Findings Of Fact Harbour Shores Hospital is a 60-bed short-term psychiatric facility, with 36 adult beds and 24 adolescent beds. The facility opened in October 1985, and had an occupancy of 62 percent at the time of the final hearing in March 1986. T. 14, 20. The hospital is an integral part of Lawnwood Medical Center, Inc., located in Ft. Pierce, Florida, and Lawnwood is owned by Hospital Corporation of America. T. 13-14. Hospital Corporation of America now operates 5,000 psychiatric beds in the United States. T. 18. Lawnwood Medical Center, Inc. (Lawnwood), submitted an application for certificate of need number 3363 on June 14, 1984, for the conversion of 12 short-term psychiatric beds to 12 short-term inpatient hospital substance abuse beds. T. 15. No construction is needed to convert these 12 beds. T. 16. The Department of Health and Rehabilitative Services (HRS) initially denied the application. T. 106. All references in this order to Harbour Shores Hospital shall include the Petitioner, Lawnwood, unless specifically stated otherwise. The parties stipulated that the only issue in this case is need and any ancillary issue which is based upon need. T. 4-5. HRS has a rule governing short and long-term hospital inpatient hospital substance abuse beds, which is rule 10-5.11(27), Florida Administrative Code. Subparagraph (f)1 of the rule contains what HRS calls bed allocations" and calculates need for a health service district as a whole. Harbour Shores Hospital is located in St. Lucie County in Health District IX. District IX is composed of St. Lucie, Indian River, Martin, Okeechobee, and Palm Beach Counties. HRS Ex. 2, p.7. The District IX Local Health Council has identified two subdistricts for purposes of allocating short term psychiatric and substance abuse beds. Subdistrict 1 is St. Lucie, Martin, Indian River, and Okeechobee Counties, and subdistrict 2 is Palm Beach County. HRS Exhibit 2, p. 7; T. 110. HRS proposes to determine need as of January 1989 using the date of the application as the starting point for the five year period specified in rule 10- 5.11(27)(f)1, Florida Administrative Code. T. 107-6. The basis of this decision is a new policy by HRS to implement the Gulf Court decision. There is a need for only 1 additional short-term substance abuse bed in District IX by January 1989 based upon Rule 10-5.11(27)(f)1, Florida Administrative Code. T. 109. HRS Exhibit 1. HRS proposes also to refer to such need for short term substance abuse beds as indicated by local health council plans, relying upon rule 10- 5.11(27)(h)3, Florida Administrative Code. T. 110. The local health plan for District IX allocates needed beds based upon the subdistricts described above. HRS Exhibit 2, p. 6. Further, the local health plan has adopted the method of HRS found in rule 10-5.11(27), supra, for calculating need, and calculates such need using the HRS rule factor of .06 substance 3. All references in this order to Harbour Shores Hospital shall include the Petitioner, Lawnwood, unless specifically stated otherwise. The parties stipulated that the only issue in this case is need and any ancillary issue which is based upon need. T. 4-5. HRS has a rule governing short and long term hospital inpatient hospital-substance abuse beds, which is Rule 10-5.11(27), Florida Administrative Code. Subparagraph (f)1 of the rule contains what HRS calls bed "allocations" and calculates need for a health service district as a whole. Harbour Shores Hospital is located in St. Lucie County in Health District IX. District IX is composed of St. Lucie, Indian River, Martin, Okeechobee, and Palm Beach Counties. HRS Ex. 2, p.7. The District IX Local Health Council has identified two subdistricts for purposes of allocating short-term psychiatric and substance abuse beds. Subdistrict 1 is St. Lucie, Martin, Indian River, and Okeechobee Counties, and subdistrict 2 is Palm Beach County. HRS Exhibit 2, p.7; T. 110. HRS proposes to determine need as of January 1989 using the date of the application as the starting point for the five-year period specified in Rule 10-5.11(27)(f)1, Florida Administrative Code. T. 107-8. The basis of this decision is a new policy by HRS to implement the Gulf Court decision. There is a need for only 1 additional short-term substance abuse bed in District IX by January 1989 based upon Rule 10-5.11(27)(f)1, Florida Administrative Code. T. 109. Exhibit 1. HRS proposes also to refer to such need for short term substance abuse beds as indicated by local health council plans, relying upon rule 10- 5.11(27)(h)3, Florida Administrative Code. T. 110. The local health plan for District IX allocates needed beds based upon the subdistricts described above. HRS Exhibit 2, p. 6. Further, the local health plan has adopted the method of HRS found in rule 10-5.11(27), supra, for calculating need, and calculates such need using the HRS rule factor of .06 substance abuse beds per 1,000 population in each of the two subdistricts. HRS Exhibit 2, pp. 5 and 8, paragraph II. Using current estimates of the populations of each subdistrict in January 1909, HRS projects that subdistrict 1 will have a surplus of 15 substance abuse beds in 1989, and all net need (16 beds) will be in subdistrict 2, which is Palm Beach County. T. 111; HRS Exhibit 1. HRS has not adopted these subdistricts by rule. T. 128-29. There was no evidence to substantiate the reasonableness of the subdistricts adopted in the local health plan. T. 131. The following is a summary of the existing and approved short-term substance abuse beds in District IX, showing county of location, and occupancy rates for 1985: Humana Hospital 16 Licensed Indian River 8509 Sebastian Lake Hospital 16 Licensed Palm Beach 3558 Palm Beaches Fair Oaks 17 Licensed Palm Beach 3807 Savannas 20 Approved St. Lucie Hospital Beds Status County Patient Days Occupancy 145.7% 60.7% 60.7% The number of patient days at Fair Oaks, however, is for four months, August, October, November, and December 1985. Thus, the actual number of patient days, 1269, has been multiplied by 3 to obtain an estimate for an entire year. T. 23- 24, 61-62. The occupancy rate is the number of patient days divided by the product of the number of days in the year (365) and the number of licensed beds. Using the statistics in paragraph 10, the average occupancy rate for the three existing facilities in District IX was 88.8 percent. If one assumes, as did Petitioner's expert, that the utilization rates for short-term substance abuse beds will at least remain the same as in 1935, with the addition of the 20 new beds at Savannas Hospital, District IX may have an occupancy rate of 63.8 percent and subdistrict 1 may have an occupancy rate of 64.8 percent . The 20 new beds at the Savannas Hospital are those granted to Indian River Community Mental Health Center, Inc., and are projected to open in November 1986. T. 83. As discussed above, Harbour Shores Hospital had been in operation about five months by the time of the March 1986 hearing, and its 60 short term psychiatric beds were averaging 62 percent occupancy, which is about 15 percent above the occupancy projected in its certificate of need application. T. 38. Harbour Shores serves patients from the four counties of subdistrict 1, St. Lucie, Martin, Indian River, and Okeechobee, and serves a significant number of patients from Palm Beach County as well; three to four percent of its patients also come from Brevard and Broward Counties. T. 19. About 80 percent of the patients at Harbour Shores in the first five months of its operation had a substance abuse problem secondary to the primary diagnosis of mental illness. T. 30, 50, 63. This is consistent with experience throughout Florida. T. 63. Most of these "dually diagnosed" patients have been through a detoxification program before entering Harbour Shores Hospital. T. 30. In its beginning months of operation, Harbour Shores has had patients referred from the courts, law enforcement agencies, community and social agencies, physicians, and from HRS. T. 21-22, 59. Harbour Shores can expect to obtain substance abuse referrals from these agencies. Staff at Harbour Shores works with the DWI Board, Students Against Drunk Driving, and school administrative personnel. T. 39-40. In October and November 1985, Harbour Shores received 38 requests from physicians, the courts, law enforcement agencies, and social agencies, for admission of patients for substance abuse treatment. T. 22, 49. There is no evidence that Harbour Shores had any such requests in December 1985 or January 1986. In February 1986, it had 14 such requests, and in March to the date of the hearing, it had 5 requests. T. 48. There is no evidence as to whether these requests were for short or long-term substance abuse services, or whether these were requests from different patients or multiple requests from the same patient. There is also no evidence that the persons requesting substance abuse treatment were not adequately treated at existing facilities. Thus, the data from these few months is not an adequate basis for determining future need for short term substance abuse beds. Ms. Peggy Cioffi is the coordinator for the Martin County Alcohol and Drug Abuse Program. Deposition, Ms. Peggy Cioffi, p. 2. Ms. Cioffi testified as to the need for substance abuse services in her area. She did not testify as an expert witness. Her program is primarily designed to assist the County Court in referrals of misdemeanants and others within the Court's jurisdiction who need substance abuse services. Id. Ms. Cioffi has difficulty placing persons needing inpatient or residential treatment. Id. at p. 3. She related an example of a county prisoner who asked to be detained in jail three months for lack of an alcohol program. Id. at p. 4. Ms. Cioffi did not state whether this person needed residential or inpatient hospital care. She also had recently reviewed a 14 page county court docket and determined that 67 percent of those charged represented alcohol or drug related offenses. Id. Ms. Cioffi did not clearly show how she was able to infer this fact. Further, Ms. Cioffi was unable to tell from this statistic how many of these defendants needed short term inpatient hospital substance abuse treatment. Id. at p. 6. She stated that a very high percentage of these could benefit from some kind of services, but did not separate the kinds of services, Id. at p.7. Ms. Cioffi stated that she often had to wait to find a place for a person in the following facilities: Dunklin, CARP, and Alcohope. Id. at p. 5. Ms. Cioffi stated that these were "residential" facilities, but she did not state whether these facilities were the equivalent of short-term inpatient hospital substance abuse facilities. These facilities are located in District IX, Id. at p. 7, but are not short-term in patient hospital substance abuse beds licensed as such. See paragraph 10 above. See also T. 96-99. In summary, although Ms. Cioffi identified a generalized need for residential or hospital substance abuse treatment, she did not draw any distinction between the two services. If there was a similarity, she did not provide evidence of the similarity. Lacking evidence in the record that need for residential treatment programs can be used to show need for inpatient hospital beds, Ms. Cioffi's testimony is insufficient to show need for the services sought by the Petitioner. The Honorable Marc Cianca is a County Judge in St. Lucie County. Deposition, Judge Marc Cianca, p. 2. Judge Cianca was of the opinion that his area attracted semi-young people with substance abuse problems in greater numbers than the retirement population. Id. at 17-18. He frequently was frustrated in his efforts to find substance abuse services for defendants in his Court. Id. at 3-5. Judge Cianca felt that most of the people he saw needed long-term therapy, beginning with inpatient services, followed by long-term follow-up programs. Id. at 12-14. Like the testimony of Ms. Cioffi (which concerned the same group of persons before the County Court), Judge Cianca did not clearly distinguish need for short-term inpatient hospital substance abuse services from need for all other forms of substance abuse treatment, and the record on this point is silent as well. For this reason, Judge Cianca's opinion that 100 short-term inpatient hospital substance abuse beds are needed must be rejected. The testimony of Ms. Cioffi and Judge Cianca is insufficient as a predicate for determining need for the inpatient hospital beds sought by the Petitioner for another reason, and that is the lack of evidence that the persons identified as needing substance abuse services will have the ability to pay for such services at Petitioner's facility, or that third party payment will be available for them. The people in need in Ms. Cioffi's testimony normally do not have funds to pay for treatment. Cioffi, p. 8. Similarly, a substantial number of the people in need seen by Judge Cianca do not have insurance coverage and would not be able to use Harbour Shores unless they qualified for Medicaid and unless Harbour Shores took all of those qualified for Medicaid. Id. at 7, 15-16. A substantial number of the persons needing substance abuse treatment do not have jobs or insurance and must rely upon "welfare" for services. Id. at 15, 17. These persons cannot afford certain programs, and must rely upon state aid through programs such as those provided by Indian River Community Mental Health Center, and for these programs there is always a waiting list. Id. at There is no evidence that any of these persons are eligible for Medicaid. Ms. Sharon Heinlen, Director of Planning and Development for Harbour Shores Hospital, who testified for the Petitioner as an expert in health planning and hospital administration, had not studied the Medicaid population in the area to determine need. T. 76. Although Harbour Shores had about 15 percent of its psychiatric patient days devoted to Medicaid patients, T. 33, the validity of this percentage for substance abuse patients, or for the reasonably near future, was not established by other evidence. Petitioner's formal application for this certificate of need projects 5 percent of its gross revenues from Medicaid and another 5 percent devoted to bad debt, indigents, and Baker Act cases. Petitioner's Exhibit 1, p. 5. But the application does not state whether this percentage will be evenly distributed among psychiatric and substance abuse patients. In any event, the percentages of indigent care and Medicaid care are too small to satisfy the need identified by Judge Cianca and Ms. Cioffi. Stated another way, the need identified by those two witnesses is not relevant to Petitioner's application except with respect to a small percentage. Harbour Shores plans to have after care for substance abuse patients. T. 40. The Savannas Hospital is the name of the hospital to be completed in November 1986 to provide, among other services, 20 short term inpatient hospital substance abuse beds under the certificate of need granted to Indian River Community Mental Health Center, Inc. T. 82-83. The primary service area of the Savannas Hospital will be the same four counties as now served by Harbour Shores Hospital, as well as Palm Beach County. T. 84. The Savannas Hospital intends to be licensed. T. 84. The Savannas Hospital is located in Port St. Lucie, in St. Lucie County. T. 95-96. The service proposed is a comprehensive substance abuse service. T. 87. Five of the twenty substance abuse beds will be devoted to detoxification. T. 92. The Savannas Hospital will be operated by the Mediplex Group in partnership with Indian River Community Mental Health Center, Inc. T. 82. The land will be owned by Mediplex. T. 95. The Savannas Hospital will be a private, for profit, hospital, while the Mental Health Center will be a not-for- profit facility. T. 86. The Savannas Hospital publicly states that it will take five percent indigent patients, which does not include Medicare. T. 87. There is no commitment to provide more indigent care. T. 89. All other patients acre expected to be fully paying. T. 93. The actual figure for free or nonpaying patients has not yet been calculated. T. 94. The Savannas Hospital will not serve Medicaid substance abuse patients because it is a freestanding facility. T. 36, 86. Humana Hospital Sebastian is the closest facility to Harbour Shores currently in operation providing inpatient short-term hospital substance abuse services, and Humana Sebastian can accept Medicaid patients. T. 59. Ms. Elizabeth Dudek testified for HRS as an expert in health planning and certificate of need review in Florida. Ms. Dudek has reviewed all of the applications made in District IX for substance abuse beds since November 1983, and as a supervisor, has reviewed all of the applications in the state for substance abuse beds. T. 104. She has been in contact with the District Alcohol, Drug Abuse, and Mental Health Program Office and has attended public hearings, as well as administrative hearings, concerning substance abuse beds in District IX. T. 104-05. She also listened to all of the evidence presented at the final hearing. It was Ms. Dudek's opinion that there was no need for the substance abuse beds sought by the Petitioner. T. 127-28. Ms. Sharon Heinlen was also qualified as an expert in health planning, as well as hospital administration. T. 13. Ms. Heinlen has only recently moved to Florida, T. 11, 66, and stated that she did not know Florida well enough to know what might be the best thing to advocate in Florida with respect to whether all hospitals should provide all services. T. 65-66. She had conducted studies of District IX, however. T. 66. The average occupancy rates for District IX testified to by Ms. Heinlen were mathematically incorrect, and the correct lower rates do not support her opinion that additional short term substance abuse beds are needed. See FF 11. The fact that about 80 percent of the psychiatric patients now are at Harbour Shores Hospital also have a substance abuse problem does not necessarily support Ms. Heinlen's opinion as to need. See FF 12. This statistic is consistent with experience in all of Florida, and therefore should be accommodated by the HRS numeric need methodology. Moreover, it must be inferred that hospitalization of these patients as psychiatric patients was proper, rather than as substance abuse patients, and that even if additional substance abuse beds were available, these patients still would need to be in a psychiatric bed for treatment of the primary diagnosis. As discussed in FF 14, the data concerning recent requests for substance abuse services at Harbour Shores Hospital is not sufficient to conclude that a need exists for additional beds. As discussed in FF 17, Ms. Heinlen did not have an adequate basis for any opinion as to the need for short-term substance abuse beds for Medicaid patients in District IX. Finally, Ms. Heinlen testified that there was a waiting list for patients to be admitted to licensed short-term substance abuse beds at Fair Oaks and Lake Hospital, but the testimony was hearsay. T. 28. Since this evidence conflicts with the relatively low occupancy rates at these same facilities, and has not otherwise been corroborated by non-hearsay evidence, it must be rejected as a basis for a finding of fact. Further, due to the conflict with the low occupancy rates, it is rejected as a basis for Ms. Heinlen's expert opinion. In summary, Ms. Heinlen's expert opinion that there is a need for short-term, inpatient hospital substance abuse beds in District IX must be rejected. It is the position of HRS that even if the rule showed a need, the occupancy factor would be a factor in showing no need. T. 134. Conversely, if the rule showed no need, the occupancy factor would be one factor among others which night show need. Id.
The Issue Whether under Section 381.494(6)(c)-(d), Florida Statutes, and Rule 10-5.11, Florida Administrative Code, Respondent, Martin Memorial Hospital Association, Inc., is entitled to a Certificate of Need ("CON") authorizing a proposed 75-bed satellite hospital in Port Salerno, Martin County, Florida.
Findings Of Fact Background Respondent, Martin Memorial Hospital Association, Inc. ("Martin Memorial"), seeks a CON to construct and operate a 75- bed satellite hospital in Port Salerno, Florida. Of the proposed 75 acute-care beds, 50 will be new and 25 will be transferred from Mar in Memorial's existing hospital in Stuart, Florida. The proposed satellite hospital will have six intensive care beds, 69 medical-surgical beds, and two operating rooms. Respondent, Department of Health and Rehabilitative Services ("DHRS"), preliminarily issued the applied-for CON. After Petitioner, Palm Beach-Martin County Medical Center ("PBMC"), owner of nearby Jupiter Hospital in northern Palm Beach County, requested a Section 120.57 hearing to contest issuance of the CON, DHRS forwarded this case to the Division of Administrative Hearings for assignment of a hearing officer. This case, in its earlier stages, was a consolidated proceeding with numerous parties and party-applicants. In the fall of 1984, several parties withdrew or were dismissed. One of the them, American Healthcorp., dismissed its challenge of DHRS' denial of its application to construct a 120-bed new hospital in Vero Beach, Florida. Prior to that dismissal, American Healthcorp. had filed a mandamus action in Leon County Circuit Court to require DHRS to issue a CON. The writ of mandamus was issued and that order was appealed by DHRS to the First District Court of Appeal. On June 18, 1985, the First District reversed the Circuit Court's order. DHRS never issued a CON to American Healthcorp., as the writ of mandamus was stayed during the pendency of the appeal. Initially, PBMC, another party-applicant, contested DHRS' denial of a proposed 66-bed addition to its existing hospital in Jupiter, Palm Beach County, Florida. Later, PBMC dropped its opposition to the denial after concluding that, due to a dramatic drop in patient census during 1984, additional beds in the area were not needed.1 On Martin Memorial's unopposed motion to dismiss, PBMC was dismissed as a party. Other nonapplicant intervenors subsequently withdrew. In the earlier consolidated proceeding, Martin Memorial had contested the denial of its initial application (filed in 1983) for a CON to construct a 150-bed satellite hospital in Port Salerno, on the same site as now proposed for the 75-bed hospital. In October, 1984, Martin Memorial revised its application, within a deadline for submittal of amended applications set by prehearing order. This revised application, now the subject of this proceeding, reduced the number of beds in the proposed hospital from 150 to 75: 50 were to be new and 25 were to be transferred from Martin Memorial's existing hospital in Stuart.2 This 75-bed application was then preliminarily approved by DHRS, as part of an effort to settle the pending consolidated proceeding. After notice of the approval was published on December 7, 1984 in the Florida Administrative Weekly, PBMC timely requested a hearing to contest it. PBMC's position is, generally, that another hospital in the area is not needed and will result in an unnecessary duplication of services and that, if built, the hospital would draw patients who would otherwise use Jupiter Hospital, to the economic injury of PBMC. The remaining party-applicant in the consolidated proceeding was Lawnwood Medical Center, whose proposed 50- bedexpansion of its hospital in Fort Pierce (St. Lucie County), was preliminarily approved by DHRS. Martin Memorial requested a Section 120.57(1) hearing to contest the approval. By stipulation dated May 15, 1985, Lawnwood Medical Center was dismissed as a party. Martin Memorial II. The Parties The applicant, Martin Memorial, operates a not-for- profit community hospital in Stuart, Florida, which has served the health care needs of the area since 1939. At that time, it had 25 beds and the site consisted o eight acres. In subsequent years, Martin Memorial added five additional acres of land, and the hospital now has 336 beds, including 26 new beds: nearing completion. Martin Memorial is a subsidiary of Coastal Health Corporation, a not-for-profit holding company. One of the holding company's other non-profit subsidiaries, Coastal Care Corporation, provides services such as ambulatory surgery and primary or emergency care at medical treatment centers. Martin Memorial and its parent corporation, Coastal Health Corporation, are governed by boards comprised of full- time residents of Martin County who serve without compensation. Martin Memorial Hospital has a proven record of providing health care to indigents. Its policy is to provide health services without regard to race, religion, national origin, or a patient's ability to pay. It has always participated in the Medicare/Medicaid Programs and participates in the county indigent program. It proposes to follow the same policy at the proposed satellite hospital. Martin Memorial Hospital, in Stuart, is adjacent to the St. Lucie River on the north, bounded by the Heida-Brad Park residential development on the east, by the St. Mary's Episcopal Church on the south, and by various businesses and residences on the west. It would be difficult for Martin Memorial Hospital to expand to meet anticipated future demand. It has found it impractical to buy additional land adjacent to its existing facility. (It does not nave eminent domain power.) Under current zoning, its height is limited to the existing six floors. Other obstacles include problems with parking access and compliance with fire safety codes. Palm Beach-Martin (PBMC) PBMC is a non-profit corporation, organized in 1973, with the stated purpose of serving tee health care needs of residents of northern Palm Beach County and southern Martin County. It operates a community not-for-profit hospital, known as Jupiter Hospital, in Jupiter, Florida. A 156-bed acute care hospital, it is the northern most hospital in Palm Beach County and provides health care services to the residents of northern Palm Beach and southern Martin Counties. Over 10% of Jupiter Hospital's patients come from the Hobe Sound area of Martin County, and another 20% come from the Tequesta area of Martin County. The boards which operate PBMC and Jupiter Hospital are made up of volunteers; one-half of whom are doctors on the hospital's medical staff, and the other half are lay-members from the community. All policy decisions are made by the boards. The hospital is managed, under contract, by hospital Corporation of America Management Company (owned by Hospital Corporation of America) which supplies only the hospital administrator and finance director, all other personnel are employees of PBMC. Like Martin Memorial, PBMC has a practice of providing health care to indigent patients. It has a Medicaid contract at its convalescent pavilion and treats Medicaid patients requiring care. (Since it has not had a Medicaid contract with the state, PBMC "writes-off" the cost of care provided to Medicaid and indigent patients. But due to an increasing number of Medicaid patients, PBMC has applied for a Medicaid contract.) It has a current contract with Palm Beach County to treat indigents in its out-patient facility. Department of Health and Rehabilitative Services (DHRS) DHRS is designated by statute as the single state agency charged with issuing and denying CONs in accordance with district plans, DHRS rules, and state and federal statutes. See, Section 381.494, Florida Statutes (1983). Geographic Facts The proposed satellite hospital would be located in Port Salerno, Martin County, 5 1/2 miles south of Martin Memorial Hospital and 15 miles north of Jupiter Hospital. The site of the proposes hospital is 35 acres in size, and is located approximately 1/4 mile east of Highway U.S. 1, on Port Salerno Road. Jonathan Dickinson State Park, abutting Highway U.S. 1 for five miles, is situated between the site of the proposed satellite hospital and PBMC's Jupiter Hospital. The area of Hobe Sound is just north of this State Park. The proposed hospital would be adjacent to the Martin County Campus of Indian River Community College. III. Standing of PBMC: Expected Impact of Proposed Hospital on PBMC. Since it is physicians who admit patients to hospitals, the extent to which medical staffs overlap is one factor used to project how a new hospital will affect an existing one. Martin Memorial Hospital and Jupiter Hospital have distinct medical staffs and there is no material overlap. Neither has it been shown that Jupiter Hospital physicians will seek staff privileges at the proposed satellite hospital. It is reasonably expected that the proposed hospital will be staffed, for the most part, by physicians who are also on the staff of Martin Memorial Hospital. Nevertheless, the proposed satellite hospital would draw away a substantial portion of Jupiter Hospital's patient base and is intended to reduce Jupiter Hospital's market share in the Hobe Sound area to near zero. (Indeed, this is a result projected in Martin Memorial's Long Range Plan.) Martin Memorial (in its Long Range Plan) estimates Jupiter Hospital's current market share to be 65%. Jupiter Hospital's primary service area includes Hobe Sound, from which it draws approximately 10% of its patients. The northern boundary of the Hobe Sound area is 20 minutes driving time from Jupiter Hospital. Hobe Sound is also within the primary service area of the proposed satellite hospital. The proposed hospital would be in the same DHRS Service District as Jupiter Hospital and both hospitals would have overlapping primary service areas. The projected loss of 10% of its patient base to the proposed satellite hospital would have a significant adverse financial impact on PBMC. It has not been shown, however, this impact would imperil the continued financial feasibility of Jupiter Hospital. IV. Numerical Bed-need Projected by Applying DHRS Rule-Based Bed-need Methodology. The proposed satellite hospital would be located in DHRS Health District 9, which consists of Indian River, St. Lucie, 55artin, Okeechobee, and Palm Beach Counties. The state acute care bed-need methodology is a complex formula contained in Rule 10-5.11(23), Florida Administrative Code. It projects bed-need, on a district-wide basis, five years into the future, creating what is referred to as a "five-year planning horizon" for assessing acute care bed-need. The formula requires several district-specific inputs, including population forecasts in four age groups, the average fertility rate in the district for the three most recent years, the average historical utilization rate in the district for the three most recent years, together with specific factors used to determine the net flow of elderly patients. Three other input factors are applied uniformly to all districts: discharge rates by service and by age cohort for Florida residents, average length of stay by service and by age cohort, and occupancy standards by service and by age cohort. Application of the formula entails seven steps: Project patient days by service and by age cohort using the formula: Patient days = projected population x discharge rate x average length of stay Adjust the projected patient days for the 65 and over age cohort to account for patient flows. Calculate bed-need by applying service- specific occupancy standards to projected patient days. Calculate the district bed allocation by summing the beds needed by service. Calculate the projected occupancy of these beds using the district's historical utilization rate. If the projected occupancy. rate is less than 75 percent or greater than 90 percent, apply specified formulas to adjust the district bed allocation (downward or upward, respectively. Check to ensure that each district will be able to meet peak demand based on the adjusted allocation. (R-l8l/, Testimony of Kolb) Population projections used in the methodology are: "for age- specific cohorts residing in the relevant district projected five years into the future," Rule 10-5.11(23)(f)1., Florida Administrative Code. These age-specific cohort projections (of county populations) must be "those developed by the State Health Planning Agency, and will be based on the latest mid-range projections published by the Bureau of Economic and Business Research of the University of Florida [BEBR]." Id. There are currently 4,695 licensed or approved acute care beds in District 9, which includes the 50 additional beds (preliminarily) approved for the proposed satellite hospital and the 45 beds approved in a subsequent batching cycle. For July, 1989, application of the bed-need methodology shows a district wide gross need of 4,621 beds. This is based on population forecasts for July, 1939, released by the Governor's Budgeting and Planning Office on- January 1, 1985.This office interpolates and publishes population forecasts based on figures received from BEBR. Since later 1934 (when Rule 27E-2.01-.04 was adopted requiring state agencies to use, in their planning, population projections provided by the Governor's Office), DHRS's Office of Health Planning and Development has used such forecasts to project bed-need under the methodology. These forecasts are appropriate for such use since they are "developed" by the State Health Planning Agency and based on the latest mid-range projections published by BEBR. When the licensed or approved bed total of 4,645 (excluding 50 beds for the proposed satellite hospital) are subtracted from the district wide gross need, there is a net surplus of 24 beds. If the 50 beds of the proposed satellite hospital are included, the net surplus increases to 74 beds. A planning horizon of January, 1990, however, is more appropriate. It more closely conforms to the methodology's requirement that need be projected five years into the future. (At hearing, all parties agreed or acquiesced to the proposition that the five year planning horizon should begin to run, to the extent possible, from the date of final hearing in June, 1985.) The latest county-wide projections released by the Governor's Office for state agency use, projects population by age and sex cohorts for January 1, 1990 and July 1, 1990. The July 1, 1990 projections are beyond the five year horizon and so less suitable for use in the methodology. Applying, then, the bed-need methodology to project bed-need for January, 1990, shows a gross need of 4,702 beds, resulting in a total district wide net need of 57 beds (excluding the proposed 50-beds satellite hospital). Hence the methodology shows (just barely) a January, 1990 need for the 50- beds sought for the proposed satellite hospital. Because of projected increases in district population, the methodology predicts a significant growth in bed-need between July, 1989 and January, 1990: bed-need grows by 81 beds or by more than 10 beds per month. PBMC contends that a planning horizon of July 1, 1989, and no later, must be used since DHRS has, historically, updated bed-need projections only on July 1 of each year. Annual updates were limited to once a year because updated population figures were received only in July. Now, however, -he situation has greatly improved. DHRS receives updated population forecasts from the Governor's Office twice a year--in January and July. There is no reason why these updated and, presumably, more accurate population forecasts cannot be used to project bed-need Martin Memorial, on the other hand, argues for a more distant planning horizon--April 1, 1990. This horizon, however, requires use of BEBR projections recently received but not yet released or interpolated by the Governor's Office, until released, such projections are not appropriate for use by state agencies. See, Rule 27E-2.01-2.04, Fla. Admin. Code. V. Consideration of Other CON Review Criteria [A CON may be denied even though the bed- need methodology projects a need for the proposed beds five years into the future. Rule 10-5.11(23)(b): "An unfavorable Certif- icate of Need determination may be made when a calculated bed-need exists, but other criteria specified in Chapter 3Bl.494(6)(c), Florida Statutes, are not met." DHRS must consider CON applications in light of all statutory and rule criteria. See, Department of Health and Rehabilitative Services v. Johnson & Johnson, 447 So. 2d 361 (Fla. 1st DCA 1984).] Subdistrict Need: Allocation of District Wide Bed-need to|-. Relevant Subdistrict In 1983, the Local Health Council divided District9 into five subdistricts: (1) Indian River County, (2) Martin and St. Lucie Counties, (3) Okeechobee County, (4) northern Palm Beach County, and (5) southern Palm Beach County. Each subdistrict "is an area where the co-unity, by itself, uses the facilities in an area. It is supposed to be a sort of natural boundary that separates the different communities." (TR-413) The council also adopted a methodology for allocating acute care beds among the five component subdistricts. (R.-19) Although DHRS has not yet adopted, by rule, District9's subdistricts and subdistrict bed-need allocation methodology, both are part of District 9's Local Health Plan, adopted after a series of workshops and public hearings. The subdistricts were identified pursuant to a protocol furnished by DHRS which required consideration of whether an area was urban or rural, or comprised a standard metropolitan statistical area (SMSA). Under the protocol, an SMSA must be designated a separate subdistrict. Since Martin and St. Lucie Counties form a SMSA, they form a separate subdistrict. The five subdistricts of District 9 were identified in a rational manner, have a factual basis, and are useful tools for health care planning purposes. The methodology for allocating district wide bed-need to the subdistricts, also part of the Local Health Plan, has also been shown to be supported by reason and accepted health care planning concepts. DHRS cannot rationally determine the need for additional acute care beds, at least in =he context of this case without looking at subdistrict need or lack of need. In this way, local needs and conditions are considered in the decision- making process. District 9 is too large to serve as a useful unit for acute care bed planning purposes. Applying the Local Health Plan's sur5istrict bed-need allocation methodology to the July, 1989 planning horizon, indicates a net acute care bed-need for the Martin/St. Lucie-- County Subdistrict (not counting the 50 beds at issue) of 103beds. If the proposed hospital were approved, the subdistrict bed-need methodology would show a remaining subdistrict need for53 acute care beds. (R-18, TR-249) When applied to the January, 1990 planning horizon, preferred to the July, 1999 horizon., the subdistrict methodology shows a net acute care bed-need of 119 beds for the Martin/St. Lucie County Subdistrict (not counting the 50 beds at issue). Thus, the bed-need allocation methodology contained in District 9's Local Health Plan, shows a need for the proposed 50-acute care beds, with a 69-bed margin. (T-18, TR-248) Since the total number of licensed and approved beds (excluding the 50-beds at issue) for the subdistrict is 761, the projected need for 119 new acute care beds in January, 1990, is considered to be substantial. But the subdistrict bed allocation methodology assumes, incorrectly, that patients do not "cross-over" from one subdistrict to another. It fails, therefore, to consider or take to account the significant number of patients residing in the Martin/St. Lucie Counties Subdistrict who use acute care beds at Jupiter Hospital, located in the subdistrict to the south. This failure in the subdistrict methodology detracts from the weight to be given the resulting bed-need calculation. Availability, Accessibility and Adequacy of Like and Existing Facilities. Section 384.494(6) (c) 2.; Florida Statutes, requires review of CON applications in context with the "availability . . . accessibility and adequacy of like and existing health care services . . . in the district of the applicant."] Excess or under-utilized acute care bed capacity is a problem because it contributes to higher health costs. There are fixed overhead costs associated with acute care beds, whether empty or filled by a patient. These costs must ultimately be borne by the patients, or their insurers In reviewing a CON application, DHRS considers the number of available unoccupied beds at the facility and in the county or subdistrict for the most recent calendar year, determines actual occupancy rates, and compares them against an 80% occupancy standard, a standard generally accepted by health care planners. For example, one stated reason for DHRS' denial of Martin Memorial's initial 150-bed application was the availability of 20 unoccupied medical-surgical beds at Martin Memorial in 1982, on an average daily basis. Similarly, the average daily availability of 73 unoccupied medical-surgical beds in the five hospitals within PBMC's service area, plus additional approved but not licensed beds in the area, were stated reasons for DHRS' denial of PBMC's initial 1983 application for additional beds. (R-13) Applying the 80% occupancy standard to 1984 bed utilization statistics in the Martin/St. Lucie County Subdistrict, there were 111 unoccupied acute care beds on an average daily basis, not counting the 50 new beds recently approved for Lawnwood Hospital and the 26 new beds soon to be available at Martin Memorial. This is a 63.7% occupancy rate. Moreover, there were 47 unoccupied licensed beds on an average daily basis at Martin Memorial Hospital (not counting the 26 new beds under construction). The same calculation using only medical-surgical beds shows that in 1984, on an average daily basis, Martin Memorial had 36 unoccupied medical-surgical beds or an occupancy rate of 661. At Jupiter Hospital and Port St. Lucie Medical Center, the two hospitals having overlapping service areas with Martin Memorial, there were 31 (58.2% occupancy rate) and 49 (43.7% occupancy rate) unoccupied medical-surgical beds, respectively, on an average daily basis. (HRS-4) There is an ample supply of available beds: there is not a shortage of acute care hospital beds at Martin Memorial Hospital or in the Martin/St. Lucie Subdistrict. Martin Memorial has shown only that there may be, or could be, bed availability problems during certain peak months at Memorial Hospital in 1990, based on seasonal considerations. At most, it has shown that, without the proposed satellite hospital, the average occupancy for its highest occupancy month in 1990 would reach 91%. (TR-263-265) However, it is possible to operate a hospital at such an occupancy level for several months and yet maintain an acceptable level of service. (Moreover, Martin Memorial's analysis fails to take into account acute care beds which would be available in 1990 at Port St. Lucie Medical Center and Jupiter Hospital, where occupancy rates would be much lower.) Martin Memorial does not assert that in 1990 its average daily occupancy rate will exceed 80%. Indeed, assuming the validity of its average length of stay and hospital utilization assumptions (which are questionable), Martin Memorial forecasts an average daily occupancy rate of 79.4%. The State Health Plan states that "the issue of surplus beds is expected to be an even greater problem in the future because of the growth of alternative delivery systems"-- (R-20, p.22)--a proposition with which Martin Memorial's expert generally agrees. (TR-287-288) The State Health Plan concludes that "the combined effect of ambulatory surgery, HMOs, DRGs, and other innovations could reduce acute care bed-need by 15% or more." Id. Thus it becomes more likely that there will be an ample supply of available unoccupied beds in the subdistrict through 1990. The proposed satellite hospital would improve or enhance the accessibility of hospital services, since it would be located closer to some patients than either Martin Memorial or Jupiter Hospital. However, it has not been shown that geographic accessibility has been or will be a serious problem without construction of the proposed satellite hospital. The proposed hospital would be located about 5 miles from Martin Memorial and about 15 1/2 miles from Jupiter Hospital. Patients in the southern part of Martin County, residing south of the northern part of Hobe Sound, can be driven to Jupiter Hospital in 20 minutes or less. The definitive standard, commonly used by DHRS and.generally accepted by health care planners to detect geographical bed-access problems, is the 30 minute drive-time standard. Under this standard, if acute care hospital beds are available and accessible, within an automobile travel time of 30 minutes under average traffic conditions to at least 90% of the population, there is no cause for concern about geographic accessibility. It is undisputed that hospital beds are now available well within 30 minutes travel time to all residents of Martin County during all relevant periods, and will continue to be so through 1990. In short, geographic accessibility is not a current or projected problem and although the satellite hospital would make in-patient services more geographically accessible to some residents, such a result could be expected whenever a new hospital is constructed, whatever its location. As to adequacy of existing and licensed and approved facilities, there is no showing that the quality or extent of health care provided is inadequate. Extent of Utilization of Like and Existing Facilities. [Under Section 3Bl.494(6)(c)2., Florida Statutes, CON applications must also be reviewed in context with the "extent of utilization . . . of like and existing health care services . . . in the service district of the applicant."] During the last two years, the health care industry has undergone major changes resulting in a sudden and dramatic decline in the use of hospital in-patient services. The main cause of this decline was the shift, in October 1983, to the Medicare prospective payment system, otherwise known as Diagnostic Related Groupings (DRGs). The DRG payment system changed Medicare reimbursement from a cost basis to a set reimbursement based on diagnosis. It has caused a sharp decline in the average length of stay of Medicare patients as well as a decrease in Medicare admissions, and a resulting decline in hospital occupancy levels. For example, in calendar year 1984, the average length of stay for hospital patients in District 9 dropped from approximately 6.8 to 6.1. (DHRS-4) Another recent development contributing to the general decline in hospital utilization .is the increasing emphasis on providing out-patient services such as out-patient surgery and home health services. In many areas of the country, the advent of Preferred Provider Organizations (PPOs) and Health Maintenance Organizations (HMOs) have significantly impacted hospital occupancy rates, not only by encouraging shorter lengths of stay, but also by greatly lowering admission rates. (TR-604-05, 773: P-3, p.l8) According to the State Health Plan, "the emergence of [these] alternative delivery systems . . . [has] exacerbated declining occupancy rates." (R-20, Vol. 1, p.26) Martin Memorial has developed free-standing medical treatment centers in Hobe Sound and Port St. Lucie, which can provide up to 80% of the services required by patients in hospital emergency rooms. (TR-56-59 R-1, Vol. II at 73) In addition, Martin Memorial is merging with VNA in Martin County to provide home health care. These alternative delivery systems, along with the expected increasing effects of PPOs, HMOs and prospective payment systems, will contribute to further declines in hospital utilization and tend to extend the time during which the existing inventory of acute care beds will be adequate to meet the needs of a growing population. Although witnesses for Martin Memorial suggest that this steady decline in bed utilization at Martin Memorial will soon bottom out and that the average length of stay will fall no further, the weight of credible evidence shows that hospital utilization in District 9, as well as in the country as a whole, is still declining and that no one can say, with any reasonable degree of certainty, just when the decline (or "nose dive," TR- 682) in average length of stay and overall utilization will stop. The Executive Director of the District 9 Health Council predicts that the average length of stay in the district may fall from its current level of 6.1 or 6.2 to 5, and a recent survey of southeastern hospitals predicts at least a 5% further decline in average length of stay from 1984 levels. Hospital admission or discharge rates in District 9 fell slightly in 1984, but on the whole have remained relatively constant. The decline in hospital utilization has been chiefly caused by the unprecedented drop in average length of stay. Several factors causing declines in hospital admissions as well as utilization in other areas of the state and country have not yet begun to significantly affect northern Palm Beach, Martin, and St. Lucie Counties. It is, however, reasonably expected that these factors, such as out-patient or ambulatory surgical centers, and home health services (which are becoming increasingly used), will continue to grow and further decrease hospital in-patient surgery and care, admissions, and utilization. (TR-39-40, 578) As stated earlier, the most recent State Health Plan predicts that the combined effect of ambulatory surgery and other alternative health care delivery systems will be to reduce hospital discharges (or admissions) by 15% or more by 1989. (R-20, Vol. 2, p.72) Another factor which will contribute to the further decline in hospital utilization in Florida will be the required adoption of prospective payment programs by hospitals and private insurers for non-;5edicare patients. Under the Florida Health Care Access Act of 1984, hospitals are required to negotiate a prospective payment arrangement with each health insurer representing 10% or more of the hospital's private pay patients (R-20, Vol. 1, p. ll) To date, the only active HMOs in District 9 are found in southern Palm Beach County and their impact has not been felt elsewhere in the district. It is reasonably anticipated, however, that HMOs will expand throughout the district in the next several years and will contribute to a further decline in admissions. (Some commentators predict HMOs will reduce hospital admission rates by as much as 40%.) (TR-605-06, 679-80) Even without these factors, the extent of the decline in hospital utilization in District 9 has been dramatic. Overall hospital utilization i., 1984 declined from a level of 73.7% in 1983 to 65.R% in 1984. (HRS-4; R-~9: P-5) The District 9 medical-surgical utilization rate dropped from 76.3% in 1983 to 67.3% in 1984. The decline in hospital utilization in the Martin/St. Lucie Subdistrict and at specific hospitals in the area have been even more pronounced: 1983-84 Subdistrict and Specific Hospital Utilization Rates Overall Rate Med-Surg Rate 1983 1984 1983 1984 Martin/St. Lucie Subdistrict 76.9%1 63.7% 78% 63.8% Martin Memorial 74.1% 65.1%4 74.8% 66.0% Port St. Lucie5 38.9% 44.1% 37.6% 43.7% Jupiter Hospital 67.7% 55.7% 71.4% 58.2% Although experts disagree on how long the decline in hospital utilization will continue and how far it will fall, it is apparent that hospital utilization is continuing to decline in District 9 in 1985. By January and February 1985, the Martin/St. Lucie Subdistrict medical-surgical utilization declined about 1% from the same period in 1984. However, the most recent data for March, 1985, shows a decline in monthly medical-surgical utilization from 73.2% in 1984 to 68.8%. Martin Memorial's patient days in 198; are less than the corresponding number of patient days .when compared to the same periods in 1984. In addition, in no single month during Martin Memorial's 1985 fiscal year, beginning on October 1, 1984, has Martin yet achieved its budgeted patient days or admissions. In fact, Martin Memorial's bed utilization is more than 10% under budget for fiscal year 1985. (TR-66-69 P-1) Martin Memorial contends that the projected increase in the population and the aging in population in the Martin/St. Lucie Subdistrict will offset the decline in average length of stay and gradually increase the in-patient population. Although it is reasonable to expect that such factors would increase utilization, over the last year in District 9, use rates have gone down and admission rates have decreased slightly even though population increased and aged. In addition, it is reasonably anticipated that the future negative impact of HMOs on use and admissions will offset these population changes and contribute to further decline in utilization. Projected Utilization of Martin Memorial Hospital and the Proposed Satellite Hospital. In projecting- utilization for its existing and proposed facilities, Martin Memorial used 1984 District 9 use rates and a constant to increasing average length of stay. (R- 18) Use and admission rates have declined steadily for several years for the under-65 population and, in the first year of DRGs, dropped by over 3% for the over-65 population. Vet, despite this t-end and projections of decreasing use and admission rates in the future due to alternative delivery programs, Martin Memorial's utilization forecast uses admission rates slightly higher than the 1984 actual admission rates. (R- 18) In projecting the average length of stay for the proposed Port Salerno Hospital, Martin 'Memorial discounted 10% from its 1984 average length of stay. This discount, however, was due to projections of lower Medicare utilization and lower intensity of services at the new hospital, and makes no allowance for any further decreases in average lengths of stay. Similarly, in projecting utilization for Martin Memorial Hospital, Martin Memorial assumed an increase over the 1984 average length of stay of 6.0 days to 6.1 days in 1990. These assumptions are unreasonable in that they 'ail to fully into account the current and projected continuing decline in hospital admissions and utilization. Consequently, little weight is assigned to Martin Memorial's forecast of future bed utilization—that the satellite hospital would experience 58% occupancy in 1990, the first year of operation. Martin Memorial projects that without the proposed Port Salerno Hospital, Martin Memorial would achieve an occupancy rate of 79.4% in 1990. This utilization projection was based on population projections for Martin County done in 1984 by Dr. Stanley Smith it fails to take into account Dr. Smith's recent revision downward of the 1990 population projections for Martin County from 100,900 to 98,700. (TR-95-96, 294-95 R-6) The decline in average length of stay and hospital use rates will have a major impact on the number of empty beds in District 9 and, at least as applied to this District, the bed- need methodology of Rule 10-5.11(23), over-states the need for additional beds in 1990. The methodology uses a constant average length of stay derived from prior years. It is not an accurate predictor of future occupancy when, as now, use rates and utilization are declining and are reasonably expected to continue to decline. (TR-639-41: TR-614, 621-22 P-5) Martin Memorial's projected 79.4% occupancy rate in 1990 is overstated because it fails to fully take into account continuing declines in average length of stay and use rates, and because the 1984 population figures used to derive the 1984 use rate may be understated, thereby overstating the use rate. Similarly, using the same assumptions, an occupancy rate of 69.7% was projected for Jupiter Hospital (without the proposed Port Salerno Hospital). This projection is also overstated for the same reasons. If the Port Salerno Hospital were approved, the 1990 occupancy figures for both Martin Memorial and Jupiter Hospital would, in all likelihood, be much lower. (TR-627 P-5) 21 A more credible projection of ,Martin Memorial Hospital's 1990 occupancy rate was offered by Thomas W. Schultz, PSMC's health care planning expert. By reducing the 1984 Dis- trict 9 use rate by 2.9% to account for declining utilization at Martin Memorial during the first three months of 1985, as well as the general continuing decline in hospital utilization, Mr. Schultz projected Martin Memorial's 1990 occupancy (without the new facility) to be only 72.6%. Similarly, because use rates are still declining and because the 1985 population numbers used to calculate the rates may have been understated, this projected occupancy is overstated. (TR-628-30, 635-38; P-5) The State Health Plan and the District 9 Local Health Plan [Section 381.494(6) (c)l. requires that CON applications be reviewed "in relation to the applicable district plan and state health plan "] Several specific utilization and occupancy standards are contained in the State Health Plan and the District 9 Local Health Plan. A major stated goal of the State Health Plan is to promote the efficient utilization of acute care services by raising the occupancy rates or acute care hospitals. (R-20) It identifies 80% as the appropriate minimum occupancy level for acute care hospitals: an average annual occupancy rate of at least 80% is made an objective. As conceded by Debra S. Kolb, Ph.D., Martin Memorial's expert health planner, the 80% occupancy standard is an appropriate minimum standard which should be "looked at as a hurdle before beds are added." (TR- 289) She adds, however, that "there are other factors, such as . . . size of the facility, seasonality issues, age problems. . . that would warrant special cases." (T?.-289) By policy and practice, DHRS considers current occupancy levels to be an important criterion and has applied this 80% occupancy standard in reviewing CON applications. Its use of these standards is illustrated by its initial action on the various applications which were once part of this proceeding. DHRS granted Lawnwood Hospital's application for several stated reasons, one of which was a 1982 occupancy rate of 90%. Martin Memorial's initial 150-bed application and PBMC's 60-bed application (later dropped) were denied, in substantial part, because of low utilization rates in 1982 and because there was an adequate supply of beds currently available. (Interestingly, both Martin Memorial and PBMC had 1982 occupancy rates exceeding 70%.) As stated by Gene Nelson, then supervisor of DHRS' CON review section, in the State Agency Action Report denying PBMC's application: "Overall utilization for Palm Beach Martin County Medical Center for 1982 was 72.3% and medical/surgical utilization was 76.4%, neither being sufficiently high to justify additional beds." (R-13) The Acute Care portion of District 9's Local Health Plan (1984), contains "Recommendations by Priority Ranking" reflecting policies and priorities which, according to the local health council, should be used (in addition to the DHRS bed-need methodology) in planning and allocating acute care bed-need. Priority I delineates the subdistricts for purposes of allocating acute care hospital beds: Priority II establishes the_ subdistrict allocation methodology: Priority III establishes an occupancy rate which must be met before additional beds may approved: Before needed beds, as determined by Rule 10-5.11(23), may be approved, applicants requesting additional acute care beds should demonstrate that certain occupancy thresholds have been achieved relative to medical/surgical, obstetric, pediatric and ICU/CCU beds. The average annual occupancy rate (most recent calendar year) in the applying facility and its corresponding subdistrict average, should equal or exceed the following levels (inclusive of CON approved beds): Medical/Surgical 75% Obstetrical 65% Pediatric 65% ICU/CCU, Monitored & Intermediate Care 75% (e.s.) (R-10, pp.48-49) The rationale for this standard is set forth in the plan: With the advent of the Medicare prospective reimbursement system, there is literally no way to estimate the magnitude of impact that this reimbursement mechanism will have on hospital admissions, occupancy rates, and average lengths of stay. Therefore, relying upon the national standard of 4 beds/1000 population was not adequate. There is a need for an indicator based solely on utilization for the elderly. Since the reimbursement mechanism for non-Medicare patients has not changed, a resource based methodology has been utilized for this population group. Moreover, the program goals of the Local Health Plan state that the overall occupancy rate in District 9--as a whole--for licensed acute care beds as well as the occupancy rate for medical/surgical beds should equal or exceed 756. (Id. at 47) These minimum annual district and subdistrict occupancy rates take seasonality and age considerations into account. Bed utilization or occupancy standards are the only bed-need criteria that look to actual, verifiable data reflecting current conditions as opposed to forecasts, which look to the future.6 Failure to achieve the occupancy standards of Priority III A of the Local Health Plan creates, at least, a strong presumption against the approval of the project. In exceptional situations, however, additional beds may be approved even though the occupancy standard is not met. A typical example projected continuing decline in hospital admissions and utilization. Consequently, little weight is assigned to Martin Memorial's forecast of future bed utilization--that the satellite hospital would experience 58% occupancy in 1990.the first year of operation would be where there was a geographical access problem.7 Both the State and District 9 Health Plans cite the high cost of unused hospital beds which add to the cost of hospitalization. (R-20, p.70: R-10, p.10) A primary goal of both plans is to raise occupancy rates and eliminate excess beds. With this in mind, it is reasonable to give considerable weight to current utilization rates even though a numerical "need" for the beds is projected by the DHRS bed-need methodology.8 1984 bed utilization at Martin Memorial, Port St. Lucie and Jupiter Hospitals, as well as average utilization for District 9 and the Martin/St. Lucie Subdistrict, fall well below the minimum occupancy standards normally applied by DHRS and set out in the State and Local Health Plans. These minimum occupancy standards have not been met and are not reasonably projected to be met by 1990. Considerable weight should be accorded this factor since that is the effect of the State and Local Health Plans and DHRS' normal practice. Moreover, since occupancy rates are based on actual current conditions, they are less subject to manipulation, and inject a healthy measure of reality into CON decision-making during a time of great change in the health care industry. Economies and Improvement Services Derived From Operation of Joint, Cooperative or Shared Resources. [Another CON criterion is whether there will be "[p]robable economies and improvements in service that may be derived from the operation of joint, cooperative, or shared health care resources." Section 3fll.494(6)(c)5., Florida Statutes.] Because the proposed project is a satellite hospital, there will be economies and improvements in services realized from the operation of joint, cooperative or shared health care resources, as compared to the operation of a wholly separate free-standing hospital. The satellite hospital will not offer obstetrics or a defined pediatric unit. It will not have a CAT Scanner, a -- personnel office similar to Martin Memorial's, or a hospital laundry. It will have an emergency room, normal operating room suites, and radiology and lab services, although the more complex lab tests will be performed at Martin Memorial Hospital. The Need for Research and Educational Facilities. [CON applications are also reviewed in context with the "need for research and educational facilities . . .." Section 381.494 (6)(c)7, Florida Statutes.] The proposed satellite hospital will be located directly adjacent to the Martin County campus of Indian River Community College ("IRCC"). The IRCC campus has an Allied Health Building with approximately nine classrooms, a nursing 120 and an emergency medical technician lab. IRCC has a contract with Martin Memorial Hospital which allows IRCC students to use Martin Memorial facilities for clinical training. Clinical training is an important part of the allied health curriculum at IRCC. Construction of the satellite hospital next to the IRCC campus would benefit the Allied Health Programs since there could be joint use of equipment, facilities, and personnel, and a better opportunity to invite doctors, nurses, and other health care professionals to the classroom. The satellite hospital would also be more convenient to students, in terms of scheduling and transportation, than Martin Memorial Hospital, where they now receive clinical training. Although the proposed satellite hospital would enhance the IRCC health care training programs, there is no evidence that the clinical training programs now provided at Martin Memorial Hospital are inadequate. The proposed facility is not predicated on a claim that its primary purpose will be to serve as a research or educational facility. Financial Feasibility of the Proposed Satellite Hospital. [Another CON criterion is "[t]he immediate and long-term financial feasibility of the proposal." Section 381.494 (6)(c)9, Florida Statutes.] Estimated Project Costs The estimated costs of the movable equipment for the proposed satellite hospital are reasonable. (Stipulation, P.3) The real property in Port Salerno where the proposed satellite hospital would be located, is owned by Martin Memorial and is of adequate size and otherwise appropriate for the proposed project. (Stipulation, p.3) The estimated cost of construction and fixed equipment is $7,490,625.00, which amounts a cost of $117.50t per square foot. (The hospital will have 850 square feet per bed: $117.50 X 850 X 75 = $7,490,625.00.) his is a reasonable cost for bidding the project in the spring of 1986. Estimates of the architect's fee ($545,317.00), the cost for surveys and borings ($25,000.00j, the 3% contingency cost ($251,000.00), the developmental costs ($195,000.00), the site work and utilities ( $960, 000.03 ), the actual land costs.($595,000.00), and the financing and refinancing costs($3,100,770.00) have also been shown to be reasonable. Short-Term Financial Feasibility Martin Memorial has sufficient funds to make the equity contribution necessary to obtain financing. It also has the ability to raise $16,370,000.00 through the sale of tax exempt bonds, which appear to be marketable. It is likely that Martin Memorial would be able to secure the necessary funds for construction. The proposed satellite hospital would be financially feasible in the short-term. Martin Memorial has proven its ability to operate a hospital efficiently and profitably. Even with the advent of the DRG payment system, Martin Memorial Hospital has continued to operate profitably. During the initial DRG phase-in year of 1984, Martin Memorial benefited financially from the use of the new prospective payment schedule. Even if bed use at the satellite hospital is less than projected, or desired, during the start-up years, it is likely that Martin Memorial would be able to subsidize its operation until, with expected population growth, utilization increases and it becomes financially self- sustaining. Long-Term Financial Feasibility The proposed satellite hospital is also financially feasible in the long-term his conclusion is supported by a financial analysis utilizing reasonable assumptions based on Martin Memorial Hospital's historical experience an t financial costs obtained from a qualified securities analyst. The financial analyst also used bed utilization projections supplied by Martin Memorial's qualified earth care planner. Although the reliability of the 1990 utilization forecast is questionable, over the long-term--with projected increases in population9 is likely that the proposed hospital would become financially feasible, self-sustaining, and able to meet its operating expenses and debt service payments. I. Availability of Manpower and Resources [Another CON criterion is "[t]he availability of resources, including health manpower . . .." Section 381.494 (6)(c)8, Florida Statutes.] Martin Memorial has an in-place recruiting department which, in the past, has successfully recruited new employees for expansion programs. It has the capability of recruiting, training, and staffing the 175 full-time equivalent medical personnel shown in its CON application. There has been no showing that the hiring of employees for the satellite hospital will significantly impact other facilities or that there is a shortage of health manpower and resources. III. Need for the Proposed Hospital (using a planning horizon of January, 1990) Based on a Balanced Consideration of all CON Criteria_ ["Need" for a proposed facility, under CON law, is determined by "a balanced consideration of all the statutory [and rule] criteria." Department of Health and Rehabilitative Services v. Johnson & Johnson, 447 So. 2d 361, 363 (Fla. 1st DCA 1984) See, Section 381.494(6)(c), Florida Statutes. DHRS may not adopt a rule allowing it to "ignore some statutory criteria and emphasize others." Id. Nor may it adopt a methodology, in rule form, which "rigidly control[s] the granting or withholding of [CON] approval." Humana, Inc. et al. v. Department of Health and Rehabilitative Services, So. 2d (Fla. 1st DCA Case No. AY-422, Opinion filed May 16, 1985), 10 F.L.W. 1222.] (a) The foregoing evidentiary findings support an ultimate finding that the proposed satellite hospital is not needed, either now or within the planning horizon of January, 1990. When measured against all pertinent statutory and rule criteria, the factors favoring approval of Martin Memorial's application are outweighed by the factors supporting denial. The DHRS numerical bed-need methodology projects a January, 1990 "need", but barely so. Further, the methodology, as one criterion among many, is assigned less weight since it is a less accurate predictor of "need" in times, such as these, when in-patient bed use is steadily declining. This decline is pervasive, has continued, unabated, for over 13 months and has not yet bottomed-out. The methodology uses bed-need figures rooted in the past and does not adequately reflect this decline. Allocation of bed-need to the Martin/St. Lucie Subdistrict shows a more substancial "need" (103 beds), but this figure is, in part, also derived from the DHRS methodology. There is an adequate current supply of available acute care beds at existing facilities, similar in nature. No geographical access problem has been shown. The existing hospitals which serve the area proposed to be served by the satellite hospital, as well as the subdistrict and district, have 1984 occupancy rates considerably below the,30% occupancy standard generally applied by DHRS and health care planners. There is not a current shortage of beds. In January, 1990, it is likely that the supply of acute care beds will continue to be adequate. In all likelihood, daily occupancy rates at Martin Memorial and in the subdistrict will still be below the 80% standard. At best, Martin Memorial has shown that during two or three peak winter months, its own institution-specific occupancy rate will exceed 90%.10 But on a short-term basis, such a rate is doable and consistent with quality health care. Current bed utilization, a readily ascertainable criterion which reflects actual conditions, should be accorded considerable weight on the scale of criterion when, as now, the health care industry is in rapid flux and past trends have been disrupted, or even displaced. With declining average lengths of stay and anticipated growth in alternative delivery systems, it is reasonably expected that acute care bed use will continue to decline. The steady drop in bed use makes it more likely that the currently existing and licensed or approved beds in the area will be adequate through January 1990. Martin Memorial's utilization forecast failed to fully take into account the steady decline in bed use. Approval of the proposed hospital would be inconsistent with the State Health Plan, which identifies 80% as a minimum occupancy rate for acute care hospitals and, more particularly, with the Local Health Plan (Priority III) which, with few exceptions, does not allow new beds (irrespective of the DHRS methodology numbers) until specified occupancy thresholds have been met. These thresholds have not yet been met. The proposed hospital would be financially feasible in the short-and-long-term, enhance competition, and improve the education of health care students at the adjacent Indian River Community College. These benefits, however, are outweighed by the other factors which support a conclusion that the proposed hospital will not be needed by January, 1990. Construction of an unneeded hospital would have the effect of reducing occupancy rates at nearby hospitals and exacerbating the problem of excess bed capacity.
Recommendation Based on the foregoing, it is RECOMMENDED: That Martin Memorial's application or a CON to construct and operate the Port Salerno Hospital be DENIED. DONE and ORDERED this 4th day of October, 1985, in Tallahassee, Florida. R. L. CALEEN, J . Hearing Officer Division of Administrative Hearings The Oakland Building 2009 Apalachee Parkway Tallahassee, FL 32301 (904)488-9675 Filed with the Clerk of the Division of Administrative Hearings this 4th day of October, 1985.
The Issue Whether Certificate of Need Application Number 9099, filed by Columbia/JFK Medical Center, L.P., d/b/a JFK Medical Center, to convert 20 skilled nursing beds to 20 acute care beds, meets the criteria for approval.
Findings Of Fact Columbia/JFK Medical Center, L.P., d/b/a JFK Medical Center (JFK) is the applicant for Certificate of Need (CON) Number 9099 to convert a 20-bed hospital-based skilled nursing unit (SNU) to 20 general acute care or medical/surgical beds. The construction cost is approximately $117,000, of the total project cost of $151,668. JFK is an affiliate of Columbia Hospital System (Columbia), the largest for-profit hospital chain in the United States. The Agency for Health Care Administration (AHCA) is the state agency which administers the CON program for health care services and facilities in Florida. JFK is a 343-bed hospital located in Atlantis, Florida, in Palm Beach County, AHCA District 9, Subdistrict 5. Pursuant to a previously approved CON, an additional 24 acute care beds are under construction at JFK, along with 12 CON-exempt observation beds, at a cost of approximately $4 million. In August 1998, JFK was allowed to convert 10 substance abuse beds to 10 acute care beds. Other acute care hospitals in District 9 include the Petitioners: St. Mary's Hospital, Inc. (St. Mary's), and Good Samaritan Hospital, Inc. (Good Samaritan), which are located in northern Palm Beach County, AHCA District 9, Subdistrict 4, approximately 11 and 9 miles, respectively, from JFK. The remaining hospitals in District 9, Subdistrict 5, in southern Palm Beach County, and their approximate distances from JFK are as follows: Wellington (8 miles), Bethesda (7 miles), West Boca (18 miles), Delray (12 miles), and Boca Raton Community (17 miles). JFK and Delray are both "cardiac" hospitals offering open heart surgery services, with active emergency rooms, and more elderly patients in their respective service areas. The parties stipulated to the following facts: JFK's CON application was submitted in the Agency for Health Care Administration ("AHCA") second hospital batching cycle in 1998, and was the only acute care bed application submitted from acute care bed District 9, Subdistrict 5. AHCA noticed its decision to approve JFK's CON 9099 by publication in Volume 25, Number 1, Florida Administrative Weekly, dated January 8, 1999. Good Samaritan and St. Mary's each timely filed a Petition for Formal Administrative Proceeding challenging approval of JFK's CON application. By Order dated March 17, 1999, the cases arising from those petitions were consolidated for the purposes of all future proceedings. JFK has the ability to provide quality care and has a record of providing quality of care. §408.035(1)((c), Fla. Stat. JFK's CON application, at Schedule 6 and otherwise, projects all necessary staff positions and adequate numbers of staff, and projects sufficient salary and related compensation. See, §408.035(1)(h). JFK has available the resources, including health personnel, management personnel, and funds for capital and operating expenditures, for project accomplishment and operation. See, §408.035(1)(h), Fla. Stat. JFK's CON application proposal is financially feasible in the immediate term. §408.035(1)(i), Fla. Stat. JFK's CON application proposal is financially feasible in the long term, except, Good Samaritan and St. Mary's contend as it relates to projected utilization. §408.035(1)(i), Fla. Stat. Schedules 9 and 10 and the architectural schematics in JFK's application are complete and satisfy all applicable CON application requirements. Schedule 1 in the application is complete, reasonable, and not at issue. JFK's proposed construction/renovation design, costs, and methods of construction/renovation are reasonable and satisfy all applicable requirements. See, §408.035(1)(m), Fla. Stat. JFK's CON application satisfies all minimum application content requirements in Section 408.037(1), Florida Statutes; except that Good Samaritan and St. Mary's contend that subsection (1)(a), is not satisfied. JFK certified that it will license and operate the facility if its CON proposal is approved. See, §408.037(2), Fla. Stat. JFK's Letter of Intent was timely filed and legally sufficient. See, §408.039(2)(a) and (c), Fla. Stat. Good Samaritan does not provide cardiac catheterization services, angioplasty, or open heart surgery. St. Mary's does not provide elective angioplasty or open heart surgery services. JFK is one of the hospitals to which Good Samaritan and St. Mary's transfer patients in need of inpatient cardiac catheterization services, angioplasty, and open heart surgery. Neither Good Samaritan nor St. Mary's have any present plans to apply for CON approval to add skilled nursing beds or acute care beds. The parties also stipulated that Subsections 408.035(1)(e), (f), (g), (h) - as related to training health professionals, (j), (k), and (2), Florida Statutes, are not at issue or not applicable to this proposal. For the batching cycle in which JFK applied for CON Number 9099, AHCA published a fixed need of zero for District 9, acute care subdistrict 5. In the absence of a numeric need for additional acute care beds in the subdistrict, JFK relied on not normal circumstances to support the need for its proposal, including the following: delays in admitting patients arriving through the emergency room to inpatient beds, delays in moving patients from surgery to recovery to acute care beds, and seasonal variations in occupancy exceeding optimal levels and, at times, exceeding 100%. Good Samaritan and St. Mary's oppose JFK's CON application. In general, these Petitioners claimed that other problems cause overcrowding in the emergency room at JFK, that the type of beds proposed will not be appropriate for the needs of most patients, that "seasonality" is not unique to or as extreme at JFK, and that a hospital-specific occupancy level below that set by rule cannot constitute a special or not normal circumstance. If JFK achieves the projected utilization, experts for Good Samaritan and St. Mary's also projected adverse financial consequences for those hospitals. Rule 59C-1.038(5) - special circumstances During the hearing, the parties stipulated that the numeric need for new acute care beds in the subdistrict is zero. The rule for determining numeric need also includes the following provision: (5) Approval Under Special Circumstances. Regardless of the subdistrict's average annual occupancy rate, need for additional acute care beds at an existing hospital is demonstrated if the hospital's average occupancy rate based on inpatient utilization of all licensed acute care beds is at or exceeds 80 percent. The determination of the average occupancy rate shall be made based on the average 12 months occupancy rate for the reporting period specified in section (4). Proposals for additional beds submitted by facilities qualifying under this subsection shall be reviewed in context with the applicable review criteria in section 408.035, F.S. The applicable time period for the special circumstances provision is calendar year 1997. JFK's reported acute care occupancy was 76.29% in 1997, and 79.7% in 1998, not 80%, as required by the rule. JFK and AHCA take the position that other special circumstances may, nevertheless, be and have been the basis for the approval of additional acute care beds. JFK also maintained that the reported average occupancy levels understated the demand for and actual use of its inpatient beds. Due to seasonal fluctuations caused by the influx of winter residents, JFK reached or exceeded 100% occupancy on 5 or 6 days, exceeded 80% occupancy on 20 days, and averaged 90.9% occupancy, in January 1999. In February 1999, the average was 96.5%, but was over 100% on 8 days, and over 90% on 25 days. In March 1999, the average occupancy was 90.1%, but exceeded 100% on one day, and 90% on 17 days. In recent years, the "season" also has extended into more months, from approximately Thanksgiving to Easter or Passover. It also includes flu season which disproportionately affects the health of the elderly. JFK also demonstrated that occupancy varies based on the day of the week, generally highest on Mondays, Tuesdays, and Wednesdays and lowest on weekends. JFK's acute care beds were also occupied by patients who were not classified as 24-hour medical/surgical inpatients. Others included observation and 23-hour patients, covered by Medicare or health maintenance organizations (HMOs). Some of those patients were classified initially as outpatients to lower reimbursement rates, but routinely subsequently reclassified and admitted as inpatients. In fact, during the applicable time period for determining occupancy, Medicare allowed patients to be classified as outpatients for up to 72-hour hospital stays. Subsequently, Medicare reduced the allowable hospital stay to 48 hours for all "outpatients," according to AHCA's expert witness. When not classified as inpatients, patients are not counted in average occupancy rates which are based solely on the admitted inpatient census, counted each midnight. For example, in February 1999, the average daily census for 23-hour patients was 10.8 patients, which, when combined with 24-hour patients, results in an average occupancy of 99.7% for the month. Due to the Medicare classification system, some but not all of the so- called 23-hour patients affect the accuracy of the inpatient utilization data. According to AHCA's expert witness, however, numeric need cannot be determined because of JFK's failure to quantify the number of Medicare patients who actually affected the acute care bed utilization. The 23-hour or observation patients may use, but do not require CON-approved and licensed acute care beds. Instead, those patients may be held in either non-CON, non-licensed "observation" beds or in licensed acute care beds. As AHCA determined, to the extent that 23-hour patients in reality stayed longer, and adversely affected JFK's ability to accommodate acute care patients, their presence can be considered to determine if special circumstances exist. Combining 24-hour and 23-hour patients, JFK experienced an occupancy rate of 80% in 1996, and 85.7% in 1997. While some of the 23-hour patients were, in fact, outpatients who should not be considered and others stayed from 24 hours up to 3 days and should be considered, JFK's proportion of Medicare services is important to determining whether special circumstances based on acute care utilization exist. With 74% of all JFK patients in the Medicare category, but without having exact numbers, it is more reasonable than not to conclude that the occupancy level is between the range of 76.29% for acute care only and 85.7% for acute care and 23-hour patients. A reasonable inference is that JFK achieved at least 80% occupancy of patients who were in reality inpatients in its acute care beds in 1997. The expert health planner for the Petitioners conceded that bed availability declines, capacity is a constraint, and high occupancy becomes a barrier to service at some level between 80 and 83% occupancy. In a prior CON filed on behalf of Good Samaritan for a 4-bed addition to an 11-bed neonatal intensive care unit (NICU), the same expert asserted that 76% occupancy was a reasonable utilization standard. That occupancy level was based on the desire to maintain 95% bed availability. An exact comparison of the occupancy levels in this and the NICU case, however, is impossible due to the small size of the NICU unit and the fact that the applicant met the occupancy level in that rule for special circumstances. The statistical data on the number patients actually using acute care beds at JFK in excess of 24-hours despite their classification, supports its claim of overcrowding. Emergency Room Conditions JFK described overcrowding in its emergency department as another special circumstance creating a need for additional acute care beds. The emergency room at JFK has 37 bays each with a bed and another 15 to 17 spaces used for stretchers. Eighteen parking spaces are reserved for ambulances in front of the emergency department. It is not uncommon for a patient to wait in the JFK emergency room up to 24 hours after being admitted to the hospital, before being moved to an acute care bed. In February 1999, after having converted 10 substance abuse beds to acute care beds in October 1998, JFK still provided 234 patient days of acute care in the emergency department. The waiting time for patients to receive a bed after being admitted through the emergency department ranged from 10 hours to 5 days in the winter, and from an average of 6 hours up to 24 hours in the summer. While JFK claims that the quality of care is not adversely affected, it does note that patient privacy and comfort are compromised due to the noise, lights, activity, and lack of space for visitors in the emergency room. JFK's patients tend to be older and sicker than the average. As a result, more patients arriving at its emergency room are admitted to the hospital. In the winter of 1998, JFK was holding up to 35 acute care inpatients at a time in the emergency room. Nationally, from 15% to 20% of emergency room patients are admitted to hospitals. By contrast, almost twice that number, or one-third of JFK's emergency room patients become admitted inpatients. Emergency room admissions are also a substantial number of total admissions at JFK. In calendar year 1998, slightly more than 65% of all inpatient admissions to JFK arrived through the emergency room, most by ambulance. Ambulance arrivals at any particular hospital are often dictated by the patient's condition, with unstable patients directed to the nearest hospital. Once patients are stabilized in the emergency room at JFK, those requiring obstetric, pediatric, or psychiatric admissions are transferred from JFK which does not provide those inpatient services. Emergency room patients in need of acute care services provided at JFK, like the neonates at issue in the prior Good Samaritan application, are unlikely candidates for transfer The emergency room at JFK receives up to 50,000 patient visits a year, up from approximately 32,000 annual visits five years ago. JFK operates one of the largest and busiest emergency departments in Palm Beach County. Due to overcrowding in the emergency department at Delray Hospital, in southern Palm Beach County, patients have been diverted to other facilities, including JFK. In terms of square footage, JFK's emergency room does not meet the standards to accommodate the 52 to 54 bays and stretchers and related activities. JFK lacks adequate space for support services which should also be available in the emergency department. The Petitioners asserted that enlarging the emergency room will alleviate its problems. JFK demonstrated, however, that regardless of the physical size of the emergency room, optimal patient care requires more capacity to transfer patients faster to acute care beds outside the emergency department. Conditions in Other Departments Of 343 operational beds at JFK at the time of the final hearing, 290 were monitored or telemetry acute care beds, 30 were critical care beds, and 23 were non-monitored, non-critical care beds. Most of the monitored beds are in rooms equipped with antennae to transmit data from electrodes and monitors when attached to patients. When monitoring is not necessary, the same beds are used by regular acute care patients. The large number of monitored beds located throughout the hospital in various units reflects JFK's largely elderly population and specialization in cardiology. In 1998, 820 inpatient cardiac catheterizations (caths) were performed at JFK. Petitioners Good Samaritan and St. Mary's transferred 90 and 28 of those cath patients, respectively to JFK. In the first five months of 1999, 449 caths were performed, including procedures on 35 patients transferred from Good Samaritan and 16 from St. Mary's. Cath lab patients are held in the lab longer after their procedures when beds are not available in cardiac or the post- anesthesia care units. The Petitioners suggested that cath lab patients could be placed in a 12-bed holding area added to the lab in July 1999; however, that space was expected to be filled by patients being prepared for caths. Open heart surgery is available in Palm Beach County at three hospitals, Delray, JFK and Palm Beach Gardens. Patients admitted to JFK for other primary diagnoses often require cardiac monitoring even though they are not in a cardiac unit. The additional 24 beds which were under construction at the time of the final hearing will also be monitored beds. The 20 beds at issue in this proceeding will not be monitored. The Petitioners questioned whether non-monitored beds will alleviate overcrowding at JFK where so many patients require monitoring. JFK physicians in various specialties testified concerning conditions in other areas of the hospital. A nephrologist, who consults primarily in intensive care units, described the backlog and delay in moving patients from intensive care into acute care beds. A cardiologist noted that patients are taking telemetry beds they do not need because there is no other place to put them. A general and vascular surgeon described the overcrowding as a problem with the ability to move patients from more to less intensive care when appropriate. Elective surgeries have been delayed to be sure that patients will have beds following surgery. The evidence presented by JFK supports the conclusion that the additional acute care beds will assist in alleviating overcrowding in other hospital units, including backlogs in the existing monitored beds. JFK has established as factual bases for special circumstances that its high occupancy exceeds the optimal much of the year, aggravated by seasonal fluctuations; that it has relatively large emergency room admissions over which it has no control; and that its intensive care and monitored beds are not available when needed. Number of Beds Needed With the conversion, in 1998, of 10 substance abuse beds to acute care beds and the 1999 construction of 24 of 40 additional beds requested by JFK, the number of licensed and approved beds at JFK increased to 367. In addition, with CON- exemption, JFK has added observation beds. As a result of AHCA's partial approval of the previous JFK request for new construction and due to unfavorable changes in Medicare reimbursement policies for hospital-based SNUs, JFK now seeks this 20-bed conversion. JFK ceased operating the SNU in October 1998, after Medicare reimbursement changed to a system based on resource utilization groups (RUGs). JFK was unable to operate the SNU without financial losses, that is, unable to cover its patient care costs under the RUGs system. The proposal to convert the beds back to acute care, as they were previously licensed will allow JFK to reconnect existing oxygen lines in the walls and to use the beds for acute care patients. Although Good Samaritan and St. Mary's suggested that JFK can profitably operate a SNU, there was no evidence presented other than its previous occupancy levels which were very high, and the fact that Columbia is not closing all of its SNUs. The Petitioners also question JFK's ability to use its SNU beds for acute care and/or observation patients. AHCA, however, took the position that acute care licensure is required for beds in which acute care patients are routinely treated. Otherwise, the agency would not have accurate data on utilization, bed inventory, and the projected need. In order to demonstrate the number of beds needed, JFK's expert used historical increases in admissions. Some admissions data was skewed because the parent corporation, Columbia, closed Palm Beach Regional in 1996, and consolidated its activities at JFK. Excluding from consideration the increase of 3,707 admissions from 1995 to 1996, JFK's expert considered approximately 800 as reasonable to assume as an average annual increase. That represents roughly the mid-point between the 1996 to 1997 increase of 605, and the 1997 to 1998 increase of 1,076 admissions. A projected increase of 800 admissions for an average 5-day length of stay would result in an increase of 4,000 patient days a year which, at 80% occupancy, justifies an increase of 14 beds a year. Considering the closing of Palm Beach Regional, the number of beds in the subdistrict will have been reduced by 170. At the hearing, JFK's expert also relied on 3.3% annual patient day increase to project the number of beds needed, having experienced an increase of 5.8% from 1997 to 1998. Using this methodology, JFK projected a need for 20 additional acute care beds by 2002, and over 40 more by 2004. That methodology assumed patient growth in the excess of population growth and, necessarily, an increase in market share. JFK's market share increased in its primary service area from approximately 19% in 1993 to 27% in 1997. But the market share also slightly declined from 1997 to 1998. AHCA's methodology for determining the number of beds needed was based on the entire population of Palm Beach County, not just the more elderly southern area. It also assumed that JFK's market share would remain constant. Using this more conservative approach than JFK, AHCA projected a need for 383 acute care beds, or 16 beds added to the current total of 367 licensed and approved beds, at an optimal 75% occupancy by the year 2004. AHCA relied on a projection of 104,959 total patient days in 2004. Using the same methodology, JFK's expert determined that total projected patient days for 1998 would have been 94,225, but the actual total was 98,126 patient days. AHCA's methodology underestimates the number of beds needed, but does confirm that more than 16 additional beds will be needed by 2004. AHCA's reliance on 75% as an optimal future occupancy level as compared to the hospital-specific historical level of 80% was criticized, as was the use of the year 2004 as a planning horizon. The rule requires 80% occupancy for a prior reporting period and does not establish any planning horizon. Good Samaritan and St. Mary's used 80% occupancy in their analysis of bed need. At 80% occupancy, Petitioners projected an average daily census of 265 patients in 331 beds in 2001, or 268 patients in 334 beds in 2002, and 270 patients in 358 beds in 2003, as compared to 367 existing and approved beds. The Petitioners' projection is an underestimate of bed-need based on the actual average daily census of 269 patients in 1998. The Petitioners' methodology erroneously projects a need for fewer licensed beds than JFK has currently, despite the special circumstances evincing overcrowding. At 80% occupancy, based on the special circumstances rule, a hospital exceeds the optimal level and needs more beds. But, according to the Petitioners, 80% is a future occupancy target for the appropriate planning horizon of 2002. As AHCA's expert noted, it is illogical to use 80% as both optimal and as an indication of the need for additional beds. Similarly, it is not reasonable to use a planning horizon which coincides with the time when more beds will be needed. Therefore, the use of 75% for the five-year planning horizon of 2004 is a reasonable optimal target, as contrasted to the need for additional beds when 80% occupancy is reached at some future time beyond the planning horizon. AHCA's underestimate of need at 16 more beds by 2004, and JFK's overestimate of need at 40 more beds by 2004, support the conclusion that the requested addition of 20 beds in this application is in a reasonably conservative range. Rule 59C-1.038(6)(a) and Subsection 408.035(l)(n) - service and commitment to medically indigent; and Rule 59C-1.038(6)(b) - conversion of beds Rule 59C-1.038(6), Florida Administrative Code, also includes the following criteria: Priority consideration for initiation of new acute care services of capital expenditures shall be given to applicants with documented history of providing services to medically indigent patients or a commitment to do so. When there are competing applications within a subdistrict, priority consideration shall be given to the applications which meet the need for additional acute care beds in a particular service through the conversion of existing underutilized beds. Subsection (a) of the Rule, overlaps with District 9 health plan allocation factor one, which must be considered pursuant to Subsection 408.035(1)(a), and with the explicit criterion of Subsection 408.035(1)(n), Florida Statutes. All three require a commitment to and record of service to Medicaid, indigent and/or handicapped patients. JFK agreed to have its CON conditioned on 5% of the care given in the 20 new beds to Medicaid and charity patients. The commitment for the 24 beds under construction is 3% for Medicaid and charity patients. If charity patients are defined as those with family incomes equal to or below 150% of federal poverty guidelines, JFK provided $2.9 million in charity care in calendar year 1998, and $720,000 as of April for 1999. JFK provided an additional 3% to 5% in Medicaid care. The Medicaid total includes Palm Beach County Health Care District patients, who are also called welfare patients. The charity care provided by JFK is equivalent to approximately 1% of its gross revenue. JFK explained its relatively low Medicaid care as a function of its relatively limited services for people covered by Medicaid, particularly, the young who utilize obstetrics and pediatrics. JFK pointed to the differing demographics in Palm Beach County with more elderly, who have Medicare coverage, located in its primary service area. Excluding pediatric and obstetric care, Medicaid covered 6.7% of patients in southern Palm Beach County as compared to 16.3% in northern Palm Beach County. Of the Medicaid patients, 2.9% in the southern area as compared to 6% in the northern area are adults. On this basis, JFK established the adequacy of its historical Medicaid and indigent care, and of its proposed commitment. Subsection (6)(b) of Rule 59C-1.038 is inapplicable when, as in this case, there are not competing applications to compare. Subsection 408.035(1)(a) - other local health plan factors and Subsection 408.035(1)(o) - continuum of care District 9 allocation factor 2, favoring cost containment practices, is enhanced by the proposed conversion rather than the new construction of beds. Within the Columbia group of hospitals, there is an effort to avoid unnecessary duplications of services. JFK caters to an elderly population and to providing cardiology, neurology, and oncology services. Columbia's Palms West provides pediatric and obstetric care. Another Columbia facility in Palm Beach County, Columbia Hospital, specializes in inpatient psychiatric services. The elimination of the hospital-based SNU at JFK does eliminate one level of care in the system, contrary to the criteria. District 9 health plan allocation factor 3 requires favorable consideration of plans, like JFK's, to convert unused or underutilized beds. In this case, the JFK SNU was highly utilized but unprofitable. There is no evidence that alternative placements in free-standing nursing homes are inappropriate or unavailable. Minor inefficiencies result from the time lag for transfers during which skilled nursing patients remain in acute care beds. To some extent, the inefficiencies were already occurring while JFK operated the SNU due to its high average census of 18 or 19 patients in a total of 20 SNU beds. Those inefficiencies are outweighed by the low cost conversion of 20 beds for $117,000, particularly as compared to its prior 24-bed construction for $4 million. In general, the applicable local health plan allocation factors support the approval of the JFK application. Rule 59C-1.030 - needs access for low income, minorities, handicapped, elderly, Medicaid, Medicare, indigent or other medically underserved In general, the proposal is intended to increase access to JFK's services by decreasing waiting times for admissions. The services are used by a large number of elderly patients, who are primarily covered by Medicare. JFK demonstrated that the population in its service area also tends to be wealthier than the population in northern Palm Beach County. Medicaid and indigent access to care at JFK is consistent and reasonable given the demographic data presented. Access for elderly Medicare patients will be enhanced by the proposal. Subsection 408.035(1)(b) - accessibility, availability, appropriateness, and adequacy of like and existing services Good Samaritan and St. Mary's argue that hospitals below 75% occupancy are available alternatives to JFK's patients. Yet, those facilities are not viable alternatives for unstable patients admitted through the emergency room. Neither is it appropriate to transfer patients who need services provided at JFK. JFK does not allege that any problems exist at other facilities, but only that it is affected by special circumstances. From January to June 1998, the closest hospitals to JFK experienced wide-ranging occupancy levels from 92% at Delray, the hospital with services most comparable to those at JFK, to 57% at Bethesda, and 47% at Wellington. The wide range in occupancy rate is further indication of uniqueness of the need for patients to access services available only at Delray and JFK. Subsection 408.035(1)(d) - outpatient care or other alternatives Admitted inpatients have no alternatives to their need for acute care beds. Subsection 408.035(1)(h) - alternative use of resources and accessibility for residents The continued use of the 20 beds as a SNU was suggested as an alternative. As noted, however, that proved to be financially unprofitable at JFK, in comparison to the low cost conversion to acute care beds. AHCA reasonably rejected the idea that of the beds being designated "observation" beds when used for acute care patients. In addition, in 1996, JFK estimated the cost of moving patients from bed to bed in the hospital due to the shortage of appropriate beds, when needed, at up to $1 million. This project is intended to meet a facility-specific need based on the demand for services at JFK from patients who cannot reasonably initially be sent or subsequently transferred to other hospitals. As such, JFK's additional beds do not meet the criterion for accessibility for all residents of the district. Subsection 408.035(1)(i) - utilization and long-term financial feasibility Good Samaritan and St. Mary's contend that JFK's proposal includes unrealistically high utilization projections for the additional 20 beds. Using 98,000 patient days in 1998, which excludes any days attributable to skilled nursing beds, total utilization projected in the second year is 78.4%. For the additional 20 beds, projected utilization is 77.4%. The expert for Good Samaritan and St. Mary's disagreed with the allocation of patient days between the existing and additional beds. If 80% utilization is assigned to existing 367 beds, as he suggested, then the average annual occupancy of the 20 new beds would be only 50%. The financial break-even point for the project, however, is 50 to 75 patient days, or 10 to 15 patients with average lengths of stay of 5 days. Therefore, even with the lower projected occupancy of 50%, or an average of 10 beds at any time, the project is financially feasible in the long-term. In reality, a separate allocation of patient days to the 20 new beds is somewhat arbitrary. It is also less important than total projected utilization, since the 20 beds do not represent a separate unit in which specialized services will be provided. The additional beds will become a part of the total medical/surgical inventory. By demonstrating that there will be sufficient total occupancy to exceed the financial break-even point in the newly converted beds regardless of the allocation of patient days to any particular bed, JFK demonstrated the long- term financial feasibility of the proposal for CON 9099. Subsection 408.035(1)(l) - impact on costs; effects of competition If the JFK proposal is approved, Good Samaritan anticipates a loss of 255 patients, or 1,392 patient days, which is equivalent to a financial loss of over $1.5 million. St. Mary's anticipates losses of 158 patients or 973 patient days, and in excess of $1 million. Both hospitals were experiencing overall operating losses in 1999. But, the estimates of financial losses for both hospitals did not take into consideration all of the expense reductions associated with serving fewer patients. Excluding pediatrics and obstetrics, which are not available at JFK, JFK's overlapping service areas with Good Samaritan and St. Mary's are minimal. Good Samaritan's market share in JFK's primary service area is 4.8%, and St. Mary's is 9.3%. Pediatrics and obstetrics contribute 30.7% of total patients at Good Samaritan, and 49.5% at St. Mary's. Physician overlap among the hospitals is also limited. Although 357 doctors admitted patients to JFK and 464 to St. Mary's in the first two quarters of 1998, the number of overlapping doctors was 28. With a total of 379 admitting doctors at Good Samaritan for the same period of time, only 21 were included in JFK's 357 admitting physicians. In general, doctors in the northern Palm Beach County acute care subdistrict seldom admit patients to hospitals in the southern subdistrict, and vice versa. The absence of overlapping medical staff also reflected the differences in the services. Most of the top twenty doctors who admitted patients to Good Samaritan and St. Mary's were obstetricians and pediatricians. When obstetricians and pediatricians are excluded, the number of overlapping doctors for JFK and Good Samaritan is reduced to 15, and for JFK and St. Mary's to 22. In addition to providing different services, to different areas of the County, doctors who practice primarily in one or the other subdistrict served patients in different payor classification mixes. In 1997, JFK's patients were 74% Medicare, consistent with the fact that a larger percentage of elderly patients live in JFK's service area. By contrast, Medicare patients were approximately 48% of the total at Good Samaritan, and 32% of the total at St. Mary's. Historically, the addition of acute care beds at JFK has not affected other hospitals in the district or even the same acute care subdistrict. After the conversion of 10 substance abuse beds in the fall of 1998, the acute care patient days at every hospital in the same subdistrict increased in early 1999 over comparable periods of time in 1998. The assumption that additional beds at JFK will take patients from other hospitals includes the assumption that JFK will draw a larger share of an incremental increase of patients. The assumption is, in other words, that all patients will be new to JFK. The expert health planner for Good Samaritan and St. Mary's conceded that facility-specific overcrowding can justify projections that the additional beds will accommodate the existing census plus growth attributable to increasing population, and will not generate new patients. The expert assumed, nevertheless that from 1478 to 1486 new patients (depending on whether the length of stay is rounded off) would be associated with JFK's project. From that total, the proportional losses allocated were 255 patients from Good Samaritan and 158 patients from St. Mary's. Another underlying assumption increase is that all of the new patients would go to other hospitals if JFK does not add 20 acute care beds. That assumption suggests that all of the patients could receive the services they need at the other facilities, which is not supported by the facts or current utilization data. More likely, with the addition of beds due to overcrowding, some patients will come from the existing hospital census at JFK. It is not reasonable to assume that JFK will have all new patients, nor that all patients could be treated at other hospitals in the absence of JFK's expansion. The proportion of emergency room admissions at JFK is reasonably expected to continue. Patients who arrive at JFK requiring open heart surgery, angioplasties or invasive cardiac caths are reasonably expected to continue to receive those services at JFK, including patients who are transferred to JFK from Good Samaritan and St. Mary's. Based on the failure to support the assumptions, and the differences in service areas, medical staff, specialties, and patient demographics, Good Samaritan and St. Mary's have not shown any adverse impact from the JFK proposal. On balance, considering the statutory and rule criteria for reviewing CON applications, JFK established, as a matter of fact, that it meets the special circumstance criteria related to emergency room admissions, pre- and post-surgical and intensive care backlogs, and average annual occupancy projections in excess of optimal levels.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED That a final order be entered issuing CON 9099 to convert 20 skilled nursing beds to 20 acute care beds at Columbia/JFK Medical Center, L.P., d/b/a JFK Medical Center, on condition that a minimum of 5% of new acute care patient days will be provided to Medicaid and charity patients. The file of the Division of Administrative Hearings, DOAH Case No. 99-0714 is hereby closed. DONE AND ENTERED this 7th day of April, 2000, in Tallahassee, Leon County, Florida. 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 7th day of April, 2000. COPIES FURNISHED: Sam Power, Agency Clerk Agency for Health Care Administration Fort Knox Building 3, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308-5403 Julie Gallagher, General Counsel Agency for Health Care Administration Fort Knox Building 3, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308-5403 Richard A. Patterson, Esquire Agency for Health Care Administration Fort Knox Building 3, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308-5403 Thomas A. Sheehan, III, Esquire Moyle, Flanigan, Katz, Kolins, Raymond & Sheehan, P.A. Post Office Box 3888 West Palm Beach, Florida 33402 Stephen A. Ecenia, Esquire Thomas W. Konrad, Esquire Rutledge, Ecenia, Purnell & Hoffman, P.A. Post Office Box 551 Tallahassee, Florida 32302-0551 Robert D. Newell, Jr., Esquire Newell & Terry, P.A. 817 North Gadsden Street Tallahassee, Florida 32303-6313
Findings Of Fact I The Parties Charter Medical of Orange County, Inc., (Charter) is a wholly-owned subsidiary of Charter Medical Corporation, founded in Macon, Georgia in 1969. The parent corporation operates approximately 92 hospitals throughout the country, including Florida. Most of its hospitals are psychiatric or substance abuse facilities. Orlando Regional Medical Center (ORMC) is a 1,119- bed, nonprofit medical system comprised of four divisions. In downtown Orlando it operates a 630-bed tertiary care hospital and a 255-bed Arnold Palmer Hospital for women and children. A Sand Lake campus is located 10 miles southwest of Orlando, off I-4, and includes medical/surgical beds and 32 licensed short-term psychiatric beds. ORMC's St. Cloud Campus in Osceola County, south of Orlando, includes 84 medical/surgical beds. The Department of Health and Rehabilitative Services (HRS) is the state agency responsible for implementing and enforcing the certificate of need program pursuant to Sections 381.701-.715, F.S. Psychiatric Institute of Orlando, Inc., d/b/a Laurel Oaks Hospital, (Laurel Oaks) is a subsidiary of P1A Psychiatric Hospitals, Inc., which is a subsidiary of National Medical Enterprises (NME). P1A owns approximately 50 psychiatric hospitals throughout the county, including (30-bed Laurel Oaks, in southwest Orange County, a short-term psychiatric and substance abuse facility for children and adolescents Health Management Associates, Inc., (HMA) is a health management company which owns or operates 16 hospitals in the southeastern United States, including four psychiatric hospitals in Florida. HMA's Crossroads University Behavioral Center is a 100-bed free-standing psychiatric hospital in northeast Orange County. Its 60 adult beds and 40 adolescent beds opened in January 1989 as a licensed long-term facility, but it has been operating continually as a short-term facility. The Applications Charter proposes to develop a new free-standing 60- bed psychiatric hospital (40 beds for adults, 10 beds for adolescents and 10 beds for children). It plans a wide range of treatment modalities utilizing a multi-disciplinary team approach, tailored to the age and needs of the patient. Although no specific site has been selected, several have been identified in southwest Orange County. Charter anticipates the total cost for the project will be $7,783,000. Charter's patients will be primarily commercially insured (71%), with 15% Medicare and 4% indigent. Charter has committed to serve this share of indigent for the first two years of operation. As a specialty hospital, Charter is not eligible to accept Medicaid patients. ORMC proposes to build a 60-bed free-standing facility on a 7.2 acre site within 40 acres it already owns at Sand Lake and adjacent to its existing Sand Lake Hospital, for a total project cost of $6,678,935. No new licensed beds are required as ORNC will transfer its 32 short term beds from the sixth floor of the Sand Lake Hospital and will convert 28 of its licensed medical/surgical beds from its downtown hospital. The 60 beds will consist of 30 adult and 30 adolescent short term beds. Since the existing 32 beds are primarily adult beds, ORMC's project will be adding adolescent beds to the inventory in District 7. Proximity to Sand Lake Hospital will facilitate shared services, including engineering, dietary and laundry. ORMC also expects the joint use of therapists at its psychiatric facility and its existing brain injury rehabilitation unit at Sand Lake. Because the facility will be added to ORMC's general hospital license, it can and will accept Medicaid patients. ORMC has committed to serve 20% Medicare, 8% Medicaid and 8% indigent patients. ORMC will likely seek an outside management firm to operate its psychiatric facility. The Review On August 23, 1988, HRS published a need for 140 short-term psychiatric beds in District 7. Its SAAR issued in January 1989, recommended approval of a total of 137 beds. When the SAAR was amended in March 1989, to include the Charter approval, the total surged to 197 approved beds. Even after First Hospital withdrew its application for 55 beds, the total approved exceeded the published need for the 1993 horizon year by two beds. Numeric Need The short-term psychiatric bed need rule is found at Rule 10- 5.011(1)(o) , F.A.C. "Short-term" is defined as an average length of stay of 30 days or less for adults, and 60 days or less for children and adolescents under 18 years. A favorable need determination will not normally be given to an applicant unless a bed need exists according to sub-paragraph (1)(o)4 of "the rule". Rule 10-5.011(1)(0)4, F.A.C. provides as follows: Bed allocations for acute care short term general psychiatric services shall be based on the following standards: A minimum of .15 beds per 1,000 population should be located in hospitals holding a general license to ensure access to needed services for persons with multiple health These beds shall be designated as short term inpatient hospital psychiatric beds. 20 short term inpatient hospital beds per 1,000 population may be located in specialty hospitals, or hospitals holding a general license. The distribution of these beds shall be based on local need, cost effectiveness, and quality of care considerations. The short term inpatient psychiatric bed need for a Department service district shall be projected 5 years into the future based on the most recent available January or July population estimate prior to the beginning to the respective batching cycle. The projected number of beds shall be based on a bed need ratio of .35 beds per 1,000 population. These beds are allocated in addition to the total number of general acute care hospital beds allocated to each Department District under Paragraph 10-5.011(1)(m). The net need for short term psychiatric beds shall be calculated by subtracting the number of licensed and approved beds from the number of projected beds. The population estimates are based on population projections by the Executive Office of the Governor. Occupancy Standards. New Facilities must be able to project an average 70% occupancy rate for adult psychiatric beds and 60% for children and adolescent beds in the second year of operation, and must be able to project an average 80% occupancy rate for adult beds and 70% for children and adolescent short term psychiatric inpatient hospital beds for the third year of operation. No additional short term inpatient hospital adult psychiatric beds shall normally be approved unless the average annual occupancy rate for all existing adult short term inpatient psychiatric beds in a service district is at or exceeds 75% for the preceding 12 month period. No additional beds for adolescents and children under 18 years of age shall normally be approved unless the average annual occupancy rate for all existing adolescent and children short term hospital inpatient psychiatric beds in the Department district is at or exceeds 70% for the preceding 12 month period. Hospitals seeking additional short term inpatient psychiatric beds must show evidence that the occupancy standard defined in paragraph six is met and that the number of designated short term psychiatric beds have had an occupancy rate of 75% or greater for the preceding year. Unit size. In order to assure specialized staff and services at a reasonable cost, short term inpatient psychiatric hospital based services should have at least 15 designated beds. Applicants proposing to build a new but separate psychiatric acute care facility and intending to apply for a specialty hospital license should have a minimum of 50 beds. The parties do not dispute that application of the formula yields a need for 140 beds, the total published in the applicable fixed need pool. Nor do the parties dispute that the occupancy standard was met, since HRS uses the lower standard of 70% as a threshold for determining whether need should be published. The parties agree that approval of both Charter's and ORMC's applications results in an excess of two beds over the published need. There is substantial dispute as to whether that excess is justified, and as to the composition of the beds as "speciality hospital" or "general hospital" beds. The only provision in agency rules or policy for exceeding bed need calculations is when "not normal" or "special" circumstances exist in the District. HRS' Policy Manual for the Certificate of Need program, dated October 1, 1988, provides in Section 9-6 B. (3): If a qualified applicant exist but the proposed project exceeds the beds or services identified in the fixed need pool, the department may award beds or services in excess of the pool when warranted by special circumstances as defined in rule 10- 5.011(1)(b), 1-4, F.A.C. and, specifically for nursing homes Rule 10-5.011(1)(K)2.j. F.A.C. (Laurel Oaks Exhibit #10, P. 9-2) The referenced sections of Rule 10-5.011(1)(b), F.A.C., relate to the enhancement of access--primarily economic access and access by underserved groups. Access is addressed in Part VII, below. No evidence was presented regarding special problems of access in District 7. Rather, HRS asserts that its excess approval was based on "rounding up" the numbers of beds, and on the favorable occupancy rates in the district. In its SAAR, HRS calculated the following occupancy rates by age cohort in the district: Adult 75.8% Child/Adolescent 74.8%; and in Orange County: Adult 57.4% Child/Adolescent 100. The adult rate is therefore slightly above the 75% minimum in the district, and substantially below the minimum in Orange County. The child/adolescent rate is above the 70% minimum in both the district and county. HRS appropriately does not utilize occupancy in beds other than licensed short term psychiatric bed in calculating its rates as it would be difficult to compute the number of available beds (medical/surgical, long term psychiatric, etc.). The rule specifies that a minimum of .15 beds per 1000 population "should" be allocated to hospital1s holding a general license and that .20 beds per 1000 population may be located in either speciality hospitals or hospitals holding a general license. Of the 140 beds needed in District 7, 75 may be located in a speciality hospital under this formula. 30 speciality beds were awarded to West Lake and are unchallenged. The Charter application for 60 speciality beds exceeds by 15, the 45 speciality beds left to be allocated. The State and Local Health Plans The State Health Plan is dated 1985-1987. Goal 1 is the only portion of the plan that is relevant in this review. It essentially reiterates the need methodology described above, regarding the .35 beds per 1000 population and the 70% and 75% annual occupancy thresholds. The applicable local health plan is the 1988 local health plan for District 7. This plan divides the district into "planning areas": Brevard, Osceola, Seminole and Orange -- the four counties within the district. Planning areas, unlike subdistricts, are more in the nature of guidelines and do not carry the same legal weight as subdistricts. Both applicants are committed to submit data to the local health councils, as provided in recommendation #2. Both applicants have committed to provide a fair share of care to the underserved, although ORMC's commitment is substantially greater and has a proven record to support it. Recommendation #5 provides that no new short-term psychiatric or substance abuse beds shall be approved until all existing beds in the planning area are operating at or above 75% occupancy for the most recent twelve months for which data is available from the local health council. This criteria is barely met when adult and children/adolescent occupancy is combined, and is not met by the occupancy rate for adult beds in Orange County. Financial Feasibility The pro formas of both applicants, which are no more than best guess estimates, are generally reasonable, based upon the experience of the applicants' existing programs. Charter's proposal makes no provision for management fees, although such fees are remitted to the parent company by its subsidiaries and are reported to the Health Care Cost Containment Board. Charter anticipates that it would not incur additional corporate overhead to support this facility if it is built. In recent years ORMC's psychiatric unit has lost money in its operation when overhead is factored into the cost. Its Program Director, Jeffrey Oppenheim, reasonably anticipates the new facility will make a profit, as it will serve a better mix of age cohorts and will offer a more desirable setting than its limited facility now located on the sixth floor of a medical/surgical hospital. The financial feasibility of both applications depends on the programs' ability to attract patients. That ability is not seriously questioned. Both applications have substantial experience in operating financially efficient health care programs. Quality of Care and Accessibility No evidence was presented to challenge either applicant's ability to provide quality care. Nor, however, was the quality of care of existing alternative programs at issue. Geographic access in District 7 is not a problem, and none suggests that the access standard in Rule 10-5.011(1)(o)5.g., F.A.C., is not met (travel time of 45 minutes or less for 90% of the service area population). Charter's inability to provide Medicaid services and its time-limited commitment to serve even 4% indigents amount to only minimal contribution to the economically underserved population. In the past, ORMC has been a receiving facility for Baker Act patients and it anticipates it will again when the psychiatric program has its new quarters. It is only one of two hospitals in Orange County eligible to provide Medicaid services and is the fifth highest provider of charity and Medicaid in the State of Florida, according to Medical Health Care Cost Containment Board data. Impact on Existing Facilities and Competition Positive competition among providers already exists in District 7. There are eleven existing short term psychiatric programs in the four-county area, including both speciality and general hospitals, and adult, children and adolescent programs. Only three obtained an occupancy rate of more than 75% for the fiscal period ending June 1988. The Availability of Health Manpower There is a shortage of nurses, qualified social workers and counsellors in District 7. HMA has experienced problems in recruiting staff at its Orlando facility. Competition for these staff has caused salaries to rise, and consequently the cost of providing services has risen. Turnover results when staff are attracted to new facilities, causing training problems and affecting quality of care. Charter has the corporate resources to conduct effective recruiting, but has no experience recruiting in the Orlando area. ORMC, a large diverse facility, with good opportunity for lateral and upward mobility, has experienced few problems staffing its programs. The Availability of Alternatives Eight of eleven District 7 short term psychiatric facilities have operated below 75% occupancy in the last two years. These under-utilized facilities are plainly alternatives for new projects proposing the same services. Neither applicant is proposing novel or innovative services in psychiatric care. That licensed long term psychiatric facilities such as HMA, are operating short term programs does not justify the approval of new short term beds, but rather suggests these programs could be converted, with little or no capital outlay, into short term programs. Conversion of under-utilized acute are beds to short term psychiatric beds is also an alternative in District 7. Acute care bed occupancy rates in each county of District 7 failed to reach 60% in the most recent 12-month period of available data. The criterion of Rule 10-5.011(1)(o)5.f., F.A.C. favors the conversion of under-utilized beds in other hospital services unless conversion costs are prohibitive. There has been a trend in the last several years away from inpatient care and toward less restrictive treatment modalities. Both applicants acknowledge this trend with their inclusion of partial hospitalization programs in their plans. ORMC has no reasonable alternative to building a new facility if it is to maintain its inpatient psychiatric program. There is an increasing demand for the medical surgical beds it currently occupies on the sixth floor of Sand Lake Hospital. There is no appropriate space in its downtown facilities. Balancing the Criteria Comparative Review and Summary As reflected above, not all of the relevant statutory and rule criteria have been met by these applicants. There remains, however, the planning horizon numerical need for additional short term psychiatric beds. While that need could likely be met with the utilization of beds that are not licensed for the provision of short term care, such a solution frustrates state licensing requirements. Three alternative dispositions exist: to deny both ORMC and Charter applications, leaving an unmet need in this cycle for 86 beds; to grant one application only; or to approve both and exceed the need by two beds. HRS argues that the two-bed difference is of little consequence and that the excessive number of specialty beds if Charter is approved is irrelevant, as no general hospital is currently competing for the beds. It is not possible to conjecture that appropriate general hospital applicants will participate in a near future cycle, but it is certain that if those beds are awarded in this cycle to a specialty hospital, they will not be available in a future cycle. Nothing requires that all beds identified in a fixed pool must be awarded in that cycle. The converse follows when, as here, other considerations weigh against approval of additional beds. Between the two applicants, ORMC more consistently meets the rule and statutory criteria. Although it still proposes a substantial capital outlay, (ORMC) relies on conversion of existing licensed beds and results in less impact on other existing programs. Its contribution to the underserved population is more substantial; it proposes more needed adolescent, rather than adult beds; and it does not violate the .15/.20 general hospital, specialty hospital bed balance. That balance needs to be maintained in this case to insure competition among Medicaid providers. In summary, the evidence supports approval of ORMC's application and denial of Charter's.
Recommendation Based on the foregoing, it is hereby, RECOMMENDED That a Final Order be entered denying CON number 5691 to Charter Medical of Orange County, Inc.; and granting CON #5697 to Orlando Regional Medical Center. DONE AND RECOMMENDED this 15th day of February, 1990, in Tallahassee, Leon County, Florida. MARY CLARK 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 February, 1990. APPENDIX TO THE RECOMMENDED ORDER IN CASES NO. 89-1358,89-1366.89-1368,89-2039 & 89-2041 The following constitute rulings on the findings of fact proposed by each party: Charter Medical of Orange County, Inc. Adopted in substance in paragraph 1. Adopted in substance in paragraph 2. Adopted in paragraph 3. Adopted in paragraph 4. Adopted in paragraph 5. 6.-17. Adopted in statement of the issues. 18. Adopted in paragraphs 6 and 7. 19.-36. Rejected as unnecessary, except as summarized in paragraph 6. 37. Addressed in paragraph 30. 38.-5O. Rejected as unnecessary, except as summarized in paragraph 36. 51.-57. Rejected as unnecessary, except at summarized in paragraph 7. 58.-6I. Adopted in paragraph 16. 62.-69. Rejected as unnecessary and contrary to the methodology in the rule. 70. Adopted in substance in paragraph 16. 71.-77. Rejected as unnecessary. 78. Rejected as contrary to the evidence and law. "Not normal" does not include high occupancy rates in several facilities. 79.-8I. Adopted in summary in paragraph 21. 82.-83. Rejected as immaterial. The evidence in this case supports maintenance of the balance, notwithstanding past practice. 84. Adopted in paragraph 15. 85.-103. Rejected as unnecessary. 104. Rejected as contrary to the law and evidence. 105. & 106. Adopted in paragraph 22. 107.-109. Adopted in paragraph 23. 110. & 111. Rejected as unnecessary. 112. & 113. Adopted in paragraph 24. Rejected as unnecessary. Adopted in substance in paragraph 25. 116.-149. Rejected as unnecessary. 150. Adopted in substance in paragraphs 26 and 29. 151.-161. Rejected as unnecessary. 162.-164. Adopted in substance in paragraph 27. 165.-171. Rejected as unnecessary. 172. Rejected as contrary to the evidence. 173.-180. Rejected as immaterial and unnecessary. 181. Adopted in paragraph 5. 182.-190. Rejected as unnecessary. Adopted in substance in paragraph 43. Rejected as contrary to the evidence. 193.-198. Rejected as unnecessary. 199. Rejected as contrary to the evidence. 200.-206. Rejected as unnecessary. Rejected as contrary to the evidence. Rejected as unnecessary. Adopted in paragraph 8. 210.-213. Rejected as unnecessary. 214. Adopted by implication in paragraph 33. 215.-218. Rejected as unsupported by the weight of evidence. Rejected as unnecessary. & 221. Rejected as contrary to the weight of evidence. 222. Adopted in summary in paragraph :28. 223.-238. Rejected as unnecessary. Orlando Regional Medical Center Adopted in paragraph 2. Adopted in paragraph 9. 3.-7. Rejected as unnecessary. Adopted in paragraph 9. Rejected as unnecessary. Adopted in substance in paragraph 42. Rejected as unnecessary. Adopted in substance in paragraph 42. Adopted in summary in paragraph 12. Adopted in paragraph 1. Rejected as ummaterial. Adopted in paragraph 6. Adopted in paragraph 4. Adopted in paragraph 15. Addressed in the preliminary statement. Adopted in paragraph 14. Rejected as unnecessary. 22.-24. Adopted in summary in paragraph 16. Adopted in paragraph 15 and conclusion of law #7. Adopted in substance in paragraph 21. Adopted in paragraph 15. Rejected as unnecessary. 29 & 30. Adopted in paragraph 9. Adopted in paragraph 20. Adopted in paragraph 22. Rejected as unnecessary. Adopted in paragraph 23. Adopted in paragraph 23. 36 & 37. Adopted in paragraph 24. Adopted in paragraph 11 and 33. Adopted in paragraph 8. Adopted in paragraph 11. 41 & 42. Rejected as unnecessary. Adopted in summary in paragraph 25. Rejected as unnecessary. Rejected as cumulative and unnecessary. Rejected as unnecessary. Adopted in summary in paragraph 26. 48.-52. Rejected as unnecessary. Adopted in paragraph 10. Rejected as contrary to the weight of evidence (the finding as to no alternatives). The finding regarding Park Place is unnecessary. Rejected as cumulative and unnecessary. Adopted in paragraph 42. Rejected as cumulative and unnecessary. Adopted in paragraph 9. Adopted in paragraph 7. Adopted in paragraph 36. Rejected as cumulative and unnecessary. Adopted in paragraph 30. 63.-66. Rejected as unnecessary. 67. Adopted in paragraph 47. The Department of Health and Rehabilitative Services 1. & 2. Addressed in Preliminary Statement. Adopted in paragraphs 6. and 9. Adopted in paragraph 24. Adopted in paragraph 20. Adopted in paragraph 36. 7.-9. Rejected as contrary to the weight of evidence. Adopted in paragraph 30. Rejected as contrary to the weight of evidence. Rejected as unnecessary. Adopted in paragraph 26. Rejected as contrary to the evidence. Adopted in paragraph 33. Adopted in substance in paragraph 32. Adopted by implication in paragraphs 30 and 34. Rejected as contrary to the evidence. Adopted in summary in paragraph 13. Adopted in paragraph 15. Rejected as contrary to the weight of evidence. Rejected as contrary to the evidence. The policy is found in HRS' Policy Manual. Rejected as immaterial. Adopted in paragraph 21. Rejected as unnecessary. 26 & 27. Adopted by implication in 23. Rejected as unnecessary. Rejected as contrary to the evidence. Rejected as immaterial. Rejected as unnecessary. Rejected as contrary to the evidence. 33 & 34. Rejected as unnecessary. Adopted in summary in paragraph 46. Rejected as contrary to the evidence, and immaterial (as to the ratio). Rejected as contrary to the definition "not normal" and immaterial. 38 & 39. Rejected as argument. Adopted in paragraph 16. Adopted in paragraph 20. Rejected as unnecessary. 43 & 49. Rejected as argument. Laurel Oaks Hospital Adopted in paragraph 1. Adopted in paragraph 2. Adopted in paragraph 3. Adopted in paragraph 4. Adopted in paragraph 5. Rejected as unnecessary. Adopted in paragraph 13. & 9. Addressed in Preliminary Statement. 10. Adopted in paragraph 18. 11.-21. Rejected as unnecessary and immaterial. Adopted in paragraph 6. Adopted in paragraph 8. Adopted in paragraph 7. Adopted in paragraph 9. Adopted in paragraph 11. Adopted in paragraph 9. Adopted in paragraph 10. Adopted in paragraph 11. Adopted in paragraph 9. Adopted in paragraph 14. Rejected as unnecessary. Rejected as contrary to the evidence. The term is "should", not "shall". Adopted in paragraph 15. 36 Adopted in paragraph 13. 37.-40. Adopted in paragraph 16. 41 & 42. Adopted in paragraph 18. 43 & 44. Rejected as immaterial and unnecessary. 45.-47. Rejected as argument and unnecessary. 48 & 49. Adopted in paragraph 21. Rejected as unnecessary. Adopted in substance in paragraph 16. 52 - 54. Rejected as unnecessary. 55 & 56. Adopted in summary in paragraph 20. 57.-61. Rejected as unnecessary or argument. 62.-65. Adopted in summary in paragraphs 20 and 23. 66 & 67. Adopted in paragraph 22. 68. Adopted in paragraph 23. 69.-72. Rejected as unnecessary or cumulative. 73 & 74. Adopted in substance in paragraph 24. 75. Rejected as contrary to the evidence. 76.-78. Rejected as unnecessary. Rejected as contrary to the evidence. Adopted in paragraphs 30 and 31. Adopted in paragraph 37. 82.-85. Rejected as unnecessary. 86. Adopted in paragraph 30. 87 & 88. Rejected as unnecessary. Adopted in paragraph 34. Adopted in paragraph 41. Adopted in paragraphs 38 and 39. 92.-95 Rejected as immaterial and unnecessary. Adopted in substance in paragraph 39. Adopted in paragraph 35. Adopted in paragraph 30. Rejected as cumulative. Rejected as contrary to the evidence. 101-112. Rejected as unnecessary. Adopted in paragraph 27. Rejected as unnecessary. Adopted in paragraph 35. 116-121. Rejected as cumulative or unnecessary. Health Management Associates1 Inc:. (HMA) 1. & 2. Adopted in paragraph 6. Adopted in paragraph 1. Adopted in paragraph 6. 5.-6. Rejected as unnecessary. 7.-11. Adopted in paragraph 9. Adopted in paragraph 10. Adopted in paragraph 12. Rejected as unnecessary. Adopted in paragraph 16. Adopted in paragraphs 15 and 23. Adopted in paragraph 25. Adopted in paragraph 16. Adopted in paragraphs 16 and 18. 20 Adopted in paragraph 19. Adopted in paragraph 15. Adopted in paragraph 21. 23 & 24. Rejected as unnecessary. 25 & 26. Adopted in paragraph 5. 27.-51. Rejected as unnecessary. 52. Adopted in paragraph 35. 53.-55. Rejected as unnecessary. COPIES FURNISHED: Stephen A. Ecenia, Esquire Michael J. Cherniga, Esquire Roberts, Baggett, LaFace & Richard 101 East College Avenue Tallahassee, FL 32301 James M. Barclay, Esquire Cobb, Cole & Bell 315 South Calhoun Street Tallahassee, FL 32301 Steven R. Bechtel, Esquire Mateer, Harbert & Bates 225 East Robinson Street Orlando, FL 32802 Edgar Lee Elzie, Esquire MacFarlane, Ferguson, Allison & Kelly First Florida Bank Building, Suite 804 Tallahassee, FL 32401 C. Gary Williams, Esquire R. Stan Peeler, Esquire Ausley, McMullen, McGehee, Carothers & Proctor 227 South Calhoun Street Tallahassee, FL 32301 John Brennan, Jr., Esquire Bonner & O'Connell 900 17th street, Suite 1000 Washington, D.C. 20006 Robert S. Cohen, Esquire Haben & Culpepper 306 North Monroe Street Tallahassee, FL 32301 John Miller, General Counsel HRS 1323 Winewood Blvd. R. S. Power, Agency Clerk HRS 1323 Winewood Blvd. Tallahassee, FL 32399-0700
Findings Of Fact A not for profit 520-bed acute care hospital in Pensacola, Baptist primarily serves not only residents of Escambia and Santa Rosa Counties, within Florida's HRS Service District I, but also patients from Escambia and Baldwin Counties in Alabama. The other two counties in District I, Okaloosa and Walton, lie outside Baptist's primary service area, but within a secondary service area, as does Covington County, Alabama. Baptist proposes to convert twelve medical/ surgical beds to a children's psychiatric service, to complement an existing 38-bed psychiatric service housed in the Behavioral Medical Center across the street from Baptist's main campus. On average, eighty percent of Baptist's existing psychiatric beds are occupied at any one time. Of four separate, psychiatric treatment programs Baptist now offers, all accredited by the Joint Commission on Accreditation for Health Care Organizations, none is designed for children below the age of 13. Seldom, and only in an emergency, has a child below this age been accepted into Baptist's program for adolescents, which is operated separately from any adult treatment program. Baptist has treated indigent and medicaid patients in its psychiatric programs, as well as patients for whose treatment it has received greater remuneration. Hospital-wide, Baptist has had "medicaid utilization" of between six and nine percent. "Baptist was willing to do Baker Act patients." Farr deposition, p. 17. Rollins deposition, p. 30. Other Resources Pensacola and Escambia County have extensive outpatient psychiatric services for children, offering a broad range of options, short of inpatient care in a treatment facility. Rollins deposition, pp. 15-16. Lakeview Community Health Center offers outpatients treatment, as do a number of private providers. The Children's Intervention Project System conducts home visits. Day care and therapeutic foster homes are also available. Professionals distinguish between "crisis stabilization" which does not "focus on treatment" and even short-term psychiatric care. Lakeview Community Health Center has a 31-bed crisis stabilization unit, which was full as of the week before the hearing. Ten of the 31 beds are reserved for children, aged 9- 17, but children's beds are not segregated from beds for adolescents. Treating children and adolescents together (if not stabilizing their crises in the same facility) is inappropriate. They have different needs and require different structures. Adolescents require more autonomy; children need more supervision. See deposition of Cruz. Farr deposition, p. 21. Rollins deposition, p. 22. Only Harbor Oaks, a free-standing facility more than 45 minutes from Pensacola and Gulf Breeze, accepts children as psychiatric patients. Harbor Oaks has 19 children's beds but does not accept medicaid patients. The children's unit at Harbor Oaks experienced an occupancy rate of approximately 74 percent in 1988. Occasionally, girls were put on waiting lists. University hospital does not accept children as psychiatric patients. It rarely accepts adolescents. West Florida Hospital, which has a program for adolescents, refuses child psychiatric patients admission. West Florida Community Center accepts no children. Nor does Humana Hospital in Ft. Walton. Play Therapy Rather than convert a part of an existing medical or surgical ward to a children's psychiatric ward, Baptist proposes to spend $565,660 to construct a facility abutting but distinct from its Behavioral Medical Center. Lakeview Medical Health Center is nearby. Farr deposition, p. 32. The parties have stipulated that "the costs and method of proposed construction, including ... energy provision and the availability of alternative, less costly or more effective methods of construction" are not in dispute, and that "the facility design schematic is reasonable and appropriate." Baptist would hire a child psychiatrist to head up to the children's psychiatric unit. Treatment teams for existing programs also include psychologists, psychiatric social workers, occupational therapists, certified recreational therapists, and nursing staff. Dr. DeMaria recommends that "somebody in the creative arts therapy," (T.99) be hired for the children's unit, as well. The parties agree that "the availability of resources, including health manpower, management personnel, and funds ... are not at issue." The plan is to create a homelike environment where children will sleep two to a room and eat together family style in a dining room. A living room, at least one classroom, a playroom and a playground out of doors are to be the situs of art, dance, music and play therapy, individual, group, and family, all in a "therapeutic milieu." Baptist intends that the children's psychiatric unit be the least restrictive inpatient facility for children possible and has given assurances that the same rigorous review now taking place in its existing psychiatric programs would see to it that children are discharged to a still less restrictive environment as soon as their conditions permitted. In large part, Baptist is counting on medical staff at the Lakeview Community Health Center, all of whom have admitting privileges at Baptist, to identify children who will need inpatient care but cannot afford to pay. Baptist has committed to reserve two beds in the proposed unit for patients who are indigent, or eligible for medicaid benefits. Baptist has also undertaken "not [to] turn away patients," Farr deposition, p. 49, needing psychiatric care. Baptist has agreed to accept a requirement that it honor this commitment, as a condition to any certificate of need it obtains. Less than 20 percent of the children seen by 19 of the 55 child psychologists practicing within Baptist's service area who responded to a survey seemed to require inpatient care, but only 60 percent of this group actually received such care. Baptist's Exhibit No. 30. A survey of referral agencies indicated some 80 children in Baptist's service area needing inpatient psychiatric care in 1988 did not receive it. Projected daily charges of $390 in Baptist's second year of operating the children's psychiatric unit are less than the $450 a day now charged by Harbor Oaks. The parties stipulated that "the pro forma income and expense statement relating to the children's short-term psychiatric beds is reasonable and requires no further proof except for validation of the number of patients days." Assuming admission rates comparable to elsewhere in the South, children in Baptist's service area would keep ten children's psychiatric beds at 70 percent average occupancy. Baptist's Exhibit No. 26. Twelve beds would make it economically feasible to serve the medically indigent as well as other children needing inpatient care. The first seven days following a child's admission staff would devote to evaluating the child. Children not discharged to a less restrictive situation by the end of the evaluation period, Baptist projects, would have an average stay totalling 28 days, as compared to the 35- to 40-day average length of stay harbor Oaks has reported. Not Normal District I has a total of 240 short-term psychiatric beds. According to the state agency action report, short-term psychiatric bed utilization was 88.9 percent at Harbor Oaks for 1987, 73.5 percent at Ft. Walton's Humana Hospital, 59.4 percent at University Hospital and 58.1 percent at West Florida. Baptist's recent experience of psychiatric bed utilization in excess of 80 percent dates to January of 1988, and is a substantial increase over the 55.8 percent reported for the period July 1986 to June 1987. Baptist's Exhibit No. 9. Projected 1993 population for District I is 601,559. Baptist's Exhibit No. 23. The parties agree that the formula set out in Rule 10- 5.011(1)(o), Florida Administrative Code, for determining "numeric need" for acute care, short-term, general psychiatric beds does not indicate a need for additional acute care short-term general psychiatric beds in District I. But 53 percent of the District's population resides in Escambia County where no treatment facility has any children's psychiatric beds. A significant number (compare Baptist's Exhibit No. 23 with T. 133) of Baptist's psychiatric admissions are patients who reside in Alabama. Although Escambia County has 52 percent of the District I population, between ages 1-12, it has none of the children's psychiatric beds. More than half the District's population lives more than 45 minutes travel time from Harbor Oaks, complicating arrangements for family therapy, often essential in these cases, Rollins deposition, pp. 28-29, and for other conferences, including discharge conferences, where parents and community-based professionals work out details necessary to effect a smooth transition from inpatient to something less restrictive. The District I Health Plan, approved on June 1, 1988, provides: The following policies and priorities are to be used in CON review in tandem with the bed need numbers on the preceding pages. POLICIES AND PRIORITIES FOR PSYCHIATRIC AND SUBSTANCE ABUSE BEDS Psychiatric or substance abuse beds which are not used by residents of the District shall not be included in the resource inventory count of the District. [NOTE: There have in the past, been facilities in another district treating patients originating solely from outside of that district. The facility's intake policies precluded the treatment of "local" district residents. In addition, the facility's marketing effort was directed entirely out- of-state. A local marketing effort plus treatment of patients originating within the district can easily be demonstrated.] Priority will be given to applicants who can demonstrate that all existing short term inpatient psychiatric beds in the subdistrict have had an average annual occupancy rate equal to or greater than 70% for the preceding year. Priority will be given to applicants who can demonstrate that all existing short term inpatient substance abuse beds in the subdistrict have had an average annual occupancy rate equal to or greater than 80% for the preceding year. Proposals for new facilities, expansions, conversions and additional services will be given priority for applicants who agree to continue or enter into Baker Act, Medicaid, Medicare and other medically indigent contracts for the provision of services to qualifying patients. Among the goals, objectives, and recommended actions set out in the 1985-1987 State Health Plan, now expired but not replaced, is a goal that short-term inpatient hospital psychiatric beds not exceed .35 per thousand population. HRS Exhibit No. 1. In requiring that .15 (of a total of .35) short-term psychiatric beds per 1,000 population be located in general hospitals eligible for medicaid reimbursement, HRS's rules do not distinguish between children and adults. But no children's psychiatric beds in Distract I are located in a facility that accepts medicaid patients. If the ratio prescribed for psychiatric beds generally applied specifically to children's psychiatric beds, District I would already have at least eight such beds: Multiplying the 19 existing beds by .15/.35 yields 8.14. Applying the rule's .15 beds per 1,000 population methodology to the 111,211 children projected to be in District I by 1993, see Baptist's Exhibit No. 23, yields a need for 16.68 children's psychiatric beds in facilities that accept medicaid patients.
Recommendation It is, accordingly, RECOMMENDED: That HRS grant Baptist's application for certificate of need No. 5669, on condition that Baptist honor its commitments to care for medically indigent and medicaid-eligible children in need of inpatient psychiatric care. DONE and ENTERED this 2nd day of November, 1989, in Tallahassee, Florida. ROBERT T. BENTON, II Hearing Officer Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, FL 32399-1550 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 2nd day of November, 1989. APPENDIX Petitioner's proposed findings of fact Nos. 1 through 6, 8 through 21, 28 through 35, 40, 44, 45, 47 through 50, 60, 61, 64, 69, 72, 73, 75, 76, 78, 79, 81 and 82 have been adopted in substance as fare as material. With respect to petitioner's proposed finding of fact No. 7, either a music therapist or an art therapist is contemplated. With respect to petitioner's proposed finding of fact No. 22, children in Escambia and Santa Rosa counties can go to Harbor Oaks. Petitioner's proposed findings of fact Nos. 23 through 26 here not established by the evidence. With respect to petitioner's proposed finding of fact No. 27 at least one eleven-year-old was also admitted. Petitioner's proposed findings of fact Nos. 36 and 37 are immaterial. Petitioner's proposed findings of fact Nos. 37, 39, 42, 43, 46, 51 through 59, 62, 63, 65, 67, 70, 71, 74, 77 and 80 relate to subordinate matters. With respect to petitioner's proposed finding of fact No. 41, the evidence did not show that everybody living in Escambia and Santa Rosa counties was more than 45 minutes from Harbor Oaks. Petitioner's proposed finding of fact No. 66 is properly a proposed conclusion of law. Respondent's proposed findings oil fact Nos. 1, 2, 5 through 8, 11, 14 and 15 have been adopted in substance insofar as material. With respect to respondent's proposed finding of fact No. 3, the petitioner's stipulation further narrowed the issues. Respondent's proposed findings of fact Nos. 4 and 17 are properly proposed conclusions of law. Respondent's proposed findings of fact Nos. 9, 10, 21, and 23 have been reject in whole or in part as unsupported by the evidence. Respondent's proposed findings of fact Nos. 12, 13, 16, 18, 20 and 22 pertain to subordinate matters. With respect to respondent's proposed finding of fact No. 19, whether institutionalizing of children is ever a good idea is not at issue in this proceeding. The question is whether services available to others should also be available to indigent patients. COPIES FURNISHED: Sam Power Agency Clerk Department of Health and Rehabilitative Services 1323 Winewood Boulevard Tallahassee, FL 32399-0700 Gregory Coler Secretary Department of Health and Rehabilitative Services 1323 Winewood Boulevard Tallahassee, FL 32399-0700 John Miller General Counsel Department of Health and Rehabilitative Services 1323 Winewood Boulevard Tallahassee, FL 32399-0700 Stephen A. Ecenia Roberts, Baggett, LaFace, and Richard 101 East College Avenue Post Office Drawer 1838 Tallahassee, FL 32302 Richard A. Patterson Assistant General Counsel Department of Health and Rehabilitative Services 2727 Mahan Drive, Suite 103 Tallahassee, FL 32308 =================================================================
Findings Of Fact Introduction Petitioner, Leesburg Regional Medical Center ("Leesburg"), is a 132-bed acute care private, not-for-profit hospital located at 600 East Dixie Highway, Leesburg, Florida. It offers a full range of general medical services. The hospital sits on land owned by the City of Leesburg. It is operated by the Leesburg hospital Association, an organization made up of individuals who reside within the Northwest Taxing District. By application dated August 13, 1982 petitioner sought a certificate of need (CON) from respondent, Department of Health and Rehabilitative Services (HRS), to construct the following described project: This project includes the addition of 36 medical/surgical beds and 7 SICU beds in existing space and the leasing of a CT scanner (replacement). The addition of the medical/surgical beds is a cost effective way to add needed capacity to the hospital. Twenty-four (24) beds on the third floor will be established in space vacated by surgery and ancillary departments moving into newly constructed space in the current renovation project. A significant portion of this area used to be an obstetric unit in the past; and therefore, is already set up for patient care. The 7 bed SICU unit will be set up on the second floor, also in space vacated as a result of the renovation project. Twelve additional beds will be available on the third and fourth floors as a result of changing single rooms into double rooms. No renovation will be necessary to convert these rooms into double rooms. It is also proposed to replace the current TechniCare head scanner with GE8800 body scanner. Based on the high demand for head and body scans and the excessive amount of maintenance problems and downtime associated with the current scanner, Leesburg Regional needs a reliable, state-of-the-art CT scanner. The cost of the project was broken down as follows: The total project cost is $1,535,000. The construction/renovation portion of the project (24 medical/surgical and 7 SICU beds) is $533,000. Equipment costs will be approximately $200,000. Architectural fees and project development costs total $52,000. The CT scanner will be leased at a monthly cost of $16,222 per month for 5 years. The purchase price of the scanner is $750,000 and that amount is included in the total project cost. The receipt of the application was acknowledged by HRS by letter dated August 27, 1982. That letter requested Leesburg to submit additional information no later than October 10, 1982 in order to cure certain omissions. Such additional information was submitted by Leesburg on October 5, 1982. On November 29, 1982, the administrator for HRS's office of health planning and development issued proposed agency action in the form of a letter advising Leesburg its request to replace a head CT scanner (whole body) at a cost of $750,000 had been approved, but that the remainder of the application had been denied. The basis for the denial was as follows: There are currently 493 medical/surgical beds in the Lake/Sumter sub-district of HSA II. Based upon the HSP for HSA II, there was an actual utilization ratio of existing beds equivalent to 2.98/1,000 population. When this utilization ratio is applied to the 1987 projected population of 156,140 for Lake/Sumter counties, there is a need for 465 medical/surgical beds by 1987. Thus, there is an excess of 28 medical/surgical beds in the Lake/Sumter sub-district currently. This action prompted the instant proceeding. At the same time Leesburg's application was being partially denied, an application for a CON by intervenor-respondent, Lake Community Hospital (Lake), was being approved. That proposal involved an outlay of 4.1 million dollars and was generally described in the application as follows: The proposed project includes the renovations and upgrading of patient care areas. This will include improving the hospital's occupancy and staffing efficiencies by reducing Med-Surg Unit-A to 34 beds and eliminating all 3-bed wards. Also reducing Med-Surg Units B and C to 34 beds each and eliminating all 3-bed wards. This will necessitate the construction of a third floor on the A wing to house the present beds in private and semi-private rooms for a total of 34 beds. There is also an immediate need to develop back-to-back six bed ICU and a six-bed CCU for shared support services. This is being done to fulfill JCAH requirements and upgrade patient care by disease entity, patient and M.D. requests. Another need that is presented for consideration is the upgrading of Administrative areas to include a conference room and more Administrative and Business office space. However, the merits of HRS's decision on Lake's application are not at issue in this proceeding. In addition to Lake, there are two other hospitals located in Lake County which provide acute and general hospital service. They are South Lake Memorial Hospital, a 68-bed tax district facility in Clermont, Florida, and Waterman Memorial Hospital, which operates a 154-bed private, not-for-profit facility in Eustis, Florida. There are no hospitals in Sumter County, which lies adjacent to Lake County, and which also shares a subdistrict with that county. The facilities of Lake and Leesburg are less than two miles apart while the Waterman facility is approximately 12 to 14 miles away. South Lake Memorial is around 25 miles from petitioner's facility. Therefore, all three are no more than a 30 minute drive from Leesburg's facility. At the present time, there are 515 acute care beds licensed for Lake County. Of these, 493 are medical/surgical beds and 22 are obstetrical beds. None are designated as pediatric beds. The Proposed Rules Rules 10-16.001 through 10-16.012, Florida Administrative Code, were first noticed by HRS in the Florida Administrative Weekly on August 12, 1983. Notices of changes in these rules were published on September 23, 1983. Thereafter, they were filed with the Department of State on September 26, 1983 and became effective on October 16, 1983. Under new Rule 10-16.004 (1)(a), Florida Administrative Code, subdistrict 7 of district 3 consists of Lake and Sumter Counties. The rule also identifies a total acute care bed need for subdistrict 7 of 523 beds. When the final hearing was held, and evidence heard in this matter, the rules were merely recommendations of the various local health councils forwarded to HRS on June 27, 1983 for its consideration. They had not been adopted or even proposed for adoption at that point in time. Petitioner's Case In health care planning it is appropriate to use five year planning horizons with an overall occupancy rate of 80 percent. In this regard, Leesburg has sought to ascertain the projected acute care bed need in Lake County for the year 1988. Through various witnesses, it has projected this need using three different methodologies. The first methodology used by Leesburg may be characterized as the subdistrict need theory methodology. It employs the "guidelines for hospital care" adopted by the District III Local Health Council on June 27, 1983 and forwarded to HRS for promulgation as formal rules. Such suggestions were ultimately adopted by HRS as a part of Chapter 10-16 effective October 16, 1983. Under this approach, the overall acute care bed need for the entire sixteen county District III was found to be 44 additional beds in the year 1988 while the need within Subdistrict VII (Lake and Sumter Counties) was eight additional beds. 2/ The second approach utilized by Leesburg is the peak occupancy theory methodology. It is based upon the seasonal fluctuation in a hospital's occupancy rates, and used Leesburg's peak season bed need during the months of February and March to project future need. Instead of using the state suggested occupancy rate standard of 80 percent, the sponsoring witness used an 85 percent occupancy rate which produced distorted results. Under this approach, Leesburg calculated a need of 43 additional beds in 1988 in Subdistrict VII. However, this approach is inconsistent with the state-adopted methodology in Rule 10- 5.11(23), Florida Administrative Code, and used assumptions not contained in the rule. It also ignores the fact that HRS's rule already gives appropriate consideration to peak demand in determining bed need. The final methodology employed by Leesburg was characterized by Leesburg as the "alternative need methodology based on state need methodology" and was predicated upon the HRS adopted bed need approach in Rule 10-5.11(23) with certain variations. First, Leesburg made non-rule assumptions as to the inflow and outflow of patients. Secondly, it substituted the population by age group for Lake and Sumter Counties for the District population. With these variations, the methodology produced an acute care bed need of 103 additional beds within Lake and Sumter Counties. However, this calculation is inconsistent with the applicable HRS rule, makes assumptions not authorized under the rule, and is accordingly not recognized by HRS as a proper methodology. Leesburg experienced occupancy rates of 91 percent, 80 percent and 73 percent for the months of January, February and March, 1981, respectively. These rates changed to 86 percent, 95 percent and 98 percent during the same period in 1982, and in 1983 they increased to 101.6 percent, 100.1 percent and 95.1 percent. Leesburg's health service area is primarily Lake and Sumter Counties. This is established by the fact that 94.4 percent and 93.9 percent of its admissions in 1980 and 1981, respectively, were from Lake and Sumter Counties. Although South Lake Memorial and Waterman Memorial are acute care facilities, they do not compete with Leesburg for patients. The staff doctors of the three are not the same, and there is very little crossover, if any, of patients between Leesburg and the other two facilities. However, Lake and Leesburg serve the same patient base, and in 1982 more than 70 percent of their patients came from Lake County. The two compete with one another, and have comparable facilities. Leesburg has an established, well-publicized program for providing medical care to indigents. In this regard, it is a recipient of federal funds for such care, and, unlike Lake, accounts for such care by separate entry on its books. The evidence establishes that Leesburg has the ability to finance the proposed renovation. HRS's Case HRS's testimony was predicated on the assumption that Rule 10-16.004 was not in effect and had no application to this proceeding. Using the bed need methodology enunciated in Rule 10-5.11(23), its expert concluded the overall bed need for the entire District III to be 26 additional beds by the year 1988. This calculation was based upon and is consistent with the formula in the rule. Because there was no existing rule at the time of the final hearing concerning subdistrict need, the witness had no way to determine the bed need, if any, within Subdistrict VII alone. Lake's Case Lake is a 162-bed private for profit acute care facility owned by U.S. Health Corporation. It is located at 700 North Palmetto, Leesburg, Florida. Lake was recently granted a CON which authorized a 4.1 million dollar renovation project. After the renovation is completed all existing three-bed wards will be eliminated. These will be replaced with private and semi-private rooms with no change in overall bed capacity. This will improve the facility's patient utilization rate. The expansion program is currently underway. Like Leesburg, the expert from Lake utilized a methodology different from that adopted for use by HRS. Under this approach, the expert determined total admissions projected for the population, applied an average length of stay to that figure, and arrived at a projected patient day total for each hospital. That figure was then divided by bed complement and 365 days to arrive at a 1988 occupancy percentage. For Subdistrict VII, the 1988 occupancy percentage was 78.2, which, according to the expert, indicated a zero acute care bed need for that year. Lake also presented the testimony of the HRS administrator of the office of community affairs, an expert in health care planning. He corroborated the testimony of HRS's expert witness and concluded that only 26 additional acute care beds would be needed district-wide by the year 1988. This result was arrived at after using the state-adopted formula for determining bed need. During 1981, Lake's actual total dollar write-off for bad debt was around $700,000. This amount includes an undisclosed amount for charity or uncompensated care for indigent patients. Unlike Leesburg, Lake receives no federal funds for charity cases. Therefore, it has no specific accounting entry on its books for charity or indigent care. Although Leesburg rendered $276,484 in charity/uncompensated care during 1981, it is impossible to determine which facility rendered the most services for indigents due to the manner in which Lake maintains its books and records. In any event, there is no evidence that indigents in the Subdistrict have been denied access to hospital care at Lake or any other facility within the county. Lake opines that it will loose 2.6 million dollars in net revenues in the event the application is granted. If true, this in turn would cause an increase in patient charges and a falling behind in technological advances. For the year 1981, the average percent occupancy based on licensed beds for Leesburg, Lake, South Lake Memorial and Waterman Memorial was as follows: 71.5 percent, 58.7 percent, 63.8 percent and 65.7 percent. The highest utilization occurred in January (81 percent) while the low was in August (58 percent). In 1982, the utilization rate during the peak months for all four facilities was 78 percent. This figure dropped to 66.5 percent for the entire year. Therefore, there is ample excess capacity within the County even during the peak demand months.
Recommendation Based on the foregoing findings of fact and conclusions of law, it is RECOMMENDED that the application of Leesburg Regional Medical Center for a certificate of need to add 43 acute care beds, and renovate certain areas of its facility to accommodate this addition, be DENIED. DONE and ENTERED this 15th day of December, 1983, in Tallahassee, Florida. DONALD R. ALEXANDER Hearing Officer Division of Administrative Hearings The Oakland Building 2009 Apalachee Parkway Tallahassee, Florida 32301 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 15th day of December, 1983.
Findings Of Fact Upon consideration of the oral and documentary evidence adduced at the hearing, the following relevant facts are found: East Pasco Medical Center (EPMC) is a non-profit 85-bed acute care hospital facility located in the East Pasco subdistrict of HRS District V. There are only two hospitals in the subdistrict -- EPMC in Zephyrhills and Humana in Dade City, which is approximately ten miles north. Humana is a 120- bed acute care hospital facility. Both facilities offer the same services and share the same medical staff. On or about September 17, 1987, EPMC submitted an application for a Certificate of Need to add 35 medical/surgical beds via a fourth floor addition to its existing facility. Its existing 85 beds are located in private rooms, and it is proposed that the additional 35 beds will also be placed in separate rooms. The application submitted to the Department of Health and Rehabilitative Services (HRS) projected a total project cost of $4,531,000. This figure was revised at the hearing to a project cost of $2,302,900. With regard to acute care services, the State Health Plan seeks to assure geographic accessibility. All residents of East Pasco County currently have access to acute care hospital services within the travel times suggested by the State plan. The State Health Plan also seeks to promote the efficient utilization of acute care services by attaining an average annual occupancy rate of at least 80 percent. The District V Local Health Plan emphasizes that additions to inpatient acute care beds in a subdistrict should not be considered unless a numeric bed need is shown and certain occupancy thresholds have been met. The recommended occupancy thresholds for medical/surgical beds are 80% for the subdistrict and 90% for the facility seeking to add beds. Application of the bed need methodology contained in HRS's Rule 10- 5.011(1)(m), Florida Administrative Code, indicates a numeric need for 57 additional acute care medical/surgical beds in the East Pasco subdistrict for the planning horizon period of July, 1992. The rule provides that HRS will "not normally approve" additional beds unless average occupancy in the subdistrict is greater than 75 percent. However, the rule permits HRS to award additional beds when there is a calculated need, notwithstanding low occupancy in the subdistrict, if the applicant had a minimum of 75% average occupancy during the 12 months ending 14 months prior to the Letter of Intent. Rule 10- 5.011(1)(m)7.e., Florida Administrative Code. The rule also permits HRS to award additional beds where the calculated numeric need substantially exceeds the number of existing and approved beds in the subdistrict and there is an access problem related to travel time. For the relevant time period, the acute care occupancy rate for the East Pasco subdistrict was below 75% percent. Indeed, over the past few years, the average occupancy rate in that subdistrict has been 54 to 58 percent. Humana only operates at about a 55% occupancy. The East Pasco subdistrict does experience seasonal fluctuations in medical/surgical occupancy, with the season for high occupancy beginning in late October and ending in mid- to late April. In addition to tourists, it is expected that the revival of the citrus industry in East Pasco County will bring more migrant pickers to the area during the peak season months. The seasonal increase in occupancy directly corresponds with a large increase in seasonal population, particularly in the Zephyrhills area. The Zephyrhills area population is much older than the Dade City population and is also much older than the State average. The HRS acute care bed need rule includes considerations of seasonal peak demands. When considering both hospitals in the subdistrict, there has been a decline in peak seasonal occupancy rates over the past few years. While the population of the East Pasco subdistrict has grown, and is expected to increase by approximately 7,200 in 1992, there is a trend of declining utilization in the subdistrict. This decline is due to increased used of outpatient services and shorter lengths of hospital stay attributable to the current reimbursement system. The medical/surgical use rate fell from 454 patient days per 1,000 population in 1986 to 414 patient days per 1,000 population in 1988. There was a similar decline in the acute care use rate. Assuming a constant medical/surgical use rate, the projected demand for 1992 would be 2,980 additional medical/surgical patient days in the subdistrict according to population projections, and about 4,267 incremental patient days according to local health council projections. EPMC's Letter of Intent to add 35 additional beds was filed in mid- July, 1987. Its acute care occupancy rate for the period of April, 1986 through March, 1987 was 75.3 percent. Occupancy at EPMC from May, 1986 to April, 1987 was 73.6%; occupancy from June, 1986 through May, 1987 was 73%; and occupancy from July, 1986 to June, 1987 was 72.2 percent. EPMC does experience periods of high occupancy during the peak season months. High occupancy levels have a greater impact upon smaller hospitals due to their lesser degree of flexibility. On occasion, during the winter months, EPNC is required to refuse admittance to patients due to crowded conditions within its facility. Patients are sometimes transferred or referred to other facilities, including Humana, although the necessity for such transfers or referrals is occasionally due to a lack of intensive or critical care beds as opposed to a lack of medical/surgical beds. During the periods of time when EPMC had high occupancy levels, beds were available at Humana. EPMC's current payor mix includes a high level of Medicare (over 60%), and it is committed, through both its Christian mission and an agreement with the County, to treat indigent and Medicaid patients. The actual amount of indigent or charity care provided by EPNC was not established. In any event, EPMC desires to increase its bed size in order to help maintain a proper payor mix at the hospital so as to ensure the financial survival of the hospital. It is felt that a greater number of beds, given the rise in population, and particularly elderly population, would allow EPNC to serve a greater number of private and/or third party insurance paying patients. While the evidence demonstrates that EPMC may operate with a less favorable payor mix than Humana, the evidence was not sufficient to demonstrate that EPMC will suffer financial ruin without additional beds. Likewise, it was not established that the patients which EPNC must turn away in the winter months are consistently paying patients. Increasing the number of beds at EPNC to 120 beds does not necessarily mean that its profitability would be improved. Volume and payor mix are the most critical factors in determining whether a hospital will be profitable. There is currently a nursing shortage throughout the nation. Rural areas, such as the eastern portion of Pasco County, experience even greater difficulty in attracting nursing personnel to the area. Due to the shortage of nurses, as well as the seasonal demand, EPMC is required to use contract care nurses throughout the year. While it would prefer to employ its own nursing staff, EPMC will use contract staff due to the seasonal variations in its nursing requirements. The use of contract or registry nurses costs 50% to 60% more on a daily basis; however, lower occupancy during the off-peak months does not justify year- round employment for as large an in-house nursing staff. For its proposed 35 beds, EPMC projects nurse manpower requirements as follows: 1 nurse manager, 4.2 R.N. charge nurses, 15.1 R.N. staff and 14.1 L.P.N. staff, for a total of 34.4 full time equivalent nursing positions. The recruiting efforts of EPNC to fill these positions will include advertising, visiting nursing schools and colleges, utilizing student nurses at the hospital and use of the Adventist Health System international network. Humana currently has 15 vacancies, or 12 to 13% of its nursing staff. Humana's nursing salaries have increased 20% over the past eighteen months. As noted above, EPNC and Humana compete for the same nursing personnel. Humana's personnel director believes that if EPNC increases its nursing staff by 34 FTEs, Humana's nursing staff will be approached to fill those positions. As a consequence, Humana will experience additional nursing shortages and will be required to further increase salaries. It is proposed that the project cost of adding 35 beds to EPMC will be financed with 100% debt financing through a bond issue. The financing will be part of a much larger bond issuance intended to finance several other projects within the Adventist hospital system. No evidence was adduced that such a bond issuance had been prepared or approved, and there was no evidence concerning the other projects which would be financed in conjunction with this project. In 1987, EPNC was carrying about five million dollars of negative equity. The hospital is currently greater than 100% financed. As noted above, the original Certificate of Need application filed with HRS listed the total project cost to be $4,531,000. In its response to omissions, EPMC stated that the construction cost would be $175 per square foot. In the updates submitted at the hearing, EPNC proposed a project cost of $2,302,900, which included a construction cost of $85 per square foot. A more reasonable cost for the addition of a floor to an existing facility would be $125 per square foot, plus an inflation factor of 6% and architectural and engineering fees of 6 to 7%. The proposed equipment list submitted by EPNC fails to include major equipment items such as an overhead paging system, a nurse call system, examination room equipment, medication distribution equipment, bed curtains, shower curtains, patient and staff support lounge items, and IV pumps. EPNC's updated equipment cost budget fails to include tax, freight, contingency and installation costs. The projected equipment costs should be tripled to adequately and reasonably equip a 35-bed nursing unit. The projected utilization and pro formas submitted by EPMC are not reasonable and were not supported by competent substantial evidence. EPMC's projected utilization for the proposed 35-bed unit is 8,950 patient days in the first year of operation and 9,580 in the second year of operation. Applying the current use rate to the population projections submitted by EPMC's expert in demographics and population projections produces only about 2,980 additional patient days in the year 1992. Given the fact that EPMC's current market share is approximately 54%, there is no reason to believe that Humana would not absorb at least some of those projected additional patient days. There are many months of the year in which additional patient days could be filled within the existing complement of 85 beds at EPNC. Depending upon the ultimate cost of the project, the break even point for financial feasibility purposes would be approximately 3,500 to 4,000 patient days. The concept behind a pro forma is to develop a financial picture of what operations will be in the first two years of operation. EPMC stated its revenues and expenses in terms of 1988 dollars and used its current revenue- to-expense ratios for projecting operations four years into the future. This is improper because gross revenues are going up, reimbursement is not increasing as rapidly and expenses, particularly salaries and insurance, are increasing. In addition, EPMC's projected 1992 salaries in several categories were less than they are currently paying for such positions. EPMC currently provides good quality of care to its patients. The only future concern in this realm is the fact that in the winter months, its intensive and critical care unit beds are often full and there is no room for additional patients. Additional medical/surgical volume from the proposed 35- bed unit would lead to additional intensive and critical care bed demand.
Recommendation Based upon the findings of fact and conclusions of law recited herein, it is RECOMMENDED that the application of East Pasco Medical Center for a Certificate of Need to add 35 acute care beds to its existing facility be DENIED. Respectfully submitted and entered this 30 day of March, 1989, in Tallahassee, Florida. DIANE D. TREMOR 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 30th day of March, 1989. APPENDIX TO RECOMMENDED ORDER CASE NO. (Case No. 88-1227) The proposed findings of fact submitted by the parties have been carefully considered and are accepted, incorporated and/or summarized in this Recommended Order, with the following exceptions: Petitioner: Third sentence rejected as not established by competent, substantial evidence. Accepted, but not included as irrelevant to the ultimate resolution of the issues. Rejected. The Personnel Director of Humana presented testimony in this proceeding. Accepted as an accurate restatement of testimony, but rejected as an erroneous conclusion of law. 16. Second sentence rejected as an erroneous conclusion of law. A18. Rejected as contrary to the evidence. 20. First sentence rejected as an erroneous conclusion of law. First sentence rejected as an erroneous conclusion of law. Rejected as not supported by competent substantial evidence. 27 and 30. Accepted as an accurate restatement of testimony, but rejected as an erroneous conclusion of law. Rejected as immaterial to the issue of need in the year 1992. First sentence rejected as not established by competent substantial evidence. First and third sentences rejected as not established by competent substantial evidence. 37 and 38. Rejected as not established by competent substantial evidence. 44. Last sentence rejected as unsupported by competent substantial evidence. Accepted only if the factors of volume and payor mix are also considered. Partially rejected as speculative and not supported by competent substantial evidence. All but first two sentences rejected as unsupported by competent substantial evidence and an erroneous conclusion of law. Rejected as unsupported by competent substantial evidence and an erroneous conclusion of law. Last sentence rejected as unsupported by the evidence. Rejected as unsupported by competent substantial evidence. Second sentence rejected as contrary to the greater weight of the evidence. 58. Rejected as irrelevant and immaterial. 60. Rejected as not established by competent substantial evidence. 62 - 67. The actual figures regarding total costs, projected utilization and those figures utilized in the pro formas were not established by competent substantial evidence and, therefore, the findings regarding the financial feasibility of the project are rejected. 71. Rejected as not supported by competent substantial evidence. 74. Rejected as not supported by competent substantial evidence. 77. Rejected as an improper factual finding and contrary to the evidence. 78 and 79. Rejected as contrary to the greater weight of the evidence. First sentence rejected as unsupported by competent substantial evidence. Last sentence rejected as unsupported by the evidence. Rejected as contrary to the evidence. Respondent: 2 and 6. Partially accepted with the additional considerations of the applicant's occupancy levels and geographic accessibility. 9. Rejected as contrary to the evidence. 19(a) Interpretation of rule not sufficiently explicated at hearing. 56 - 58. Actual figures are not established by competent evidence due to the failure to establish with reliability the total costs of the project. Intervenor: Second sentence accepted with the additional considerations of the applicant's occupancy levels and geographic accessibility. Third sentence rejected. Interpretation of rule not sufficiently explicated at hearing. First sentence rejected, but this does not preclude a consideration of such a period. Third sentence rejected as not established by the greater weight of the evidence. 31. Second sentence rejected as speculative. 40 and 41. Accepted as factually correct, but not included due to the showing of unused capacity within the East Pasco subdistrict. 55 and 56. Actual figures are not established by competent evidence due to the failure to establish with reliability the total costs of the project. 63 and 72. Same as above with regard to second sentence. 92. Rejected as an overbroad statement or conclusion. 97. Second sentence rejected as overbroad and not supported by the evidence. COPIES FURNISHED: E.G. Boone and Jeffrey Boone 1001 Avenida del Circo Post Office Box 1596 Venice, Florida 34284 Stephen M. Presnell Macfarlane, Ferguson, Allison & Kelly Post Office Box 82 Tallahassee, Florida 323a2 James C. Hauser Messer, Vickers, Caparello, French & Madsen, P.A. Post Office Box 1876 Tallahassee, Florida 32302 Gregory L. Coler, Secretary Department of Health and Rehabilitative Services 1323 Winewood Boulevard Tallahassee, Florida 32399-0700 Sam Power, Clerk Department of Health and Rehabilitative Services 1323 Winewood Boulevard Tallahassee, Florida 32399-0700 John Miller, Esquire General Counsel Department of Health and Rehabilitative Services 1323 Winewood Boulevard Tallahassee, Florida 32399-0700
The Issue Kindred Hospitals East, LLC ("Kindred") and Select Specialty Hospital-Palm Beach, Inc. ("Select-Palm Beach"), filed applications for Certificates of Need ("CONs") with the Agency for Health Care Administration ("AHCA" or the "Agency") seeking approval for the establishment of long-term care hospitals ("LTCHs") in Palm Beach County, AHCA District 9. Select-Palm Beach's application, CON No. 9661, seeks approval for the establishment of a 60-bed freestanding LTCH in "east central" Palm Beach County about 20 miles south of Kindred's planned location. Kindred's application, CON No. 9662, seeks approval for the establishment of a 70-bed LTCH in the "north central" portion of the county. The ultimate issue in this case is whether either or both applications should be approved by the Agency.
Findings Of Fact Long Term Care Hospitals Of the four classes of facilities licensed as hospitals by the Agency, "Class I or general hospitals," includes: General acute care hospitals with an average length of stay of 25 days or less for all beds; Long term care hospitals, which meet the provisions of subsection 59A-3.065(27), F.A.C.; and, Rural hospitals designated under Section 395, Part III, F.S. Fla. Admin. Code R. 59A-3.252(1)(a). This proceeding concerns CON applications for the second of Florida's Class I or general hospitals: LTCHs. A critically ill patient may be admitted and treated in a general acute care hospital, but, if the patient cannot be stabilized or discharged to a lower level of care on the continuum of care within a relatively short time, the patient may be discharged to an LTCH. An LTCH patient is almost always "critically catastrophically ill or ha[s] been." (Tr. 23). Typically, an LTCH patient is medically unstable, requires extensive nursing care with physician oversight, and often requires extensive technological support. The LTCH patient usually fits into one or more of four categories. One category is patients in need of pulmonary/respiratory services. Usually ventilator dependent, these types of LTCH patients have other needs as well that requires "complex comprehensive ventilator weaning in addition to meeting ... other needs." (Tr. 26). A second category is patients in need of wound care whose wound is life-threatening. Frequently compromised by inadequate nutrition, these types of LTCH patients are often diabetic. There are a number of typical factors that may account for the seriousness of the wound patient's condition. The job of the staff at the LTCH in such a case is to attend to the wound and all the other medical problems of the patient that have extended the time required for care of the wound. A third category is patients with some sort of neuro-trauma. These patients may have had a stroke and are often elderly; if younger, they may be victims of a car accident or some other serious trauma. They typically have multiple body systems that require medical treatment, broken bones and a closed head injury for example, that have made them "very sick and complex." (Tr. 27). The fourth category is referred to by the broad nomenclature of "medically complex" although it is a subset of the population of LTCH patients all of whom are medically complex. The condition of the patients in this fourth category involves two or more body systems. The patients usually present at the LTCH with "renal failure ... [and] with another medical condition ... that requires a ventilator ..." Id. In short, LTCHs provide extended medical and rehabilitative care to patients with multiple, chronic, and/or clinically complex acute medical conditions that usually require care for a relatively extended period of time. To meet the definition of an LTCH a facility must have an average length of inpatient stay ("ALOS") greater than 25 days for all hospital beds. See Fla. Admin. Code R. 59A-3.065(34). The staffs at general acute care hospitals and LTCHs have different orientations. With a staff oriented toward a patient population with a much shorter ALOS, the general acute care hospital setting may not be appropriate for a patient who qualifies for LTCH services. The staff at a general acute care hospital frequently judges success by a patient getting well in a relatively short time. It is often difficult for general acute care hospital staff to sustain the interest and effort necessary to serve the LTCH patient well precisely because of the staff's expectation that the patient will improve is not met in a timely fashion. As time goes by, that expectation continues to be frustrated, a discouragement to staff. The LTCH is unlike other specialized health care settings. The complex, medical, nursing, and therapeutic requirements necessary to serve the LTCH patient may be beyond the capability of the traditional comprehensive medical rehabilitation ("CMR") hospital, nursing home, skilled nursing facility ("SNF"), or, the skilled nursing unit ("SNU"). CMR units and hospitals are rarely, if ever, appropriate for the LTCH patient. Almost invariably, LTCH patients are not able to tolerate the minimum three (3) hours of therapy per day associated with CMR. The primary focus of LTCHs, moreover, is to provide continued acute medical treatment to the patient that may not yet be stable, with the ultimate goal of getting the patient on the road to recovery. In comparison, the CMR hospital treats medically stable patients consistent with its primary focus of restoring functional capabilities, a more advanced step in the continuum of care. Services provided in LTCHs are distinct from those provided in SNFs or SNUs. The latter are not oriented generally to patients who need daily physician visits or the intense nursing services or observations needed by an LTCH patient. Most nursing and clinical personnel in SNFs and SNUs are not experienced with the unique psychosocial needs of long-term acute care patients and their families. An LTCH is distinguished within the healthcare continuum by the high level of care the patient requires, the interdisciplinary treatment model it follows, and the duration of the patient's hospitalization. Within the continuum of care, LTCHs occupy a niche between traditional acute care hospitals that provide initial hospitalization care on a short-term basis and post-acute care facilities such as nursing homes, SNFs, SNUs, and comprehensive medical rehabilitation facilities. Medicare has long recognized LTCHs as a distinct level of care within the health care continuum. The federal government's prospective payment system ("PPS") now treats the LTCH level of service as distinct with its "own DRG system and ... [its] own case rate reimbursement." (Tr. 108). Under the LTCH PPS, each patient is assigned an LTC- DRG (different than the DRG under the general hospital DRG system) with a corresponding payment rate that is weighted based on the patient diagnosis and acuity. The Parties The Agency is the state agency responsible for administering the CON Program and licensing LTCHs and other hospital facilities pursuant to the authority of Health Facility and Services Development Act, Sections 408.031-408.045, Florida Statutes. Select-Palm Beach is the applicant for a free-standing 60-bed LTCH in "east Central Palm Beach County," Select Ex. 1, stamped page 12, near JFK Medical Center in AHCA District 9. Its application, CON No. 9661, was denied by the Agency. Select-Palm Beach is a wholly owned subsidiary of Select Medical Corporation, which provides long term acute care services at 83 LTCHs in 24 states, four of which are freestanding hospitals. The other 79 are each "hospitals-in-a- hospital" ("HIH" or "LTCH HIH"). Kindred is the applicant for a 70-bed LTCH to be located in the north central portion of Palm Beach County in AHCA District 9. Its application, CON No. 9662, was denied by the Agency. Kindred is a wholly owned subsidiary of Kindred Healthcare, Inc. ("Kindred Healthcare"). Kindred Healthcare operates 73 LTCHs, 59 of which are freestanding, according to the testimony of Mr. Novak. See Tr. 56-57. Kindred Healthcare has been operating LTCHs since 1985 and has operated them in Florida for more than 15 years. At the time of the submission of Kindred's application, Kindred Healthcare's six LTCHs in Florida were Kindred-North Florida, a 60-bed LTCH in Pinellas County, AHCA District 5; Kindred-Central Tampa, with 102 beds, and Kindred-Bay Area- Tampa, with 73 beds, both in Hillsborough County, in AHCA District 6; Kindred-Ft. Lauderdale with 64 beds and Kindred- Hollywood with 124 beds, both in Broward County, ACHA District 10; and Kindred-Coral Gables, with 53 beds, in Dade County, AHCA District 11. The Applications and AHCA's Review The applications were submitted in the first application cycle of 2003. Select-Palm Beach's application is CON No. 9661; Kindred's is CON No. 9662. Select-Palm Beach estimates its total project costs to be $12,856,139. Select-Palm Beach has not yet acquired the site for its proposed LTCH, but did include in its application a map showing three priority site locations, with its preferred site, designated "Site 1," located near JFK Medical Center. At $12,937,419, Kindred's estimate of its project cost is slightly more than Select-Palm Beach's. The exact site of Kindred's proposed LTCH had not been determined at the time of hearing. Kindred's preference, however, is to locate in the West Palm Beach area in the general vicinity of St. Mary's Hospital, in the northern portion of Palm Beach County along the I-95 corridor. This is approximately 15 to 20 miles north of Select's preferred location for its LTCH. There is no LTCH in the five-county service area that comprises District 9: Indian River, Okeechobee, St. Lucie, Martin, and Palm Beach Counties. There are two LTCHs in adjacent District 10 (to the south). They have a total of 188 beds and an average occupancy of 80 percent. The Agency views LTCH care as a district-wide service primarily for Medicare patients. At the time of the filing of the applications, the population in District 9 was over 1.6 million, including about 400,000 in the age cohort 65 and over. About 70 percent of the District 9 population lives in Palm Beach County. More than 70 percent of the District's general acute care hospitals are located in that county. Kindred's preferred location for its LTCH is approximately 40 to 50 miles from the closest District 10 LTCH; Select-Palm Beach is approximately 25 to 35 miles from the closest District 10 LTCH. The locations of Select Palm-Beach's and Kindred's proposed LTCHs are complementary. The SAAR Following its review of the two applications, AHCA issued its State Agency Action Report ("SAAR"). Section G., of the report, entitled "RECOMMENDATION," states: "Deny Con #9661 and CON #9662." Agency Ex. 2, p. 43. On June 11, 2003, the report was signed by Karen Rivera, Health Services and Facilities Consultant Supervisor Certificate of Need, and Mr. Gregg as the Chief of the Bureau of Health Facility Regulation. It contained a section entitled "Authorization for Agency Action" that states, "[a]uthorized representatives of the Agency for Health Care Administration adopted the recommendations contained herein and released the State Agency Action Report." Agency Ex. 2, p. 44. The adoption of the recommendations is the functional equivalent of preliminary denial of the applications. In Section F. of the SAAR under the heading of "Need," (Agency Ex. 2, p. 40), the Agency explained its primary bases for denial; it concluded that the applicants had not shown need for an LTCH in AHCA District 9. The discussions for the two, although not precisely identical, are quite similar: Select Specialty Hospital-Palm Beach, Inc.(CON #9661): The applicant's two methodological approaches to demonstrate need are not supported by any specific discharge studies or other data, including DRG admission criteria from area hospitals regarding potential need. The applicant also failed to provide any supporting documentation from area physicians or other providers regarding potential referrals. It was further not demonstrated that patients that qualify for LTCH services are not currently being served or that an access problem exists for residents in District 9. Kindred Hospitals East, L.L.C. (CON #9662): The various methodological approaches presented are not supported by any specific DRG admission criteria from area hospitals suggesting potential need. The applicant provided numerous letters of support for the project from area hospitals, physicians and case managers. However, the number of potential referrals of patients needing LTCH services was not quantified. It was further not demonstrated that patients that qualify for LTCH services are not currently being served or that an access problem exists for residents in District 9. Id. At hearing, the Agency's witness professed no disagreement with the SAAR and continued to maintain the same bases contained in the SAAR for the denials of the two applications The SAAR took no issue with either applicant's ability to provide quality care. It concluded that funding for each applicant was likely to be available and that each project appeared to be financially feasible once operating. The SAAR further stated that there were no major architectural concerns regarding Kindred's proposed facility design, but noted reservations regarding the need for further study and revision of Select Palm-Beach's proposed surgery/procedure wing, as well as cost uncertainties for Select Palm Beach because of such potential revisions. By the time of final hearing, however, the parties had stipulated to the reasonableness of each applicant's proposed costs and methods of construction. The parties stipulated to the satisfaction of a number of the statutory CON criteria by the two applicants. The parties agreed that the applications complied with the content and review process requirements of sections 408.037 and 409.039, Florida Statutes, with one exception. Select reserved the issue of the lack of a Year 2 of Schedule 6, (Staffing) in Kindred's application. The form of Schedule 6 provided by AHCA to Kindred (unlike other schedules of the application) does not clearly indicate that a second year of staffing data must be provided. The remainder of the criteria stipulated and the positions of the parties as articulated in testimony at hearing and in the proposed orders that were submitted leave need as the sole issue of consequence with one exception: whether Kindred has demonstrated that its project is financially feasible in the long term. Kindred's Long Term Financial Feasibility Select-Palm Beach contends that Kindred's project is not financially feasible in the long term for two reasons. They relate to Kindred's application and are stated in Select Palm Beach's proposed order: Kindred understated property taxes[;] Kindred completely fails to include in its expenses on Schedule 8, patient medical assistance trust fund (PMATF) taxes [citation omitted]. Proposed Recommended Order of Select-Palm Beach, Inc., p. 32, Finding of Fact 97. Raised after the proceeding began at DOAH by Select- Palm Beach, these two issues were not considered by AHCA when it conducted its review of Kindred's application because the issues were not apparent from the face of the application. AHCA's Review of Kindred's Application Kindred emerged from a Chapter 11 bankruptcy proceedings on April 20, 2001, under a plan of reorganization. With respect to the events that led to the bankruptcy proceeding and the need to review prior financial statements, AHCA made the following finding in the SAAR: Under the plan [of reorganization], the applicant [Kindred] adopted the fresh start accounting provision of SOP 90-7. Under fresh start accounting, a new reporting entity is created and the recorded amounts of assets and liabilities are adjusted to reflect their estimated fair values. Accordingly, the prior period financial statements are not comparable to the current period statements and will not be considered in this analysis. Agency Ex. 2, p. 30. The financial statements provided by Kindred as part of its application show that Kindred Healthcare, Kindred's parent, is a financially strong company. The information contained in Kindred's CON application filed in 2003 included Kindred Healthcare's financial statements from the preceding calendar year. Kindred Healthcare's Consolidated Statement of Operations for the year ended December 31, 2002, showed "Income from Operations" to be more than $33 million, and net cash provided by operating activities (cash flow) of over $248 million for the period. Its Consolidated Balance Sheet as of December 31, 2002, showed cash and cash equivalents of over $244 million and total assets of over $1.6 billion. In light of the information contained in Kindred's CON application, the SAAR concluded with regard to short term financial feasibility: Based on the audited financial statements of the applicant, cash on hand and cash flows, if they continue at the current level, would be sufficient to fund this project as proposed. Funding for all capital projects, with the support of its parent, is likely to be available as needed. Agency Ex. 2, p. 30 (emphasis supplied). The SAAR recognized that Kindred projected a "year two operating loss for the hospital of $287,215." Agency Ex. 2, p. Nonetheless, the SAAR concludes on the issue of financial feasibility, "[w]ith continued operational support from the parent company, this project [Kindred's] is considered financially feasible." Id. The Agency did not have the information, however, at the time it reviewed Kindred's application that Kindred understated property taxes and omitted the Public Medicaid Trust Fund and Medical Assistance Trust Fund ("PMATF") "provider tax" of 1.5 percent that would be imposed on Kindred's anticipated revenues of $11,635,919 as contended by Select-Palm Beach. Consistent with Select Palm-Beach's general contentions about property taxes and PMATF taxes, "Kindred acknowledges that it likely understated taxes to be incurred in the operation of its facility." Kindred's Proposed Recommended Order, paragraph 50, p. 19. The parties agree, moreover, that the omitted PMATF tax is reasonably projected to be $175,000. They do not agree, however, as to the impact of the PMATF tax on year two operating loss. The difference between the two (approximately $43,000) is attributable to a corporate income tax benefit deduction claimed by Kindred so that the combination of the application's projected loss, the omitted PMATF tax, and the deduction yields a year two operating loss of approximately $419,000. Without taking into consideration the income tax benefit, Select-Palm Beach contends that adding in the PMATF tax produces a loss of $462,000. Kindred and Select-Palm Beach also disagree over the projection of property taxes by approximately $50,000. Kindred projects that the property taxes in year two of operation will be approximately $225,000 instead of the $49,400 listed in the application. Select-Palm Beach projects that they will be $50,000 higher at approximately $275,000. Whether Kindred's or Select-Palm Beach's figures are right, Kindred makes two points. First, if year two revenues and expenses, adjusted for underestimated and omitted taxes, are examined on a quarterly basis, the fourth quarter of year two has a better bottom line than the earlier quarters. Not only will the fourth quarter bottom line be better, but, using Kindred's figures, the fourth quarter of year two of operations is profitable. Second, and most importantly given the Agency's willingness to credit Kindred with financial support from its parent, Kindred's application included in its application an interest figure of $1.2 million for year one of operation and $1.03 million for year two. Kindred claims in its proposed recommended order that "[i]n reality ... this project will incur no interest expense as Kindred intends to fund the project out of cash on hand, or operating capital, and would not have to borrow money to construct the project." Id., at paragraph 54, p. 20. Through the testimony of John Grant, Director of Planning and Development for Kindred's parent, Kindred Healthcare, Kindred indicated at hearing that its parent might, indeed, fund the project: A ... Kindred [Healthcare] would likely fund this project out of operating capital. Like I said, in the first nine months of this year Kindred had operating cash flow of approximately $180 million. So it's not as if we would have to actually borrow money to complete a project like this. Q And what was the interest expense that you had budgeted in Year Two for this facility? A $1,032,000. Q ... so is it your statement then that this facility would not owe any interest back to the parent company? A That's correct. Tr. 221-222 (emphasis supplied). If the "financing interest" expense is excluded from Kindred's statement of projected expenses in Schedule 8 of the CON application, using Kindred's revised projections, the project shows a profit of approximately $612,0002 for the second year of operation. If Select-Palm Beach's figures and bottom line loss excludes the "finances interest" expense, the elimination of the expense yields of profit for year two of operations in excess of $500,000. If the support of Kindred's parent is considered as the Agency has signaled its willingness to do and provided that the project is, in fact, funded by Kindred Healthcare rather than financed through some other means that would cause Kindred to incur interest expense, Kindred's project is financially feasible in the long term. With the exception of the issue regarding Kindred's long term financial feasibility, as stated above, taken together, the stipulation and agreements of the parties, the Agency's preliminary review contained in the SAAR, and the evidence at hearing, all distill the issues in this case to one overarching issue left to be resolved by this Recommended Order: need for long term care hospital beds in District 9. Need for the Proposals From AHCA's perspective prior to the hearing, the only issue in dispute with respect to the two applications is need. This point was made clear by Mr. Gregg's testimony at hearing in answer to a question posed by counsel for Select-Palm Beach: Q. ... Assuming there was sufficient need for 130 beds in the district is there any reason why both applicants shouldn't be approved in this case, assuming that need? A. No. (Tr. 398). Both applicants contend that the application each submitted is superior to the other. Neither, however, at this point in the proceeding, has any objection to approval of the other application provided its own application is approved. Consistent with its position that both applications may be approved, Select-Palm Beach presented testimony through its health care planner Patricia Greenberg3 that there was need in District 9 for both applicants' projects. Her testimony, moreover, rehabilitated the single Kindred methodology of three that yielded numeric need less than the 130 beds proposed by both applications: Q ... you do believe that there is a need for both in the district. A I believe there's a need for two facilities in the district. Q It could support two facilities? A Oh, absolutely. Q And the disagreement primarily relates to the conservative approach of Kindred in terms of not factoring in out-migration and the narrowing the DRG categories? A Correct. ... Kindred actually had three models. Two of them support both facilities, but it's the GMLOS model that I typically rely on, and it didn't on the surface support both facilities. That's why I reconciled the two, and I believe that's the difference, is just the 50 DRGs and not including the out-migration. That would boost their need above the 130, and two facilities would give people alternatives, it would foster competition, and it would really improve access in that market. Tr. 150-51. Need for the applications, therefore, is the paramount issue in this case. Since both applicants are qualified to operate an LTCH in Florida, if need is proven for the 130 beds, then with the exception of Kindred's long term financial feasibility, all parties agree that there is no further issue: both applications should be granted. No Agency Numeric Need Methodology The Agency has not established a numeric need methodology for LTCH services. Consequently, it does not publish a fixed-need pool for LTCHs. Nor does the Agency have "any policy upon which to determine need for the proposed beds or service." See Fla. Admin. Code R. 59C-1.008(2)(e)1. Florida Administrative Code Rule 59C-1.008(2), which governs "Fixed Need Pools" (the "Fixed Need Pools Rule") states that if "no agency policy exist" with regard to a needs assessment methodology: [T]he applicant will be responsible for demonstrating need through a needs assessment methodology which must include, at a minimum, consideration of the following topics, except where they are inconsistent with the applicable statutory or rule criteria: Population demographics and dynamics; Availability, utilization and quality of like services in the district, subdistrict or both; Medical treatment trends; and Market conditions. Fla. Admin. Code R. 59C-1.008(2)(e)2. The Fixed Need Pools Rule goes on to elaborate in subparagraph (e)3 that "[t]he existence of unmet need will not be based solely on the absence of a health service, health care facility, or beds in the district, subdistrict, region or proposed service area." Population, Demographics and Dynamics The first of the four topics to be addressed when an applicant is responsible for demonstrating need through a needs assessment methodology is "population, demographics and dynamics." The Agency has not defined service areas for LTCHs. Nonetheless, from a health planning perspective, it views LTCH services as being provided district-wide primarily for Medicare patients. Consistent with the Agency's view, Select-Palm Beach identified the entire district, that is, all of AHCA District 9, as its service area. It identified Palm Beach County, one of the five counties in AHCA District 9, as its primary service area. In identifying the service area for Select-Palm Beach, Ms. Greenberg drew data from various sources: population estimates for Palm Beach County and surrounding areas; the number of acute care hospital beds in the area; the number of LTCH beds in the area; the types of patients treated at acute care hospitals; and the lengths of stay of the patients treated at those hospitals. AHCA District 9 has more elderly than any other district in the State, and Palm Beach County has more than any other county except for Dade. Palm Beach County residents comprise 71% of the District 9 population. It is reasonably projected that the elderly population (the "65 and over" age cohort) in Palm Beach County is projected to grow at the rate of 8 percent by 2008. The "65 and over" age cohort is significant because the members of that cohort are most likely to utilize hospital services, including LTCH services. Its members are most likely to suffer complications from illness and surgical procedures and more likely to have co-morbidity conditions that require long- term acute care. Persons over 65 years of age comprise approximately 80 percent of the patient population of LTCH facilities. Both Select-Palm Beach and Kindred project that approximately 80 percent of their admissions will come from Medicare patients. Since 90 percent of admissions to an LTCH come from acute care facilities, most of the patient days expected at Select-Palm Beach's proposed LTCH will originate from residents in its primary service area, Palm Beach County. When looking at the migration pattern for patients at acute care facilities within Palm Beach County, the majority (90 percent) come from Palm Beach County residents. Thus, Select- Palm Beach's projected primary service area is reasonable. Just as Select-Palm Beach, Kindred proposes to serve the entire District. Kindred proposes that its facility be based in Palm Beach County because of the percentage of the district's population in the county as well as because more than 70% of the district's general acute care hospitals are in the county. Its selection of the District as its service area, consistent with the Agency's view, is reasonable. Currently there are no LTCHs in District 9. Availability, Utilization and Quality of Like Services The second topic is "availability, utilization and quality of like services." There are no "like" services available to District residents in the District. Select-Palm Beach and Kindred, therefore, contend that they meet the criteria of the second topic. There are like services in other AHCA Districts. For example, AHCA District 10 has at total of 188 beds at two Kindred facilities in Fort Lauderdale and Hollywood. The Agency, however, did not present evidence of their quality, that they were available or to what extent they are utilized by the residents of AHCA District 9. Medical Treatment Trends The third topic is medical treatment trends. Caring for patients with chronic and long term care needs is becoming increasingly more important as the population ages and as medical technology continues to emerge that prolongs life expectancies. Through treatment provided the medically complex and critically ill with state of the art mechanical ventilators, metabolic analyzers, and breathing monitors, LTCHs meet needs beyond the capability of the typical general acute care hospitals. In this way, LTCHs fill a niche in the continuum of care that addresses the needs of a small but growing patient population. Treatment for these patients in an LTCH, who otherwise would be cared for without adequate reimbursement to the general acute care hospital or moved to an alternative setting with staff and services inadequate to meet their needs, is a medical trend. Market Conditions The fourth topic to be addressed by the applicant is market conditions. The federal government's development of a distinctive prospective payment system for LTCHs (LTC-DRG), has created a market condition favorable to LTCHs. General acute care hospitals face substantial losses for the medically complex patient who uses far greater resources than expected on the basis of individual diagnoses. Medicare covers between 80 and 85 percent of LTCH patients. The remaining patients are covered by private insurance, managed care and Medicaid. LTCH programs allow for shorter lengths of stay in a general acute care facility, reduces re-admissions and provide more discharges to home. These benefits are increasingly recognized. Numeric Need Analysis Kindred presented a set of needs assessment methodologies that yielded numeric need for the beds applied for by Kindred. Select-Palm Beach did the same. Unlike Kindred, however, all of the needs assessment methodologies presented by Select-Palm Beach demonstrated numeric need in excess of the 130 beds proposed by both applications. Select-Palm Beach's methodologies, overall, are superior to Kindred's. Select-Palm Beach used two sets of needs assessment methodologies and sensitivity testing of one of the sets that confirmed the methodology's reasonableness. The two sets or needs assessment methodologies are: (1) a use rate methodology and (2) length of stay methodologies. The use rate methodology yielded projected bed need for Palm Beach County alone in excess of the 130 beds proposed by the two applicants. For the year "7/05 - 6/06" the bed need is projected to be 256; for the year "7/06 - 6/07" the bed need is projected to be 261; and, for the year "7/07 - 6/08" the bed need is projected to be 266. See Select Ex. 1, Bates Stamp p. 000036 and the testimony of Ms. Greenberg at tr. 114. If the use rate analysis had been re-computed to include two districts whose data was excluded from the analysis, the bed need yielded for Palm Beach County alone was 175 beds, a numeric need still in excess of the 130 beds proposed by both applicants. The use rate methodology is reasonable.4 The length of stay methodologies are also reasonable. These two methodologies also yielded numeric need for beds in excess of the 130 beds proposed. The two methodologies yielded need for 167 beds and 250 beds. Agency Denial The Agency's general concerns about LTCHs are not without basis. For many years, there were almost no LTCH CON applications filed with the Agency. A change occurred in 2002. The change in the LTCH environment in the last few years put AHCA in the position of having "to adapt to a rapidly changing situation in terms of [Agency] understanding of what has been going on in recent years with long-term care hospitals." (Tr. 358.) "... [I]n the last couple of years long-term care hospital applications have become [AHCA's] most common type of application." (Tr. 359.) At the time of the upsurge in applications, there was "virtually nothing ... in the academic literature about long- term care hospitals ... that could [provide] ... an understanding of what was going on ... [nor was there anything] in the peer reviewed literature that addressed long-term care hospitals" id., and the health care planning issues that affected them. Two MedPAC reports came out, one in 2003 and another in 2004. The 2003 report conveyed the information that the federal government was unable to identify patients appropriate for LTCH services, services that are overwhelmingly Medicare funded, because of overlap of LTCH services with other types of services. The 2004 report gave an account of the federal government decision to change its payment policy for a type of long-term care hospitals that are known as "hospitals-within- hospitals" (tr. 368) so that "hospitals within hospitals as of this past summer [2004] can now only treat 25 percent of their patients from the host hospital." Id. Both reports roused concerns for AHCA. First, if appropriate LTCH patients cannot be identified and other types of services overlap appropriately with LTCH services, AHCA cannot produce a valid needs assessment methodology. The second produces another concern. In the words of Mr. Gregg, The problem ... with oversupply of long-term care hospital beds is that it creates an incentive for providers to seek patient who are less appropriate for the service. What we know now is that only the sickest patient ... with the most severe conditions are truly appropriate for long-term care hospital placement. * * * ... [T]he MedPAC report most recently shows us that the greatest indicator of utilization of long-term care hospital services is the mere availability of those services. Tr. 368-369. The MedPAC reports, themselves, although marked for identification, were not admitted into evidence. Objections to their admission (in particular, Kindred's) were sustained because they had not been listed by AHCA on the stipulation required by the Pre-hearing Order of Instructions.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that a final order be issued by the Agency for Health Care Administration that: approves Select-Palm Beach's application, CON 9661; and approves Kindred's application CON 9662 with the condition that financing of the project be provided by Kindred Healthcare. DONE AND ENTERED this 18th day of April, 2005, in Tallahassee, Leon County, Florida. S 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 18th day of April, 2005.
