in central Florida. The resources Nemours offers in the greater Orlando area are especially significant with 17 Primary Care Clinics, five Urgent Care Clinics, 10 Specialty Care Clinics, nine Nemours Hospital partners, and, of course, NCH itself. These clinics are located throughout OTSA 3 where Nemours determined access to pediatric care was lacking, including Orlando, Melbourne, Daytona Beach, Titusville, Kissimmee, Lake Mary, and Sanford, as well as neighboring Lakeland. The clinics are fully staffed with hundreds of Nemours-employed physicians who live in the clinic communities. Through these satellite locations, as well as the Nemours CareConnect telemedicine platform, Nemours is able to bring access to its world-class subspecialists located at NCH to children throughout the State of Florida who otherwise would not have access to such care. Nemours was established to provide state of the art medical care to children through its integrated model. Nemours’ development has been and continues to be driven by its mission and objective to be a top-tier, world-class pediatric healthcare system. NCH is the first completely new “green field” children’s hospital in the United States in over 40 years, allowing Nemours to integrate cutting-edge technology and a patient-centered approach throughout. Nemours has created a unique integrated model of care that addresses the needs of the child across the whole continuum, connecting policy and prevention, to the highest levels of specialized care for the most complex pediatric patients. From its inception, Nemours envisioned the development of a comprehensive cardiothoracic transplant program as proposed by the CON applications at issue in this proceeding. NCH is located in the Lake Nona area, just east of downtown Orlando in a development known as Medical City. Medical City is comprised of a new VA Hospital, the University of Central Florida (UCF) College of Medicine and School of Biomedical Sciences, the University of Florida (UF) Research and Academic Center, the Sanford Burnham Medical Research Institute, and a CON-approved hospital, which is a joint venture between UCF and AHCA, which will serve as UCF’s teaching hospital. Medical City is intended to bring together life scientists and research that uses extraordinarily advanced technology. Co- location in an integrated environment allows providers and innovators of healthcare, “the brightest minds,” so to speak, to interact and to share ideas to advance healthcare and wellness efforts. Agency for Health Care Administration AHCA is the state health-planning agency that is charged with administration of the CON program as set forth in sections 408.031-408.0455, Florida Statutes. Context of the Nemours Applications Pursuant to Florida Administrative Code Rule 59C-1.044, AHCA requires applicants to obtain separate CONs for the establishment of each adult or pediatric organ transplantation program, including: heart, kidney, liver, bone marrow, lung, lung and heart, pancreas and islet cells, and intestine transplantations. “Transplantation” is “the surgical grafting or implanting in its entirety or in part one or more tissues or organs taken from another person.” Fla. Admin. Code R. 59A-3.065. Heart transplantation, lung transplantation, and heart/lung transplantation are all defined by rule 59C-1.002(41) as “tertiary health services,” meaning “a health service which, due to its high level of intensity, complexity, specialized or limited applicability, and cost, should be limited to, and concentrated in, a limited number of hospitals to ensure the quality, availability, and cost effectiveness of such service.” AHCA rules define a “pediatric patient” as “a patient under the age of 15 years.” Fla. Admin. Code R. 59C-1.044(2)(c). AHCA rules divide Florida into four OTSAs, corresponding generally with the northern, western central, eastern central, and southern regions of the state. Fla. Admin. Code R. 59C-1.044(2)(f). The programs at issue in this proceeding will be located in OTSA 3, which is comprised of Brevard, Indian River, Lake, Martin, Okeechobee, Orange, Osceola, Seminole, and Volusia Counties. Currently, there are no providers of PHT in OTSA 3, and there are no approved PHLT programs statewide. The incidence of PHT in Florida, as compared to other types of solid organ transplants, is relatively small. The chart below sets forth the number of pediatric (ages 0-14) heart transplant discharges by year for the four existing Florida PHT programs during Calendar Years (CY) 2013 through 2016, and the 12-month period ending June 2017: HOSPITAL HEART TRANSPLANT CY 2013 CY 2014 CY 2015 CY 2016 12 MONTHS ENDING JUNE 2017 All Children’s Hospital 7 14 9 8 7 UF Health Shands Hospital 6 8 15 15 9 Memorial Regional Hospital 5 5 5 7 4 Jackson Health System 2 2 1 4 1 Total 20 29 30 34 21 The above historic data demonstrates that the incidence of PHT statewide is relatively rare, and does fluctuate from program to program and from year to year. As can be seen, the most recent available 12-month data reflects that only 21 PHTs were performed during that time, for an average program volume of only 5.25 cases. Florida has more existing and approved PHT programs than every other state in the country except California, which has more than double the pediatric population of Florida. And like Florida, two of the California programs are extremely low- volume programs. Additionally, evidence regarding the number of PHLT patients demonstrated just how rare this procedure is. From 2013 to 2016, there was an annual average of only four PHLTs nationally, with only one actual transplant on a Floridian. Nemours’ health planner stated that although Nemours projected in its application that it would perform one heart/lung procedure each year, it is a “very low-volume service,” and Nemours in actuality expects that there will be years with zero volume of PHLT. The CON Applications Nemours filed its applications for heart transplantation, heart/lung transplantation, and lung transplantation in the second Other Beds and Programs Batching Cycle of 2016. Nemours is proposing the development of a comprehensive cardiothoracic transplant program, which will be the only such program in Florida. This will be achieved by combining three types of transplant services (heart, lung, and heart/lung) in one comprehensive cardiothoracic transplant program. Each application was conditioned on the development of all three transplantation programs. Nemours is located in OTSA 3, where there is currently no PHT provider, PLT provider, or PHLT provider. There are, however, three providers of pediatric open-heart surgery and pediatric cardiac catheterization, and a large, growing pediatric population. Unlike any other facility in Florida, the Nemours Cardiac Center (Cardiac Center) is uniquely organized to treat all forms of congenital heart disease. The Cardiac Center employs a “programmatic approach” to offer the most beneficial environment and the finest care available for pediatric patients. The Cardiac Center, physically located at NCH, throughout Florida, is organized as a single Department of Cardiovascular Services to house Cardiac Surgery, Cardiac Anesthesia, Cardiac Intensive Care Unit (ICU), and Cardiology. Cardiac Center physicians throughout Florida are organized as a single entity with the goal of providing the highest quality, patient-centered care to all patients without the usual barriers created by the departmental “silos.” The entire Cardiac Center clinical team, including nurses and physicians, is dedicated solely to the special challenges of congenital heart abnormalities and makes the care of children with heart disease the life’s work of team members. The fully integrated organizational structure permits the team to take shared responsibility for all aspects of the delivery of quality care to these pediatric patients from admission to discharge. The Cardiac Center holds weekly patient consensus conferences, where all providers, including physicians, nurses, and the patients’ caregivers, participate in case reviews of all inpatients and those patients scheduled for surgery or catheterization. The Cardiac Center is “state of the art” with a designated cardiovascular operating room, a designated cardiovascular lab that includes an electrophysiology lab, and a dedicated comprehensive care unit. In addition, The Foundation has furthered the commitment to the Cardiac Center by funding an additional $35 million expansion to the sixth floor of NCH, adding an additional 31 inpatient beds, an additional operating room, and a comprehensive cardiovascular intensive care unit. Dr. Peter D. Wearden joined Nemours in 2015 as the chief of cardiac surgery, chair of the Department of Cardiovascular Services, and director of the Cardiac Center at Nemours. Dr. Wearden will serve as director of the Comprehensive Cardiothoracic Transplant Program at Nemours and will be instrumental in the development and implementation of the program. Dr. Wearden was recruited from the Children's Hospital of Pittsburgh (CHP), where he served as the surgical director of Heart, Lung, and Heart/Lung Transplantation. He was also the director of the Mechanical Cardiopulmonary Support and Artificial Heart Program. CHP rose to a US News and World Report top 10 program during Dr. Wearden’s tenure. CHP is at the forefront of organ transplantation and is where the first pediatric heart/lung transplantation was performed. Dr. Wearden is a trained cardiothoracic surgeon who completed fellowships in both cardiothoracic surgery (University of Pittsburgh) and Pediatric and Congenital Heart Surgery (Hospital for Sick Children, Toronto, Canada). He is certified by the American Board of Thoracic Surgery and holds additional qualifications in Congenital Heart Surgery from that organization. In his tenure as a board-certified pediatric transplant specialist, he has participated in over 200 pediatric cardiothoracic transplantations, of which he was the lead surgeon in over 70. In addition, he has procured over $20 million in National Institutes of Health research funding since 2004 specific to the development of artificial hearts and lungs for children and their implementation as a live-saving bridge to transplantation. Dr. Wearden was a member of the clinical team that presented to the Food and Drug Administration (FDA) panel for approval of the Berlin Heart, the only FDA-approved pediatric heart ventricular assist device (VAD)1/ currently available, and he proctored the first pediatric artificial heart implantation in Japan in 2012. A VAD is referred to as “bridge to transplant” in pediatric patients because the device enables a patient on a waiting list for a donated heart to survive but is a device on which a child could not live out his or her life. Both utilization of VADs and heart transplantation procedures are in the “portfolio of surgical interventions” that can save the life of a child with heart failure. Dr. Wearden is an international leader in the research and development of VADs. Victor Morell, an eminent cardiac surgeon and chief of Pediatric Cardiac Surgery at CHP, testified that Dr. Wearden’s presence in Orlando alone and the work that he will be able to do with VADs and a PHT program will likely save lives. Many of the physicians that comprise the Nemours Cardiac Center transplant team not only have significant transplant experience, but also have experience performing transplants together. These physicians came with Dr. Wearden from CHP, were trained by Dr. Wearden, or otherwise worked with Dr. Wearden at some point in their careers. The physicians recruited to the Nemours transplantation team were trained at or hail from among the most prestigious programs in the country. For example, Dr. Kimberly Baker, a cardiac intensivist, was trained by Dr. Wearden in the CHP ICU. Dr. Constantinos Chrysostomou, Nemours’ director of cardiac intensive care, worked with Dr. Wearden at CHP, and has experience starting the pediatric ICU in Los Angeles at Cedar Sinai Hospital. Dr. Steven Lichtenstein, chief of cardiac anesthesia, held the same position at CHP for 12 years before he was recruited to Nemours. Dr. Karen Bender, a cardiac anesthesiologist, was recruited by Dr. Wearden from the Children’s Hospital of Philadelphia – one of the leading programs in the country. Dr. Michael Bingler, a cardiac interventionalist, was at Mercy Children’s Hospital in Kansas City for eight years. Dr. Adam Lowry of the Nemours cardiac intensive care center previously trained at both Texas Children’s Hospital (the number one program in the country) and Stanford. The 11 physicians that comprise the Cardiac Center’s Cardiothoracic Physician Team have collectively participated in 1,146 cardiothoracic transplantations. These physicians came to Nemours to care for the most acute, critically ill patients, including those requiring PHT. In addition to the physician team, the expertise and skill of the non-physician staff in the catheterization lab, the operating room, and the cardiac ICU are crucial to a successful program. Dr. Dawn Tucker is the administrative director of NCH’s Cardiac Center and heads the nursing staff for NCH’s Cardiac Center, which includes 23 registered nurses with transplant experience. Dr. Tucker holds a doctorate of Nursing Practice and was formerly the director of the Heart Center at Mercy Children’s Hospital in Kansas City, where she oversaw the initiation of a PHT program. The average years of experience for total nursing care in cardiac units across the nation is two years. The average years of experience in the Nemours Cardiac Center is eight years. Medical literature shows the greater the years of nursing staff experience, the lower the mortality and morbidity rates. The nursing staff at Nemours, moreover, has extensive experience in dealing not only with pediatric cardiac patients, but with pediatric heart transplants as well. The Cardiac Center’s cardiothoracic nursing staff has over 220 years of collective cardiothoracic transplant experience. Nemours operates a “simulation center” that allows the Cardiac Center to simulate any type of cardiac procedure on a model patient before performing that procedure on an actual patient. The model patient’s “heart” is produced using a three- dimensional printer that creates a replica of the heart based on MRI’s or other medical digital imaging equipment. These replica hearts are printed on-site, using the only FDA-approved software for such use, and are ready for use in the simulation center within a day after medical imaging. Nemours Cardiac Center currently performs what the Society of Thoracic Surgeons has coined “STAT 5” cardiac procedures. STAT 5 cardiac procedures are the most complex; STAT 1 procedures are the least complex. A PHT is a STAT 4 procedure. Since Dr. Wearden’s arrival at the Nemours Cardiac Center, there have been no patient mortalities. The uncontroverted evidence established that Nemours has assembled a high-quality, experienced, and unquestionably capable team of physicians and advanced practitioners for its cardiothoracic transplantation programs and is capable of performing the services proposed in its applications at a high level. UF Health Shands While not a party to this proceeding,2/ UF Health Shands’ (Shands) presence at the final hearing was pervasive. AHCA called numerous witnesses affiliated with Shands in its case-in-chief. The scope of the testimony presented by Shands- affiliated witnesses was circumscribed by Order dated December 13, 2017 (ruling on NCH’s motion in limine) that: At hearing, the Agency may present evidence that the needs of patients within OTSA 3 are being adequately served by providers located outside of OTSA 3, but may not present evidence regarding adverse impact on providers located outside of OTSA 3. Baycare of Se. Pasco, Inc. v. Ag. for Health Care Admin., Case No. 07-3482CON (Fla. DOAH Oct. 28, 2008; Fla. AHCA Jan. 7, 2009). UF Health Shands Hospital is located in Gainesville, Florida. UF Health Shands Children’s Hospital is an embedded hospital within a larger hospital complex. Shands Children’s Hospital has 200 beds and is held out to the public as a children’s hospital. The children’s hospital has 72 Level II and III NICU beds. Unlike Nemours, Shands offers obstetrical services such that babies are delivered at Shands. It also has a dedicated pediatric intensive care unit (PICU) as well as a dedicated pediatric cardiac intensive care unit. The Shands Children’s Hospital has its own separate emergency room and occupies four floors of the building in which it is located. It is separated from the adult services. Shands Children’s Hospital is nationally recognized by U.S. News & World Report as one of the nation’s best children’s hospitals. The children’s hospital has its own leadership, including Dr. Shelley Collins, an associate professor of pediatrics and the associate chief medical officer. As a comprehensive teaching and research institution, Shands Children’s Hospital has virtually every pediatric subspecialty that exists and is also a pediatric trauma center. The children’s hospital typically has 45 to 50 physician residents and 25 to 30 fellows along with medical students. Over $139 million has been awarded to Shands for research activities. As a teaching hospital, Shands is accustomed to caring for the needs of patients and families that come from other parts of the state or beyond. Jean Osbrach, a social work manager at Shands, testified for AHCA. Ms. Osbrach oversees the transplant social workers that provide services to the families with patients at Shands Children’s Hospital. Ms. Osbrach described how the transplant social workers interact with the families facing transplant from the outset of their connection with Shands. They help the families adjust to the child’s illness and deal with the crisis; they provide concrete services; and these social workers help the families by serving as navigators through the system. These social workers are part of the multidisciplinary team of care, and they stay involved with these families for years. Shands is adept at helping families with the issues associated with getting care away from their home cities. Shands has apartments specifically available in close proximity to the children’s hospital and relationships with organizations that can help families that need some financial support for items such as lodging, transportation, and gas. Ms. Osbrach’s ability to empathize with these families is further enhanced because her own daughter was seriously ill when she was younger. Ms. Osbrach testified that, while she was living in Gainesville, she searched out the best options for her child and decided that it was actually in Orlando. Despite the travel distance, she did not hesitate to make those trips in order to get the care her child needed at that time. The Shands Children’s Hospital is affiliated with the Children’s Hospital Association, the Children’s Miracle Network, the March of Dimes, and the Ronald McDonald House Charities. Shands operates ShandsCair, a comprehensive emergency transport system. ShandsCair operates nine ground ambulances of different sizes, five helicopters, and one fixed-wing jet aircraft. ShandsCair does over 7,000 transports a year, including a range of NICU and other pediatric transports. ShandsCair is one of the few services in the country that owns an EC-155 helicopter, which is the largest helicopter used as an air ambulance. This makes it easier to transport patients that require a lot of equipment, including those on extracorporeal membrane oxygenation (ECMO). Patients on ECMO can be safely transported by ground and by air by ShandsCair. Shandscair serves as a first responder and also provides facility-to- facility transport. It has been a leader in innovation. The congenital heart program at Shands includes two pediatric heart surgeons, as well as pediatric cardiologists Dr. Jay Fricker and Dr. Bill Pietra, both of whom testified for AHCA. Dr. Fricker did much of his early work and training at the Children’s Hospital of Pittsburgh, and came to the University of Florida in 1995. He is a professor and chief of the Division of Cardiology in the Department of Pediatrics at Shands. He is also the Gerold L. Schiebler Eminent Scholar Chair in Pediatric Cardiology at UF. He has been involved in the care of pediatric heart transplant patients his entire career. Dr. Bill Pietra received his medical training in Cincinnati and did his early work at several children’s hospitals in Colorado. He came to the University of Florida and Shands in July 2014 and is now the medical director, UF Health Congenital Heart Center. Shands performed its first pediatric heart transplant in 1986. Shands treats the full range of patients with heart disease and performs heart transplants on patients, from infants through adults, with complex congenital heart disease. Shands provides transplants to pediatric patients with both congenital heart defects and acquired heart disease (cardiomyopathy). Shands will accept the most difficult cases, including those that other institutions will not take. Data presented by AHCA dating back to the beginning of 2014 demonstrate that Shands has successfully transplanted numerous patients that were less than six months old at the time of transplantation. This data also demonstrates that Shands serves all of central and north Florida, as well as patients that choose to come to Shands from other states. PHT patients now survive much longer than in the past, and in many cases, well into adulthood. Because Shands cares for both adult and pediatric patients, it has the ability to continue to care for PHT patients as they transition from childhood to adulthood. Managed care companies are now a significant driver of where patients go for transplantation services. Many managed- care companies identify “centers of excellence” as their preferred providers for services such as PHT. Shands is recognized by the three major managed-care companies that identify transplant programs as a center of excellence for PHT services. AHCA’s Preliminary Decision Following AHCA’s review of Nemours’s applications, as well as consideration of comments made at the public hearing held on January 10, 2017, and written statements in support of and in opposition to the proposals, AHCA determined to preliminarily deny the PHT and PHLT applications, and to approve the PLT application. AHCA’s decision was memorialized in three separate SAARs, all dated February 17, 2017. Marisol Fitch, supervisor of AHCA’s CON and commercial-managed care unit, testified for AHCA. Ms. Fitch testified that AHCA does not publish a numeric need for transplant programs, as it does for other categories of services and facilities. Rather, the onus is on the applicant to demonstrate need for the program based on whatever methodology they choose to present to AHCA. In addition to the applicant’s need methodology, AHCA also looks at availability and accessibility of service in the area to determine whether there is an access problem. Finally, an applicant may attempt to demonstrate that “not normal” circumstances exist in its proposed service area sufficient to justify approval. Statutory Review Criteria Section 408.035(1) establishes the statutory review criteria applicable to CON Applications 10471 and 10472. The parties have stipulated that each CON application satisfies the criteria found in section 408.035(1), (d), (f), and (h), Florida Statutes. The only criteria at issue essentially relate to need and access. However, the Agency maintains that section 408.035(1)(c) is in dispute to the extent that center transplant volume as a result of Nemours’ approval would lead to or correlate with negative patient outcomes. AHCA believes that there is no need for the PHT or PHLT programs that Nemours seeks to develop because the needs of the children in the Nemours service area are being met by other providers in the state, principally Shands and Johns Hopkins All Children’s Hospital. Section 408.035(1)(a) and (b): The need for the health care facilities and health services being proposed; and the availability, quality of care, accessibility, and extent of utilization of existing health care facilities and health services in the district of the applicant. Florida Administrative Code Rule 59C-1.044(6)(b).3/ The criteria for the evaluation of CON applications, including applications for organ transplantation programs, are set forth at section 408.035 and rule 59C-1.044. However, neither the applicable statutes nor rules have a numeric need methodology that predicts future need for PHT or PHLT programs. Thus, it is up to the applicant to demonstrate need in accordance with rule 59C-1.044. There are four OTSAs in Florida, numbered OTSA 1 through OTSA 4. NCH is located in OSTA 3, which includes the following counties: Seminole, Orange, Osceola, Brevard, Indian River, Okeechobee, St. Lucie, Martin, Lake, and Volusia. (See § 408.032(5), Fla. Stat; Fla. Admin. Code R. 59C- 01.044(2)(f)3.) OTSA 3 also generally corresponds with the pediatric cardiac catheterization and open-heart surgery service areas defined by AHCA rule. (See Fla. Admin. Code R. 59C- 1.032(2)(g) and 59C-1.033(2)(h)). Currently, there is no provider of PHT in OTSA 3, but there are three providers of pediatric cardiac catheterization and pediatric open-heart surgery: Orlando Health Arnold Palmer Hospital for Children; Florida Hospital for Children; and Nemours. There are no licensed providers of PHLT anywhere in the State of Florida. There are four existing providers and one approved provider of PHT services in Florida: UF Shands in OTSA 1; Johns Hopkins All Children’s Hospital in OTSA 2; Jackson Memorial Hospital in OSTA 4; and Memorial Regional Hospital, d/b/a Joe DiMaggio’s Hospital in OTSA 4; and a third approved program in OTSA 4, Nicklaus Children’s Hospital, which received final approval from AHCA in August 2017. As noted above, there is no fixed-need pool published for PHT, PHLT, or PLT programs. Alternatively, AHCA follows rule 59C-1.008(2)(e)2., which requires consideration of population demographics and dynamics; availability, utilization and quality of like services in the district, subdistrict, or both; medical treatment trends; and market conditions. To quantify the need for a new PHT program in District 7, OTSA 3, Nemours created and presented a methodology that started with the statewide use rate in its projected first year. Then for the second year, Nemours aggressively increased the use rate to the highest rate in any of the other transplant service areas in the state. Then, in an even more aggressive (and unreasonable) assumption, Nemours projected that it would essentially capture all of the cases in OTSA 3 by the second year of the program. In its application, the assumptions resulted in a projection that Nemours would do four transplants in the first year of operation and eight in the second. These projections fall short of the rule requirement that the applicant project a minimum of 12 transplants per year by the second year of operation. Fla. Admin. Code R. 59C-1.044(6)(b)2. At hearing, Nemours updated (increased) those first and second year projections to 7 and 13 cases, respectively. However, these updated projections included one child, aged 15 to 17, in year one, and two in year two. There are several reasons these projections lack credibility. First, as noted, Nemours assumed a near- 100 percent market share based on the highest use rate in the state by just year two. Second, when Nemours prepared its update, it used the most recent calendar year data. However, this was not the most current data. Calendar Year 2016 reflected 34 cases statewide, but that number had dropped to 21 for the most recent 12-month period available at the time of the hearing. Use of this most recent 12-month data would have significantly decreased the Nemours PHT volume projections. In addition, the projection of 13 cases by year two would place Nemours at a higher PHT case volume than three of the four established programs in the state, and would be at a level that is nearly equivalent to the much more established Shands program. This is not credible, especially considering that Nemours also admitted at hearing that only two OTSA 3 residents received pediatric heart transplants in 2016. The existence of unmet need cannot be based solely on the absence of an existing service in the proposed service area. Fla. Admin. Code R. 59C-1.008(2)(e)3. While Nemours’ own health planner agreed that the absence of a PHT program in OTSA 3 is not itself a basis for finding need, Nemours nevertheless argues that this rule is inapplicable in this proceeding because the title to this subsection of the rule is “Comparative Review” and a portion of this subsection addresses competing applications in the same cycle. As detailed further in the Conclusions of Law section herein, this interpretation is unconvincing and rejected. AHCA interprets this rule provision to apply to those batched applications submitted without the submission of a competing application in the same batching cycle, as with Nemours in this proceeding. Nemours initiated its cardiac catheterization and cardiac surgery program in June/July 2016. In its PHT application, Nemours projected that it would meet or exceed the rule minimum required volumes of 200 cardiac catheterizations and 125 open-heart surgery cases by the end of 2017. Actual volumes achieved by Nemours in CY 2017 were 97 open-heart cases and 196 cardiac catheterizations. The incidence of PHLT is extremely low. During the four calendar years, 2013 through 2016, there were only 16 PHLT transplants performed nationwide. Only one Florida resident received a PHLT during that four-year period, and that was performed in Massachusetts. Also during that four-year period, only three Florida residents were registered for PHLT. There is no evidence in this record as to why two of the three registered Florida residents did not obtain a PHLT. Based on the national use rate for PHLTs from CY 2013 through CY 2016, Nemours projects that it will perform an average of one PHLT per year. Nemours acknowledges that due to the extremely low incidence of PHLTs, there may be some years that no PHLTs are performed at Nemours. Geographic Access There is no evidence of record that families living in central Florida are currently being forced to travel unreasonable distances to obtain PHT services. Indeed, there are five existing or approved programs within the state, with at least two located very reasonably proximate to OTSA 3. According to the analysis of travel distances for PHT patients living in OTSA 3 contained in the Nemours application (Exhibit 15), only some residents located in Brevard and Indian River Counties are not within 120 miles of an existing PHT program. There was agreement that patients that need a PHT are approaching the end-stage of cardiac function, and in the absence of a PHT will very likely die. Accordingly, it is reasonable to infer that the parents of a child living in central Florida and needing a PHT will travel to St. Petersburg or Gainesville for transplant services rather than let their child die because the travel distance is too far. To the contrary, the evidence in this record from witnesses on both sides, as well as common sense, is that families will go as far as necessary to save their child. The notion that there is some pent-up demand for PHT services among central Florida residents (especially when there is no evidence of a single OTSA 3 patient being turned down or unable to access a PHT) is without support in this record. The parents of four pediatric patients testified at the final hearing. Two testified for Nemours. The other two testified for AHCA and were parents of children that received PHTs at Shands. One of the Nemours witnesses was the parent of a child that has not received a transplant. The other received transplant services at Johns Hopkins All Children’s Hospital in St. Petersburg. The parents of the two Shands patients were representative of the two broad categories of PHT patients. One was a patient with a congenital heart defect that lives in Cocoa Beach (Brevard County). The patient likely had the heart defect since birth, but it was not diagnosed until she was six years old. That patient was asymptomatic at the time of diagnosis but deteriorated over a period of years. While she was first seen at Shands, the family had the time and researched other prominent institutions, including Texas Children’s Hospital, Boston Children’s Hospital, Children’s Hospital of Pittsburgh, and the Mayo Clinic in Rochester, Minnesota. They did this because, like all of the parents that testified, they “would have gone to the ends of the earth” to save their child. This family researched the volumes and experience of the programs they considered and looked for what they felt was the best program for their child, and ultimately chose Shands. It was clear that they felt Shands was the right choice. Their daughter received her heart transplant at Shands, is doing well, and is now considering what college to attend. Additionally, this family did not find the two hours and 35 minute travel time from their home in Brevard County to Shands to be an impediment, and actually consider Shands as being relatively close to their home. This testimony supports the obvious truism that obtaining the best possible outcome for a sick child is the paramount goal of any parent. The other parent witness called by AHCA has a daughter that, on Christmas Eve in 2008, went from perfectly healthy to near death and being placed on life support within a 24-hour period. As opposed to a congenital heart defect, this patient had cardiomyopathy. This family lives in Windermere, a suburb of Orlando. She acquired a virus that attacked her heart. She was initially treated at Arnold Palmer Children’s Hospital where she had to be placed on ECMO. From there, she was safely airlifted to Shands while still on ECMO where, upon arrival, the receiving team of physicians informed the family that she was one of the most critically ill children they had ever seen. After an 11-hour open-heart surgery, a Berlin Heart was successfully implanted and kept her alive for four months until an appropriate donor heart became available. This patient also had an excellent outcome and is now a student at the University of Florida. The following exchange summarizes how the child’s mother felt about the inconvenience of having to travel from the Orlando area to Gainesville: Q If a family in Orlando told you, or in your city of residence told you that their child was critically ill and they were worried about having to travel and potentially spend time in Gainesville to get care, what would you tell them? A Well, I would tell them to just take it a day at a time and – when your child is critically ill, convenience never really comes into your mind. What comes into your mind is how do I help my child live. And so you will go anywhere. And it’s just an hour and a half, it just doesn’t matter. When you are talking about saving your child, it means nothing. It literally means nothing. It is clear from the testimony of these two parents that nothing about having a gravely ill child is “convenient.” It creates great stress, but it was also clear that having an experienced provider was more important than just geographic proximity. The mothers of the two Shands patients persuasively spoke of their concerns about further diluting the volumes of the existing programs that could result from approval of a sixth pediatric heart transplant program in Florida, particularly when there are two other programs that are not that far from the Orlando area.4/ While transplantation is not an elective service, it is not done on an emergent basis. As noted, the number of families affected is, quite fortunately, very small. While having a child with these issues is never “convenient,” the travel issues that might exist do not outweigh the weight of the evidence that fails to demonstrate a need for approval of either application. The Orlando area, being centrally located in Florida, is reasonably accessible to all of the existing providers. Most appear to go to Shands, which is simply not a substantial distance away. The credible evidence is that families facing these issues are able to deal with the travel inconvenience. In addition, Nemours presented evidence regarding the various locations at which they provide services, ranging from Pensacola to Port St. Lucie. Clearly, Nemours sees itself as providing some cardiac services to patients in these locations, but it would also suggest that patients seen at these locations may be referred to NCH for transplant services, which would mean that some patients would be bypassing closer facilities. As observed by AHCA, for Nemours to posit that it is appropriate for patients to travel from Pensacola or Jacksonville to Orlando while asserting that it is not acceptable for patients in Orlando to go to Gainesville or St. Petersburg is an illogical inconsistency. Financial Access Nemours asserts that approval of its proposed programs will enhance financial access to care. Nemours currently serves patients without regard to ability to pay and will extend these same policies to transplant recipients. Approximately half of Nemours’ projected PHTs are to be provided to Medicaid recipients, the other half to commercially insured patients.5/ However, there was no competent evidence of record that access to PHT or PHLT services was being denied by any of the existing transplant providers because of a patient’s inability to pay. Transplant Rates at Shands In its need methodology, Nemours utilized the use rate from OTSA 1 where Shands is located because it is the highest use rate in the state. Despite this, Nemours then asserted that Shands is not performing as many PHTs as it could or should. The Nemours CON applications are not predicated on any argument that their proposed programs are needed because of poor quality care at any of the existing pediatric transplant programs in Florida. Indeed, Dr. Wearden stated his belief that Shands provides good quality care in its transplant programs, and he respects the Shands lead surgeon, Dr. Mark Bleiweis. As evidence of his respect for the Shands PHT program, Dr. Wearden has referred several transplant patients to Dr. Bleiweis at Shands. Despite that position, Nemours argued that the Shands program is unduly conservative and cautious in its organ selection and may have some “capacity” issues due to a few cited instances of apparent surgeon unavailability. These assertions, made by Nemours witnesses with no first-hand knowledge of the operations of the Shands program, are not persuasive. With regard to whether the Shands program is unduly “cautious,” “conservative,” or “picky,” Nemours relied on a document produced by Shands in discovery. Nemours also relied on data reported by Shands to the Scientific Registry of Transplant Recipients (SRTR). The data included a list of all of the organs offered to Shands since the beginning of 2015, the sequencing of the offer of that organ to Shands, whether the organ was transplanted at Shands or elsewhere, the primary and secondary reasons the organ was refused (if refused) and other information. The SRTR exhibit demonstrates that a high number of the organs that are offered are not acceptable for transplant on patients waitlisted at Shands. It also shows that organs that are accepted may have to be examined by many different centers before being deemed potentially acceptable. This demonstrates the extensive level of complexity, nuance, and clinical judgment involved in the decision to accept an organ for transplant in a pediatric patient. Indeed, Dr. Wearden agreed that the decision by a program to accept or turn down an organ involves both clinical expertise and judgment, and that there are many reasons an organ might be turned down, which helps explain why the transplanted percentage of total organs offered nationally is on average, so small. Dr. Wearden chose a few examples of organs that were not taken by Shands to express an opinion that Shands may be unduly conservative in its organ selection. However, this assertion was credibly refuted by Dr. Pietra, a transplant cardiologist and the medical director of the UF Health Congenital Heart Center. Dr. Pietra discussed the complexity of these cases and how simply looking at the SRTR data does not provide enough information to reach Dr. Wearden’s conclusion. An organ that might be acceptable for one patient would not be acceptable for another for a host of reasons. Many more organs are rejected by transplant centers than are accepted. Dr. Pietra credibly opined that being conservative and cautious are important traits for a transplant surgeon, particularly for one that wants the accepted organ to work well for the patient long-term. That does not mean that Shands is rejecting organs when it should have taken them, nor does the SRTR data support the proposition that the Nemours program should be approved because its program may have accepted an organ for a particular patient that Shands might have rejected. Nemours also argues that Shands performs PHTs at a rate lower than the region and the country, and that this should mitigate for the approval of another program. This assertion is predicated on waitlist information reported in the SRTR data. Patients that are placed on the waitlist have different status designations, depending on the severity of their condition. That status may change, up or down, over time. Due to the shortage of organs, until a patient reaches status 1A, he or she is unlikely to be offered an organ. The evidence reflected that Shands puts patients on the PHT organ waitlist at a time earlier than the moment they require the transplant surgery under what is called the “pediatric prerogative.” This helps those patients maintain their status on the list but does not result in organs being provided to less severely ill patients to the detriment of those in greater need. Further, the record evidence supports the finding that Shands waitlists patients because the clinical determination has been made that the child will ultimately require a transplant. This was corroborated by the parent of a Shands PHT patient who testified that when her daughter was placed on the waitlist, Dr. Fricker concluded at that time that her daughter would ultimately need a PHT, even though she was placed on a lower status initially, and it was a few years before the transplant occurred. Transplant surgeon Dr. Victor Morell, of the Children’s Hospital of Pittsburgh, testified that he waitlists his PHT patients not only when they need the procedure performed immediately, but rather when, in his clinical judgment, he determines the patient will ultimately need a PHT. This testimony supports the finding that there is nothing clinically unusual or inappropriate about how the Shands program waitlists patients. Shands realizes that its philosophy, which is contemplated within and permitted under the United Network for Organ Sharing (UNOS) rules, makes its statistics, both in terms of percent of patients transplanted and waitlist mortality, look worse. While Shands’ waitlist mortality may be higher than expected as reflected in the SRTR data, it is still significantly lower than in the UNOS region or the United States. Shands advocates for its patients by their waitlist practices because it believes it helps secure the best outcomes for its patients. It does not indicate need for a new PHT program. Nemours also suggests that there may be a “capacity” problem at Shands because the organ rejection information provided by Shands shows that, during the 3-year period of CY 2015 through CY 2017, there were seven entries showing as either a primary or secondary reason for organ rejection that the surgeon was unavailable. However, this included both adult and pediatric hearts, and further investigation revealed that in only four instances were there potential PHT recipients at Shands. Of those four hearts that were rejected, two were not accepted by any PHT provider, and the two that were accepted were placed with adult transplant patients, not PHT patients. Shands has two PHT transplant surgeons. In very few instances at Shands, an organ was offered but not accepted because the surgeon was not available for one of several reasons. In one instance, there was another transplant scheduled. A surgeon could be ill, could be gone, or may have just completed another long surgery and be too fatigued to safely perform another. Like Shands, Nemours also has two experienced PHT surgeons. Although Dr. Wearden believes that Nemours would endeavor to not reject an organ for this reason, this ambition ignores reality. He cannot guarantee that the same could not or would not happen at Nemours for the same reasons it occasionally occurs at Shands. As explained by Dr. Pietra, when there are only small to medium volume programs, there is not likely to be a sufficient number of surgeons such that this scenario can be avoided entirely. Not Normal Circumstances In both its heart and heart/lung applications, Nemours articulated the following “not normal circumstances” in seeking approval: Florida does not have any approved pediatric heart/lung transplant programs. Florida's only two approved pediatric lung transplant programs have not performed any lung transplant programs in the last two reporting years according to AHCA reporting data. Significantly, there are no pediatric heart transplant or lung transplant programs in AHCA's Organ Transplant Service Area OTSA 3 in which NCH is located-an area of the State with one the fastest growing and youngest populations. Florida has no other pediatric comprehensive, multi-organ thoracic transplant program. Florida has no other pediatric comprehensive, multi-organ thoracic transplant program that is part of a pediatric specific integrated delivery system such as Nemours offers. NCH offers a unique, dedicated model of cardiothoracic care developed at its Alfred I. duPont Hospital for Children (AIDHC) in Wilmington, Delaware and implemented upon the opening of the program at NCH. The key and differentiating element of this Model of Care is a unified team of cardiac clinical and administrative professionals who serve children with cardiac problems in dedicated facilities (the "Cardiac Team"). The Cardiac Team only cares for children with cardiac diagnoses. As such, the Cardiac Team of anesthesiologists, surgeons, cardiologists, nurses, and other support personnel do not "float" to other hospital floors or departments as in a typical hospital setting. This dedicated model of cardiac care allows the Cardiac Team to develop highly specialized knowledge and relationships to provide the best treatment protocols for patients with cardiac conditions. NCH has developed state-of-the art facilities and innovative clinical pathways for the care of the most complex pediatric thoracic patients. NCH has and will bring new opportunities for research in pediatric cardiology, cardiac surgery, and pulmonary medicine, particularly clinical translational and basic research into the linkages between childhood obesity and cardiac conditions. Nemours operates a regional network of clinics in Florida, with primary locations in Pensacola, Jacksonville, and Orlando, that will operate in partnership with NCH for the appropriate regional referral of patients in Florida for pediatric thoracic care. NCH can reduce the out-migration of pediatric, thoracic transplant patients from OTSA 3 to other parts of the State as well as the out-migration of these patients to other out-of-state transplant programs. Similarly, NCH will reduce the outmigration of organs donated in Florida to other states ensuring that Florida recipient patients are first priority for organs donated in Florida. NCH has in place the infrastructure, facilities, and resources to seamlessly add thoracic transplant services to its existing comprehensive cardiac surgery program. Additional needed staff are already being recruited to this program. As a result, the project has minimal incremental cost that will need to be incurred. Total project costs are, therefore, estimated to be $715,425.00. In addition, according to Nemours, an additional “not normal” circumstance has emerged since the filing of the applications: the approval of Nemours’ PLT application in the absence of a PHT program at the facility, which it contends is “a very unusual situation.” Noteworthy about these purported reasons for approval are that: (1) none of them are specifically directed at a unique circumstance relating to a need for another PHT program; and (2) most of them are either a recitation of the fact that there is no existing program in the service area or are about Nemours’ capability to provide these services. They are not directed at whether there is a need for its proposed programs. In fact, the main thrust of Nemours’ case was directed at proof regarding its capabilities. But the flaw in this theme is best demonstrated in the testimony of Dawn Tucker, the last witness called by Nemours. Ms. Tucker is the cardiac program administrative director for Nemours. When asked why she supported the proposed program, she talked about the experience of the team, a desire to care for sick patients, an organization (Nemours) that financially supports the program, and the network of centers that Nemours has in Florida. These factors address why Nemours “wants” these CONs. None of them addresses the threshold issue of whether there is a “need” for these programs in OTSA 3. More specifically, the first, third, and fourth bullet points are all based on the absence of a program in OTSA 3. By rule, that is not a basis for establishing need. Fla. Admin. Code R. 59C-1.009(2)(e)3. AHCA appropriately rejected the absence of a program in OTSA 3 as the sole basis upon which need for the proposed projects could be established. The second bullet point relates to the pediatric lung transplant application that is not at issue in this matter. The fifth and sixth bullet points relate to the Nemours integrated model of care. But again, this does not address whether there is a need for the proposed programs. The fact that Nemours has an employed-physician model is not unique or “not normal.” AHCA considered the information regarding the model of care and correctly noted that the model of care does not itself enhance access or improve outcomes. It should be noted that Shands’ doctors are employed by the University of Florida. In addition, the reliance on this model does not guarantee a robust program. This bullet point references the much older and more established Alfred I. duPont Hospital for Children in Wilmington, Delaware, that is touted as the model for Nemours. Nemours presented evidence relating to its more established hospital in Delaware that also provides PHT services. However, the PHT program at duPont is a low-volume program, performing only one PHT in 2016. None of the managed- care companies that recognize Shands as a center of excellence also recognizes the duPont Hospital as such. One of the companies--Lifetrac--acknowledges duPont as a “supplemental” program, whereas Shands is one of its “select” programs. This demonstrates that simply having the financial resources of the duPont Foundation or the model of care used by that organization does not guarantee high volumes or success. The “not normal circumstance” bullet points regarding Nemours’ facilities, research, and other infrastructure similarly do not demonstrate need. Otherwise, a hospital could obtain a CON for a new program by spending the money in advance and then demanding approval based upon those expenditures. AHCA recognized that Nemours had recruited some very qualified clinicians, but correctly noted that that does not create or evidence need for the proposed programs. The remaining bullet point asserts that approval of the PHT and PHLT programs could reduce outmigration of both patients and organs. By definition, because neither of these transplant programs exists in OTSA 3, all patients leave OTSA 3 for these services. Again, that alone does not establish need, nor is it automatically a “not normal” circumstance. As discussed herein, Nemours has not demonstrated a sufficient need or an access problem that justifies approval of either application. With regard to the outmigration of organs from Florida, Nemours has argued that Florida is a net exporter of organs and that this is a “not normal” circumstance justifying approval of its application. However, organs harvested in one state are commonly used in another. There is nothing unusual or negative about that fact. Indeed, Dr. Wearden agreed that in his experience, this is a common occurrence. There is a national allocation system through UNOS and this sharing, as explained by Dr. Pietra, facilitates the best match for organs and patients. UNOS divides the country into regions for the purpose of allocation of donor organs, with Florida being one of six states in Region 3. The evidence of record did not establish that approval of the Nemours applications would result in the reduction of organs leaving Florida, or even that such would be a desirable result. Nemours also argued at hearing that approving their applications would increase the number of donor organs that are procured and transplanted in Florida. Nemours suggested that its programs would increase public awareness and implied that it would accept organs for future patients that surgeons at other programs turn down. However, these arguments are purely conjectural and are rejected. No record evidence exists which demonstrates that a Nemours program would increase the supply of organs in Florida. Indeed, Nemours presented no such relevant data or statistical evidence in its applications to demonstrate that this will occur. Finally, Nemours argues that its PHT and PHLT applications should be approved because it does not make sense for AHCA to have approved the PLT program but denied the other two applications. Nemours goes on to note that while there are hospitals in the country that do PHTs but not PLTs, there are no hospitals that do lungs but not hearts. Regardless of whether that is true, Florida law separates these three services into separate CON applications, which are reviewed independently. The wisdom of the rule is not at issue in this proceeding. Regardless of any overlap in the skill sets required to perform these procedures, approval of the pediatric lung transplant application does not determine need for pediatric heart or pediatric heart/lung programs. Nemours failed to establish that “not normal” circumstances currently exist that would warrant approval of either the PHT or PHLT programs. Nor did Nemours credibly demonstrate any other indicators of need for its proposed programs. Section 408.035(1)(c): The ability of the applicant to provide quality of care and the applicant’s record of providing quality of care. The parties stipulated that Nemours is a quality provider. However, AHCA maintains that this criterion is in dispute to the extent that center transplant volume as a result of Nemours’ approval would lead to or correlate with negative patient outcomes. Nemours failed to demonstrate that it would achieve the volumes it projected unless it takes significant volumes from other Florida providers.6/ Approval of Nemours will not create transplant patients that do not exist or are not currently able to reasonably access services. While Nemours has assembled a team of professionals with varying levels of transplant experience, it has not been demonstrated that it will achieve volume sufficient to reasonably assure quality care.7/ Section 408.035(1)(e): The extent to which the proposed services will enhance access to health care for residents of the service district. Approval of the Nemours PHT and PHLT programs would unquestionably improve geographic access to those services for the very few residents of OTSA 3 that need them. However, given the extreme rarity of pediatric heart and heart/lung transplants, approval of the Nemours programs would not result in enhanced access for a significant number of patients. Moreover, there was no credible non-hearsay evidence presented at hearing that any resident of OTSA 3 that needed PHT or PHLT services was unable to access those services at one of the existing PHT programs in Florida or, for PHLT, at a facility elsewhere. Based upon persuasive evidence at hearing, there is also clearly a positive relationship between volume and outcomes. As with any complex endeavor, practice makes perfect. In this instance, maintaining a minimum PHT case volume provides experience to the clinicians involved and helps maintain proficiency. According to the credible testimony of Dr. Pietra, programs should perform no fewer than 10 PHTs per year. “If you can stay above 10, then your program is going to be exercised at a minimum amount to keep everybody sort of at a peak performance.” The clear intent of the minimum volume requirement of 12 heart transplants per year contained in rule 59C- 1.044(6)(b)2. is to ensure a sufficient case volume to maintain the proficiency of the transplant surgeons and other clinicians involved in the surgical and post-surgical care of PHT patients. In addition, pediatric transplant programs are measured statistically based on outcomes, such as mortality and morbidity. Because of this, the loss of even one patient in a small program can be devastating to that hospital’s mortality statistics. As such, small programs may become less willing to take more complicated patients. In a perverse sort of way, adding more programs that dilute volumes may decrease, rather than increase, access because of the fear a small program might have for taking more complex patients. Adequate case volume is also important for teaching facilities, such as Shands, to benefit residents of all the OTSAs by being able to train the next generation of transplant physicians. The mothers of the two Shands patients that testified made note of the complexity of their daughters’ conditions and how their cases were used for training purposes. There was no persuasive evidence of record that approval of the Nemours applications would meaningfully and significantly enhance geographic access to transplant services in OTSA 3. The modest improvement in geographic access for the few patients that are to be served by the two programs is not significant enough to justify approval in the absence of demonstrated need. There is no evidence that approval of the Nemours applications will enhance financial access nor that patients are not currently able to access PHT or PHLT services because of payor status. Section 408.035(1)(g): The extent to which the proposal will foster competition that promotes quality and cost- effectiveness. It is clear that establishing and maintaining a transplant program is expensive. Given the limited pool of patients, the added expense of yet a sixth Florida program is not a cost-effective use of resources. This criterion also relates to the Nemours position that AHCA should approve the PHT and PHLT applications simply because the PLT application was approved, and it would not be cost-effective for Nemours unless the PHT and PHLT applications were also approved. However, each of these applications must rise or fall on its own merit. As of the hearing, Nemours had not yet implemented its PLT program. Given the absence of need for either the PHT or PHLT programs, the cost-effective solution might be for Nemours to reconsider implementation of the PLT program. 408.035(1)(i): The applicant’s past and proposed provision of health care services to Medicaid patients and the medically indigent. AHCA agreed at hearing that Nemours satisfies section 408.035(1)(i). Nonetheless, Nemours provides a very high level of Medicaid services, and projects a high-level volume related to Medicaid patients and charity care patients. As noted, approximately half of the PHTs projected by Nemours will be performed on Medicaid patients. Conformance with this criterion would mitigate toward approval had there been persuasive evidence that Medicaid and medically indigent patients are currently being denied access to PHT and PHLT services. However, no such evidence was presented.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that a final order be entered denying CON Application Nos. 10471 and 10472 filed by The Nemours Foundation, d/b/a Nemours Children’s Hospital. DONE AND ENTERED this 31st day of July, 2018, in Tallahassee, Leon County, Florida. S W. DAVID WATKINS Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 31st day of July, 2018.
Findings Of Fact The Agency For Health Care Administration ("AHCA") is the state agency responsible for the administration of certificate of need ("CON") laws in Florida. On February 5, 1993, AHCA published a need for one additional adult open heart surgery program in District 9. AHCA defines open heart surgery as a "tertiary health service" which, due to complexity, cost, and the relationship between volume and quality of care should be concentrated in a limited number of hospitals. Rule 59C-1.002(66), Florida Administrative Code. District 9 is located generally along the southeast coast of Florida and includes Palm Beach, Indian River, Martin, St. Lucie, and Okeechobee Counties. Palm Beach is the county at the southern end of District 9. The parties have referred to the counties other than Palm Beach, as the four northern counties. Martin County is north of Palm Beach, and St. Lucie, Okeechobee, and Indian River are further north. The applicants in this proceeding, seeking to establish an additional District 9 adult open heart surgery program, are Lawnwood Medical Center, Inc., d/b/a Lawnwood Regional Medical Center, Inc. ("Lawnwood"), St. Mary's Hospital, Inc. ("St. Mary's"), and Martin Memorial Medical Center, Inc. ("Martin Memorial"). Lawnwood Regional Medical Center Lawnwood is a 335-bed for-profit hospital located in Ft. Pierce, in St. Lucie County. Lawnwood has CON approval for the construction of an additional 18 skilled nursing beds and 10 level II NICU beds. In addition to the 335 licensed beds, Lawnwood has 16 unlicensed bassinets for a total of 351. Lawnwood's 335 licensed beds include 60 psychiatric beds, located one and a half blocks away from the main Lawnwood building, at a facility called Harbor Shores. Lawnwood has 260 general acute care beds. When Lawnwood filed its application, its parent corporation was HCA, Inc., a subsidiary of the Hospital Corporation of America. HCA was also the parent corporation of the Medical Center of Port St. Lucie, the only other hospital in St. Lucie County, and of Raulerson Hospital in Okeechobee County. After the application was filed and prior to hearing, a subsidiary of Columbia Health Care Corporation merged with HCA. As a result of the merger, the administrator of Lawnwood also serves as the market manager assigned to coordinate the services offered at the three hospitals. Lawnwood is classified by the State as a disproportionate share provider of Medicaid-reimbursed services for financially needy patients. In 1993, 21 percent of its total patient days were attributable to Medicaid and 4 percent to charity. Lawnwood operates an outpatient cardiac catheterization ("cath") laboratory and, in 1992, received CON approval to perform inpatient cardiac caths in a lab which was scheduled to open in October 1994. The outpatient lab opened in 1988 at Lawnwood. In 1989, 561 cardiac cath lab procedures were performed at Lawnwood, 494 in 1990, 362 in 1991, and 468 procedures in 1993. Although 602 procedures were reported to the local health council in 1993, these were performed on 468 patients, which is the number consistent with reporting methods of other cath labs. As a result of the diagnostic caths, 45 patients were referred for open heart surgery, and 98 for angioplasties. Of the 45 patients referred for open heart surgery, 26 were actually scheduled for the procedure. Lawnwood proposes to establish an adult open heart surgery program for a total project cost of $4.99 million. The project includes construction of two dedicated operating rooms, renovations to provide a 4-bed dedicated recovery room, and conversion of 12 acute care beds to construct a 12-bed cardiovascular intensive care unit ("CVICU"). St. Mary's Hospital St. Mary's is a 430-bed not-for-profit hospital, which has been operated 55 years by the Franciscan Sisters, currently through a parent organization called the Allegheny Health System. St. Mary's is the largest hospital in District 9, and the largest provider of womens' and childrens' medical services in the district. St. Mary's is a designated regional perinatal intensive care center with level II and III neonatal intensive care units, and is the designated level II trauma center for the northern area of Palm Beach County. Like Lawnwood, St. Mary's is recognized by the State as a disproportionate share provider of services to Medicaid reimbursed and indigent patients. It is approximately sixth in the state in the provision of services to financially needy patients. St. Mary's cardiac cath lab began operation in February 1988. There were 267 inpatient and 116 outpatient cardiac caths at St. Mary's lab in 1991, 240 and 118 respectively in 1992, and 171 and 115 respectively from January to November 1993. St. Mary's operates a 10-bed coronary care unit. St. Mary's proposes to establish an adult open heart surgery program for a total of $2,166,351, funded by private donors. The project will include renovations to two existing operating rooms and to a recovery room area. Martin Memorial Medical Center Martin Memorial is a 336-bed not-for-profit acute care hospital, with an additional 17 nursery/bassinets which are not required to be in the total licensed beds. The ultimate parent corporation for the Martin Memorial facilities and its foundation is Martin Memorial Health Systems, a not-for- profit corporation with a volunteer community board of directors. Martin Memorial's beds are divided between two campuses, with 236 beds in Stuart, and 100 in Port Salerno. The Port Salerno hospital opened in September, 1992 and is approximately 8 miles south of Stuart. Included in the 236 beds at Martin Memorial in Stuart are 5 level II neonatal intensive care beds, 23 intensive care unit beds, 45 ventilator, telemetry or other monitored beds, and 134 medical/surgical beds. Martin Memorial's existing cardiac services include a cardiac cath lab which opened in 1989 and, that year, reported 250 procedures. Caths at Martin reached the highest volume, 905 in 1991, followed by 799 in 1992, and 867 in 1993. Martin Memorial proposes to establish an adult open heart surgery program in Stuart for a total project cost of $3,594,720. Martin's project includes a newly constructed open heart surgery suite adjacent to the cardiac cath lab and, as a back-up, renovation of an existing operating room. As a part of an approved, separate CON application, Martin proposes to renovate and expand to accommodate a 13-bed surgical intensive care unit ("SICU") with four private rooms dedicated as a cardiovascular intensive care unit ("CVICU"). The expenses associated with the four CVICU rooms are included in the total open heart surgery project costs. Existing Open Heart Surgery Providers In Or Adjacent To District 9 All of the existing adult open heart surgery programs in District 9 are in Palm Beach County, at Delray Community Hospital ("Delray"), JFK Medical Center, Inc. ("JFK"), and AMI Palm Beach Gardens Community Hospital, Inc. d/b/a Palm Beach Gardens Medical Center ("Palm Beach Gardens"). The same services are also available in the adjacent districts to the north in District 7 at Holmes Regional Medical Center in Brevard County, and to the south in District 10 at AMI North Ridge General Hospital in Broward County. In addition, established referral patterns exist from District 9 to Miami Heart Institute in Dade County and Holy Cross Hospital in Broward County. All residents of District 9 have access to open heart surgery within two hours average drive time, which exceeds the geographic access standard of Rule 59C-1.033(4)(a), Florida Administrative Code. Delray is located in southern Palm Beach County and is a level II trauma center for that area. JFK is a 369-bed not-for-profit hospital located in Atlantis, Florida, approximately midway between Boca Raton and West Palm Beach, in north central Palm Beach County. The corporation which owns and operates JFK, also is the parent of a fund-raising foundation, and other subsidiaries, some of which are for-profit corporations. JFK has had an open heart surgery program since 1987. JFK's two operating rooms are equipped and sized identically, and located in close proximity to the two room cardiac cath lab and the intensive care unit. JFK has the capacity to perform up to 1000 cases annually, while actual annual volumes at JFK have ranged from 350 to 370 cases. Palm Beach Gardens is a 204-bed for-profit hospital located in the northern part of Palm Beach County. It operates the oldest open heart surgery program in the district, having started in 1982 or 1983. In fiscal year 1992- 1993, there were 477 open heart surgery patients at Palm Beach Gardens, of which 173 resided in the four northern counties of the District. Palm Beach Gardens has 11 operating suites, 7 capable of being used for open heart surgeries, and 4 dedicated solely to open heart surgeries. The current capacity of Palm Beach Gardens is 900 open heart procedures a year. By adding staff, Palm Beach Gardens could reach a volume of 1100 cases a year. While Palm Beach Gardens has excess capacity in its operating rooms, at the peak of its seasonal demand, delays occur in scheduling non-emergency surgeries due to inadequate capacity in its 24-bed intensive care unit. Occupancy levels in the 24 beds were 112.5 percent in 1993, according to Treasure Coast Health Council data. Although Palm Beach Gardens also suggested that an 8-bed overflow unit supplemented the 24 beds, accounting reports do not reflect billings for their use as intensive care services. Comparison of Applicants and Applications Subsection 408.035(1)(a) -- need in relation to state and local plans The 1989 state health plan, Healthy Floridians, includes six preferences for the review of open heart surgery applications. The first preference favors applicants establishing programs in counties with a population over 100,000 and a higher percentage than statewide average of 18.8 percent elderly persons. All the experts in health planning testified that the term "elderly" in this preference means persons 65 years of age and older, which is consistent with the age group with the greatest demand for open heart surgery. St. Mary, Lawnwood, and Martin meet the preference. The 1993 population of Palm Beach County was 900,000, St. Lucie's was 162, 598, and Martin's was 108,089. The population age 65 and over as a percentage of total population was 24 percent in Palm Beach, 21.2 percent in Lawnwood, and 27.5 percent in Martin County. The second state preference is for applicants who can demonstrate the ability to perform at least 350 annual procedures within 3 years of initiating an open heart program. Lawnwood reasonably projected a total of 314 open heart surgery procedures in year one, 350 in year two, and 386 in year three. Lawnwood's utilization projections are conservatively based on the assumption that, by the third year, 70 percent of its open heart patients will come from St. Lucie and Okeechobee Counties, which are already in its primary service area. Martin Memorial's expert questioned Lawnwood's projected open heart volumes from Martin and Indian River Counties, based on its acute care and cath lab patient origins. In addition, traditional referral patterns show Indian River patients going north to Brevard and Orange Counties, while Martin County patients go south to Palm Beach, Broward, and Dade Counties. Considering the acute care and cath lab competition within the four northern counties, the absence in that area of any competition for an open heart surgery program, the relative success of Lawnwood's outpatient cath lab despite its limitations and competition, and its affiliation with Port St. Lucie and Raulerson hospitals, Lawnwood established the reasonableness of its projected utilization. Lawnwood also reasonably expects to reverse some of the 73.5 percent out-migration for open heart surgery by residents of the northern four counties. See, Findings of Fact 27, infra. Martin Memorial's projections of 249 cases in year one, 317 in year two, and over 350 in year three are also reasonable. Martin Memorial's underlying assumptions, that its open heart surgery market share will at least equal that of its acute care, that it will keep some patients previously referred from its cath lab, and that, it, like Lawnwood, would reverse some district out-migration, are also reasonable. Martin Memorial referred 172 patients from its cath lab for open heart surgery in 1993, in contrast to 45 from St. Mary's, and 41 from Lawnwood. Martin Memorial's projections are based on 1991-1992 use rates which declined in 1993. Despite the one year decline and some expert predictions of a continuing downward trend in use rates, Martin Memorial's projections are bolstered by the fact that its open heart surgery primary service area includes Port St. Lucie, which contains 40 percent of the population of St. Lucie County and is the fastest growing area of District 9. That area, which is closer to Stuart, but is located in the St. Lucie County community in which Lawnwood has an affiliate hospital, supports both the projections of Lawnwood and Martin Memorial, and could be served by an open heart surgery program at either facility. Although Martin Memorial's projected volumes are higher than and inconsistent with other projections made by Martin Memorial, the reasonableness of the projections was established. St. Mary's projected 171 open heart surgeries in year one, 265 in year two, and 363 in year three. The projections are based on the use of a gravity model designed to determine potential volume "attracted" to the program by using the size of the hospital and the proximity of patients as factors. The model used a zip code level analysis to take into consideration the fact that St. Mary's expects a sub-county primary service area, as a result of sharing the county with the three existing District 9 providers. The projected utilization was reduced, by St. Mary's expert, to take into consideration an expected start- up factor. There is, however, substantial expert testimony that the variables and/or the weight attributed to each variable included in this gravity model are inadequate to explain actual or potential volumes. There is substantial evidence that the size of a hospital is not reliable enough to be one of only two variables in a model. For example, JFK although larger than Palm Beach Gardens, only exceeded 350 cases in 1991-1992 by 16, when smaller Palm Beach Gardens with an older open heart surgery program reached 499 cases. The model also fails to consider actual physician referral patterns. St. Mary's projections and its ability to exceed 350 cases also depend on its ability to attract Medicaid patients over and above the patients projected by the gravity model. See, Findings of Fact 35, infra. The volume of diagnostic cardiac caths at St. Mary's is low and has declined over the past three years. In part, the volume is low because there is no open heart surgery back-up available in the event the diagnostic cardiac cath indicates that need. Cath patients suspected of needing more invasive procedures are diverted by referring physicians to hospitals with angioplasty and open heart programs. But that explanation of St. Mary's volumes apparently is incomplete, since, by contrast Boca Raton Community Hospital and Martin Memorial, which also have no open heart surgery back-up, have had more steadily increasing cardiac cath volumes. The fact that St. Mary's cath volumes are low and its open heart surgery projections unreliable is also attributable to the fact that St. Mary's is located 11 miles north of JFK and 5 1/2 miles south of Palm Beach Gardens, therefore, at a competitive disadvantage with these established programs. The third state health plan preference applies to proposals, for improving access for persons currently leaving the district. With almost half of Palm Beach County open heart surgery patients receiving the service outside the county, St. Mary's claims to be in the best location to reverse that trend if geographical access is the problem. St. Mary's also points to the convenience of access to its hospital, which is 2 miles from Interstate 95, the main north-south transportation corridor through the district. Approval of St. Mary's proposal will not, however, reverse out-migration to the extent that it is attributable to factors such as seasonal residency, established physician referral practices from northern areas of District 9 to providers in adjacent districts, and managed care contractual arrangements. Lawnwood is located in the largest, fastest growing, and most centrally located county of the northern four counties. St. Lucie County is adjacent to each of the other three northern counties, with Martin to the south, Okeechobee to the west, and Indian River to the north. The level of "out-migration," defined as those patients leaving the district to receive the service, increases dramatically from south to north in District 9, from 55 percent in Martin, 70 percent in St. Lucie, 80 percent in Okeechobee, to 100 percent in Indian River County. Considering growth in western St. Lucie County, the needs of St. Lucie and Okeechobee County residents, and the alternative to out-migration provided for both Indian River and Martin County residents, the Lawnwood location is superior to that of Martin Memorial in terms of the ability to improve access to the service. See, also Findings of Fact 23-24, supra. The fourth state preference for applicants with a history of providing disproportionate share Medicaid and charity care favors the applications of St. Mary's and Lawnwood, in that order. Martin Memorial argues that it also meets the disproportionate share criteria, which the preference requires, although it has not been designated by the State, which the preference does not require. Relying on the criteria in subsection 409.911(2), Florida Statutes, Martin claims to meet or exceed the disproportionate share requirements for 1990, despite the agency's reliance on 1989 data. Assuming, arguendo, that Martin is entitled to the preference, the comparative ranking of St. Mary's first, Lawnwood second, and Martin third remains the same. In addition, the preference looks at a history of disproportionate service, as does subsection 408.035(1)(n), in part, which Martin failed to establish. For 1991, St. Mary's provided 15.8 percent of total District 9 Medicaid, Lawnwood provided 11.7 percent, and Martin Memorial, 1.7 percent. Martin Memorial established that it treated a larger number of Medicaid patients with circulatory diseases as a proportion of Medicaid patients in Martin County, as compared to St. Lucie County residents treated at Lawnwood. However, the absolute number of circulatory disease Medicaid patients treated at Lawnwood was approximately two and half times the number treated at Martin Memorial. Statistical indicators, including per capita income and low income patients diagnosed with circulatory diseases, demonstrate that residents of St. Lucie and Okeechobee Counties are less affluent, and more medically needy than those in Palm Beach and Martin Counties. The fifth state preference favors the applicant offering a service with the highest quality of care at the least expense. The preference includes an explanation that larger facilities usually have more available resources to meet the preference. As the largest hospital with the lowest cost per case by the second year of the program, $22,659, St. Mary's best meets the preference. Martin's projected cost is $26,909 and Lawnwood's is $27,085. Martin Memorial's expert calculated total expenses per case at $23,221 for Martin Memorial, $22,615 for St. Mary's, and $23,645 for Lawnwood. St. Mary's projected charges of $50,600 in year one and $53,100 in year two. Lawnwood projected charges of $55,199 in year one, and $58,133 in year two. Martin Memorial projected charges of $55,594, in year one, $58,955 in year two. Total project costs were estimated at $2,166,351 for St. Mary's, $3,594,720 for Martin Memorial, and $4,995,039 for Lawnwood. Using either set of cost data or the projected charges, St. Mary's best meets this preference based on size, the lowest total project costs, and the lowest projected charges for open heart surgery services. Martin Memorial and Lawnwood have, as described by one expert, remarkably similar costs, and the same is true of projected average charges per case. The final state preference favors applicants who will include protocols for the use of innovative therapeutic alternatives to surgery for appropriate patients, including streptokinase and tissue plaminogen activator therapies. Lawnwood and Martin Memorial currently use streptokinase. St. Mary's performs emergency angioplasties, and uses streptokinase therapy. All three applicants meet the preference for providing and/or planning to provide alternative therapies to open heart surgery. The first District 9 local health plan allocation factor gives a priority for established cardiac cath programs. Based on expert testimony, a cardiac cath program exceeding 150 annual procedures is established. All the applicants exceed the minimum volume and, therefore, comply with the allocation factor. Martin Memorial has the highest volume in an operational inpatient and outpatient lab and, meets the allocation factor better than Lawnwood and St. Mary's. The other District 9 factor favors applicants with a documented commitment to provide services regardless of patient's ability to pay. Lawnwood projects 2.51 percent Medicaid and 1.5 percent charity care in year two. St. Mary's projects providing 5 percent Medicaid and 3.5 percent charity care in year two. Martin Memorial projects 2 percent Medicaid and 1.9 percent charity care in year two. St. Mary's best meets the factor, followed by Lawnwood, and then Martin Memorial. More Medicaid residents live in the primary service area of Lawnwood than that of Martin Memorial. Martin has filed CON compliance reports demonstrating difficulty in meeting prior CON Medicaid conditions due to the demographics of its service area. Subsections 408.035(1)(b) - availability, quality of care, efficiency, accessibility, extent of utilization of like and existing programs; 408.035(2)(b) - appropriate and efficient use of existing inpatient facilities; and 408.035(2)(d) - serious problems in obtaining care without proposed new program(s). With the exception of seasonal excess demand for Palm Beach Gardens' ICU beds, the evidence demonstrates there is excess capacity in existing District 9 providers. Geographic access to existing providers in or adjacent to the district is also reasonable. The quality of care at existing providers is excellent. St. Mary's asserts that its proposal will best assist in alleviating access barriers to open heart surgery for low income persons with limited geographic mobility. One expert estimated that 38 District 9 Medicaid patients needed, but did not receive, open heart surgeries in 1991, based on the use rates for commercially insured patients. In general, the highest density of population with a demand for invasive heart therapies and open heart surgeries is concentrated in southern and central Palm Beach County. However, expert testimony established that Medicaid patients are underserved for reasons, other than the policies of the existing providers. The evidence does not show that St. Mary's proposal can overcome these financial barriers. St. Mary's is a level II trauma center, and maintains that trauma patients in need of open heart surgery are at risk of death from having to wait for transfers. Transfers of patients from St. Mary's to Palm Beach Gardens or JFK for open heart surgery take from three hours to three days, averaging 8 to 12 hours, in approximately 30 percent of the cases. From May 1991 through January 1994, over 2600 trauma patients were treated at St. Mary's. Expert testimony, after review of medical records, indicates that from one to six patients needed open heart surgery, an insufficient number to constitute a not normal circumstance for the establishment of an open heart program at St. Mary's. Palm Beach Gardens' position that an additional adult open heart surgery program is not needed in District 9 is rejected. Open heart surgery use rates are not increasing nationally or in Florida. However, District 9 population is increasing, as is open heart surgery utilization for District 9 as a whole, and for Palm Beach, St. Lucie and Okeechobee Counites, while remaining static in Martin County and decreasing in Indian River. Palm Beach Gardens and JFK have demonstrated that in Palm Beach County, an additional open heart surgery program is not needed, and would be detrimental to existing programs. See, Findings of Fact 51-52. Subsection 408.035(1)(c) - quality of care The applicants, like the existing providers, are accredited by the Joint Commission on Accreditation of Healthcare Organizations. All of the applicants provide excellent quality care, as indicated by their accreditations and proposals, compromised only by their ability to achieve the projected volumes. See, Findings of Fact 23-26. Subsection 408.035(1)(d) - alternatives or outpatient facilities and 408.035(2)(a) - alternatives to inpatient services There are no alternatives or facilities other than acute care hospitals in which open heart surgeries can be performed. The criterion is inapplicable to this case. Subsections 408.035(1)(e) - economies of joint or shared facilities and 408.035(2(k) - modernization or sharing arrangements as alternatives to new construction. Martin Memorial is a part of a network of hospitals planning a more formalized affiliation to attract managed care contracts. Lawnwood is a part of a large corporate group, which can offer experience in establishing an open heart surgery program. Neither of these arrangements entitles the applicants to special consideration under the statutory criterion, as it has been construed by AHCA. In this case, each applicant is a separate acute care hospital. An alternative arrangement for a shared program was considered by Martin Memorial, but there is no showing that any proposal which improves access for the northern four counties could avoid the necessity for new construction. Subsection 408.035(1)(f) - needs for equipment and services not accessible in adjoining areas There is no evidence that any applicant proposes to provide a service not readily available in adjoining areas. On the contrary, each applicant proposes to offer an alternative within the district for residents who currently use providers in adjoining areas. See, Finding of Fact 27. Subsection 408.035(1)(g) - need for research and educational programs There is no evidence that any of the applicants will meet research or educational needs, or is a teaching hospital. AHCA has strictly construed the statutory criterion to apply to teaching hospitals. Subsection 408.035(1)(h) - availability of resources, including staff, management, and funds for capital and operating expenditures, including personnel required in Rule 59C-1.033(5)(b). The Cleveland Clinic has expressed an interest in providing surgeons for Martin Memorial's program, but no agreement has been formalized. Martin Memorial was criticized for not having a full-time infectious disease specialist, inadequate pulmonary and nephrology specialists, and for being unable to perform transesophageal echocardiology, all of which are necessary to support an open heart surgery program. St. Mary's was criticized for not planning to have nurses assigned exclusively to its open heart surgery team. Lawnwood has been unable to attract full-time coverage in thoracic, orthopedic, and neurosurgery. Despite these specific criticisms, each applicant has successful recruitment mechanisms and affiliations which will be enhanced by the presence of an open heart surgery program. The applicants' staffing and equipment proposals are reasonable. Both St. Mary's and Lawnwood are subsidiaries of larger organizations which include hospitals with open heart surgery programs. Subsection 408.035(1)(i) - immediate and long term financial feasibility St. Mary's has the ability to establish an adult open heart surgery program for a total of $2,166,351, funded by private donors. St. Mary's provided a pro forma of expected revenues and expenses to establish financial feasibility based on two factors which were challenged, the average length of stay ("ALOS") and the mix of payer classifications for patients. St. Mary's projected 10.3 days as the ALOS. JFK's experts suggested that a 13-day ALOS is more reasonable, particularly for a new program. JFK's actual experience was an ALOS of 16.1 days in 1988, 14.5 days in 1992, and 12.6 days by the year ending June 1993. Mature programs generally have lower ALOS than newer ones. Currently, ALOS in the District are 10.9 for Palm Beach Gardens, 12.5 for Delray, and 14.5 for JFK. JFK's assertion that St. Mary's initial ALOS will more likely be 13 days not 10.3 is reasonable. The fact that the ALOS will be longer than that projected in the pro forma means that expenses for the care of each patient will be greater, while revenues will not increase proportionately. Revenues are limited in fixed Diagnostic Related Group ("DRG") reimbursement categories, such as Medicare and managed care, which are the dominant payer groups, in contrast to the more flexible per diem reimbursement of commercial insurers. St. Mary's failed to include revenues and expenses for the construction period, anticipating only capital expenditures and start-up costs for implementing a new service. St. Mary's pro forma was based on a first year payer mix which includes 12.4 percent managed care and 11.6 percent commercial insurance in 1995. At JFK, the open heart surgery payor mix was 33 percent managed care and 9 percent commercial in 1993. St. Mary's underestimated the proportion of patients in the DRG-based managed care category, as compared to the per diem arrangements typical of commercial insurance. Taking into consideration increased expenses of $251,000 in year one and $409,000 in year two, due to adjustments from 10.3 to 13 days in the ALOS, and reduced revenues of $350,000 in year two, St. Mary's proposal is not financially feasible. The conclusion is also compelled by St. Mary's failure to establish the reasonableness of its utilization projections for the program. See, Finding of Fact 25. Martin Memorial has the funds necessary to establish an open heart surgery program for $3,594,720. Its pro forma shows revenues and expenses for the construction period, which are identical with or without the open heart surgery program. Martin Memorial's pro forma is flawed by double counting revenues from patients currently spending some time and revenues at Martin Memorial prior to transfers for open heart surgery. Revenues associated with pre-transfer stays must be deducted from revenues for open heart surgeries of average total lengths of stay. The amounts of over-stated revenues were not calculated by Palm Beach Gardens expert, and other criticisms of Martin Memorial's pro forma are rejected. Lawnwood, like St. Mary's, failed to include any construction period revenues and expenses in its pro forma. Lawnwood, as a separate legal entity, does not have the funds to establish its open heart surgery program, without relying on its parent, Hospital Corporation of America. The commitment of funds, represented by a letter dated April 30, 1993, indicated the source as either internally generated cash or available lines of credit. Lawnwood demonstrated its financial feasibility, in part, by showing that its open heart program's break-even point, at which expenses and revenues would be equal is 182 cases, well below projected utilization. See, Findings of Fact 23. Subsection 408.035(1)(j) - special needs and circumstances of health maintenance organizations The applicants do not propose to provide any different or special services for health maintenance organizations, nor is any applicant in this batch itself a health maintenance organization, as required by AHCA's interpretation to the statutory criterion. NME Hospitals, Inc., d/b/a West Boca Medical Center v. HRS, DOAH Case Nos. 90-7037 and 91-1533 (F.O. 4/8/92). Subsection 408.035(1)(k) - substantial, specialty services to non-residents of the service district Although the applicants propose to provide open heart surgery, which is one of the specialty services listed in the statute, they do not project that they will serve residents of other districts. The applications are not distinguishable on the basis of Subsection 408.035(1)(k), Florida Statutes. Subsection 408.035(1)(l) - impact on costs and effects of competition with existing providers. If St. Mary's proposal is approved and, as St. Mary's projects, two- thirds of its patients come from existing district providers, the program at JFK will be adversely affected. As the result of JFK's loss of approximately 106 cases, its net income could also be reduced up to $2.6 million. By contrast, programs at Lawnwood or Martin Memorial would have a negligible impact on JFK. The existing program at Palm Beach Gardens would suffer an adverse impact from the approval of programs at either St. Mary's or Martin Memorial. The adverse impact of a program at Martin Memorial is greater. Palm Beach Gardens could lose from 128 to 142 cases in the first year and from 179 to 198 cases in the third year in the worst case scenarios, depending on whether the use rate declines or remains constant. In addition, the further development of the VHA Network proposed by some District 9 hospitals, including Martin Memorial, as a means to attract managed care contracts, would enhance referrals to an open heart surgery program at Martin Memorial. Reasonable estimates of the financial loss to Palm Beach Gardens range between $2.8 and $3.1 million, although Palm Beach Gardens, with $9 million in annual income, would still be profitable. While the numeric calculations required in Rule 59C-1.033(7)(c), Florida Administrative Code, indicate that there will be enough total open heart surgeries to allow each of the existing providers to continue to exceed 350 operations, Palm Beach Gardens would be disproportionately, adversely affected by a program at Martin Memorial, as would JFK by a program at a St. Mary's. As the lowest volume provider, JFK is also at greater risk of dropping below the 350 minimum level established as indicative of the quality of care. Subsection 408.035(1)(m) - costs and methods of construction With total project costs of $4.99 million, Lawnwood's proposal to construct two new, dedicated operating rooms is the most expensive. Martin Memorial's cost of $3.59 million includes new construction of one and renovation of another operating room. St. Mary's low project cost of $2.16 reflects the fact that renovations rather than new construction is planned. The advantages of new construction, however, are that the size of the operating rooms will exceed general state requirements, and comply with recommendations developed specifically for open heart surgery. See, Findings of Fact 58, infra. Subsection 408.035(1)(n) - past and proposed service to Medicaid and medically indigent patients Based on history and proposed service, the applicants rank, in order, St. Mary's, Lawnwood, and Martin Memorial in complying with the criterion. See, Findings of Fact 28 and 32, supra. Subsection 408.035(1)(o) - continuum of care in multilevel system, including acute, skilled nursing, and home health care The applicants failed to distinguish their proposals on the basis of this statutory criterion. Other Criticisms of the Applications St. Mary's has a 16-bed intensive care unit, 4 of those beds will require no additional equipment to be used to provide post-operative care for open heart surgery patients. The 4 beds are located adjacent to the intended open heart surgery operating suite. The proposed 4-bed ICU was criticized for being too crowded, and inadequately designed to allow adequate patient observation and monitoring, and for not being dedicated solely to open heart surgery patients. The 16-bed unit has experienced over 90 percent occupancy rates, but some of those patients have required the staffing, but not the equipment available in the intensive care unit. St. Mary's acknowledged potential capacity problems, but has the ability to create additional step-down unit beds to relieve the ICU unit, when necessary. In addition, outpatient surgeries were scheduled to be performed in a separate facility beginning in July 1994. While some clinicians may prefer a separate ICU, there was no evidence of any requirements that open heart surgery patients receive post-operative care in a separate ICU, nor that the lack of a specialized unit means a lack of staff capable of caring for such patients. St. Mary's project involves the renovation of a total of 1731 square feet, 764 net square feet of that in the main operating room on the first floor. The back-up operating room at St. Mary's is 480 square feet, below the American College of Cardiologists' recommendation and 1992 Federal Guidelines of a minimum of 600 and up to 800 square feet. Despite the term "back-up," expert testimony established the need for regular use of both operating rooms, one for regularly scheduled procedures and one for emergencies which occur within the cardiac cath lab or the post-operative intensive care unit. The size of St. Mary's back-up operating room meets state requirements for operating rooms, which do not differentiate on the basis of the type of surgery. St. Mary's also demonstrated that open heart surgeries are performed in comparably sized or smaller operating rooms at JFK. The space allocated to Lawnwood's 4-bed open heart surgery recovery room was criticized as inadequate to accommodate the equipment and personnel required to monitor and, if necessary, to revive post-operative patients. The space allocated complies with state licensure requirements. Reconfiguration of the beds and equipment in the space is permissible, if necessary, in final construction documents which must be approved by AHCA. Lawnwood's proposal was also criticized because the CVICU will be located three stories above the surgical area and recovery rooms. There was no evidence that the location of the CVICU violated licensure requirements or compromised the quality of care. The use of restricted elevator access between the surgical/recovery area and the CVICU is reasonable. AHCA favored the applications of both Lawnwood and Martin over that of St. Mary's due to their locations outside Palm Beach County. Having been told by staff that it was then a "toss up" between the two, AHCA's Division Director selected Martin Memorial. The Division Director, Dr. James Howell, is a former Deputy District Administrator for AHCA District 9 and former County Health Director for Palm Beach County. In explaining his decision, Dr. Howell testified as follows: Q. Ultimately, sir, you recommended to Ms. Dudek that Martin be approved rather than Lawnwood; isn't that correct, sir? A. Yes, sir. In our mutual discussions we had a discussion about two. To be straightforward, the reason that I'd recommended Martin was that Martin is a long-term community hospital with local community responsiveness or local community board of directors, as far as I know, and that AHCA owned - now I believe, it's part of the Columbia system, was in St. Lucie County and was a newer hospital, and that, you know, I felt more comfortable with giving the first CON in the area to a group that had a long heritage and commitment to the area, even though I can tell you I can't say anything negative about AHCA in dealings with them. Q. Or Columbia? A. Or Columbia; right. I can't say anything. That's not meant to be prejudicial with them. They did a good job with us, with maternity/child health. Q. You did approach this batch, did you not, sir, with a bias towards Martin Memorial because you knew the institution had been there a long, long time and was a very stable institution; isn't that correct? A. That is quite correct, yes, sir. See, Transcript, p. 251. The court reporter's references to "AHCA" are corrected and understood, in this context, to refer to HCA or Hospital Corporation of America. The statutory and rule criteria, on balance, demonstrate that open heart surgery programs at Martin Memorial or Lawnwood are more likely to improve access, to meet projected volumes, and to be financially feasible. Of these two, however, Lawnwood is better situated to reverse district out-migration, and has to be preferred, under the state and local health plans and subsection 408.035(1)(n), Florida Statutes, for its history of providing a disproportionate share of its services to Medicaid and charity patients. Finally, the most significant distinction between the applicants is that the quality of care at existing providers, as measured by their volumes of open heart surgeries, will not be adversely affected by the approval of a new program at Lawnwood. Application Content AHCA accepted Martin Memorial's application, although two different letters of intent for mutually exclusive open heart surgery programs were filed simultaneously by Martin Memorial, one for a program shared with Indian River Memorial, and one for a separate program. Martin Memorial's application also, arguably exceeds the scope of its Board approval by including renovation of a portion of the surgical intensive care unit ("SICU"). AHCA accepted Martin Memorial's proposal to allocate the cost of 4 of 13 SICU beds to the open heart surgery project. As a practical matter, Martin Memorial's witnesses concede, the 4 beds cannot be constructed independently. The Board separately authorized the filing of an expedited CON for the SICU construction and renovations. In an Additional Motion For Summary Recommended Order Palm Beach Gardens' submitted correspondence between AHCA and Martin Memorial attempting to establish that the separate SICU CON has expired. AHCA accepted Lawnwood's application without a construction period pro forma, and without identification of the ultimate parent corporation of the subsidiary, Lawnwood Medical Center, Inc.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Agency For Health Care Administration enter a Final Order issuing Certificate of Need 7245 to Lawnwood Medical Center, Inc., denying Certificate of Need 7244 to St. Mary's Hospital, Inc., and denying Certificate of Need 7243 to Martin Memorial Medical Center, Inc. DONE AND ENTERED this 13th day of March, 1995 in Tallahassee, Leon County, Florida. ELEANOR M. HUNTER Hearing Officer Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-1550 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 13th day of March, 1995. APPENDIX TO RECOMMENDED ORDER, CASE NO. 93-4908 To comply with the requirements of Section 120.59(2), Fla. Stat. (1991), the following rulings are made on the parties' proposed findings of fact: Petitioner, Lawnwood's Proposed Findings of Fact. Accepted in Findings of Fact 2. Accepted in Findings of Fact 3. Accepted in Findings of Fact 16. Accepted in Findings of Fact 4. Accepted in Findings of Fact 5. 6-12. Accepted in or subordinate to Findings of Fact 27. 13. Accepted in Findings of Fact 13. 14-20. Accepted in or subordinate to Findings of Fact 27. Accepted in or subordinate to Findings of Fact 16 and 19. Accepted in Findings of Fact 24. 23-39. Accepted in or subordinate to Findings of Fact 27. 40-47. Accepted in or subordinate to Findings of Fact 34. 48-61. Accepted in or subordinate to Findings of Fact 28. 62-64. Accepted in relative terms or subordinate to Findings of Fact 27. 65-70. Accepted in or subordinate to Findings of Fact 28. 71-73. Accepted in part or subordinate to Findings of Fact 24 and 28. 74-86. Accepted in part or subordinate to Findings of Fact 2, 23 and 27. 87. Issue not reached. 88-94. Accepted in or subordinate to Findings of Fact 23. 95. Rejected in part and accepted in part in Findings of Fact 24. 96-100. Accepted in Findings of Fact 23. 101-105. Accepted in general in Findings of Fact 24. 106-111. Accepted in Findings of Fact 47. Accepted in general in Findings of Fact 22-29. Accepted in Findings of Fact 22. Accepted in Findings of Fact 23. Accepted in Findings of Fact 27. Accepted in Findings of Fact 28. Accepted in relevant part in Findings of Fact 28. Accepted in Findings of Fact 29. Accepted in Findings of Fact 30. 120-122. Accepted in or subordinate to Findings of Fact 31. 123-131. Accepted in or subordinate to Findings of Fact 5, 43, and 48. 132-133. Accepted in or subordinate to Findings of Fact 43. 134-146. Accepted in or subordinate to Findings of Fact 48. Accepted in or subordinate to Findings of Fact 45, 48 and conclusions of law 66. Accepted in or subordinate to Findings of Fact 48. 139-141. Accepted in or subordinate to Findings of Fact 29. 142-147. Accepted in or subordinate to Findings of Fact 48. 148-152. Accepted in Findings of Fact 8 and 53. Accepted in Findings of Fact 43. Subordinate to Finding of Fact 53. 155-164. Accepted in or subordinate to Findings of Fact 59. 165-173. Accepted in or subordinate to Findings of Fact 43. 174. Accepted in or subordinate to Findings of Fact 38 and 43. 175-176. Accepted in Findings of Fact 7. Accepted in or subordinate to Findings of Fact 38 and 43. Accepted in or subordinate to Findings of Fact 53. Accepted in or subordinate to Findings of Fact 43. 180-181. Accepted in Findings of Fact 38. 182-187. Accepted in Findings of Fact 61. 188. Rejected in Findings of Fact 61. Petitioner, Palm Beach Gardens' Proposed Findings of Fact. 1-3. Accepted in Findings of Fact 16-19. Accepted in Findings of Fact 5 and 8. Accepted in Findings of Fact 13 and 15. Accepted in Findings of Fact 9 and 12. Accepted in preliminary statement. Accepted in Findings of Fact 3. Accepted in Findings of Fact 3 and 16. 10-15. Accepted in or subordinate to Findings of Fact 61. Rejected in conclusions of law 69. Rejected in Findings of Fact 53. Rejected in Findings of Fact 2 and 27. 19-25. Accepted in or subordinate to Findings of Fact 16 and 33. 26. Accepted in or subordinate to Findings of Fact 34. 27-44. Accepted in or subordinate to Findings of Fact 52. 45-48. Accepted in or subordinate to Findings of Fact 18-19 and 27-28. 49-52. Accepted in or subordinate to Findings of Fact 51 and 52. 53. Accepted in general in Findings of Fact 27. 54-55. Accepted in or subordinate to Findings of Fact 33. 56-60. Accepted in or subordinate to Findings of Fact 34. 61. Rejected "substantially" in Findings of Fact 52. 62-72. Accepted in or subordinate to Findings of Fact 16, 27, and 33. 73-76. Accepted in or subordinate to Findings of Fact 27. 77-84. Accepted in or subordinate to Findings of Fact 52. 85-92. Accepted in or subordinate to Findings of Fact 27,28 and 34. 93-103. Accepted in or subordinate to Findings of Fact 28. 104-105. Accepted in Findings of Fact 31. 106. Accepted in Findings of Fact 32. 107-109. Accepted in or subordinate to Findings of Fact 28 and 32. 110-111. Accepted in Findings of Fact 22. 112-125. Accepted in or subordinate to Findings of Fact 23. 126. Accepted in or subordinate to Findings of Fact 5. 127-141. Accepted in Findings of Fact 23 and 24. 142. Rejected in Findings of Fact 7 and 23. 143-145. Accepted in or subordinate to Findings of Fact 7 and 23. 146-151. Issue not reached. 152-158. Accepted in or subordinate to Findings of Fact 24. 159-160. Accepted in Findings of Fact 27. 161-162. Accepted in Findings of Fact 28. Accepted in part in Findings of Fact 28. Accepted in or subordinate to Findings of Fact 29. 165-167. Accepted in or subordinate to Findings of Fact 29. 168-169. Accepted in Findings of Fact 30. 170. Accepted in Findings of Fact 21-30. 171-172. Rejected in general in Findings of Fact 47. 173. Accepted in Findings of Fact 47. 174-184. Rejected or subordinate to Findings of Fact 47. 185-187. Rejected or subordinate to Findings of Fact 43 and 47. 188-193. Accepted in Findings of Fact 47. 194-199. Subordinate to Finding of Fact 47. 200. Accepted in Findings of Fact 29. 201-208. Accepted in or subordinate to Findings of Fact 52. 209. Rejected. 210-218. Accepted in or subordinate to Findings of Fact 52. 219. Rejected conclusion as to "substantial" in Findings of Fact 52. 220-229. Accepted in or subordinate to Findings of Fact 52. 230. Rejected conclusion as to "substantial" in Findings of Fact 52. Petitioner, St. Mary's, Proposed Findings of Fact. 1-3. Accepted in or subordinate to Findings of Fact 9. Accepted in Findings of Facts 3 and 22. Accepted in or subordinate to Findings of Fact 9. Accepted in or subordinate to Findings of Fact 12. Accepted in or subordinate to Findings of Fact 10. Accepted in or subordinate to Findings of Fact 27. Accepted in Findings of Fact 2. 10-12. Accepted in or subordinate to preliminary statement and Finding of Fact 12. 13-14. Accepted in or subordinate to Findings of Fact 58. 15-17. Accepted in or subordinate to Findings of Fact 43. 18-24. Accepted in Findings of Fact 30. 25-26. Rejected in Findings of Fact 44-46. 27-29. Accepted in Findings of Fact 30. Rejected in Findings of Facts 44-46. 31-32. Accepted in or subordinate to Findings of Fact 44. 33-35. Accepted in or subordinate to Findings of Fact 43. 36. Accepted in or subordinate to Findings of Fact 56. 37 Accepted in or subordinate to Findings of Fact 9. Accepted in Findings of Fact 38. Accepted in or subordinate to Findings of Fact 58. Accepted in or subordinate to Findings of Fact 60. 41-44. Accepted in or subordinate to Findings of Fact 56 and 57. 45-54. Accepted in or subordinate to Findings of Fact 35. 55. Rejected in Findings of Fact 35. 56-57. Accepted in or subordinate to Findings of Fact 34. 58. Conclusion rejected, although access is limited by comparison to commercially insured patients, See, Findings of Fact 34. 59-66. Accepted in or subordinate to Findings of Fact 34. 67-73. Accepted in Findings of Facts 9, 28 and 32. Accepted in Findings of Fact 9. Accepted in Findings of Fact 32. Accepted in Findings of Fact 34. Rejected as significant benefit in Findings of Fact 34. Accepted (as both interests can be better accomplished) in Findings of Fact 27. Accepted in or subordinate to Findings of Fact 25. 80-81. Rejected in Findings of Fact 25. Accepted in Findings of Fact 25. Rejected in Findings of Fact 25. Rejected in Findings of Fact 25. Rejected as valid in Findings of Fact 34. 86-88. Accepted in Findings of Facts 27 and 36. 89-91. Accepted in part or subordinate to Findings of Fact 26. Rejected in Findings of Fact 26. Rejected in Findings of Fact 25 and 26. Accepted in Findings of Fact 22. Rejected in Findings of Fact 25-26. Rejected in general in Findings of Fact 27. 97-98. Accepted in or subordinate to Findings of Fact 28. Accepted in Findings of Fact 29. Accepted in Findings of Fact 30. Rejected conclusion in Findings of Fact 35. Accepted in Findings of Fact 31 and 32. 103-104. Accepted in Findings of Fact 27. 105. Accepted in Findings of Fact 37. 106-107. Rejected in Findings of Fact 51. Accepted in or subordinate to Findings of Fact 51. Accepted in Findings of Fact 35. Rejected in Findings of Fact 47. Rejected in Findings of Fact 48. Rejected in Findings of Fact 24. Intervenor, JFK Medical Center, Inc.'s Proposed Findings of Fact. Accepted in Findings of Fact 9. Accepted in Findings of Fact 18. Accepted in or subordinate to Findings of Fact 12. 4-6. Accepted in or subordinate to preliminary statement. 7-9. Accepted in or subordinate to Findings of Fact 2. Accepted in or subordinate to Findings of Fact 16-19. Accepted in Findings of Fact 27. Accepted in relevant part in Findings of Fact 16 and 27. 13-19. Accepted in or subordinate to Findings of Fact 34. Accepted in Findings of Fact 18 Accepted in Findings of Fact 35. Accepted in Findings of Fact 19. 23 Accepted in relevant part in Findings of Fact 33. 24. Accepted in Findings of Fact 36. 25-27. Accepted in or subordinate to Findings of Fact 33. 28-31. Accepted in or subordinate to Findings of Fact 27. 32-34. Accepted in or subordinate to Findings of Fact 27 and 34. 35-44. Accepted in or subordinate to Findings of Fact 35. 45-48. Accepted in or subordinate to Findings of Fact 25. 49-50. Accepted in or subordinate to Findings of Fact 26. 51. Accepted in or subordinate to Findings of Fact 25. 52-57. Accepted in Findings of Fact 44-46. 58. Subordinate to Finding of Fact 44-46. 59-66. Accepted in or subordinate to Findings of Fact 51. 67-75. Accepted in or subordinate to Findings of Fact 59. 76-78. Rejected in Findings of Fact 59. 79-80. Accepted in or subordinate to Findings of Fact 25. 81-82. Rejected in or subordinate to Findings of Fact 57. 83-84. Accepted in or subordinate to Findings of Fact 52. 85. Accepted in Findings of Fact 43. 86-89. Accepted in or subordinate to Findings of Fact 58. Respondent, AHCA's Proposed Findings of Fact. 1. Accepted in general or subordinate to Findings of Fact 5-8. 2. Accepted in or subordinate to Findings of Fact 9-12. 3. Accepted in or subordinate to Findings of Fact 13-15. 4. Accepted in Findings of Fact 16 and 18. 5. Accepted in Findings of Fact 6 and 19. 6. Accepted in preliminary statement and Findings of Fact 2. 7. Accepted in Findings of Fact 31 and 32. 8. Accepted in Findings of Fact 31. 9. Accepted in or subordinate to Findings of Fact 7. Subordinate to Findings of Fact 7. Accepted in Findings of Fact 11 and 26. Accepted in or subordinate to Findings of Fact 14. 13,14. Accepted in or subordinate to Findings of Fact 6, 28 and 32. 15. Accepted in or subordinate to Findings of Fact 10, 28 and 32. 16,17. Accepted in or subordinate to Findings of Fact 28 and 32. Accepted in Findings of Fact 21-30. Accepted in Findings of Fact 22. 20,21. Accepted in part in Findings of Facts 23 and 24. Accepted in Findings of Fact 24. Accepted in or subordinate to Findings of Fact 25, 26 and 27. Accepted in Findings of Fact 27. Accepted in Findings of Fact 28. Subordinate to Findings of Fact 29. Accepted in Findings of Fact 5, 9 and 13. Rejected conclusion in terms of other indicators in Findings of Fact 29. Accepted in or subordinate to Findings of Fact 5, 9, 13 and 29. Accepted in Findings of Fact 29. 30-33. Accepted in or subordinate to Findings of Fact 23-26. Accepted in Findings of Fact 30. Accepted in Findings of Fact 27 and 34-37. 36-37. Accepted in or subordinate to Findings of Fact 27. 38. Accepted in Findings of Fact 35. 39-42. Accepted in or subordinate to Findings of Fact 32 and 34. Accepted in Findings of Fact 18. Accepted in Findings of Fact 51. Accepted conclusion in Findings of Fact 52. 46-48. Accepted in Findings of Fact 23-26 and 38. Accepted in Findings of Fact 14 and 24. Accepted if last line changed from "St. Mary's" to "Lawnwood" in Findings of Fact 27, 36 and 37. 51-52. Accepted in Findings of Fact 40 and 61. Accepted in Findings of Fact 42. Accepted in Findings of Fact 29. Accepted in Findings of Fact 48. Accepted in Findings of Fact 44-46. Accepted in Findings of Fact 47. Accepted in Findings of Fact 29 and 51. Respondent, Martin Memorial's Proposed Findings of Fact. Accepted in Findings of Fact 2. Accepted in preliminary statement. Accepted in or subordinate to Findings of Fact 13. Accepted in or subordinate to Findings of Fact 9. 5-6. Accepted in or subordinate to Findings of Fact 5. Accepted in or subordinate to Findings of Fact 7, 11 and 14. Accepted in or subordinate to preliminary statement and Findings of Fact 19. Accepted in or subordinate to preliminary statement and Findings of Fact 18. Accepted in preliminary statement and Finding of Fact 1. Accepted in Findings of Fact 3 and 16. Accepted in or subordinate to Findings of Fact 23 and 24. Accepted in or subordinate to Findings of Fact 20. 14-15. Accepted in or subordinate to Findings of Fact 14. Accepted in or subordinate to Findings of Fact 15. Subordinate to Finding of Fact 13. Accepted in Findings of Fact 12. Accepted in relevant part or subordinate to Findings of Fact 8 and 49. 20-21. Accepted in Findings of Fact 61. Accepted in Findings of Fact 62. Accepted in preliminary statement and Finding of Fact 2. Accepted in Conclusions of Law 74. Accepted in Findings of Fact 52. Accepted in Findings of Fact 38. 27-28. Rejected conclusion that program is superior in terms of quality of care in Findings of Fact 38. 29-30. Accepted in or subordinate to Findings of Fact 43. Accepted in general or subordinate to Findings of Fact 43. Rejected in or subordinate to Findings of Fact 59. 33-34.. Accepted conclusion in Findings of Fact 43. 35-37. Accepted in or subordinate to Findings of Fact 23-26. 38-40. Conclusion rejected in substantial part in Findings of Fact 23. 41-43. Accepted in substantial part in Findings of Fact 24. 44. Accepted in Findings of Fact 47. 45-48. Accepted in or subordinate to Findings of Fact 48. 49-50. Rejected in Findings of Fact 66 and 67. 51-52. Accepted in or subordinate to Findings of Fact 29. Rejected conclusion in part in Findings of Fact 23 and 24. Accepted in Findings of Fact 27. 55-59. Accepted in or subordinate to Findings of Fact 28 and 34. 60. Conclusion rejected in Findings of Fact 32. 61-62. Accepted in Findings of Fact 27. 63. Rejected in general in Findings of Fact 23 and 27. 64-65. Rejected as to alternatives for "residents most likely" to the extent that is inconsistent with need in relation to state plan, in Findings of Fact 27. Accepted in Findings of Fact 51 and 52. Accepted in or subordinate to Findings of Fact 52. Rejected in Findings of Fact 52. 69-70. Accepted in or subordinate to Findings of Fact 52. Accepted in Findings of Fact 22 and 30. Rejected conclusion in Findings of Fact 23 and 24. Accepted except last sentence in Findings of Fact 27. 74-75. Accepted in Findings of Fact 28. 76-77. Accepted in or subordinate to Findings of Fact 29. Rejected conclusion or subordinate to Findings of Fact 29. Accepted in Findings of Fact 32. 80-82. Accepted in or subordinate to Findings of Fact 31. COPIES FURNISHED: W. David Watkins, Esquire 2700 Blair Stone Road, Suite C Post Office Box 6507 Tallahassee, Florida 32314-6507 (Counsel for St. Mary's Hospital) Michael J. Cherniga, Esquire David C. Ashburn, Esquire Greenberg, Traurig, Hoffman, et al. Suite 2000 111 South Monroe Street Tallahassee, Florida 32302 (Counsel for Palm Beach Gardens Community Hospital) Elizabeth McArthur, Esquire 101 North Monroe Street, Suite 1000 Post Office Drawer 11307 Tallahassee, Florida 32302 (Counsel for Lawnwood Medical Center) Leslie Mendelson, Esquire Senior Attorney Agency for Health Care Administration 325 John Knox Road, Suite 301 Tallahassee, Florida 32303-4131 Byron B. Mathews, Jr., Esquire 201 South Biscayne Boulevard Suite 2200 Miami, Florida 33131 Robert A. Weiss, Esquire John M. Knight, Esquire The Perkins House, Suite 200 118 North Gadsden Street Tallahassee, Florida 32301 R. S. Power, Agency Clerk Agency for Health Care Administration Atrium Building, Suite 301 325 John Knox Road Tallahassee, Florida 32303 Harold D. Lewis, Esquire The Atrium, Suite 301 325 John Knox Road Tallahassee, Florida 32303
The Issue At issue in these proceedings is whether there exists a need for a new open heart surgery program in HRS District IX and, if so, whether the applications of St. Mary's Hospital, Inc. (St. Mary's), Boca Raton Community Hospital, Inc. (Boca), and Martin Memorial Hospital Association, Inc. (Martin), or any of them, for a certificate of need to establish such a program should be approved.
Findings Of Fact Case status In September 1989, Boca Raton Community Hospital, Inc. (Boca), St. Mary's Hospital, Inc. (St. Mary's), and Martin Memorial Hospital Association, Inc. (Martin), filed timely applications with the Department of Health and Rehabilitative Services (Department or HRS) for a certificate of need (CON) to establish a new open heart surgery program in HRS District IX. That district is comprised of Palm Beach, Martin, St. Lucie, Indian River, and Okeechobee Counties. Boca's and Martin's applications sought authorization to establish an adult open heart surgery program, whereas St. Mary's application sought authorization to establish an adult and pediatric open heart surgery program. On January 26, 1990, the Department published notice in the Florida Administrative Weekly of its intent to grant Boca's application, and to deny the applications of St. Mary's and Martin. St. Mary's and Martin filed timely protests to the Department's proposed action, and three existing providers of open heart surgery services in the district, NME Hospitals, Inc., d/b/a Delray Community Hospital (Delray), JFK Medical Center, Inc. (JFK), and AMI/Palm Beach Gardens Medical Center, Inc. (Palm Beach Gardens), timely protested the Department's intention to grant Boca's application or intervened to oppose the approval of any new open heart surgery program in the district. The applicants Boca, a 394-bed not-for-profit community hospital, is the southernmost hospital in Palm Beach County and HRS District IX, being located in Boca Raton, Florida, just two miles north of the Broward County/HRS District X line. It was established in the 1960's, and is a comprehensive hospital providing adult cardiac catheterization services, as well as most services available in an acute care facility, with the exception of a designated psychiatric unit, burn unit, and neonatal intensive care. During the period of April 1988 through March 1989, Boca performed 656 adult inpatient cardiac catheterizations, and referred 192 patients for open heart surgery between July 1988 and June 1989. By its application, Boca proposes to establish an adult open heart surgery program to enhance its cardiology services. Boca's primary service area covers a radius of approximately ten miles around the hospital, and it routinely serves patients from Boynton Beach, Palm Beach County, on the north to Pompano Beach, Broward County, on the south. Presently, three providers of open heart surgery services are located proximate to Boca: approximately 11 miles north of Boca, an average drive time of 17 minutes, is Delray, a current provider of open heart surgery services in District IX; approximately 21 miles north of Boca, an average drive time of 32 minutes, is JFK, a current provider of open heart surgery services in District IX; and approximately 15 miles south of Boca, an average drive time of 19 minutes, is North Ridge General Hospital (North Ridge), a current provider of open heart surgery services in District X and the recipient of the vast majority of referrals for open heart services from Boca. St. Mary's, a 378-bed not-for-profit community hospital located in West Palm Beach, Florida, is owned by the Franciscian Sisters of Allegheny, and has served the community for more than 50 years. In addition to the full range of medical surgical services, St. Mary's offers obstetrics, a Regional Perinatal Intensive Care Center (RPICC) -- levels II and III, blood bank, dialysis center, substance abuse center, hospice center, free-standing cancer clinic, adult inpatient cardiac catheterization laboratory, and children's medical services clinic. Upon the opening of its 40-bed psychiatric center, which is currently under construction, St. Mary's will be the largest hospital in District IX. During the period of April 1988 through March 1989, St. Mary's performed 254 adult inpatient cardiac catheterziations. By its application, St. Mary's proposes to enhance its existing services by establishing an adult and pediatric open heart surgery program. Currently, there are no pediatric open heart surgery programs in District IX. There are, however, two current providers of adult open heart surgery services located in Palm Beach County and proximate to St. Mary's: approximately 6 miles north of St. Mary's is Palm Beach Gardens, and approximately 11 miles south of St. Mary's is JFK. Martin, a 336-bed not-for-profit community hospital established in 1939, is located in Stuart, Martin County, Florida. As with the other applicants, Martin offers a full range of acute care services, as well as adult inpatient cardiac catheterization services, a non-invasive cardiology laboratory, and cardiac rehabilitation and support services for cardiac patients and their families. No significant data is, however, available on Martin's adult inpatient cardiac catheterization program since it is a new service. By its application, Martin proposes to establish an adult open heart surgery program. Currently, there are no open heart surgery programs located in the four northern counties of District IX (Martin, St. Lucie, Indian River, and Okeechobee Counties), and Martin is currently the only hospital located in those four counties that provides in-patient cardiac catheterization services. Accordingly, to access open heart surgery services within the district, residents of the northern four counties must avail themselves of the current programs existent in Palm Beach County. The protestants As heretofore noted, open heart surgery services are currently available at three facilities within District IX; Delray, JFK and Palm Beach Gardens, each of which is located in Palm Beach County. Delray is a 211-bed acute care hospital, sited in the southern portion of Palm Beach County, and located in Delray Beach, Florida. It is a comprehensive hospital providing all services normally available in an acute care facility, with the exception of obstetrics, pediatrics and radiation ontology, and is part of a larger medical campus, operated by the same parent company, that includes a 60-bed inpatient rehabilitation hospital that is physically attached to Delray, a 120-bed psychiatric hospital, and a 120-bed skilled nursing facility. In addition to its other services, Delray provides inpatient cardiac catheterization services and has, since 1986, provided adult open heart surgery services. With a recent addition, Delray has two dedicated open heart operating rooms (ORs) and one back up, as well as three separate intensive care units for coronary care, medical intensive care and surgical intensive care. For calendar year 1989 Delray reported to the local health counsel that it performed 338 open heart cases. Delray is located approximately 11 miles north of Boca, an average drive time of approximately 17 minutes. Between Delray and Boca, there is more than a 50 percent overlap in the medical staffs of the two hospitals, and almost 70 percent overlap in the areas of cardiology and internal medicine. Considering the overlap in the facilities' service areas, it is reasonable to conclude that if Boca's application is approved Delray would lose 122 open heart and 84 angioplasty cares in Boca's first year of operation and 130 open heart and 93 angioplasty cases in Boca's second year of operation. Such losses would translate into a after-tax income loss to Delray of approximately $645,000 in the first year of operation alone. Such loss of revenue and patients could adversely impact Delray's existing program. JFK is a 369-bed community hospital located in Atlantis, Florida; a small town just south of West Palm Beach. It provides a full range of medical- surgical services, with the exception of OB-GYN and nursery services, including cardiac, cancer, orthopedic, and medical/surgical intensive care and coronary care. It established its inpatient cardiac catheterization and open heart surgery program in February 1987, and currently has ten operating rooms, two of which are devoted exclusively to open heart surgery, and a 16-bed cardiac care unit (CCU), 10 beds of which are dedicated to open heart patients. For calendar year 1989, JFK reported to the local health council that it performed 262 open heart cases. As sited, JFK is located just south of West Palm Beach and within 10 miles of St. Mary's. Currently, there is an 83 percent overlap in the MDC-5 service areas (the service area closest to the open heart surgery program) of St. Mary's and JFK, and a substantial overlap between cardiologists on the staffs of both facilities. During the period of January 1988 - May 1990, 43 percent of the patients St. Mary's referred for open heart and angioplasty services were referred to JFK. Assuming St. Mary's could achieve the volumes it projected in its application, it is reasonable to assume that JFK would lose 75 open heart and 83 angioplasty cases in St. Mary's first year of operation, and 91 open heart and 100 angioplasty cases in St. Mary's second year of operation. Such lose in the first year of St. Mary's operation would translate into a net reduction of $1,200,000 in JFK's income. Such loss of revenue and patients could adversely impact JFK's existing program. Palm Beach Gardens is a 205-bed acute care hospital sited in north Palm Beach County. It provides inpatient cardiac catheterization services and has, since 1983, provided open heart surgery services. Currently, Palm Beach Gardens maintains two operating rooms dedicated to open heart surgery, and has a third operating room available for open heart surgery should the demand arise. For calendar year 1989, Palm Beach Gardens was the largest provider of open heart surgery services in the district, having reported to the local health council that it performed 491 open heart cases. Palm Beach Gardens is located approximately 10 miles south of the Palm Beach County/Martin County line or a straight line distance of approximately 25 miles south of Martin and approximately 10 miles north of St. Mary's. During the period of July 1988 - June 1989, 229 residents of St. Mary's primary service area had open heart surgery at Palm Beach Gardens, and 142 residents of Martin's primary service area obtained such services at that facility. If Martin's proposal is approved and its utilization projections realized, Palm Beach Gardens would lose approximately 84 cases in year one of Martin's operation and 101 cases in year two. Such losses in year two would translate into a $1,400,000 pretax reduction in Palm Beach Gardens' net revenues. Such reduction in revenues and patients was not, however, considering Palm Beach Garden's financial condition and open heart surgery volume, shown to have any significant adverse impact to Palm Beach Gardens, or any identifiable program within its facility. Likewise, should St. Mary's application be approved, volumes at Palm Beach Gardens would not be reduced below optimal levels, and it would not suffer any significant adverse impact to existing programs. The parties' stipulation The parties have agreed that the following facts are admitted: Boca, St. Mary's, and Martin Memorial timely filed their Letters of Intent and CON applications at issue in this proceeding. Further, the parties stipulate that the Letter of Intent complied with all statutory and rule requirements. The construction costs of $100,000 as set forth in Table 25 of St. Mary's application is a reasonable construction costs estimate for the renovation of one special procedures room to perform open heart surgery as proposed in St. Mary's schematic plans. The parties admit that adult open heart surgery services are currently available within a maximum automobile travel time of two hours under average travel conditions for at least 90 percent of HRS Service District IX's population. This stipulation is not meant to preclude other relevant evidence regarding travel times within or without District IX. All existing providers of open heart surgery in District IX are JCAHO accredited; all applicants in this proceeding are JCAHO accredited. Each of the applicants, if approved, have the ability to implement and apply circulatory assist devices such as intra-aortic balloon assist and prolonged cardiopulmonary partial bypass for adult open heart surgery. Each of the applicants, if approved, will be capable of fulfilling the requirements of an adult open heart surgery program to provide the following services: medicine, for example, cardiology, hematology, nephrology, pulmonary medicine and infectious diseases; pathology, for example, anatomical, clinical, blood bank and coagulation lab; anesthesiology, including respiratory therapy; radiology, for example, diagnostic nuclear medicine lab; neurology; adult cardiac catheterization laboratory services; non-invasive cardiographics lab, for example, electrocardiography including cardiographics lab, for example, electrocardiography including exercise stress testing, and echocardiography; intensive care; and emergency care available 24 hours per day for cardiac emergencies. This stipulation relates only to the provision of medical services, not that the applicants have sufficient capacity to provide those services in connection with an open heart surgery program. The redesignation of acute care beds from medical/surgical beds to any type of critical care unit beds, except for neonatal intensive care beds, does not require a certificate of need unless the hospital incurs a capital expenditure in excess of the capital expenditure threshold in accomplishing this redesignation. The Department's open heart surgery and methodology and the "fixed need" pool. On August 11, 1989, the Department, pursuant to Rule 10-5.008(2)(a), Florida Administrative Code, published notice of the fixed need pool for open heart surgery programs for the July 1992 planning horizon in the Florida Administrative Weekly. Pertinent to this case, such notice established a net need for zero new adult open heart surgery programs in District IX. There was, however, no publication of any fixed need pool for pediatric open heart surgery. Following publication of the fixed need pool, the Department received protests contending that its calculation of net need was erroneous. Upon review, the Department concluded that its initial calculation was in error, and on September 1, 1989, the Department published a notice of correction in the Florida Administrative Weekly, and established a new net need for one open heart surgery program in District IX. On September 5, 1989, St. Mary's challenged the Department's corrected need assessment, claiming the Department had underestimated the need in District IX for adult open heart surgery services, and on September 8, 1989, Palm Beach Gardens challenged the Department's assessment, claiming the Department had overestimated the need for open heart services in the district. These challenges were forwarded by the Department to the Division of Administrative Hearings, along with a request for the assignment of a hearing officer to conduct all necessary proceedings required under law. Pertinent to the derivation of the fixed need pool, the Department has established by rule an adult and pediatric open heart surgery methodology that must normally be satisfied before any new open heart surgery programs will be approved. That methodology, codified in Rule 10-5.011(1)(f), Florida Administrative Code, forms the premise for the Department's calculation of net need in the instant case. Pertinent to this case, Rule 10-5.011(1)(f), Florida Administrative Code, provides: 2. Departmental Goal. The Department will consider applications for open heart surgery programs in context with applicable statutory and rule criteria. The Department will not normally approve applications for new open heart surgery programs in any service area unless the conditions of Sub-paragraphs 8. and 11., below are met. * * * 8. Need Determination. The need for open heart surgery programs in a service area shall be determined by computing the pro- jected number of open heart surgical pro- cedures in the service area. The following formula shall be used in this determination: Nx = Uc X Px Where: Nx = Number of open heart procedures projected for year X; Uc = Actual use rate (number of procedures per hundred thousand popu- lation) in the service area for the 12 month period beginning 14 months prior to the Letter of Intent deadline for the batching cycle; Px = Projected population in the service area in Year X; and Year X = The year in which the proposed open heart surgery program would initiate service, but not more than two years into the future. * * * 11.a. There shall be no additional open heart surgery programs established unless: the service volume of each existing and approved open heart surgery program within the service area is operating at and is expected to continue to operate at a minimum of 350 adult open heart surgery cases per year or 130 pediatric heart cases per year; and, the conditions specified in Sub-paragraph 5.d., above, will be met by the proposed program. No additional open heart surgery programs shall be approved which would reduce the volume of existing open heart surgery facilities below 350 open heart procedures annually for adults and 130 pediatric heart procedures annually, 75 of which are open heart. Sub-subparagraph 5d, referenced in subparagraph 11a(II), provides: Minimum Service Volume. There shall be a minimum of 200 adult open heart procedures performed annually, within 3 years after initiation of service, in any institution in which open heart surgery is performed for adults. There shall be a minimum of 100 pediatric heart operations annually, within 3 years of initiation of service, in any insti- tution in which pediatric open heart surgery is performed, of which at least 50 shall be open heart surgery. Essentially, the subject methodology contemplates that three conditions must be satisfied before an application for a new adult open heart surgery program in the district would normally be approved: (1) a calculated net numeric need under the Department's mathematical methodology; (2) a determination that "the service volume of each existing and approved open heart surgery program within the service area is operating at and is expected to continue to operate at a minimum of 350 open heart surgery cases per year"; and (3) a demonstration that the applicant could perform "a minimum of 200 open heart procedures (cases) annually within 3 years after service is initiated." The first two conditions are utilized by the Department to initially establish the fixed need pool for open heart surgery services. The third condition is, by rule, related to an applicant's ability to provide quality care, and will be discussed infra. As a threshold for calculating need, and the fixed need pool, the Department's mathematical need methodology contains the formula for deriving the gross number of open heart surgical cases anticipated two years into the future. This methodology is based on the actual use rate in the district for the 12- month period beginning 14 months prior to the letter of intent deadline for the batching cycle. The number of cases is then divided by 350, which is consistent with the minimum service volume mandates of subparagraph 11 of the rule, to derive an actual gross need for open heart surgery programs at the horizon year. Existing and approved programs are then substracted to determine if there is a net need for a new open heart surgery program. While there was some dispute among the parties as to what the appropriate underlying data was to drive the Department's numerical need methodology, the parties agreed and the proof demonstrated a fractional need greater than .5, under the formula. 1/ The second step in establishing a need for open heart surgery programs, and the fixed need pool, is a determination, as required by subparagraph 11(2)I of the rule, of whether "each existing and approved open heart surgery program within the service areas is operating at and is expected to continue to operate at 350 adult open heart surgery cases per year." Here, based on the data available to the Department when it established the fixed need pool, the three existing providers had operated at the following case levels for the preceding year: Palm Beach Gardens - 494 cases; Delray - 328 cases; and JFK - 275 cases. Consequently two of the three existing providers were not operating at 350 cases per year. 2/ Based on the foregoing data, the Department initially published a net need for zero new open heart surgery programs in District IX. However, following the receipt of protests to the fixed need pool it had established, the Department, based on the same data, concluded its initial decision was erroneous, and published a notice of correction which established a net need for one new open heart surgery program in the district. This decision was timely challenged. The Department's ultimate decision to publish a need for one new program was based on two factors. First, the Department had historically rounded the numerical need up where fractional need, as calculated by its methodology, was .5 or higher. Second, although of questionable validity at the time, the Department had for several years "interpreted" the 350 case level, referred to in subparagraph (11) of the rule, to require that the average of the existing programs be at 350 before a new program would be approved, as opposed to the literal rule requirement that "each existing and approved open heart surgery program ... [be] ... operating at ... a minimum of 350 adult open heart surgery cases per year." Accordingly, with differing views then pending in the Department, it elected to recalculate the utilization level by applying the averaging approach, as opposed to applying the rule as written which it had done in initially determining zero need, and therefore published a corrected need for one new program. On January 23, 1990, the Department issued final orders in three cases, each of which involved CON applications for open heart surgery services filed in the September 1988 batching cycle, Hillsborough County Hospital Authority v. Department of Health and Rehabilitative Services, 12 FALR 785 (1990), Humana of Florida, Inc. v. Department of Health and Rehabilitative Services, 12 FALR 823 (1990), and Mease Health Care v. Department of Health and Rehabilitative Services, 12 FALR 853 (1990). In each final order the Department's Secretary stated, with regard to the Department's averaging interpretation, that: I conclude that the rule should be applied as written and that numeric need should be found only where each existing and approved open heart surgery program within the service district is operating at a minimum level of 350 open heart cases per year .... I am not unmindful that the conclusion reached here departs from an established practice of interpreting subparagraph 11 of the need rule by averaging the number of cases done by the existing providers and finding subparagraph 11 to be satisfied if the average was 350 cases or more. As previously stated, I am now satisfied that application of the rule as written is more consistent with sound health planning .... Consequently, the averaging practice that resulted in the Department's corrected notice of need for the September 1989 batching cycle at issue in this case was specifically rejected by the Department as being contrary to the rule as written before it published its notice of intent to grant Boca's application. Even though the corrected need published by the Department was erroneous, as being derived contrary to the express language of the rule methodology, the Department and the applicants contend that such error is not subject to correction in this case because of the Department's fixed need pool rule and the Department's incipient policy regarding when it will correct errors in a fixed need pool that has already been published. Such contentions are, however, unpersuasive as a matter of law, discussed infra, and as not supported by any compelling proof. The Department's fixed need pool rule, codified at Rule 10- 5.008(2)(a), Florida Administrative Code, provides: Publication of Fixed Need Pools. The depart- ment shall publish in the Florida Administra- tive Weekly, at least 15 days prior to the letter of intent deadline for a particular batching cycle the fixed need pools for the applicable planning horizon specified for each service ... These batching cycle specific fixed need pools shall not be changed or adjusted in the future regardless of any future changes in need methodologies, popu- lation estimates, bed inventories, or other factors which would lead to different projections of need, if retroactively applied. In this case there has been no change in the Department's need methodology that leads to a different projection of need, as proscribed by the fixed need pool, but, rather, an identified failure of the Department to properly apply its rule when it assessed need. While the Department may have consistently misapplied its rule in the past, such consistency does not cloth it past action with any propriety where, as here, such action is properly challenged or, stated differently, because the rule was misapplied in the past does not lead to the conclusion that its proper application constitutes a change in need methodologies. Accordingly, it is found that the fixed need pool rule does not, under the circumstances of this case, preclude correction of the need established through the Department's publication of its notice of correction. 3/ The Department and the applicants also contend that the Department's policy on how it will treat corrections to a fixed need pool that has already been published, and errors in a published fixed need pool which are discovered after the cycle has begun, precludes any correction of the need published for this batching cycle. Pertinent to this point, the Department points to its policy, which was published in the Florida Administrative Weekly contemporaneously with its initial assessment of zero need, that provides: Any person who identifies any error in the fixed need pool numbers must advise the agency of the error within ten (10) days of publica- tion of the number. If the agency concurs in the error, the fixed need pool number will be adjusted prior to or during the grace period for this cycle. Failure to notify the agency of the error during this ten day period will result in no adjustment to the fixed need pool number for this cycle and a waiver of the person's right to raise the error at subsequent proceedings. Any other adjustments will be made in the first cycle subsequent to identification of the error including those errors identified through administrative hearings or final judicial review. Any person whose substantial interest is affected by this action and who timely advised the agency of any error in the action has a right to request an administrative hearing pursuant to Section 120.57, Florida Statutes. In order to request a proceeding under Section 120.57, Florida Statutes, your request for an administrative hearing must state with specifi- city which issues of material fact or law are in dispute. All requests for hearings shall be made to the Department of Health and Rehab- ilitative Services and must be filed with the agency clerk at 1323 Winewood Blvd. Building 1, Room 407, Tallahassee, Florida 32301. All requests for hearings must be filed with the agency clerk within 30 days of this publication or the right to a hearing is waived. According to the Department, its policy is to correct computational errors in the fixed need pool only if they are brought to its attention during the grace period which is triggered by the filing of a letter of intent, and if there is sufficient time to publish a corrected fixed need pool prior to the CON application deadline so that all potential competing providers will have notice of the changes. Errors brought to the Department's attention after the grace period will only be considered in the development of the subsequent batching cycle's fixed need pool, regardless of the nature or magnitude of the error. Errors brought to the Department's attention during the grace period, but not reviewed by the Department until after the grace period would only be corrected for subsequent batches. Errors identified in administrative hearings or upon judicial review, even though predicated upon a timely notice of error to the Department, would be corrected in subsequent batches, but not for the batch in which the error occurred. The Department's enunciated rational for the foregoing policy is to instill "predictability" in the CON process, which it suggests promotes competition and affords the Department an opportunity to select from a broader field the best qualified applicants to "meet the need." Such rationale lacks, however, any reasonable basis in fact where, as here, there is no need to be met, and affronts sound health planning principles. The 350 minimum procedure level established for existing providers, before a new program can be approved, is an important threshold bearing on quality of care. In this regard, it has been demonstrated that there is a direct relationship between volume of procedures and mortality, with better results being obtained at facilities operating at a minimum level of 200-350 procedures annually. Accordingly, precision in assessing the need for new open heart surgery programs is crucial to assure that any new program could reasonably be expected to achieve a sufficient level of service, and to assure that the level of service provided by existing facilities would not fall below the optimum threshold. The Department's policy ignores this relationship, would recognize a need where none exists and thereby adversely impact existing programs, and would impinge on future planning horizons. As importantly, the Department's policy would supplant its own rule methodology for calculating need, and render illusory any decision based on a balanced review of statutory criteria. Accordingly, it is concluded that the Department has failed to explicate its policy choice in the instant case, and that numeric need under the Department's methodology is a viable issue in these proceedings. The need for the services being proposed in relationship to the district plan and state health plan. Applicable to this case is the 1989 Florida State Health Plan, which contains the following preferences to be considered in comparing applications for open heart surgery programs: Preference shall be given to applicants estab- lishing new open heart surgery programs in larger counties in which the percentage of elderly is higher than the statewide average and the total population exceeds 100,000. Preference for new open heart surgery programs shall be given to applicants clearly demonstra- ting an ability to perform more than 350 adult procedures annually within three years of initiating the program. Quality of care has been demonstrated to be directly related to volume; thus, facilities are expected to perform a minimum of 350 adult procedures annually. Preference shall be given to applicants who will improve access to open heart surgery for persons who are currently seeking the service outside of their HRS district. This will improve accessibility and reduce travel time for the residents in the district. Preference shall be given to an applicant with a history of providing a disproportionate share of charity care and Medicaid patient days in the respective acute care subdistrict. Qualifying hospitals shall meet Medicaid disproportionate share hospital criteria. Priority should be given to an applicant who provides services to all persons, regardless of their ability to pay. Preference shall be given to an applicant that can offer a service at the least expense yet maintain high quality of care standards. The physical plant of larger facilities can usually accommodate the required operating and recovery room specifications with lower capital expendi- tures than smaller facilities. Larger facilities also have a greater pool of the specialized personnel needed for open heart surgical procedures. Preference shall be given to an applicant that performs percutaneous transluminal angioplasty, streptokinase, or other innovative techniques as alternatives to surgery for low-risk patients. The applicant shall include in its application a protocol regarding the selection of patients for surgery or alternative non-surgical therapeutic cardiac procedures. All three applications are reasonably consistent with the state health plan's preference for establishing open heart surgery programs in counties in which the percentage of elderly is higher than the statewide average and the total population exceeds 100,000. In 1989, Palm Beach County had a population of 873,347, 23.4 percent of which were age 65 and over, which was higher than the statewide average of 17.9 percent. The next most populous counties in the district fell within Martin's primary service area, and were St. Lucie County, with a population of 142,440, 18.3 percent of which were age 65 and over, and Martin County, with a population of 96,336, 25.1 percent of which were age 65 and over. In all, the northern four counties had a population of 360,644, 21.2 percent of which were age 65 and over. The state health plan also accords a preference to applicants who clearly demonstrate an ability to perform more than 350 adult procedures within three years of initiating the program. Of the three applicants, Boca is in the best position to achieve the preference based on the number of diagnostic cardiac caths performed at this facility, and the number of patients it has referred for open heart surgery. Comparatively, Martin and St. Mary's are unlikely to achieve such level of service within three years of initiating a program. The third objective of the state health plan accords a preference for the applicant that will more clearly improve access to open heart surgery for persons who are currently seeking the service outside the district. Currently, while there is no access problem in the district, it is apparent that many district residents leave the district for open heart surgery. During the period of July 1988 - June 1989, open heart procedures were performed on 782 people residing in Boca's primary service area. Of those, 316 received treatment in a District IX facility, 383 received treatment in a District X (Broward County) facility, and the balance received treatment elsewhere, but predominately in Dade County (District XI). While there was a substantial outmigration from Boca's primary service area for open heart services, the vast majority of such outmigration, 325 people, was serviced at North Ridge, a mere fifteen mile/nineteen minute trip from the Boca area. With regard to St. Mary's primary service area, the proof demonstrated that during the same period 566 people sought open heart services, with 455 of those people receiving treatment within District IX. Of the 111 who sought service outside the district, 41 received treatment in Broward County and 61 received treatment in Dade County. Finally, with regard to Martin's primary service area, 316 people sought open heart services, with 148 of those people receiving treatment within the district. Of the 168 who sought service outside the district, 90 received treatment in Broward County, 29 in District VII hospitals, and 39 in Dade County. As heretofore noted, access is not a problem within District IX. However, to the extent this preference seeks to address the issue of outmigration, the proof demonstrates that Martin is the superior applicant. Clearly, the 15 mile/19 minute trip from the Boca area to North Ridge is not a barrier to access, and the number of people from St. Mary's primary service area seeking services outside the district are small in comparison to the other applicants. The residents of Martin's primary service area who seek treatment outside the district are, however, disproportionately large when one considers the aggregate travel time they incur when accessing services in the Orlando or Melbourn areas, or Dade and Broward Counties. The fourth objective of the state health plan accords a preference for the applicant with a history of providing a disproportionate share of charity care and Medicaid patient days in the district. Among the applicants, St. Mary's is the only disproportionate share provider and provides the largest number of Medicaid patient days in the district. As between Boca and Martin, the proof demonstrates that Martin is more committed to, and has historically been a greater provider of, care to the medically indigent. The fifth objective of the state health plan accords a preference to the applicant that can offer a service at the least expense yet maintain high quality of care standards. Here, each of the applicants are large facilities, with demonstrated commitments to maintaining high quality of care standards. Martin has, however, demonstrated that it can offer the proposed service at the least expense. 4/ The last objective of the state health plan accords a preference to the applicant that will perform percutaneous transluminal angioplasty, strepokinase, or other innovative techniques as alternatives to surgery. Here, all applicants propose to offer such services. District IX's 1988 Health Plan was in effect at the time the CON applications were at issue in this case were filed; however, that plan had not been adopted as a rule. Accordingly, such plan is not pertinent to this proceeding. Venice Hospital, Inc. v. Department of Health and Rehabilitative Services, Case Nos. 90-2383R, et seg., (DOAH 1990). The availability, quality of care, efficiency, appropriateness, accessibility, extent of utilization, and adequacy of like and existing health care services in the district. Open heart surgery is a specialized, tertiary health care service. A tertiary health service is defined by Section 381.702(20), Florida Statutes, as: ... a health service which, due to its high level of intensity, complexity, specialized or limited applicability, and cost, should be limited to, and concentrated in, a limited number of hospitals to ensure the quality, availability, and cost-effectiveness of such service.... As a tertiary service, planning for open heart surgery services is done on a regional basis and concentrated in a limited number of hospitals to insure the quality, availability and cost effectiveness of the program. Essentially, the concept of regionalization creates a distinction between hospitals; some hospitals offer routine acute care services, while special high risk services are concentrated in a limited number of hospitals. Encompassed within such concept is the expectation that patients will be transferred from one facility to another to obtain tertiary care services. As a touchstone for assessing need within a service district, the Department has adopted the open heart surgery need methodology, discussed supra, that must normally be satisfied before a new open heart surgery program will be approved. Under that methodology, further need for adult open heart surgery programs is determined based on the projected increase in the number of open heart surgery procedures two years into the future and the open heart surgery volume of existing providers. The rule provides that, regardless of the projected growth in the number of open heart procedures, no additional adult open heart programs are granted unless each existing adult open heart program performs a minimum of 350 procedures annually. Application of the rule methodology to the facts of this case projects a growth in the projected number of open heart procedures sufficient to support a fractional need greater than .5, which the Department reasonably rounded to 1. However, two of the existing three providers were not performing a minimum of 350 procedures annually. Therefore, there is no need under the Department's methodology for a new open heart surgery program in District IX. While no need under the Department's methodology, the applicants have advanced several factors which they contend reflect negatively on the availability, quality of care, efficiency, appropriateness, accessibility, extent of utilization or adequacy of existing open heart programs in the district, and which they suggest warrant a finding of need based on special or not normal circumstances. Foremost among the factors pressed by the applicants as indicitive of an abnormal circumstance is the high number of District IX residents who seek open heart surgery services outside the district; referred to in this case as outmigration. Outmigration is, however, simply an observation of patient flow patterns and does not, in and of itself, constitute an abnormal circumstance that would demonstrate need in the district. Rather, to demonstrate a not normal circumstance, such outmigration must be demonstrated to be a consequence of some failing of existing programs, i.e., accessibility or quality of care, to be pertinent to any abnormal need assessment. 5/ In this case, there is no such failing in the existing programs. The three existing adult open heart surgery programs in the district are currently available to 90 percent of the population of the district within a maximum automobile travel time of two hours. Under such circumstances there is no geographic access problem within the district. Moreover, only Martin would actually enhance accessibility, were it a problem, because the residents of the four northern counties it proposes to serve must currently travel to Palm Beach County to access services within the district. In contrast, Boca is within approximately 30 minutes travel time of two existing providers in the district and an additional provider in District X. Likewise, St. Mary's is located less than 10 miles from two of the existing providers in the district. As with geographic access, there is likewise no economic access problem in the district. While the Medicaid use rate within the district for calendar year 1989 was .1 percent, well below the statewide average of approximately 2 percent, such raw statistic does not demonstrate that there is a Medicaid access problem in the district. To persuasively demonstrate such fact from use statistics would require a demonstration that Palm Beach County's use rate was significantly lower than counties with similar demographics. Here, there was no such showing. Moreover, St. Mary's, the largest provider of Medicaid services in the district, was only shown to have transferred three Medicaid patients for open heart or angioplasty services from January 1988, through May 1990. Finally, each of the existing providers have contracted with the Palm Beach County Health Care District to provide care to indigent patients, and have not refused service to anyone regardless of their ability to pay. Accordingly, it is concluded that there is no economic access problem within the district. With two of the three existing providers operating below 350 procedures when this cycle commenced, there is clearly excess capacity within the district when one considers the fact that a single operating room has the capacity to handle at least 500 cases annually. In reaching this conclusion, the applicants' assertion that delays may have been encountered in gaining admission to some facilities during the season because of a lack of critical care beds has not been overlooked. However, any such delays were not reasonably quantified in terms of number or duration, and were not shown to be significant. As importantly, existing facilities have increased their critical care bed capacity, and can increase it further by merely redesignating acute care beds from medical/surgical beds to any type of critical care beds needed as the exigency arises. Although two of the three existing providers offer relatively new programs, the proof is compelling that each provides a quality surgical and post surgical open heart surgery program, appropriately staffed, and that there is no want of quality care within the district. The use of agency nurses, as suggested by one applicant, was not persuasively demonstrated to reflect adversely on quality of care. Succinctly, simply because one is an agency nurse does not suggest substandard performance, and the use of agency nurses, as needed, to staff a facility does not, of itself, aversely impact patient care. Here, the staffs of existing facilities are appropriately trained and supervised, and offer their patients a quality program. While there is certainly a significant outmigration from the district for open heart surgery services, such outmigration was not shown to be related to any infirmity in existing programs. Rather, such outmigration is most reasonably attributable to physicians' established referral patterns or patient preference. 6/ Finally, regarding special circumstances, St. Mary's suggests that its designation as a trauma center and the lack of pediatric open heart services to 90 percent of the population within a maximum automobile travel time of two hours warrant approval of its application. Such suggestions are, however, not supported by compelling proof. While it is true that St. Mary's has been selected by the Palm Beach County Health Care District, along with Delray, for designation as a Level II trauma center, such designation has not been contractually finalized and St. Mary's has not applied for such designation with the Department. As importantly, on October 1, 1990, a new law regarding trauma centers became effective which will reopen the county trauma center designation process, and require facilities to be designated by the state as trauma centers. Under such circumstances, it is speculative whether St. Mary's will become a trauma center, and until such event actually occurs such factor is not significant to these proceedings. St. Mary's quest for a pediatric open heart surgery program is premised on special circumstances, not numeric need, and finds it basis on the fact that no pediatric open heart surgery program exists in the district and that such pediatric services are not available to 90 percent of the population within two hours travel time. While such may be the case, St. Mary's application, on balance, fails to support such an award for a number of reasons. First, St. Mary's application projects that it will perform 10 pediatric open heart surgery cases in its first year of operation, and 20 in its second year of operation. It contains no projection for the third year of operation, but St. Mary's consultant, Michael Schwartz, opined that St. Mary's would perform 50 pediatric open heart surgery cases by the third year based on his belief that St. Mary's would capture 80 to 100 percent of the potential pediatric referrals from District IX and the northern portion of District X. Mr. Schwartz's opinions are not, however, credible. During the period July 1, 1988 to June 30, 1989, there were 40 pediatric open heart surgery cases performed on patients residing throughout District IX, with 22 receiving treatment at Jackson Memorial (Dade County), 14 at Miami Children's Hospital, and 4 at Shands in Gainesville. During the same period, there were 24 open heart pediatric patients in northern District X, an area equi-distant in travel time from the Miami facilities and St. Mary's, with 15 receiving treatment at Jackson Memorial, 8 at Miami Children's Hospital and 1 at Shands. Each of these facilities are either teaching hospitals or specialty pediatric hospitals, are among the top four facilities in the state that perform over 100 pediatric open heart surgery cases each year, and each enjoys an excellent reputation for providing quality pediatric care. Given existent referral patterns and the quality of existing pediatric programs, it is improbable that St. Mary's could reach its projected utilization for years one and two, much less attain a level of 50 pediatric open heart surgery cases during its third year of operation. In 1994, the third year of St. Mary's program, there would be approximately 53 pediatric open heart surgery cases performed on patients residing throughout District IX. To attain a level of 50 cases in its third year, St. Mary's would have to attract almost 100 percent of all cases arising within the district, an improbable occurrence. Equally improbable is St. Mary's ability to penetrate the pediatric open heart surgery market in northern Broward County, an area defined by Mr. Schwartz as being equi-distant in travel time from the Miami facilities and St. Mary's, given existent referral patterns and physicians' satisfaction with existing programs. In sum, the proof demonstrates that St. Mary's could not reasonably be expected to perform 50 pediatric open heart surgery cases within three years of initiating service. In addition to its inability to generate sufficient volume to maintain service quality in a pediatric open heart surgery program, St. Mary's also lacks a pediatric cardiac cath program which is required of any facility proposing pediatric open heart surgery services. Notably, with regard to pediatric cardiac services, Rule 10-5.011(1)(e), which relates to cardiac catheterization services, and Rule 10-5.011(1)(f), which relates to open heart services, are mutually dependent. The cardiac catheterization rule, as it relates to pediatrics, provides: 6. Coordination of Services. * * * Pediatric cardiac catheterization programs must be located in a hospital in which pediatric open heart surgery is being performed. * * * 8. Need Determination. * * * f. Pediatric cardiac catheterization programs shall be established on a regional basis. A new pediatric cardiac catheterization program shall not normally be approved unless the numbers of live births in the service planning area, minus the number of existing and approved programs multiplied by 30,000, is at or exceeds 30,000. (Emphasis added) Also pertinent to this issue, the open heart surgery rule provides: 3. Service Availability. * * * c. The following services must be provided in the health care facility within which the open heart surgery program is located and must be capable of fulfilling the requirements of an open heart surgery program: * * * (VI) Cardiac catheterization laboratory.... The Department reasonably interprets the foregoing provisions as mandating that a pediatric cardiac catheterization program or pediatric open heart surgery program may not be approved independent of the other but, rather, they must coexist. Since the proof is clear that St. Mary's only operates and is only approved by the Department to operate an adult cardiac cath program, and it has not applied for a pediatric cardiac cath program, its proposal is deficient. 7/ In view of the foregoing, it is concluded that, while pediatric open heart services are not currently available within District IX and are not available to 90 percent of the population within two hours travel time, that St. Mary's application to initiate such services should be denied. It is further found that the provisions of the open heart surgery rule relating to the two- hour access standard, which does not specifically state whether such standard applies to adult, pediatric or both, is not applicable to pediatrics. Rather, the Department interprets such rule provision to apply only to adult programs, because such standard is not necessarily pertinent to pediatric open heart surgery since it is more specialized or tertiary in nature than adult open heart surgery programs. Given the close relationship between the cardiac cath rule and the open heart surgery rule, the Department's position is reasonable. In this regard, the cardiac cath rule establishes a travel standard for adult programs, but not pediatric. Rather, it provides for establishment of such programs on a "regional basis," and provides that a new pediatric cardiac cath program should not normally be approved unless the number of live births exceeds 30,000. Here, there were only 16,500 live births in District IX in 1988, a number that is insufficient to warrant a pediatric cardiac cath program. Given such fact, and the relationship between the two rules, the Department's interpretation is reasonable and the two-hour travel time standard does not apply to pediatric open heart surgery. Finally, as to adult open heart surgery services, it is concluded that there exist no special circumstances within the district that would warrant approval of a new open heart surgery program, and that existing facilities are providing appropriate quality care that is accessible to all residents of the district regardless of their ability to pay. The ability of the applicant to provide quality of care and the applicant's record of providing quality of care. Each of the applicants in this case has established an excellent record for providing quality care to their patients, and would be generally expected to provide high quality care for open heart surgery patients notwithstanding some failings in their applications. During the course of the proceeding, some protestants contended that because an applicant failed to detail some particular item of equipment essential to an open heart program, that such failing reflected adversely on their ability to provide quality care. While such could be the case in the abstract, it does not, where, as here, the applicants have sound records, with a demonstrated ability to attract quality personnel to staff their programs. Such failings are, however, germane to the feasibility of the applicant's proposals, discussed infra. Other failings pointed to by the protestants, included: St. Mary's proposal to utilize a call team composed of nurses who customarily assist at thoracic surgery and to recover its open heart patients in a mixed intensive care unit; St. Mary's inability to achieve a 200 and 350 case level per year; Martin's inability to achieve a 350 case level per year; and Martin's failure to document in its application the manner in which it could rapidly mobilize an open heart surgery team 24-hours a day, or how it would treat emergency patients within a two-hour period. Again, considering the quality of the applicants, and the quality personnel they will attract, as well as the parties' stipulation, these failings are minor and do not reflect adversely on their proposals with but one exception. 8/ The only significant factor presented that could bear on an applicant's ability to provide quality care is its ability to achieve optimal utilization levels. In this regard, it has been demonstrated that a relationship exists between the volume of open heart surgical procedures performed at a hospital and the quality of care rendered at those facilities, as measured by patient outcomes. Overall, facilities performing more than 350 cases per year experienced the lowest in-hospital death rate, with those performing more than 200 cases per year being next in line. Pertinent to this issue, the Department has adopted Rule 10-5.011(f)5, Florida Administrative Code, which addresses service quality for open heart surgery programs. That rule, as heretofore noted under the findings related to the Department's need methodology, requires that a minimum of 200 adult open heart surgery cases be performed annually within 3 years of initiating the service, and that at least 50 pediatric open heart surgery cases be performed within 3 years of initiating such service. Here, St. Mary's has failed to demonstrate that it can achieve such level of utilization, and its ability to offer a quality program is therefore suspect. As importantly, Rule 10- 5.011(f)11.a.(II) precludes the approval of St. Mary's application under such circumstances. Boca and Martin could reasonably expect to perform at least 200 cases within 3 years. The need in the service district of the applicant for special equipment and services which are not reasonably and economically accessible in adjoining areas, and the needs and circumstances of those entities which provide a substantial portion of their services or resources to individuals not residing in the service district in which the entities are located. As heretofore noted, North Ridge is located in northern Broward County, a mere 15 mile/19 minute drive time from Boca. North Ridge is a 395-bed hospital that provides all services with the exception of obstetric and radiation therapy, and has for 15 years provided open heart surgery services. It currently has two cardiac catheterization laboratories, and two dedicated and two backup open heart operating rooms. At an average of 750 cases per year, over the last few years, North Ridge has additional capacity, and could comfortably accommodate 1,000 cases per year. North Ridge's primary service area is, and has been for sometime, northern Broward County and southern Palm Beach County, although prior to the initiation of other services in Palm Beach County it serviced the entire area. North Ridge markets extensively in southern Palm Beach County, has follow-up activities for its Palm Beach County residents, and has strong ties with the physician community in southern Palm Beach County. Accordingly, North Ridge has an established presence in southern Palm Beach County, with approximately 30-40 percent of its patients coming from that area. North Ridge's mortality statistics, along with its utilization and reputation, mark it as an excellent facility with a quality open heart surgery program. Moreover, its charges for open heart surgery services are significantly below those of Palm Beach County facilities, as well as those proposed by Boca. North Ridge's location makes it easily accessible to the patients of southern Palm Beach County, and physicians have not experienced any significant problems gaining access to that facility. Moreover, Boca's patients have been accorded first priority at North Ridge. With new technology and the development of various drug therapies, it is extremely rare for a patient to have such an urgent need for open heart surgery that transportation becomes a significant issue. When urgently needed, North Ridge, as well as Delray, can adequately serve the needs of southern Palm Beach County. In sum, there is a viable alternative for residents of southern Palm Beach County to Boca's application, and that is their continued referral to North Ridge. That program is easily accessible, reasonably priced, and historically sound. On the other hand, to approve Boca's application would significantly adversely impact North Ridge, since their service areas in southern Palm Beach County and northern Broward County overlap in most material respects. The availability of resources, including health manpower, management personnel, and funds for capital and operating expenditures, for project accomplishment and operations. Each applicant has demonstrated that it either has or can obtain all resources, including health manpower, management personnel and funds for capital and operating expenditures. Boca and Martin each have the funds on hand for project accomplishment, and St. Mary's has demonstrated its ability to acquire such funds through donations, as needed, for project accomplishment. Each applicant is a quality provider of acute care services, and has demonstrated through its existing programs its ability to attract and retain appropriate management and health manpower for project accomplishment, notwithstanding the current nursing shortage being experienced locally and nationally. Accordingly, while the cost of skilled personnel to staff their open heart surgery programs may exceed their initial estimates in some cases, any of the applicants should be able to appropriately staff their program through the use of existing staff, national or local recruitment, or a combination thereof. While each applicant has adequate resources, the viability of Boca's application has been challenged based on its failure to provide a complete list of all capital projects in its application, as required by Section 381.707(2)(a), Florida Statutes. In this regard, the proof demonstrates that the only item listed in its application was for an "expansion/upgrade" of the physical plant at a proposed cost of $6.2 million. That information was an accurate financial description of that project at the time, but did not include other items relating to other construction and equipment purchases to which Boca was committed. In this regard, as of September 1989, Boca had committed itself to an additional $1,261,400 for projects relating to its 1989 fiscal year and $1,380,039 for projects relating to its 1990 fiscal year, for a total of $2,641,439. All of these items will be capitalized by Boca, and it could have provided a list or summary of such projects at the time of filing its application in September 1989. Boca's failure to do so, failed to comply with section 381.707(2)(a), and prevented the Department from having a complete picture of Boca's financial resources to complete the project. The extent to which the proposed services will be accessible to all residents of the service district, and the applicant's past and proposed provision of health care service to Medicaid patients and the medically indigent. Of the proposed programs, only those advanced by St. Mary's and Martin would be reasonably accessible to all residents of the service district. In this regard, the geography and population densities of the district demonstrate that Palm Beach County, at 1,993 square miles, is the single most populous county in the district, with a 1989 population of 873,347. The northern four counties are geographically larger than Palm Beach County, at 2,404 square miles, and contained a 1989 population of 360,664, nearly one-third of the total population of the district. The most dense population in the northern four counties is the Martin County/Port St. Lucie area. The district itself measures 100 miles in length, north to south, in a straight line. Martin is located approximately 60 miles from the southern boarder of the district, St. Mary's is approximately 30 miles, and Boca is 2.1 miles Considering Boca's geographic location, it would not be readily accessible to all residents of the district. Martin and St. Mary's are, on the other hand, sited such that they could, geographically, address the needs of the district as a whole. However, St. Mary's, like Boca, is proximate to a number of open heart surgery providers and would not improve geographic accessibility within the district, as would Martin. Further bearing on the issue of accessibility, is the applicants' commitment to service Medicaid and the medically indigent. In this regard, the proof demonstrates that Boca has not been an historic provider of Medicaid or indigent care, and for its fiscal 1989 dedicated less than 1 percent of its total admissions to Medicaid and indigent care. On the other hand, St. Mary's patient mix has included 15 percent Medicaid and 5 percent indigent, and it is the highest Medicaid provider in the district. Martin has, although to a lesser degree than St. Mary's, also demonstrated a commitment to the underserved by historically serving 5 1/2 percent Medicaid and indigent patients. In its application, Boca "committed" to provide at least 2 percent of gross revenue generated by the open heart surgery program for the provision of charity or indigent care on an annual basis. Considering Boca's nominal historic commitment to indigent care, its location in an affluent area of Palm Beach County, and its closed staff, Boca could not reasonably achieve such level of care, and would not increase accessibility for underserved groups. Comparatively, St. Mary's and, to a lesser extent, Martin, would increase accessibility for underserved groups should the need exist. Here, St. Mary's has projected that 7 percent of its total patient days will be devoted to Medicaid patients and 3 percent to indigent patients, and Martin has projected 5 percent Medicaid and indigent. The costs and methods of the proposed construction. In its application, Boca estimated a total project cost of $7,499,856 to construct and equip a new addition to house its open heart surgery program. That figure included a $6,147,900 construction fund and $783,056 for equipment costs to complete the two operating suites, recovery areas and ten-bed surgical intensive care unit proposed. Its estimates were, however, deficient. Boca's equipment budget, as it appeared in its application, was prepared by an individual who had no expertise in this area, and was deficient in terms of the actual equipment listed and its cost. To properly equip and furnish the two operating room suites, recovery room areas and a ten-bed surgical intensive care unit proposed by Boca would require an expenditure in excess of $1,690,000. Adding necessary instrumentation and a backup pump could add an additional $50-60,000. At hearing, Boca sought to minimize the significance of its underestimation by offering the testimony of an expert in medical equipment planning, cost estimating and procurement. That expert, Richard Drinkwine, was most credible and found, upon review of the Boca proposal that it was wanting in both equipment and cost. In his opinion a more reasonable cost to purchase moverable equipment would be $1,027,267, and a reasonable estimate for the furniture needs of Boca would be $92,257. This estimate was based on the assumption that Boca would not initially equip its second operating room, exam rooms or recovery rooms. To do so, would add an additional cost of $411,329 (movable and fixed equipment) for the second operating room and $160,000 to equip the recovery areas. Adding needed instrumentation and a back up pump would bring Boca's equipment costs to over $1,740,000. 9/ While Boca underestimated its equipment costs, the proof demonstrates that its construction estimate of $6,147,900 was overstated. The major factor which accounts for the overstatement by Boca in its application was an over estimate of the cost to construct the first floor of its addition, which is a covered parking area. In fact, Boca will be able to construct its proposed addition for approximately $5,226,397, or $921,503 less than it estimated in its application. Although Boca could realize a significant savings on construction costs, and those savings would be adequate to almost offset the deficiencies in its equipment budget, the restructuring of its application at this time is not appropriate under the Department's Rule 10-5.010(2)(b). Notably, while the total cost figures might be the same, the additional equipment that is needed to equip Boca's program, and that was omitted from its application, is significant. In addition to Boca's failure to demonstrate the reasonableness of its cost proposal, it is also found that Boca's proposal is oversized and overpriced to meet any demands Boca could reasonably expect to fulfill at any time in the foreseeable future. First, each of the two operating rooms proposed by Boca are over 1,100 square feet in size. Such size is more than twice the size reasonably needed to accommodate open heart surgery. Second, areas in the central core and lounges are also larger then needed. More significantly, Boca is proposing a four-bed recovery area and ten dedicated SICU beds. Even assuming there is a need for an additional open heart surgery program in the district, Boca could never reasonably expect to capture sufficient market share to justify the capital expenditure necessary to warrant a 10-bed SICU. Ten SICU beds could handle between 900 and 1400 open heart patients in a year. There are no programs anywhere in South Florida, no matter how mature or well respected, that have achieved utilization close to that level, and it is not reasonable for Boca to expect to achieve such volumes. Significantly, a portion of the capital cost for Boca's project would, under the present system, be passed along to the federal government by the capital cost pass through. By this mechanism, over $3,500,000 of Boca's project would ultimately be reimbursed to the hospital in the form of Medicare payments. Compared to Boca's cost proposal, St. Mary's is modest. Here, the schematics submitted by St. Mary's with its application and omissions response depict the existing surgical suites at St. Mary's and the minor renovations necessary to convert an existing room into the proposed open heart surgery suite. As proposed, St. Mary's program would have a dedicated open heart surgery suite, as well as a backup operating room. Recovery would be accommodated in its existing 16-bed ICU. In its application, St. Mary's estimated a maximum project cost of $850,000 to remodel its existing facility and equip its proposed open heart surgery program. That figure included up to $100,000 for remodeling costs, and up to $700,000 for equipment costs. St. Mary's estimates are reasonable and cost effective whether its program is dedicated to adult and pediatric open heart surgery service or simply adult services. Significantly, the equipment needed to perform open heart surgery on adults and pediatrics is the same except for some special instruments. That cost, at less than $25,000, is nominal and does not affect the reasonableness of St. Mary's estimates. As proposed in its application, Martin would construct 2,800 square feet of new space at its facility for the purpose of implementing an open heart surgery program. The location of the project is the hospital's first floor adjacent to both the cardiac catheterization laboratory and the existing surgical suites. This location will provide rapid access for cardiac catheterization emergencies requiring open heart intervention and will share common areas with the existing surgical suites, minimizing additional construction and project cost. It is also proximate to a 9-bed surgical intensive care unit. Of the eight existing operating rooms at Martin, two are large enough to serve as backup open heart operating rooms in the event of an emergency, but Martin has not proposed to establish, or budgeted the necessary equipment to establish, a backup operating room. Martin, like St. Mary's, proposes a modest expenditure, compared to Boca, for the initiation of its open heart surgery program. In this regard, Martin's application estimates its total project cost at $1,239,029. That figure includes a total construction cost budget of $796,669, and an equipment budget at $375,360. Martin's costs and methods of proposed construction are reasonable. While the proof demonstrates that approximately $411,000 is a reasonable cost to equip an open heart surgery suite, it also demonstrated that Martin currently has on hand some necessary equipment, such as cell-savers and heating-cooling machines. Under such circumstances, Martin could reasonably equip its program within its $375,360 budget. It could not, however, equip a backup operating room within such budget, and without a backup operating room could not reasonably expect to be able to handle 500 open heart cases a year, as required by rule 10-5.011(f)3d, given the need to back up its cardiac cath program. The immediate and long-term financial feasibility of the proposal. To assess the financial feasibility of the project, Boca's pro forma of income and expense, contained within its application, projects 192 patients during the first year of operation of its open heart surgery program and 211 patients during the second year. Projected charges for both years are based on $55,430 for DRG 104 and $41,942 for DRG 106 with an average length of stay of 10 days. Payor class mix is estimated to be as follows: Medicare 70 percent, Medicaid 0 percent (nominal), insurance 25 percent, other 3 percent, and indigent 2 percent. Net revenue over expenses for year one is projected to be $1,303,312, and for year two to be $1,597,959. Boca's proposed charges, utilization levels, and payor mix are reasonable. However, its pro forma contained unreasonable assumptions regarding average length of stay, total deductions and expenses. 10/ At hearing, Boca made no effort to defend the unreasonable assumptions it had presented to the Department through the pro forma contained in its application. Rather, conceding the unreasonableness of its assumptions, it sought to minimize their import through the testimony of Rufus Harris, an expert in health care finance and accounting. Such objective was not, however, attained. Mr. Harris, employed during the course of these proceedings, actually prepared a completely new pro forma for the Boca program. That pro forma significantly changed Boca's average length of stay from 10 to 16 days; significantly reduced the number of full time equivalents (FTEs) in open heart surgery, recovery and the surgical intensive care unit (SIC) from 39.3 to 24.1; increased the number of support FTEs from 25 to 30 or 32; increased the cost per FTE in the open heart surgery program by $800; increased the cost for each support FTE by $14,000; included the indigent care assessment ($68,000), utility cost ($108,000) and malpractice insurance cost ($17,000) that had been omitted from the application; increased the supply cost by $618,000; and reduced deductions from revenue by $186,000. But for the charges, utilization levels, and payor mix, Mr. Harris' pro forma is a complete revision of Boca's application pro forma, and demonstrates that such pro forma was not based on reasonable assumptions. Although not based on reasonable assumptions, Mr. Harris opined that such failing is not material since Boca's pro forma, like his pro forma, calculated a profit. Mr. Harris' opinion is rejected. The bottom line profit he derived was based on a substantial change in Boca's proposed program. Such slight of hand does not address the financial feasibility of the program Boca proposed in its application. Boca's proposal, developed through the testimony of its construction, equipment and financial experts, bears little resemblance to its initial application, and must be rejected as an impermissible amendment. Boca's application proposed two operating rooms. As such, Boca could facially handle at least 500 open heart surgery cases per year. As amended, with one operating room, Boca could not reasonably expect to attain such level of operations, given the need to back up its cardiac catheterization program, contrary to Rule 10- 5.011(1)(f)3d. As proposed, Boca's open heart surgery program would include recovery areas and a 10-bed SICU, fully staffed. As amended, the SICU would be staffed with one FTE and other staffing substantially reduced. Through downsizing, Boca would presume to significantly alter its proposal, and thereby demonstrate the reasonableness of its cost and financial feasibility projections. Such was not, however, the proposal submitted to the Department for review, and it cannot be permitted, at this stage of the proceedings, to amend its proposal in such material respects. Accordingly, based on the record, Boca has failed to demonstrate the financial feasibility of its proposal. 11/ St. Mary's pro forma of income and expenses projects 200 adult and 10 pediatric open heart surgery cases during its first year of operation, and 240 adult and 20 pediatric during its second year of operation. Separate pro formas describe the adult and pediatric parts of St. Mary's proposal. Actual charges proposed by St. Mary's will vary by DRG, as will average length of stay. The weighted average charges are, however, projected to be $38,000 for adult services and $43,025 for pediatric services during its first year of operation, and $39,900 for adult services and $45,176 for pediatric services during its second year of operation, based on a 10 day average length of stay. Payor class mix for adults is estimated as follows: Medicare 50 percent, Medicaid 7 percent, self pay/commercial 40 percent, and indigent 3 percent. Payor class mix for pediatrics is estimated to be as follows: Medicare 0 percent, Medicaid 50 percent, self pay/commercial 40 percent, and indigent 10 percent. Net revenue over expenses for its adult program is projected, on an incremental cost basis, to be $2,297,566 for year one, and $2,885,102 for year two. Net revenue for its pediatric program is projected, on an incremental cost basis, to be $62,326 for year one, and $224,797 for year two. St. Mary's proposed charges, average length of stay, utilization levels, payor mix, as well as its assumptions regarding total deductions and expenses are not reasonable. St. Mary's proposed charges were not shown to be reasonably achievable. Rather, where, as here, a facility's charge structure is based on consumption of services, the increased costs associated with an open heart program, discussed infra, would translate into significantly higher charges than those proposed by St. Mary's. St. Mary's application contains no data to reasonably support its conclusions that it will achieve 200 adult cases in year one and 240 adult cases in year two, nor did the proof it offered at hearing demonstrate such potential. Rather, the persuasive proof demonstrated that St. Mary's could not reasonably expect to attract more than 80 adult open heart cases in its first year of operation, and that it would not even be able to attract 200 open heart cases during its third year of operation. Notably, the area St. Mary's proposes to serve is currently adequately served by two open heart surgery programs. St. Mary's pro forma contains several other serious flaws. First, its gross patient revenues are driven by an average length of stay of 10 days. Such assumption is unreasonable, and St. Mary's could more reasonably expect an average length of stay of 15-17 days, with significantly higher expenses associated with the greater consumption of resources occasioned by such increased length of stay. Second, St. Mary's payor mix is significantly understated for Medicare. Here, the proof demonstrates that St. Mary's could reasonably expect to achieve a 68-70 percent Medicare utilization rate, as opposed to the 50 percent it projected. Such increase would significantly reduce its self pay/commercial, assuming its Medicaid and indigent utilization levels are to be accorded any credence, and significantly increase its deductions from revenue. Third, St. Mary's pro forma significantly understated expenses, primarily with regard to supplies and FTEs. Had St. Mary's reasonably calculated its average length of stay at 15-17 days, its expenses for supplies and FTEs would have been substantially higher. Additionally, St. Mary's application only addresses the need to tap incremental FTEs in the nursing area, whereas initation of an open heart program would have a tremendous impact on all services in the hospital, such as lab, pharmacy and social services, with attendant higher costs. Based on the opinion of Richard Cascio, an expert in health care finance, which is credited, St. Mary's proposal is not financially feasible in the long term. 12/ Regarding St. Mary's pediatric open heart program, the proof, as heretofore found, fails to support is utilization projection of 10 cases in year one and 20 cases in year two. Therefore, St. Mary's has failed to demonstrate the long term financial feasibility of that program operated, as proposed, concurrently with an adult program. As a stand alone program, neither St. Mary's application nor the proof at hearing reasonably address such a prospect. However, since the pediatric program was not shown to be financially feasible with the adult program bearing a significant portion of operating expenses, it must also be concluded that the pediatric program would not be financially feasible were it to carry all operating expenses. Martin's pro forma of income and expenses is predicated upon 148 adult open heart surgery cases during its first year of operation, and 195 cases during its second year of operation. Actual charges proposed by Martin will vary by DRG, as will average length of stay. Projected average charges are, however, projected to be $41,000 during its first year of operation and $43,080 during its second year of operation, based on a 15.7 day average length of stay. Payor class mix is estimated as follows: Medicare 63.0 percent, Medicaid 2.5 percent, private pay/commercial insurance 32.5 percent, and free care 2 percent. Net revenue over expenses is projected to be $260,000 for year one and $337,000 for year two. Martin's utilization levels, proposed charges, payor mix, and average length of stay are reasonable. Martin's pro forma did, however, contain some unreasonable assumptions regarding expenses, primarily staffing costs. 13/ Martin's pro forma estimates staffing costs based on the manpower requirements (FTEs) and salaries set forth in Table 11 of its application. It further calculates fringe benefits at 20 percent of salaries. Notably, however, the number of people needed to staff a program at a given FTE level is significantly higher than the raw FTE number. Accordingly, since Martin projected its salary expense and fringe benefits based on FTE's, its expenses associated with those items are understated. Further, the salaries Martin proposed in Table 11 for its operating room nurses are entry level salaries and Martin could not reasonably expect to recruit experienced open heart surgery personnel at such rates. Nor is its projected salary for a perfusionist, at $59,551 reasonable. A more reasonable figure would be in excess of $75,000. Even though the proof offered in opposition to Martin's application did demonstrate that Martin's assumptions regarding salary expenses were understated, it failed to demonstrate that Martin could not meet current market demands and still be profitable. Rather, Martin's proposal, while generating a lower bottom line, will still be profitable if such increased expenses are considered, and it is financially feasible in the long term. While each of the applicant's have demonstrated the immediate financial feasibility of their projects, by demonstrating the availability of funds for project accomplishment and operation, only Martin has demonstrated the long term financial feasibility of its proposal. Other criteria bearing on capital expenditure proposals for the provision of new health services to inpatients. In cases of capital expenditure proposals for the provision of new health services to inpatients, Section 381.705(2), Florida Statutes, requires that the Department reference each of the following in its findings of fact: That less costly, more efficient, or more appropriate alternatives to such inpatient services are not available and the development of such alternatives has been studied and found not practicable. That existing inpatient facilities pro- viding inpatient services similar to those proposed are being used in an appropriate and efficient manner. In the case of new construction, that alternatives to new construction, for example, modernization or sharing arrangements, have been considered and have been implemented to the maximum extent practicable. That patients will experience serious problems in obtaining inpatient care of the type proposed, in the absence of the proposed new service. In the instant case, none of the foregoing criteria can be answered in the affirmative. Rather, the proof demonstrates that less costly, more efficient or more appropriate alternatives currently exist through increased utilization of existing facilities. It further demonstrates that two of the existing three providers have not yet attained a 350 case per year level of operation, and that their services are therefore not yet being used at an appropriate level. Existing utilization levels and capacity further demonstrate that patients will not experience any serious problems in accessing such services. Finally, the applicants further failed to demonstrate that they had considered alternatives to new construction and had implemented them to the maximum extent possible. In the case of all applicants' there is no proof of any effort to initiate sharing arrangements. On the matter of Boca's complaints regarding delays experienced in effecting patient transfers by ambulance, as well as the inadequacy of such ambulances and their breakdowns, it offered no proof that it had investigated other ambulance services or its ability to operate its own service and found them impractable. Notably, such services are an item over which Boca has significant control, and its failure to investigate alternatives in this regard evidences the insignificance of any such problem. The criteria on balance. In evaluating the applications at issue in this proceeding, none of the criteria established by Section 381.705, Florida Statutes, or Rule 10- 5.011(1)(f), Florida Administrative Code, has been overlooked. The applicants' failure to demonstrate need, either numeric or not normal circumstances, as well as their failure to demonstrate compliance with Section 381.705(2), Florida Statutes, is, however, dispositive of their applications, and such failure is not outweighed by any other or combination of any other criteria. Further, even were the fixed need pool accorded the deference suggested by the Department, the other indicators of need subsumed within other criteria would dispel such illusion, and again compel the conclusion that there is no need in this case. Had numeric need been demonstrated, and the need requirements encompassed within section 381.705(2) satisfied, the proof would still fail to support an award to Boca or St. Mary's. Rather, among the competing applicants, Martin was shown to best satisfy the pertinent review criteria on balance and would, under such circumstances, be the favored applicant.
Recommendation Based on the foregoing findings of fact and conclusions of law, it is recommended that a final order be entered denying the applications of Boca, St. Mary's and Martin for a certificate of need to establish an open heart surgery program in District IX. RECOMMENDED in Tallahassee, Leon County, Florida, this 15th day of March 1991. WILLIAM J. KENDRICK Hearing Officer Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-1550 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 15th day of March 1991.
The Issue Whether any of the applications of Oak Hill Hospital, Citrus Memorial Hospital, or Brooksville Regional Hospital for adult open heart surgery programs should be granted?
Findings Of Fact District 3 Extended across the northern half of the state with a reach from central Florida to the Georgia line, District 3 is the largest in land area of the eleven health service planning districts created by the Florida Legislature. See Section 408.032(5), Florida Statutes. Sites of the three hospitals whose futures are at issue in this proceeding are in two of the sixteen District 3 counties: Citrus County and at the southern tip of the district, Hernando County. The three hospitals aspire to join the ranks of District 3's six existing providers of adult open heart surgery programs. Three of the existing providers are in Alachua County, all within the incorporated municipality of Gainesville: Shands at Alachua General Hospital, Shands at the University of Florida, and North Florida Regional Medical Center. Two of the existing providers are in Marion County: Munroe Regional Medical Center and Ocala Regional Medical Center. The sixth provider, opened in November of 1998 as the most recently approved by AHCA in the district, is in Lake County: the Leesburg Regional Medical Center. The CON status of the two Ocala providers is somewhat unusual. Located across the street from each other in downtown Ocala, they share virtually the same medical staff. Pursuant to a Stipulation and Settlement Agreement with the State of Florida, the two have offered adult open heart surgery services since 1987 under a single certificate of need issued for a joint program that reflects their proximity and identity of medical staff. The Agency's view of the arrangement has evolved over the years. It now holds the position that Munroe Regional and Ocala Regional operate independent programs. Accordingly, AHCA lists each as separate programs on its inventory of adult open heart services in District 3. Nonetheless, the two operate as a joint program pursuant to the Settlement Agreement and under state sanction reflected in the agreement, that is, they derive their authority to offer adult open heart surgery services from a single certificate of need. Other than a change of attitude by the Agency, there is nothing to detract from the status they have enjoyed since the agreement reached with the state in 1987: two hospitals operating a joint program under a single certificate of need. The three Gainesville providers all operated at an annual volume of less than 350 procedures during the reporting period that was most current at the time of the filing of the applications by the three competitors in this case. Those competitors are: Citrus Memorial, Oak Hill, and Brooksville Regional. Citrus Memorial, Oak Hill, Brooksville Regional Citrus Memorial Health Foundation, Inc., is a 171-bed, not-for-profit community hospital located in Inverness, Florida. HCA Health Services of Florida, Inc., d/b/a Oak Hill Hospital is a 204-bed hospital located in Oak Hill, Florida. Hernando HMA, Inc., d/b/a Brooksville Regional is a 91- bed hospital located in Brooksville, Florida. Hernando HMA, Inc. (the applicant for the program to be sited at Brooksville Regional) also operates a second campus under a single hospital license with Brooksville Regional. The 75-bed campus is in southern Hernando County in Spring Hill. Citrus and Hernando Counties Citrus Memorial is in Citrus County to the south of the cities of Gainesville and Ocala, the sites of five of the existing providers of adult open heart surgery in the district. Further south, Oak Hill and Brooksville Regional are in Hernando County. Although adjacent to each other along a boundary running east-west, the county line is a natural divide, north and south, with regard to service areas for open heart surgery. Substantially all Citrus County residents, including Citrus Memorial patients, receive open heart surgery and angioplasty services at one of the two Ocala providers to the north. In contrast, almost all Hernando County residents (94 percent) receive open heart services at Bayonet Point, a provider in Health Planning District 5 to the south of Hernando County. The neatness of this divide would be disrupted by the approval of the application of Brooksville Regional. Brooksville's application includes part of south Citrus County in its designated primary service area, an appropriate choice because of Brooksville Regional's location on Route 41 with good access to Citrus County. At present, however, the divide between north and south along the Citrus/Hernando boundary remains a Mason-Dixon line of open heart surgery service areas. During the year ended September 1999, for example, 408 Citrus County residents received open heart surgery in Florida. Of these, 85 percent received them in Ocala at one of the two providers there. During the same period, 618 Citrus County residents underwent angioplasty, with 89.7 percent of them going to the two Ocala providers. During the year ended March 1999, 698 Hernando County residents underwent open heart surgery at Florida Hospitals. Of the 663 residents of Oak Hill's primary service area, 94.3 percent received services at Bayonet Point in District 5. Similarly, of the 779 Oak Hill primary service area residents receiving angioplasty, 93.8 percent went south to Bayonet Point. Brooksville Regional projects that 10 percent of its OHS/angioplasty volume will be from Citrus County. Still, 90 percent of the volume is projected to be from Hernando County. Thus, even with the threat posed by Brooksville's application to the divide at the Citrus/Hernando boundary, the overwhelming percentage of Brooksville's patients will be from south of the Citrus-Hernando boundary. In sum, there is de minimis competition between would- be-provider Citrus Memorial and the providers to the north vis- a-vis would-be-providers Oak Hill and Brooksville Regional and the providers to the south in the arena of open heart surgery services needed by residents of the district. Bayonet Point Under the umbrella of HCA Health Services of Florida, Inc., Bayonet Point is a provider of open heart surgery services in Pasco County. Only thirty minutes by road from its sister HCA facility Oak Hill and 45 minutes from Brooksville Regional, Bayonet Point captures approximately 94 percent of the open heart surgery patients produced among the residents of Hernando County. Although its location is in a county that is only one county to the south of the two Hernando County hospitals, Bayonet Point is in a different health planning district. It is in District 5 on its northern edge. The residents of Hernando County who receive open heart surgery services at Bayonet Point, a premier provider of adult open heart surgery services in the state of Florida, are well served. Operating at far from capacity, the quality of its open heart program is excellent to the point of being outstanding. Position of the Parties re: "not normal" circumstances The Agency's Open Heart Surgery Rule, Rule 59C-1.033, Florida Administrative Code (the "Rule") establishes a need methodology and criteria applicable to review of certificate of need applications for the establishment of adult open heart surgery programs. The Rule also governs a hospital's ability to offer therapeutic cardiac catheterization interventional services (i.e., coronary angioplasty). Pursuant to Rule 50C- 1.032, Florida Administrative Code, a cardiac catheterization program that includes the provision of coronary angioplasty must be located within a hospital that provides open heart services. Applying the methodology of Rule 50C-1.033 (the "Rule"), AHCA determined that a "fixed need pool" of zero existed in District 3 for the July 2002 planning horizon. Calculation under the formula in the Rule produced a fixed need pool of one. Several District 3 programs, however, did not have an annual case volume of 350 or more procedures. The Rule's methodology requires that calculated numeric need be zeroed out whenever there are existing programs in a district with a sub- 350 annual volume. (See Section (7)(a)2., of the Rule.) As required, therefore, the Agency published a numeric need of zero for the applicable planning horizon. The determination of zero numeric need was not challenged and so became final. Their aspirations confronted with a numeric need of zero, Citrus Memorial, Oak Hill and Brooksville Regional, nonetheless, each filed applications seeking the establishment of adult open heart surgery programs. As evidenced by the Agency's initial decision to grant Citrus Memorial's application and by its change of position with regard to Oak Hill's application, the Agency is in agreement that "not normal" circumstances exist to justify granting the applications of both Citrus Memorial and Oak Hill. Thus, while the parties may differ as to the precise identification of those circumstances, all agree that there are circumstances that support the approval of at least one application (and perhaps two) for an adult open heart surgery in District 3 for the July 2002 planning horizon. It is undisputed that a new OHS program in Hernando County would have no effect on the three existing programs located in Gainesville that perform less than 350 procedures annually. This circumstance is a "not normal" circumstance, as previously found by the Agency. It allows an application's approval in the face of the Rule's dictate that the Agency will not normally approve an application when an existing provider falls below the 350 watermark. It is not, however, a circumstance that compels the award of a CON to any of the parties as in the case of "not normal" circumstances typically recognized by the Agency. (An example of such a circumstance would be an access problem for a specific population.) Rather, it is a circumstance that allows the Agency to overcome the zeroing-out effect of the Rule that demanded a fixed-need pool of zero. It is a circumstance that allows AHCA to award an adult open heart surgery CON to one of the Hernando County hospitals provided there is a demonstration of need. There are no typical "not normal" circumstances that support any of the applications. There are no geographic, economic or clinical access problems for the residents of the any of the primary service areas of the three applicants that rise to the level of "not normal" circumstances. Nor would granting the applications of any of the three support cost efficiencies. In the case of Oak Hill, moreover, granting its application would both reduce the operating efficiencies at Bayonet Point and increase the average operating cost per case at Bayonet Point. Approval of an application is not compelled by the "not normal" circumstance that exists in this case. The "not normal" circumstance simply clears the way for approval provided there is a demonstration of need. Stipulated Matters The parties stipulated that all applicants have a good record of providing quality of care and that all sections of the respective applications addressing that issue be admitted into evidence without further proof so as to establish record of quality of care. Accordingly, the parties stipulated that each application satisfies Section 408.035(1)(c) as to "the applicant's record in providing quality of care." The parties stipulated that, subject to proving their ability to generate the open heart surgery and angioplasty volumes projected in their respective applications, each applicant has the ability to provide adequate and reasonable quality of care for those proposed services. Accordingly, subject to the proof involving service volume levels, each application satisfies Section 408.035(1)(c) as the "ability of the applicant to provide quality of care . . .". The parties stipulated that all applicants have available and adequate resources, including health manpower, management personnel, and funds for capital and operating expenditures in order to implement and operate their proposed projects. Furthermore, they stipulated that all sections of their respective applications relating to those proposed projects and all sections of their respective applications relating to those issues were to be admitted into evidence without proof. Accordingly, all applications satisfy that portion of Section 408.035(1)(h), Florida Statutes (1999) related to the availability of resources. The parties stipulated that all applications satisfy, and no further proof is required to demonstrate, immediate financial feasibility as referenced in Section 408.035(1)(i), Florida Statutes (1999). The parties stipulated that the costs and methods of proposed construction, including schematic design, for each proposed project were not in dispute and were reasonable, and that all sections of each application related to those issues were to be admitted into evidence without further proof. (Stip., p.3.) Accordingly, each application satisfies Section 408.035(l)(m), Florida Statutes (1999). The parties stipulated that each application contained all documentation necessary to be deemed complete pursuant to the requirements of Section 408.037, except that Section 408.037(b)3. is still at issue regarding operational financial projections (including a detailed evaluation of the impact of the proposed project on the cost of other services provided by the applicant). The parties stipulated that each applicant satisfied all of the operational criteria set forth in the Rule (those operational criteria being encompassed in subsections 3, 4, and 5). Accordingly, it is undisputed that each applicant will have the support services, operational hours, open heart surgery team mobilization, accreditation, availability of health personnel necessary for the conduct of open heart surgery, and post- surgical follow-up care required by the Rule in order to operate an adult open heart surgery program. The Hernando County Hospitals Oak Hill Oak Hill is located on Highway 50, in the southern part of Hernando County, between the cities of Brooksville and Springhill. Oak Hill's licensed bed compliment includes 123 medical/surgical beds, 24 ICU beds, 50 telemetry beds, and 7 beds for obstetrics. Oak Hill provides an array of medical services and specialties, including: cardiology, internal medicine, critical care medicine, family practice, nephrology, pulmonary medicine, oncology/hematology, infectious disease treatment, neurology, pathology, endocrinology, gastroenterology, radiation oncology, and anesthesiology. Board certification is required to maintain privileges on the medical staff of Oak Hill. Oak Hill's six-story facility is situated on a large campus, and has been renovated over time so that the hospital's physical plant permits the provision of efficient care for patients. Oak Hills's surgery department has five operating rooms, plus a cystoscopy room. The department performs approximately 7,800 surgeries annually, a figure that demonstrates functional efficiency. Oak Hill is JCAHO accredited, with commendation. Recently named one of the nation's top 100 hospitals for stroke care by one organization, it has also received recognition for the excellence of its four intensive care units. Oak Hill's cancer program is the only one to have received full accreditation from the American College of Surgeons within a six-county contiguous area. Oak Hill recently expanded its emergency department and implemented a fast track program called Quick Care. The program is designed to treat lower acuity patients more rapidly. Gallup Organization surveys reflect a 98 percent patient satisfaction rate with the emergency department, the eighth best rate among the approximately 200 HCA-affiliated hospitals. During 1999, the emergency department treated 24,678 patients. During the same period, 376 patients presented to Oak Hill's emergency department with an acute myocardial infarction, and there were 258 such patients during the first eight months of 2000. Oak Hill operates a mature cardiology program with ten Board-certified cardiologists on staff. Eight of the ten perform diagnostic cardiac catheterizations in the hospital's cath laboratory. Oak Hill's program is active with regard to both invasive and non-invasive cardiology. The non-invasive cardiology laboratory offers a variety of services, including echocardiography, holter monitoring, stress testing, electrocardiography, and venous, arterial and carotid artery testing. The invasive cardiology laboratory has been providing inpatient and outpatient cardiac catheterization services since 1991. During calendar year 1999, Oak Hill saw 1,671 diagnostic cardiac catheterization procedures and transferred 619 cardiac patients to Bayonet Point, 258 for open heart surgery, 311 for angioplasty, and 50 patients for cardiac catheterization. The volume of catheterization procedures at Oak Hill has led to the construction of a second "cardiac cath" laboratory suite, scheduled for completion in May of 2001. The cath lab's medical director (Dr. Mowaffek Atfeh, the first interventional cardiologist in Hernando County) has served in that capacity since inception of the lab in 1991. The cath lab equipment is state-of-the-art. Oak Hill's cath lab provides excellent quality of care through its Board-certified cardiologists and the dedication and experience of its well- trained nursing and technical staff. Brooksville Regional Originally a 166-bed facility operated by Hernando County, 75 of the beds at Brooksville Regional were moved in 1991 to create a second facility at Spring Hill. A few years later, the facilities went into bankruptcy. The bankruptcy proceeding concluded in 1998, with operational control of both facilities being acquired by Hernando HMA, Inc. ("Hernando HMA"). The CON applicant for the adult open heart surgery program to be sited at Brooksville Regional, Hernando HMA is a wholly-owned subsidiary of Health Management and Associates, Inc. ("HMA"), a corporation located in Naples, Florida, and whose shares are traded publicly. Under the arrangement produced by the bankruptcy proceeding, Hernando County retained ownership of the buildings and the land. Hernando HMA, in turn, operates the facilities per a long-term lease with the County. Hernando HMA operates the Brooksville Regional and Spring Hill Campuses under a single hospital license issued by AHCA. The two campuses therefore share key administrative staff, including their chief executive officer. They share a single Medicare provider number and they have a common medical staff. HMA (Hernando HMA's parent) operates 38 hospitals throughout the country, many in the State of Florida. Among the 38 is Charlotte Regional Medical Center in Charlotte County, an existing provider of adult open heart surgery and recently recognized as one of the top 100 OHS programs in the country. Charlotte Regional will be able to assist Brooksville Regional with staff training and project implementation if its application is approved. An active participant in managed care contracting, Hernando HMA is committed to serving all payer groups, including Medicaid and indigent patients. It recently qualified as a Medicaid disproportionate share provider. It also serves patients without ability to pay. In fiscal year 2000, it provided $5 million of indigent care. Under the lease agreement Hernando HMA has with Hernando County, it must continue the same charity care policies as when the facilities were operated by the County. Hernando HMA must report annually to the County to show compliance with this charity care obligation. Also under the lease, Hernando HMA is obliged to invest $25 million in renovations and improvements to the two facilities over a 5-year period. About $10 million has already been invested. If the adult open heart surgery program is granted this would nearly satisfy the $25 million obligation. The County reserves to itself certain powers under the lease. For example, the County reserves the authority to pre- approve the discontinuation of any services currently offered at these facilities. Also, if Hernando HMA seeks to relocate either of the two, the County retains the authority whether to approve the relocation. The Spring Hill facility is located in the southwest portion of Hernando County, very near the Pasco County line. It is a general acute care facility, offering a full range of cardiology and other acute care services. Spring Hill was recently approved to add the tertiary service of Level II Neonatal Intensive Care. The Brooksville facility is located in the geographic center of Hernando County. Its service area is all of Hernando County and southern Citrus County. Brooksville is a full- service, general acute care facility. It offers services in cardiology, orthopedics, general surgery, pediatrics, ICU, telemetry, gynecology, and other acute services. Brooksville Regional has 91 acute care beds. Normally, the beds are used as 12 ICU beds, 24 telemetry beds, and 55 medical/surgical beds. During its peak annual period of occupancy, Brooksville has the capability to use up to 40 beds for telemetry purposes. The hospital has ample unused space and facilities associated with its 91 beds that resulted from the move of the 75 beds to create the Spring Hill campus. Brooksville Regional offers full scope cardiology services and technologies, including diagnostic cardiac catheterization. Just as in the case of Oak Hill, the cardiac cath lab is state-of-the-art. The only cardiac services not offered at the hospital are open heart surgery and angioplasty. The quality of cardiology and related services at Brooksville Regional are excellent. The equipment, the nursing staff, the allied health professional staff, and the technology support services are very good. The medical staff is broad- based and highly qualified. Brooksville Regional offers substantial educational and training programs for its nursing staff and other personnel on staff. Brooksville Regional routinely treats patients in need of OHS or angioplasty services. Nearly 400 patients per year receive a diagnostic cardiac cath at Brooksville Regional and are then transferred for open heart surgery or angioplasty. The vast majority of these patients are transferred to Bayonet Point, about 45 minutes away. In addition to transfers of patients following diagnostic catheterization, Brooksville Regional transfers about 120 patients per year to Bayonet Point who have not had such services. These patients fall into two categories: (1) high- risk patients, and (2) persons presenting at Brooksville's emergency room in need of angioplasty or open heart surgery. The Proposals Citrus Memorial By its application, Citrus Memorial proposes to establish a program that will provide adult open heart surgery and angioplasty services. There is no dispute that Citrus Memorial has the ability to provide adequate and reasonable quality of care for the proposed project (just as per the stipulation of the parties, there is no dispute that all of the applicants have such ability.) There is also no dispute that each applicant, including Citrus Memorial, will have all of the staff, equipment and other resources necessary to implement and support adult open heart surgery and angioplasty services. The ability to provide high quality care stems, in part, from Citrus Memorial's contract with the Ocala Heart Institute. Under the contract the Institute will provide supervision of the implementation and ongoing operations of the Citrus Memorial program. This supervision will be provided under the leadership of the president of the Institute, cardiovascular surgeon Michael J. Carmichael, M.D. The contract between Citrus Memorial and the Ocala Heart Institute is exclusive. Citrus Memorial will not extend medical staff privileges to any cardiovascular surgeon not affiliated with the Ocala Heart Institute unless approved by the Institute. The Ocala Heart Institute (whose physician members include not only cardiovascular surgeons, but also cardiovascular anesthesiologists and invasive cardiologists) has similar exclusive contracts for the operation of adult open heart surgery programs at Monroe Regional Medical Center and at Ocala Regional Medical Center and at Leesburg Regional Medical Center. At these three hospitals, the Institute's physicians have consistently produced excellent outcomes. The Ocala Heart Institute produces these results not just through the skills of its physicians but also through the use of the same clinical protocols at each hospital governing the provision of open heart surgery. Citrus Memorial proposes to follow identical protocols at its facility. Excellent open heart surgery outcomes for the Institute's physicians are also the product of standardized facility design, equipment and supplies. The standardization of design, equipment, supplies, and protocols has the added benefit of clinical efficiencies that reduce costs and shorten lengths of stay. Beyond supervision of the initial implementation of the program, the Ocala Heart Institute will provide the medical directorship for Citrus Memorial's program. In cooperation with Munroe Regional, the directorship's 24-hour-a-day, 7-days-a-week coverage of the program will include scheduled case, emergency case, and backup coverage by cardiovascular surgeons, cardiovascular anesthesiologists, perfusionists, and interventional cardiologists. The Ocala Heart Institute will provide education and training to Citrus Memorial's medical staff and other hospital personnel as appropriate. The Institute's obligations will include continually working to improve the quality of, and maintain a reasonable cost associated with, the medical care furnished to Citrus Memorial's open heart surgery and angioplasty patients, consistent with recognized standards of medical practice in the field of cardiovascular surgery. The contract with the Ocala Heart Institute ensures to the extent possible that Citrus Memorial will have a high- quality adult open heart surgery program. Oak Hill Through approval of its application to establish an adult open heart surgery program at its facility, Oak Hill hopes Hernando County residents who now must travel outside the county to receive open heart and angioplasty services will be better served. In particular, Oak Hill hopes to provide these services to the residents of the six zip code area that comprise its primary service area ("PSA"). Containing 75 percent of the county's population, Oak Hill's PSA also encompasses the county's concentration of recent growth. Oak Hill's administration is committed to the proposal contained in its application. It has the support of the hospital's Board of Trustees and medical staff. Not surprisingly, the proposal enjoys a measure of popularity in the county. A petition in support of a program at Oak Hill drew 7,628 signatures from residents of Hernando County. This popularity is based in the fact that residents now must leave District 3 (albeit Bayonet Point in District 5 is close to Oak Hill and closer for many residents of south Hernando County) to receive open heart and angioplasty services. The number of affected residents is substantial. In 1999, for example, over 600 cardiac patients were transferred by ambulance from Oak Hill to Bayonet Point. A greater number of patients traveled on a scheduled basis to Bayonet Point for cardiac care. The vast majority of Hernando County residents and Oak Hill primary service area residents in need of OHS services receive them at Regional Medical Center-Bayonet Point. HCA Health Services of Florida, a subsidiary of HCA-The Healthcare Company ("HCA") holds the Bayonet Point license. It also is the licensee of Oak Hill and other hospitals in Florida including North Florida Regional and Ocala Regional. Bayonet Point (Regional Medical Center-Bayonet Point) is an acute care hospital in Hudson. Hudson is in Pasco County, the county immediately to the south of Hernando County. Although in a separate health planning district (District 5), Bayonet Point is relatively close to Oak Hill, 17 miles to the south. Bayonet Point's open heart surgery program experiences the fourth highest case volume in the state. The program is recognized as one of the top two programs in the state. It enjoys a national reputation. For example in July of 1999, it was ranked 50th in the nation in cardiology and heart surgery in U.S. News and World Report's list of "America's Best Hospitals." Oak Hill, as a sister hospital of Bayonet Point under the aegis of HCA, plans to develop its program in cooperation with Bayonet Point and its cardiovascular surgeons so as to bring the high quality program at Bayonet Point to Oak Hill's community and patients. A prospective operational plan for the adult open heart surgery program has been initiated by Oak Hill with assistance from Bayonet Point. Oak Hill, unlike Citrus Memorial, did not present evidence concerning the specific duties to be imposed on each physician group under contract. Nor did Oak Hill present evidence as to whether and how those groups would create and implement the type of standardization of protocols, facility design, equipment, and supplies that Citrus Memorial's program will rely upon for high quality and reduced costs. Nonetheless, it can be expected that the cooperation of Oak Hill and Bayonet Point, as sister HCA hospitals, will continue through the development and implementation of appropriate staff training, policies, procedures and protocols in the establishment of a high quality program at Oak Hill. Oak Hill's achieved volume in its open heart surgery program, if approved, will be at the direct expense of Bayonet Point. Its approval will increase the operating costs per case at Bayonet Point. Patients transferred from Oak Hill to Bayonet Point for OHS and angioplasty receive excellent outcomes. Patients are transferred to Bayonet Point for OHS and angioplasty smoothly and without delay particularly because Bayonet Point operates a private ambulance system for the transport of cardiac patients to its hospital. Two groups of cardiovascular surgeons are the exclusive cardiovascular/thoracic surgeons at Bayonet Point. Although, at present, there are no capacity constraints at Bayonet Point, both groups support a program at Oak Hill and are committed to participate in an open heart surgery program at Oak Hill. If approved, Oak Hill will enter similar exclusive contracts with the two groups. Raymond Waters, M.D., a cardiovascular surgeon, heads one of the groups. He has performed open heart surgery at Bayonet Point since its inception and is largely responsible for the development of the surgery protocols used there. Dr. Waters has consulting privileges at Oak Hill. In addition to consulting there, Dr. Waters presents medical education programs at Oak Hill. Forty to 50 percent of Dr. Waters' patients come from Hernando County and Oak Hill Hospital. Dr. Waters and his group strongly support initiation of an open heart surgery ("OHS") program at Oak Hill. Their support is based, in part, on the excellence of the institution, including its physical structure, cath labs, intensive care units, nursing staff, medical staff, and the state of its cardiology program. Dr. Waters and his group are prepared to assist in the development of an open heart surgery program at Oak Hill, and to assure appropriate surgery coverage. Oak Hill will create a Heart Center at the hospital to house its OHS program. All diagnostic and invasive cardiac services will be located in one area of the hospital to ensure efficient patient flow and access to support services. The center will occupy existing space to be renovated and newly constructed space on the first floor of the facility. Two new cardiovascular surgery suites, with all support spaces necessary, will be constructed, along with an eight-bed cardiovascular intensive care unit. The hospital's two state- of-the-art cardiac catheterization laboratory suites are available for diagnostic procedures and angioplasty procedures. A large waiting area and cardiac education/therapy room will also be constructed. Open heart surgery patients will progress from the OR to the new CVICU for the first 24-28 hours after surgery. From the CVICU, the patient will be admitted to a thirty-bed telemetry monitored progressive care unit, located on the second floor. Currently a 38-bed medical/surgical unit, thirty of the beds will remain as PCU beds. Eight beds will be relocated to create the CVICU. The PCU will provide continued care, education and discharge planning for post open heart surgery and angioplasty patients. Oak Hill will also implement a comprehensive cardiac rehabilitation program for both inpatients and outpatients. Brooksville Regional Like Oak Hill, part of the purpose of the Brooksville Regional proposal is to provide more convenient OHS and angioplasty services to Hernando County residents in need of them, 94 percent of whom now travel to Bayonet Point in Pasco County for such services. In addition to proposing improvements in patient convenience and access, Brooksville Regional sees its application as increasing patient choice and competition in the delivery of the services. Indeed, patient choice and competition for the benefit of patients, physicians and payers of hospital services are the cornerstone of Brooksville Regional's application. There is support for the proposed program from the community and from physicians. For example, Dr. Jose Augustine, a cardiologist and Chief of the Medical Staff at Oak Hill since 1997, wrote a letter of support for an open heart program at Brooksville Regional. Although he believes Hernando County would be better served by a program at Oak Hill, he wrote the letter for Brooksville Regional because, "if Oak Hill didn't get it, [he] wanted the program to be here in Hernando County." (Oak Hill No. 12, p. 43.) Consistent with his position, Dr. Augustine finds Brooksville Regional to be an appropriate facility in which to locate an open heart program and he would do all he could to support such a program including providing support from his cardiology group and encouraging support other physicians. But Brooksville Regional offered no evidence regarding the identity of its cardiovascular surgeons. Hernando HMA proposes to construct a state-of-the-art building of 19,500 square feet at Brooksville Regional to house its OHS program. Two OHS operating rooms will be built. Eight CVICU beds will be used for the program, to be converted from other licensed beds. A second cath lab will be added. The total project cost is nearly $12 million. Brooksville Regional proposes to serve all of Hernando County. In addition, 10 percent of its volume is expected to come from Citrus County. Brooksville Regional commits to serving all payer groups with the vast majority projected to be Medicare, Medicare HMO/PPO and non-Medicare managed care. Brooksville lists two specific CON conditions in its application. First, it commits to over 2 percent for charity care and 1.6 percent for Medicaid. Second, it commits to establishing the OHS program at Brooksville's existing facility, located at 55 Ponce de Leon Boulevard in the City of Brooksville. The second of these two was reaffirmed unequivocally at hearing when Brooksville introduced testimony that if Brooksville's CON application is approved, its OHS program will be located at Brooksville's existing facility. Need In Common One "not normal" circumstance exist that supports all three applications: the lack of effect any approval will have on the sub-350 performers in the district. Which, if any, of the three applicants should be awarded an adult open heart surgery program, therefore, is determined on the basis of need and that determination is to be made in the context of comparative review. Benefits of Increased Blood Flow Lack of blood flow to the heart caused by narrowed arteries or blood clots during a heart attack, results in a loss heart of muscle. The longer the blood flow is disrupted or diminished, the more heart muscle is lost. The more heart muscle lost, the more likely the patient will either die or, should the patient survive, suffer a severe reduction in the quality of life. The key to prevent the loss of heart muscle in a heart attack is to restore blood flow to the heart through a process of revascularization as quickly as possible. Cardiovascular surgeons and cardiologists make reference to this phenomenon through the maxim, "time is muscle." The faster revascularization is accomplished the better the outcome for the patient. Those who treat heart attack patients seek to restore blood flow within a half hour of the onset of the attack. Revascularization within such a time frame maximizes the chance of reducing permanent damage to the heart muscle from which the patient cannot recover. Achievement of revascularization between 30 minutes and 90 minutes of the attack results in some damage. Beyond 90 minutes, significant permanent damage resulting in death or severe reduction in quality of life is likely. The three primary treatment modalities available to a patient suffering from a heart attack are: 1) thrombolytics; 2) angioplasty and 3) open heart surgery. Thrombolytic therapy is the standard of care for the initial attempt to treat a heart attack. Thrombolytic therapy is the administration of medication, typically tissue plasminogen ("TPA") to dissolve blood clots. Administered intravenously, the thrombolytic begins working within minutes in an attempt to dissolve the clot causing the heart attack and, therefore, to prevent or halt damage to the heart muscle. Thrombolytic therapies are successful in restoring blood flow to the affected heart muscle about 60 to 75 percent of the time. In the event it is not successful or the patient is not appropriate for the therapy, the patient is usually referred for primary angioplasty, a therapeutic cardiac catheterization procedure. Cardiac catheterization is a medical procedure requiring the passage of a catheter into one or more cardiac chambers with or without coronary arteriograms, for the purpose of diagnosing congenital or acquired cardiovascular diseases, and includes the injection of contrast medium into the coronary arteries to find vessel blockage. See Rule 59C-1.032(2)(a), Florida Administrative Code. Primary angioplasty is defined as a therapeutic cardiac catheterization procedure in which a balloon-tipped catheter inflated at the point of obstruction is used to dilate narrowed segments of coronary arteries in order to restore blood flow to the heart muscle. Rule 59C-1.032(2)(b), Florida Administrative Code. More often now, in the wake of cardiac care advances, a "stent" is also placed in the re-opened artery. A stent is a wire cylinder or a metal mesh-sleeve wrapped around the balloon during an angioplasty procedure. The stent attaches itself to the walls of the blocked artery when the balloon is inflated, acting much like a reinforced conduit through which blood flow is restored. Its advantage over stentless angioplasty is improved blood flow to the heart and a reduction in the likelihood that the artery will collapse in the future. In other words, a stent may prevent substantial re-occlusion. The development of stent technology has led to dramatically increased angioplasty procedure volumes in recent years and the trend is continuing. Based on mortality rates, studies suggest that immediate angioplasty, rather than thrombolytic treatment, is the preferred treatment for revascularization. When thrombolytic therapy is inappropriate or fails and a patient is determined to be not a candidate for angioplasty, the patient is referred for open heart surgery. Under the Open Heart Surgery Rule, Rule 59C-1.032, Florida Administrative Code, a cardiac catheterization program that includes the provision of angioplasty must be located within a hospital that also provides open heart surgery services. Open heart surgery is a necessary backup in the event of complications during the angioplasty. The residents of Citrus Memorial's primary service area (and those of Oak Hill's and Brooksville Regional's), therefore, do not have immediate access (that is access to a hospital in their county of residence) to not just open heart surgery services but to angioplasty services as well. In addition to increased benefits to the residents of the proposed service areas, much of the need in this case is based on a demonstration of geographic access problems. For example, population concentration and historical utilization of open heart surgery services in the district demonstrate that the open heart surgery programs in the district are maldistributed. At the same time, the Bayonet Point program's service by virtue of both superior quality and proximity to Hernando County ameliorates the effect of the maldistribution of the programs intra-district particularly with regard to the residents of Hernando County. The four southernmost of the 16 counties in the district (Citrus, Hernando, Sumter and Lake) account for approximately 41 percent of the total adult population and 53.5 percent of the population aged 65 and over within District 3 as a whole. The super majority of aged 65 and over population in these counties is of great significance since that population is the primary base of those in need of adult open heart surgery and angioplasty. This same base accounts for 57 percent of the total annual open heart surgeries performed on district residents. For District 3 as a whole, 27 percent of the adult population is aged 65 and older. In comparison, 38.2 percent of Citrus County residents fall within that age cohort, 37.2 percent of Hernando County residents and 33.3 percent of residents in Lake and Sumter Counties combined fall within that age cohort. In contrast, in the northern part of the district, the counties closest to the three Gainesville open heart surgery programs (Columbia, Hamilton, Suwanee, Alachua, Bradford, Dixie, Gilchrist, Lafayette, Levy, and Union) contain a combined basis of 32.4 percent and Putnam County contains 24.7 percent of the District 3 population aged 65 and over. The overall District 3 open heart surgery use rate (number of surgeries per 1,000 population age 15 and over) is 3.47. Yet, the combined use rate for Columbia, Hamilton, and Suwanee Counties is 1.96, the combined use rate for Alachua, Bradford, Dixie, Gilchrist, Lafayette, Levy, and Union Counties is 1.55, and the Putnam County use rate is 2.05. More specifically, the northern county use rates are significantly below the use rates for the remainder of District 3 counties. Marion County is 4.12. Citrus County is at 4.26. Hernando County is at 6.41. Lake and Sumter Counties are at 4.31. Transfers Drive time is but one component of the total time necessary to effectuate a patient transfer. Additional time is consumed in making transfer and admission arrangements with the receiving hospital, awaiting arrival of an ambulance to begin transport, and preparing and transferring the patient into and out of the ambulance. Time delays that necessarily accompany hospital-to-hospital transfers can be critical, clinically. The fact that a facility-to-facility transfer is required means that the patient is at relatively high risk. Otherwise, the patient would be sent home and electively scheduled later. The need to travel outside the community carries other adverse consequences for patients and their families. Continuity of care is disrupted when patients cannot receive hospital visits from their regular and trusted physicians. Separation from these physicians increases stress and anxiety for many patients, and patients heal better with lower levels of stress and anxiety. Further, most OHS patients are elderly, and travel by their spouses to another community to visit is stressful and difficult at best, sometimes impossible. The elderly loved ones of the patient also tend to have health problems and, even when able, the drive to the hospital is stressful. District 3 Out-migration A high volume of OHS patients leave District 3 for OHS services. During the year ended March 1999, there were a total of 3,520 District 3 residents discharged from Florida hospitals following OHS. Only 2,428 of those OHS cases were reported by hospitals located within District 3. An outmigration rate of 31 percent, on its face, is indicative of a district geographic access problem. The problem is mitigated, however, by an understanding that most of the outmigration is of Hernando County residents who are able to travel or are transferred to Bayonet Point, a provider within 30 to 45 minutes driving time from the two Hernando County applicants in this proceeding. Citrus Memorial Volume Projections and Financial Feasibility Citrus Memorial reasonably projects an open heart surgery case volume of 266 for the first year of operation, 313 for the second year, and 361 for the third year. Citrus Memorial reasonably projects an angioplasty case volume of 409 for the first year of operation, 481 for the second year, and 554 for the third year. The Citrus Memorial program is financially feasible in the long term. It will generate approximately $1 million in not-for-profit income by the end of the second year of operation ($327,609 from open heart surgery cases, and $651,323 from angioplasty cases). Increased Access in Citrus County The two Ocala hospitals are approximately 30 miles from Citrus Memorial. With traffic, the normal driving time from Citrus Memorial to the hospitals is 60 minutes. The driving time from Oak Hill to Bayonet Point is normally 29 minutes or about half the time it takes to get from Citrus Memorial to one of the Ocala providers. The drive time from Brooksville Regional to Bayonet Point is approximately 45 minutes, 25 percent faster than the driving time from Citrus Memorial to the Ocala hospitals. Myocardial infarction patients for whom thrombolytic therapy is inappropriate or ineffective who present to the emergency room at Citrus Memorial, on average, therefore, are exposed to greater risk of significant heart muscle damage than those who present to the emergency rooms at either Oak Hill or Brooksville Regional. The delay in transfer for a Citrus Memorial patient in need of angioplasty or open heart surgery can be compounded by the ambulance system in Citrus County. There are only 7 ambulances in the system. If one is out of the county, the provider of ambulance services will not allow another to leave the county until the first has returned. Citrus Memorial presented medical records of 17 cases in which transfers took more than an hour and in some cases more than 3 hours from when arrangements for transfers were first made. There was no testimony to explain the meaning of the records. Despite the status of the records as admissible under exceptions to the hearsay rule and therefore the ability to rely on them for the truth of the matters asserted therein, the lack of expert testimony diminishes the value of the records. For example in the first case, the patient presented at the emergency room on June 14, 1999. Treatment reduced the patient's chest pain. In other words, thrombolytics appeared to be beneficial. The patient was admitted to the coronary care unit after a diagnosis of unstable angina, and cardiac catheterization was ordered. On June 15, the next day, at about 11:40 a.m., "just prior to going down to Cath Lab, patient developed severe chest pain." (Citrus Memorial Ex. 16, p. 1017.) Following additional treatment, the chest pains were observed half an hour later to be "better." (Id.) Several hours later, at 1:45 p.m., that day, transfer to Ocala Regional was ordered. (Id., p. 1043). The patient's progress notes show that the transfer took place at 3:45 p.m., two hours after the order for transfer was entered. Whether rapid transfer was required or not is questionable since the patient appears to have been stabilized and had responded to thrombolytics and other therapy. In contrast, the second of the 17 cases is of a patient whose "risk of mortality [was] . . . close to 100%." The physician's notes indicate that at 1:10 p.m. on August 8, 1999, "emergency cardiac cath [was] indicated [with] a view toward revascularization." (Citrus Memorial Ex. 16, p. 1093). The same notes indicate after discussion between the physician and the patient and his spouse "that transfer itself is risky, but that risk of mortality [if he remained at Citrus Memorial] . . . is close to 100 percent." Although these same notes show that at 1:10 p.m., the patient's transfer had been accepted by the provider of open heart surgery, it was not until 3:30 p.m., that the "Ocala team" (id., at 1113) was shown to be present at Citrus Memorial and not until 3:45 p.m., that the patient was "transferred to Ocala." (Id.) Given the maxim that "time is muscle," it may be assumed that the 2-hour and 45- minute delay in transfer from the moment the patient was accepted for transfer until it occurred and the ensuing time thereafter for the drive to Ocala contributed to significant negative health consequences to the patient. Whatever the value of the 17 sets of medical records, they demonstrate that transfers from Citrus Memorial on occasion take up time that is outside the 30-minute and 90-minute timeframes for avoiding significant damage to heart muscle or minimizing such damage to heart attack patients for whom angioplasty or open heart surgery procedures is indicated. Citrus Memorial also presented twenty sets of records from which the "emergent" nature of the need for angioplasty or open heart intervention was more apparent from the face of the records than in the 17 cases. (Compare Citrus Memorial Ex. No. 16 to No. 17). These records reveal transport delays in some cases, lack of immediate bed ability at the Ocala hospitals in others, and in some cases both transport delays and lack of bed availability. In 16 of the cases, it took over 90 minutes for the patient to reach the receiving hospital and in 13 of the cases, it took 2 hours or more. It would be of significant benefit to some of those who present to Citrus Memorial's emergency room with myocardial infarctions to have access to open heart surgery services on site should thrombolytic therapy be inappropriate or prove ineffective. Other Access Factors Besides time considerations, there are other factors that provide comparisons related to access by Citrus Memorial service area residents on the one hand and Hernando County residents to be served by either Oak Hill or Brooksville Regional on the other. Among the other factors relied on by Citrus Memorial to advance its application is a comparison of use rate. The use rate per 1,000 population aged 15 and over for Hernando County is 6.08, compared to 4.13 for Citrus County. "[B]y definition" (tr. 458), the use rates show need in Hernando County greater than in Citrus County. But the use rates could indicate an access problem financially or geographically. In the end, there are a lot of components that make up the use rate. One is obviously the age of the population and underlying heart disease, two, . . . is the physician practice patterns in the county. [S]tudies . . . show that [in] two equivalent populations, . . . one with a very conservative medical community that . . . hospitalizes more frequently . . . [versus] another . . . where the physicians hospitalize less frequently for the same situation or who use a medical approach versus a surgical approach. (Id.) While there may be one possible explanation for the lower use rate in Citrus County than in Hernando County that favors Citrus Memorial, a comparison of use rates on the state of this record is not in Citrus Memorial's favor. Other factors favor Citrus Memorial. In support of its open heart surgery and angioplasty volumes, for example, Citrus Memorial reasonably projects an 80 percent market share for such services from its primary service areas. In contrast, Oak Hill projected a much lower market share from its primary service area: 58 percent. The lower market share projection by Oak Hill is due to the proximity of the Bayonet Point program to Hernando County. The difference in the two projections reveals greater demand for improved access in Citrus County than in Hernando County. This same point is revealed by projected county outmigration. Statewide data reveals that the introduction of open heart surgery services within a county causes a county resident generally to stay in the county for those services. Yet with a new program in Hernando County, Bayonet Point is still projected reasonably to capture one-half of the open heart surgeries and angioplasties performed on Hernando County residents, further support for the notion that Hernando County residents have adequate access to open heart surgery services through Bayonet Point's program. As to angioplasty demand, Oak Hill projected an angioplasty/open heart surgery ratio of 1.3. Citrus Memorial's ratio is 1.5. Geographic access limitations also adversely affect continuity of care. To have open heart surgery performed at another hospital, the patient will have to travel for pre- operative, operative, and post-operative follow-up services and duplication of tests. This lack of continuity of care often results in the patient's primary and specialty care physicians not following the patient and not being involved with all phases of care. In assessing travel time and access issues for open heart surgery and angioplasty services, travel time and distance present not only potential hardship to the patient, but also to the patient's family and friends who accompany and visit the patient. These issues are of particular significance to elderly persons (be they the patient, family member or friend) who do not drive and must rely on others for transport. Financial Access - Indigent Care Consistent with its mission as a community not-for- profit hospital, Citrus Memorial will accept any patient who comes to the hospital regardless of ability to pay. In 1999, Citrus Memorial provided approximately $4.9 million in charity care, representing 3.6 percent of its gross revenues. Citrus County provided Citrus Memorial with $1.2 million dollars in subsidization, part of which was allotted to capital construction and maintenance, part of which was allotted to charity care. Subtracting all $1.2 million, as if all had been earmarked for charity care, from the charity care, the dollar amount of Citrus Memorial's out-of-pocket charity care substantially exceeds the dollars for the same period provided by Oak Hill ($1.3 million) and by Brooksville Regional ($935,000). The percentage of gross revenue devoted to charity care is also highest for Citrus Memorial; Brooksville Regional's is 1.1 percent and tellingly, Oak Hill's, at 0.6 percent is less than one-quarter of Citrus Memorial's percentage of out-of- pocket charity care. "[C]learly Citrus has a much stronger charity care credential than does either Oak Hill or Brooksville Regional." (Tr. 241). But this credential does not carry over into the open heart surgery arena. As a condition to its CON, Citrus Memorial committed to a minimum 2.0 percent of total open heart surgery patient days to Medicaid/charity patients. The difference between Citrus Memorial's commitment and that of Oak Hill's and Brooksville Regional's, both standing at 1.5 percent, is not nearly as dramatic as past performance in charity care for all services. The difference in the comparison of Citrus Memorial to the other applicants between past overall charity care and commitment to future open heart services for Medicaid and charity care is explained by the population that receives open heart and angioplasty services. That population is dominated by those over 65 who are covered by Medicare. Competition Citrus Memorial's current charges for cardiology services are significantly lower than comparable charges at Oak Hill or Brooksville Regional. A comparison of the eight cardiology-related DRGs that typically have high volume utilization reveals that Oak Hill's gross charges are 62 percent greater than Citrus Memorial's gross charges. A comparison of gross charges is not of great value, however, even though there are some payers that pay billed charges such as "self-pay" and indemnity insurance. When managed care payments are a function of gross charges then such a comparison is of more value. On a net revenue per case basis for those DRGs, Oak Hill's net revenues are 10 percent greater than Citrus Memorial's. A 10 percent difference in net revenues, a much narrower difference than the difference in gross charges, is significant. Furthermore, it is not surprising to see such a narrowing since most of the utilization is covered by Medicare which makes a fixed payment to the provider. A comparison of projections in the applications reveals that Oak Hill's gross revenue per open heart surgery cases will be 164 percent greater than Citrus Memorial's gross revenue per such case. Oak Hill's net revenue per open heart surgery case will be 32 percent greater than Citrus Memorial's net revenue per such case. A comparison of projections in the applications also reveals that Oak Hill's gross revenue per angioplasty case will be 74 percent greater than Citrus Memorial's and that Oak Hill's net revenues per angioplasty case will be 13 percent greater than Citrus Memorial's. If a program is established at Oak Hill, there will be a hospital within District 3 with a new open heart surgery program. But what Oak Hill, under the umbrellas of HCA, proposes to do in reality is to take a quarter of the volume from [Bayonet Point, a] premier facility to set up in a sense a satellite operation at a facility . . . 16 miles away . . . [when] those patients already have an established practice of going to the premier tertiary facility . . . [ and when the two enjoy] a very strong positive relationship. (Tr. 1434). Such an arrangement will do little to nothing to enhance competition. Comparing Citrus Memorial and Brooksville Regional gross revenues on the basis of the same cardiology-related DRGs reveals that Brooksville's gross charges are 83 percent greater than Citrus Memorial's charges. A comparison of projections in the applications reveals that Brooksville Regional's gross revenue per open heart surgery case will be 147 percent greater than Citrus Memorial's and the Brooksville's net revenue per open heart surgery case will be 45 percent greater than Citrus Memorial's. A comparison of projections in the applications reveals that Brooksville's gross revenue per angioplasty case will be 36 percent greater than Citrus Memorial's and that Brooksville's net revenue per angioplasty case will be 7 percent lower than Citrus Memorial's. Impact of a Citrus Memorial Program on Existing Providers Citrus Memorial reasonably projected that by the third year of operation, a Citrus Memorial program will take away 100 cases from Ocala Regional. In 1999 Ocala Regional had an open heart surgery volume of 401 cases. In 2000, its annual volume was 18 cases more, 419. This is a decline from both the immediately prior two-year period, 1997 to 1998 and the two-year period before that of 1995 to 1996. The volume decline for the two-year period 1999 to 2000 compared to the previous two-year period, 1997 to 1998 is not at all surprising because of "two big factors." (Tr. 97). First, in 1997 and 1998, Ocala Regional was used as a training site for the development of Leesburg Regional's open heart surgery program that opened in December of 1998. In essence, Ocala Regional enjoyed an increase in the volume of cases in 1997 and 1998 when compared to previous years and a spike in volume when compared to both previous and subsequent two-year periods because of the 1997-98 short-term "windfall.) (Id.) Second, Ocala Regional was a Columbia-owned facility. In 1999 and thereafter, "Columbia developed a lot of bad publicity because of some federal investigations that were going on of the Columbia system." (Id.) The publicity negatively affected the hospital's open heart surgery volume in 1999 and 2000. The second factor also helps to explain why Ocala Regional's volume in 1999 and 2000 was lower than in 1995 and 1996. There are other factors, as well, that help explain the lower volume in 1999 and 2000 than in 1995 and 1996. In any event if impact to Ocala Regional, alone, were to be considered for purposes of the prohibition in Rule 59C- 1.033(7)(c), that a new program will not normally be approved if approval would reduce 12-month volume at an existing program below 350, then the impact might result in veto by rule of approval of a program at Citrus Memorial. But Ocala Regional is but one hospital under a single certificate of need shared with another hospital across the street from its facility: Munroe Regional. Annualization for 1999 of discharge data for the 12 months ending September 30, 1999 shows that Munroe Regional enjoyed a volume of 770 cases. There is no danger that the program carried out by Ocala Regional and Munroe Regional jointly under a single certificate of need will fall below 350 procedures annually should Citrus Memorial be approved. Oak Hill Need for Rapid Interventional Therapies and Transfers A high number of residents of Oak Hill's proposed service area present to its emergency room with myocardial infarctions. Many of them would benefit from prompt interventional therapies currently made available to them at Bayonet Point. Over 600 patients annually, almost two patients every day, must be transferred by ambulance from Oak Hill to Bayonet Point for cardiac care. A significant number of them would benefit from interventional therapy more rapidly available. The travel time from Oak Hill to Bayonet Point is the least amount of time, however, of the travel time from any of the three applicants in this proceeding to the nearest existing open heart provider; Brooksville Regional to Bayonet Point or Citrus Memorial to one of the Ocala providers. The extent of the benefit, therefore, is difficult to quantify and is, most likely, minimal. As with the other two applicants, thrombolytic therapy is the only method of revascularization currently available to Oak Hill's patients because Oak Hill is precluded by Agency rule and clinical standards from offering angioplasty without on-site open heart surgery backup. The percentage of MI patients who are ineligible for thrombolytic therapy, coupled with the percentages of patients for whom thrombolytic therapy is ineffective, are extremely significant given the high number of MI patients presenting to Oak Hill's emergency room. During 1998, 418 patients presented to Oak Hill's ER with an MI, and 376 MI patients presented in 1999. During the first eight months of 2000, 255 MI patients presented to Oak Hill's ER, an annualized rate of 384. Conservatively, thrombolytic therapy is not effective for at least 10 percent of patients suffering from an acute MI, either because patients are ineligible to receive the treatment or the treatment fails to clear the blockage. Accordingly, it may be conservatively projected that at least 104 patients who presented to Oak Hill's ER between 1998 and August 2000 (10 percent of 1049) suffering an MI were in need of angioplasty intervention for which open heart surgery backup is required. Most patients are diagnosed as in need of OHS or angioplasty as a result of undergoing a diagnostic cardiac catheterization. Oak Hill performs an extremely high volume of cardiac cath procedures for a hospital that lacks an OHS program. In 1999, for example, it performed 1,641 cardiac catheterizations. This is a higher volume than experienced by any of six hospitals during the year prior to which they recently implemented new OHS programs. If Oak Hill had an OHS program, most of the patients at Oak Hill determined to be in need of angioplasty or OHS could receive those procedures at Oak Hill. Such an arrangement would avoid the inevitable delay and stress occasioned by a transfer to Bayonet Point or elsewhere. Furthermore, if Oak Hill had an OHS program then those patients in need of diagnostic cardiac catheterization and angioplasty sequentially would have immediate access to the interventional procedure. The need is underscored for those patients presenting to Oak Hill's ER with myocardical infarctions who do not respond to thrombolytics because, as stated earlier in this order, access to angioplasty within 30 minutes of onset is ideal. Oak Hill transfers an extremely high number of cardiac patients for angioplasty and open heart surgery. In 1999, Oak Hill transferred 258 patients to Bayonet Point for open heart surgery, and 311 for angioplasty/stent procedures. Of course, most OHS patients are scheduled on an elective basis for surgery, rather than being transferred between hospitals, as is evident from the fact that during the 12-month period ending March 1999, 698 Hernando County residents underwent OHS. For now, Oak Hill patients determined to be in need of urgent angioplasty or open heart surgery must be transferred by ambulance to an OHS provider which for the vast majority of patients is Bayonet Point. Approximately 17 miles south, the average drive time to Bayonet Point from Oak Hill is 30 minutes but it can take longer when on occasion there is traffic congestion. Once the transfer is achieved and patient receives the required procedure, the drive can be difficult for the patient's family and loved ones. Community members often express to physicians and hospital staff their support and desire for an OHS program at Oak Hill. Many believe travel outside Hernando County for those services is cumbersome for loved ones who are important to the patient's healing process. The community support and demand for these services is evidenced by the 7,628 resident signatures on petitions in support of Oak Hill's efforts to obtain approval for an OHS program. While a program at Oak Hill would be more convenient, Oak Hill did not demonstrate a transfer problem that would rise to the level of "not normal" circumstances. Because of Oak Hill's relationship with Bayonet Point, Bayonet Point's proximity and excess capacity, coupled with the high quality of the program at Bayonet Point, Oak Hill's case is more in the nature of seeking a satellite. As one expert put it at hearing, [Oak Hill] is, in fact, a satellite. And my question is, [']What's the wisdom of doing that if you don't have the problems that normally are being addressed when you grant approval of a program?['] In other words, if you don't have transfer issues [that rise to the level of "not normal" circumstances], if you don't have access issues, if you're not achieving any price competition, if it's not particularly cost effective, why would you [approve Oak Hill]? (Tr. 1537-38). Oak Hill's Projected Utilization Oak Hill projected a range of 316 to 348 OHS cases during its first year, and by its third year a range of between 333 and 366 cases. Those volumes are sufficient to ensure excellent quality of care from the beginning of the program, particularly with the involvement of the Bayonet Point surgeons. Oak Hill defined its primary service area (PSA) for OHS based on historic MDC-5 cardiology related diagnosis discharges from its hospital. For the 12-month period ended March 1999, over 90 percent of Oak Hill's MDC-5 discharges were residents of six zip codes, all in the vicinity of Oak Hill Hospital and within Hernando County. Accordingly, that area was chosen as the PSA for projecting OHS utilization. Out-of-PSA residents accounted for only 8.9 percent of Oak Hill's MDC-5 discharges, and of these, 1.5 percent were out-of-state patients, and 4.9 percent were residents from other parts of District 3. For the year ending ("YE") March 1999, Oak Hill had an MDC-5 market share of 40.9 percent within its PSA, without excluding angioplasty, stent, and OHS cases. If angioplasty, stent, and OHS cases are excluded, Oak Hill's PSA market share was 52.7 percent. In order to project OHS service demand, Oak Hill examined the population projections for 1999 and 2004 for District 3, and for Oak Hill's PSA. The analysis was based on age-specific resident populations and use rates, to serve as a contrast to the Agency's projections. The numeric need formula in the OHS Rule utilizes a facility based use rate derived by totaling all of the reported OHS cases performed by hospitals within a District during a given time period, and then dividing those cases by the adult population aged 15 and over. While a facility-based use rate measures utilization in those District hospitals, however, it does not measure out-migration. Nor does it reflect the residence of the patients receiving those services. On the other hand, a resident-based use rate identifies where patients needing OHS actually come from, and permits development of age specific use rates. For example, the resident-based use rates reflects that the southern portion of District 3 has a much higher concentration of elderly persons than does the northern portion of the District, and reveals extremely high migration out of the District for OHS services. Oak Hill's PSA is more elderly than the District 3 population as a whole. In 1999, 32.8 percent of the Oak Hill PSA population was aged 65 or over, as opposed to only 21.5 percent for District 3 as a whole, with similar results projected for the population in 2004, the projected third year of operation of Oak Hill's program. Based on the district-wide use rate resulting from the OHS Rule need methodology, Hernando County would be expected to generate 276 OHS cases in the planning horizon of July 2002 (use rate of 2.3 per 1000 adult population). Application of this OHS Rule use rate to Hernando County clearly understates need if resources to meet the need are considered within the isolation of the boundaries of District 3. For example, the OHS Rule based projection of 276 OHS cases in 2002, is far below the actual 664 Hernando County resident OHS discharges during YE March 1998, and the 698 OHS cases during YE March 1999. While the facility-based district-wide use rate was 2.3, the Hernando County resident-based use rate was 6.45 per 1000 population. The fact of increasing use rates with age is demonstrated by the Hernando County resident use rate of 6.95 for ages 55-64, increasing to 12.01 for ages 65-74, and increasing again to 14.95 for age 75 and over. But focusing on Hernando County use rates within District 3 ignores the reality of the proximity of an excellent program at Bayonet Point. Oak Hill reasonably projected OHS demand in its PSA by examining the age-specific use rates of residents in the southern portion of District 3, which experienced an overall use rate of 4.55 for the year ending March 1999. Those age-specific use rates were then applied to the age-specific population forecast for each of the three horizon years of 2002 through 2004, resulting in an expected PSA demand for OHS of 547 cases in 2002, 561 cases in 2003, and 575 cases in 2004. Those projections are conservative given that 663 actual open heart surgeries were reported among PSA residents during the YE March 1999. The same methodology was used to project angioplasty service demand in the PSA, resulting in an expected demand ranging from 721 cases in 2002 to 758 cases in 2004. Oak Hill then projected its expected OHS case volume by assuming that its first year OHS market share within its PSA would be the same as its MDC-5 market share, being 52.7 percent. Oak Hill next assumed that by the third-year operation its market share would increase to equal its current cardiac cath PSA market share of 57.9 percent. It further assumed that it would have a non-PSA draw of 8.9 percent, which is equal to its current non-PSA MDC-5 market share. Oak Hill reasonably expects that 91.1 percent of its OHS cases would come from within its six zip code PSA, with the remaining 8.9 percent expected to come from outside that area. Oak Hill then projected an expected range of OHS discharges during its first three years of operation by using both a low estimate and a high estimate. The resulting utilization projections reflect a low range of 316 OHS cases in 2002, 324 cases in 2003, and 333 cases in 2004. The high range estimate for the same years respectively would be: 348, 357, and 366 cases. The same methodology was used to project angioplasty cases, resulting in the following low range: 417 cases in 2002; 428 in 2003; and 438 in 2004. The expected high range for the same respective years would be: 458, 470, and 482. Oak Hill's OHS and angioplasty utilization projections are reasonable. Long-term Financial Feasibility Long-term financial feasibility is defined as a demonstration that the project will achieve and maintain financial self-sufficiency over time. Oak Hill's projected gross charges were based on Bayonet Point's charge structure. The projected payer mix was based on Oak Hill's cardiac cath experience. Projected net reimbursement by payor source was based on Oak Hill's experience for Medicare, Medicaid, and contractual adjustment history. Oak Hill's expenses were projected on a DRG specific basis using information generated by the cost accounting system at Bayonet Point. The use of Bayonet Point's expense experience is a reasonable proxy for a number of reasons. Its patient base is comprised of patients who are reasonably expected to be the base of Oak Hill's patients. Management there is similar to what it will be at an Oak Hill program. And, as stated so often, the two facilities are relatively close in location. To account for differences between Bayonet Point's expenses and Oak Hill's project costs, interest and depreciation, adjustments were made by Oak Hill as reflected in its application. As a means of compensating for fixed costs differentials between the two hospitals, Oak Hill added its salary costs projected in Schedule 6 to the salary expenses already included in Bayonet Point's costs. (Schedule 6 nursing, administration, housekeeping, and ancillary labor costs exceeded $3 million in the first year of operations.) This counting of two sets of salary expenses offsets any economies of scale cost differential that may exist between the OHS programs at Bayonet Point and Oak Hill. A reasonable 3 percent annual inflation factor was applied to both projected charges and costs. The reasonableness of Oak Hill's overall approach is supported by Citrus Memorial's use of a substantially similar pro forma methodology in modeling its proposed program on Munroe Regional Medical Center. Oak Hill reasonably projects a profit of $1.38 million in the first year of operation, and that profitability will increase as the case volumes grow thereafter. An Oak Hill program will cost Bayonet Point (a sister HCA hospital) patients and may diminish the corporate profits of the two hospital's parent corporation, HCA Health Services of Florida, Inc. It is clear from the parent's most recent audited financial statements, however, that it has ability to absorb a lower level of profit from Bayonet Point without jeopardizing the financial viability of Oak Hill. Brooksville Regional argues that the financial impact to Bayonet Point of an Oak Hill program demonstrates that the Oak Hill application is nothing more than a preemptive move to stifle competition. Oak Hill, in turn, characterizes its proposal as a sound business judgement to compete with non-HCA hospitals in District 3. Whatever characterization is applied to the Oak Hill proposal, it is clear that it is financially feasible in the long term. Other Statistics The AHCA population estimates for January 1, 1999, show a Hernando County population of 108,687 and a Citrus County population of 98,912. The same data sources show the "age 65 and over" population (the "elderly") in Hernando to be 40,440 and in Citrus to be 37,822. During the year 2000, there were 2,545 more people aged 65 and over in Hernando County than in Citrus County. By the year 2005, the difference is expected to be 3.005. The total change in the elderly population between 2000 and 2005 is projected to be 4,109 in Citrus County and 4,614 in Hernando County. Generally, the older the population, the older the OHS use rate. Comparatively, then, Hernando County has the larger population to be served both now, and in all probability, in the foreseeable future. Oak Hill has the largest cardiology program among the applicants. For the 12-month period ending September 1999, MDC- 5 discharges were 1,130 at Brooksville Regional, 2,077 at Citrus Memorial and 2,812 at Oak Hill. The combined Brooksville and Spring Hill Regional Hospital MDC-5 case volume of 2,238 is below Oak Hill's MDC case volume for the same period. Oak Hill is the largest cardiac cath provider among the applicants. For the 12-month period ending September 2000, Citrus Memorial reported 646 cardiac catheterization procedures and Brooksville Regional reported 812. Oak Hill reported 1,404 such procedures, only sixty shy of a volume double the combined volume at the other two applicants. The level of ischemic heart disease in an area is indicative of the level of open heart surgery needed by residents of the area. The number of ischemic heart disease cases by county during the 12-month period ending September 1999 were: 1,038 for Alachua; 1,978 for Citrus; 2,816 for Marion; and, Hernando, 3,336. During the 12-month period ending September 1999, 657 Hernando County residents underwent OHS at Florida hospitals, while only 408 residents of Citrus County did so. Similarly, 948 Hernando County residents had angioplasty, while only 617 Citrus County residents underwent angioplasty. For the year ending June 30, 1999, the Citrus County OHS use rate was 4.26 per 1,000 population, substantially lower than the Hernando County use rate of 6.41. A comparison of the use rates for the year ending September 30, 1999, again shows Hernando County's use rate to be higher: 4.13 for Citrus, 6.08 for Hernando. Hernando County also experiences a higher cardiovascular mortality rate than does Citrus County. During 1998, the age-adjusted cardiovascular mortality rate per 100,000 population for Citrus was 330.88 and 347.40 for Hernando. During 1999, those mortality rates were 304.64 in Citrus and 313.35 in Hernando (consistent with the decline between 1998 and 1999 for the state as a whole). The Hernando mortality rates greater than Citrus County's indicate a greater prevalence of heart disease in Hernando County than in Citrus County. Most importantly, during 1999, Oak Hill transferred 619 patients to Bayonet Point for cardiac intervention - 258 for open heart surgery, 311 for angioplasty/stent, and 50 for cardiac cath. Brooksville Regional transferred a combined 383 patients after diagnostic cardiac catheterization to other hospitals for either angioplasty or OHS. Brooksville Regional has 91 licensed beds, Citrus Memorial has 171 beds and Oak Hill has 204 beds. Although with Spring Hill one could view Brooksville Regional as "two hospital systems with 166 beds under common ownership and control" (Tr. 1544), at 91 beds, Brooksville would become the smallest OHS program in the state in terms of licensed bed capacity, Hospitals of less than 100 beds are not typically of a size to accommodate an OHS program. There might be dedicated cardiovascular hospitals of 100 beds or less with capability to support an open heart surgery program, but "open heart surgical services in [a general, surgical-medical hospital of less than beds] would overwhelm the hospital as far as the utilization of services." (Tr. 126). Oak Hill's physical plant, hospital size, number of beds, medical staff size, number of cardiologists, cath lab capacity, number of cath procedures, number of admissions, and facility accessibility to the largest local population are all factors in its favor vis-à-vis Brooksville Regional. In sum, Oak Hill is a hospital more ready and appropriate for an adult open heart surgery program than Brooksville. Alternatives As an alternative to its CON application, Oak Hill considered the possibility of seeking approval of a program to be shared with Bayonet Point. Learning that the Agency looks with disfavor on inter-district shared adult open heart surgery programs, Oak Hill decided to seek approval of a program independent of Bayonet Point but one that would rely on Bayonet Point's experience and expertise for development, implementation and operation. Bed Capacity Brooksville contends that Oak Hill lacks sufficient bed capacity to accommodate the implementation of an OHS program in conjunction with its projected-related increased admissions. Brooksville relied on an Oak Hill daily census document, focusing on the single month of January, arguing that the document reflected that Oak Hill exceeded its licensed bed capacity on 5 days that month. The licensed bed capacity, however, was not exceeded. Observation patients, who are not inpatients, and not properly included in the inpatient count, were included in the counts provided by Brooksville. Seasonal peaks in census during the winter months, particularly January, are common to all area hospitals. Similarly, all hospitals experience a higher census from Monday through Thursday, than on other days. Oak Hill has adequate capacity and flexibility to accommodate those rare occasional days during the year when the number of patients approaches its number of beds. Patients are sometimes hospitalized for "observation," and when so classified are expected to stay less than 24 hours. Typically, Oak Hill places such patients in a regular "licensed" bed, so long as such beds are available. There are other areas in the hospital suitable for observation patients, including: 12 currently unused and unlicensed beds adjacent to the cardiac cath recovery area; six beds in the ER holding area; eight beds in the ER Quick Care Unit; and additional beds in the same day surgery recovery area. Observation patients can be cared for appropriately in these other areas, a routine hospital practice. Peak season census is "a fact of life" for hospitals, including Oak Hill and Brooksville. Oak Hill has never been unable to treat patients due to peak season demands. January is the only month during the year when bed capacity presents a challenge at Oak Hill. If necessary, Oak Hill could coordinate patient admissions with Bayonet Point to ensure that all patients are appropriately accommodated. Oak Hill can successfully implement a quality OHS program with its current bed capacity. In fact, all parties have stipulated to Oak Hill's ability to do so. Moreover, should it actually come to pass in future years that Oak Hill's annual average occupancy exceeds 80 percent, it may add up to 20 licensed beds on a CON exempt basis. Brooksville Regional Factors favoring Brooksville over Oak Hill Bayonet Point is the dominant provider of OHS/angioplast to residents of Hernando County. As a non-HCA hospital, a Brooksville program (in contrast to one at Oak Hill) would enhance patient choice in Hernando County for hospitals and physicians, and would create an environment for price and managed care competition. Other health planning factors that support Brooksville Regional over Oak Hill are the locations of the two Hernando County hospitals and the ability of the two to transfer patients to Bayonet Point. Patient Choice and Competition Of the OHS/angioplasty services provided to Hernando County residents, Bayonet Point provides 94 percent, the highest county market share of any hospital that provides OHS services to residents of District 3. Indeed, it is the highest market share provided by any OHS provider in any one county in the state. The importance of patient choice and managed care competition has been acknowledged by all the parties to this proceeding. If Brooksville Regional's program were approved, Hernando County residents would have choice of access to a non- HCA hospital for open heart and angioplasty services and to physicians and surgeons other than those who practice at Bayonet Point. This would not be the case if Oak Hill's program was approved instead of Brooksville's. Price Competition Although Brooksville is not a "low-charge provider for cardiovascular services" (tr. 1347), approving Brooksville creates an environment and potential for price competition. A dominant provider in a marketplace has substantial power to control prices. Adding a new provider creates the motivation, if not the necessity, for that dominant provider to begin pricing competitively. A dominant provider controls prices more than hospitals in a competitive market. Bayonet Point's OHS charges illustrate this. Approving Brooksville's application creates an environment for potential price competition with Bayonet Point, whereas approving Oak Hill's application, whose charges are expected to be the same as Bayonet Point's, does not. Managed Care Contracting Just as competitive effects on pricing are reduced in an environment in which there is a dominant provider, so managed care contracting is also affected. Managed care competition depends not just on competition between managed care companies but also on payer alternative within a market. If a managed care company is forced to deal with one health care provider or hospital in a marketplace, its competitive options are reduced to the benefit of the hospital that enjoys dominance among hospitals. "[T]he power equation moves much more strongly in that type of environment towards the provider [the dominant hospital] and away from the managed care companies." (Tr. 1471). Managed care companies who insure Hernando County residents have no alternative when it comes to open heart surgery and angioplasty services but to deal with Bayonet Point. With a 94 percent share of the Hernando County residents in need of open heart and angioplasty services, there is virtually no competition for Bayonet Point in Hernando County. The managed care contracting for both Bayonet Pont and Oak Hill is done at HCA's West Florida Division office, not at the individual hospital level. Approving Oak Hill will not promote or provide competition for managed care. Approving Brooksville, on the other hand, will provide managed care competition over open heart and angioplasty services in Hernando County. Ability to Transfer Patients While transfers of Hernando County patients always produce some stress for the patient and are cumbersome as discussed above for the patient's loved ones, there is no evidence of transfer problems for Oak Hill that would rise to the level of "not normal" circumstances. Outcomes for patients transferred from Oak Hill to Bayonet Point on the basis of morbidity statistics, mortality statistics, length of stay, patient satisfaction, and family satisfaction are excellent. It is not surprising that sister hospitals situated as are Oak Hill and Bayonet Point would enjoy minimal transfer delays and access problems encountered when patients are transferred. Transfers between unaffiliated hospitals are not normally as smooth or efficient as between those that have some affiliation. Unlike Oak Hill's patients, Brooksville patients, for example, are never transported for OHS/angioplasy by Bayonet Point's private ambulance. Other than in emergency cases, Bayonet Point decides the date and manner when the patient will be transferred. But just as in the case of Oak Hill, there is no evidence of transfer problems between Brooksville Regional and Bayonet Point that would amount to an access problem at the level of "not normal" circumstances. Outmigration As detailed earlier, there is extensive outmigration of Hernando County residents to District 5 for open heart and angioplasty procedures. The outmigration pattern on its face is in favor of both applications of Oak Hill and Brooksville. The outmigration from Hernando County, however, is of minimal weight in this proceeding since Bayonet Point is so close to both Oak Hill and Brooksville. The patients at the two Hernando hospitals have good access to Bayonet Point, a facility that provides a high level of care to Hernando County residents in need of open heart surgery and angioplasty services. The relationship is inter-district so that it is true that there is outmigration from District 3. Outmigration statistics showing high outmigration from a district have provided weight to applications in other proceedings. They are of little value in this case. Location of the Two Hernando Hospitals Brooksville is located in the "dead center" (Tr. 1290) of Hernando County. With good access to Citrus County via Route 41, it is convenient to both Hernando County residents and some residents of Citrus County. It reasonably projects, therefore, that 90 percent of its open heart/angioplasty volume will be from Hernando County with the remaining 10 percent from Citrus. Oak Hill is located in southwest Hernando County, closer to Bayonet Point than Brooksville. Oak Hill's primary service area is substantially the same as that part of Bayonet Point's that is in Hernando County. Oak Hill does not propose to serve Citrus County. Brooksville, then, is more centrally located in Hernando County than Oak Hill and proposes to serve a larger area than Oak Hill. Financial Feasibility (long-term) Brooksville has operated profitably since its bankruptcy. In its 1999 fiscal year, the first year out of bankruptcy, Hernando HMA earned a profit of $3 million. In fiscal year 200, Brooksville's profit was $6 million. OHS programs are generally very profitable. There is no OHS program in Florida not generating a profit. Brooksville's projected expenses and revenues associated with the program are reasonable. Schedule 5 in the Brooksville application contains projected volumes for OHS/angioplasty. The payer mix and length of stay were based on 1998 actual data, the most recent data for a full year available. The projected volumes are reasonable. The projected volumes are converted to projected revenues on Schedule 7. These projections were based on actual 1998 charges generated for both Hernando and Citrus County residents since Brooksville proposes to serve both. These averages were then reasonably projected forward. Schedule 7 and the projected revenues are reasonable. These projected volumes and revenues account for all OHS procedures performed in Hernando and Citrus Counties in 1998 even though effective October 1, 1998, the DRG procedure codes for OHS procedures were materially redefined. Thus, when Brooksville's schedules were prepared using 1998 data, only 3 months of data were available using the new DRG codes. Brooksville opted to use the full year of data since using a full year's worth of data is preferable to only 3 months. Similarly, the DRGs for angioplasty both as to balloon and with stent were re-classified. Again, Brooksville opted to use the full year's worth of data. Brooksville's expert explained the decision to use the full year's worth of data and the effect of the DRG reclassification on Brooksville's approach, "We've captured all the revenues and expenses associated with these open heart procedures and just because the actual DRGs have changed, doesn't . . . impair the results because both revenues and expenses are captured in these projections." (Tr. 1651). Schedule 8 includes the projected expenses. It included the health manpower expenses from Schedule 6 and the project costs from Schedule 1. The remaining operating expenses were based upon the actual costs experienced by all District 3 OHS providers generated from a publicly-available data source, and then projected forward. As to these remaining operating costs, consideration of an average among many providers is far preferable to relying on just one provider. Schedule 8 was reasonably prepared. It accounts for all expense to be incurred for all types of OHS and angioplasty procedures. It is based on the best information available when these projections were prepared and are based on 12 months of actual data. Even if the projections of the schedules are not precise because of the re-classification of DRGs, they contain ample margins of error. Brooksville's financial break-even point is reached if it performs 199 OHS and 100 angioplasty procedures. This low break-even point provides additional confidence that the project is financially feasible. Brooksville demonstrated that its proposed program will be financially feasible.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Agency for Health Care Administration enter a final order that grants the application of Citrus Memorial (CON 9295) and denies the applications of Oak Hill (CON 9296 )and Brooksville Regional (CON 9298). DONE AND ENTERED this 4th day of October, 2001, in Tallahassee, Leon County, Florida. DAVID M. MALONEY Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 4th day of October, 2001. COPIES FURNISHED: Diane Grubbs, Agency Clerk Agency for Health Care Administration 2727 Mahan Drive Fort Knox Building Three, Suite 3431 Tallahassee, Florida 32308-5403 William Roberts, Acting General Counsel Agency for Health Care Administration 2727 Mahan Drive Fort Knox Building Three, Suite 3431 Tallahassee, Florida 32308-5403 Michael J. Cherniga, Esquire Seann M. Frazier, Esquire Greenberg Traurig, P.A. East College Avenue Post Office Box 1838 Tallahassee, Florida 32302-1838 Stephen A. Ecenia, Esquire Rutledge, Ecenia, Purnell and Hoffman, P.A. 215 South Monroe Street, Suite 420 Tallahassee, Florida 32302-0551 James C. Hauser, Esquire Metz, Hauser & Husband, P.A. 215 South Monroe Street, Suite 505 Post Office Box 10909 Tallahassee, Florida 32302 John F. Gilroy, III, Esquire Agency for Health Care Administration 2727 Mahan Drive Fort Knox Building Three, Suite 3431 Tallahassee, Florida 32308-5403
Findings Of Fact Based upon all of the evidence, the following findings of fact are determined: Petitioner, Lakeland Regional Medical Center (LRMC), is a 897-bed private, not-for-profit, general acute care hospital located at 1324 Lakeland Hills Boulevard, Lakeland, Florida. It is considered a major regional referral hospital and provides a wide range of tertiary services, including open heart surgery. The facility is located in District 6 and is one of six facilities in the district having an existing open heart surgery program. Respondent, Department of Health and Rehabilitative Services (HRS), is the state agency charged with the responsibility of administering the Health Facility and Services Development Act, also known as the Certificate of Need (CON) law. On September 26, 1988 intervenor, Winter Haven Hospital, Inc. (WHH), filed with HRS an application for a CON seeking authority to establish an open heart surgery program at its facility in Winter Haven, Florida. After reviewing the application, on February 3, 1989, HRS published notice of its intent to issue the requested CON. If approved, this program would be in competition with similar programs operated by LRMC and intervenor, Hillsborough County Hospital Authority d/b/a Tampa General Hospital (TGH). Those two parties have initiated formal proceedings in Case Nos. 89-1286 and 89-1287 to contest the proposed grant of authority. Intervenor, Venice Hospital, Inc. (Venice), has a pending application for authority to establish an open heart surgery program in a separate administrative proceeding and has intervened in opposition to LRMC's rule challenge. It is noted that LRMC, WHH and TGH are located in District 6 while Venice is located in an adjoining, but separate, district. All parties have standing in this proceeding. In order for HRS to grant a certificate of need, it is necessary for an applicant to satisfy all relevant rule and statutory criteria. In this vein, the agency has promulgated Rule 10-5.011(1)(f), Florida Administrative Code (1987), which contains certain criteria pertaining to open heart surgery programs. That rule provides in relevant part as follows: (f)2. Departmental Goal. The Department will consider applications for open heart surgery programs in context with applicable statutory and rule criteria. The Department will not normally approve applications for new open heart surgery programs in any service area unless the conditions of Sub-paragraphs 8. and 11., below, are met. * * * 11.a. There shall be no additional open heart surgery programs established unless: (1) the service volume of each existing and approved open heart surgery program within the service area is operating at and is expected to continue to operate at a minimum of 350 adult open heart surgery cases per year or 130 pediatric heart cases per year, (Emphasis added) * * * The requirements of this rule, which are unambiguous, and other pertinent statutory and rule criteria, are to be applied by HRS to all applicants, including WHH, during the CON review process. Although the rule itself is not being challenged by LRMC, subparagraph 11.a. of the rule is at the heart of this controversy. Petitioner and TGH contend that the clear language of the rule requires that, absent the existence of not normal circumstances, HRS may not award a CON unless each existing and approved open heart surgery program in the service area is operating at and is expected to continue to operate at 350 procedures per year. Because there are now six approved and existing open heart surgery programs in the district, petitioner argues that the rule mandates that, before a new program can be authorized, each of the six programs must meet the required level of 350 procedures per year. They contend further that the particular policy applied by HRS to WHH's application is not apparent on the face of rule 10-5.011(1)(f)2. and thus it constitutes an unpromulgated rule. In preliminarily approving WHH's application, HRS admits that it used a so-called averaging policy which it agrees may be described in the following manner: HRS has formulated and is applying in reviews of Certificate of Need ("CON") applications for new open heart surgery services a policy of general applicability that is uniformly and consistently applied, which calls for the averaging of the utilization of existing and approved adult open heart surgery programs in the applicable service area, and which deems subparagraph 11.a.(I) of Rule 10-5.011(1)(f), Fla. Admin. Code, to be met if the average utilization of all such existing and approved programs in that service area is at least 350 cases (the "Averaging Policy"). Pursuant to its Averaging Policy, HRS will approve a CON application for a new adult open heart surgery program under Rule 10- 5.011(1)(f), Fla. Admin. Code, even if each existing and approved program in the proposed service area is not operating at a minimum of 350 adult cases per year, and even if no "not normal" circumstances are presented in the application or found to exist in the State agency Action Report. Stated another way, HRS deemed subparagraph 11.a. to have been met in WHH's case because, after dividing the total number of procedures performed district wide by the number of existing and approved programs, there were an average number of procedures in excess of 350 for each program in the district. It used this averaging process even though two programs were not operational at the time the review process took place, and only two (LRMC and TGH) of the six programs had actually performed more than 350 procedures during the specified time period being measured. 1/ Thus, the averaging policy used by HRS allows approval of a CON application for open heart surgery even if only some programs in a district, rather than each, have the required 350 case volume. The averaging technique has not been reduced to writing in a memorandum, manual or agency policy directive, and it has not been formally adopted as a rule. In this regard, HRS, but not WHH and Venice, has admitted that the policy is indeed a rule. The results of applying that "rule" are contained in the state agency action report issued by HRS and made a part of this record. HRS has consistently and uniformly applied this averaging technique in every open heart surgery case except one since the rule was adopted in substantially its present form on February 14, 1983. 2/ It has been applied without discretion by those HRS personnel who have the responsibility of administering the CON law and regulations. The proponents of the averaging policy argued first that the language in subparagraph 11.a. authorized its use. However, nothing in the language of the existing rule expressly refers to an averaging process. They also contended that when other provisions within the rule are read, the use of the policy becomes apparent. More particularly, they pointed to subsection (7) of the rule which requires that the provision of open heart surgery be consistent with the state health plan. That plan provides in part that one of its objectives is to maintain an average volume of 350 procedures at all programs in the state. However, the state health plan is not mentioned in subparagraph 11.a., subsection (7) does not track or mirror the averaging technique, and the same subsection does not alert the user of the rule to the fact that an averaging process will be applied.
