The Issue The issue is whether the Agency for Health Care Administration should approve Petitioner's application for a Certificate of Need to establish a freestanding 44-bed long-term acute care hospital in Sarasota County.
Findings Of Fact Based upon the testimony and evidence received at the hearing and the parties' stipulations, the following findings are made: Parties Petitioner is a wholly-owned subsidiary of Select Medical Corporation (Select), which owns and operates 79 LTACHs in 24 states including a 40-bed LTACH in Miami-Dade County that was licensed in December 2002. The Agency is the state agency responsible for administering the CON process and licensing LTACHs and other hospital facilities. Petitioner’s Proposed LTACH In the first batching cycle of 2003 for “other beds and programs,” Petitioner timely filed an application for a CON to establish a freestanding 44-bed LTACH in Sarasota County. Sarasota County is located in District 8 for health planning purposes. The other counties in District 8 are DeSoto, Charlotte, Lee, Glades, Hendry, and Collier. Petitioner's proposed LTACH will be located in the city of Sarasota, which is in northern Sarasota County, close to the boundary between Sarasota and Manatee Counties. Petitioner projected in the application that its proposed LTACH would be operational by June 2005. The utilization projections in the application focused on the facility's third year of operation, which is the 12-month period ending June 2008. The specific mix of services to be provided at Petitioner’s proposed LTACH has not yet been determined; however, it is expected that the services will include the same "core" services found at other Select LTACHs. Those services are the treatment of pulmonary and ventilator patients, neuro- trauma and stroke patients, medically complex patients, and wound care. Petitioner’s facility will include a four-bed “high observation” unit in which the most unstable and highest acuity patients will be located. The nurse-to-patient ratio in that unit will be two-to-one, and the level of monitoring will be similar to that of an intensive care unit (ICU) in a general acute care hospital. Application Review and Denial Petitioner's application was designated CON 9657, and was reviewed along with the CON application filed by Petitioner for a 60-bed LTACH in Lee County. The Lee County application, CON 9656, is not at issue in this proceeding. On June 11, 2003, the Agency issued its State Agency Action Report (SAAR), which recommended denial of both CON applications filed by Petitioner. The primary basis for denial of the Sarasota County application described in the SAAR was Petitioner's failure to demonstrate a need for its proposed 44- bed LTACH. The parties stipulated that Petitioner's CON application satisfied all of the applicable statutory and rule criteria except those related to "need," and that the only issue to be determined in this proceeding is whether Petitioner established a need for its proposed facility.1 LTACHs, Generally LTACHs are health care facilities that provide extended medical and rehabilitative care to patients with multiple, chronic, and/or clinically complex acute medical conditions. To qualify as an LTACH, the facility must serve a patient population whose average length of stay exceeds 25 days. LTACH services are most highly utilized by persons in the 65 and older (“65+”) age cohort because those persons are more likely to have complex and/or co-morbid medical conditions that require long-term acute care. The typical LTACH patient is still in need of considerable acute care, but a traditional acute care hospital is no longer the most appropriate or lowest cost setting for that care. Most LTACH admissions are patients transferred from a traditional acute care hospital. It is not uncommon for an LTACH patient to be transferred directly from the hospital's critical care unit or ICU after the patient has been diagnosed and stabilized. Traditional post-acute care settings -- nursing homes, skilled nursing facilities (SNFs), skilled nursing units (SNUs), comprehensive medical rehabilitation (CMR) hospitals, and home health care -- are not appropriate for the typical LTACH patient because the patient's acuity level and medical/therapeutic needs are higher than those generally treated in those settings. Indeed, unlike traditional post-acute care settings which typically do not admit patients who still require acute care, the core patient-group served by LTACHs are patients who require considerable acute care through daily physician visits and intensive nursing care which can average as much as nine hours per day. LTACH patients are often discharged to a traditional post-acute care setting such as a nursing home, SNF, SNU, CMR, or home health care. Thus, those facilities cannot be considered as "substitutes" for LTACHs, even though there is overlap between the diagnoses and services provided to lower acuity LTACH patients and higher acuity patients in those traditional post-acute care settings. The federal government has recently developed a Medicare payment system specifically for LTACHs. That system recognizes the LTACH patient population as being distinct from the patient populations treated by traditional acute care hospitals and post-acute care providers such as nursing homes, SNFs, SNUs, and CMRs, even though there may be some overlap between the patient populations served by LTACHs and those other types of facilities. LTACH services are reimbursed by Medicare at a predetermined rate that is weighted based upon the patient's diagnosis and acuity, regardless of the cost of care. This reimbursement system is similar to, but uses Diagnosis Related Groups (DRGs) that are different than the DRGs used in the traditional acute care hospital setting. LTACHs fit into the continuum of care between traditional acute care hospitals and traditional post-acute care facilities. LTACHs are designed to serve patients that would otherwise have to be maintained in a traditional acute care hospital, where the standard reimbursement rates may be insufficient to cover the costs associated with a lengthy stay, or be moved to an alternative setting such as a nursing home, SNF, SNU, or CMR, where the patient may not receive the level of curative care needed. The recently-adopted, LTACH-specific system for Medicare reimbursements is expected to enhance the status of LTACHs as part of the continuum of care. LTACHs in Florida Currently, there are only nine LTACHs operating in Florida with a total of 683 licensed beds. The facilities are concentrated in six counties, Dade, Broward, Hillsborough, Pinellas, Duval, and Clay. There are an additional 182 beds which have been approved by the Agency but which are not yet operational. Those beds include a new 40-bed facility in Sarasota County (discussed below) and an additional 22 beds at the existing 60-bed Pinellas County facility, which is in the health planning district (District 5) immediately to the north of District 8. The Pinellas County facility is located in St. Petersburg, which is approximately 25 to 30 miles north of Petitioner’s proposed facility. The Florida LTACH facilities are well utilized. The occupancy rates at the facilities range from 54.6 percent to 99.2 percent. Four of the nine facilities have occupancy rates higher than 80 percent, and the average occupancy rate for all of the facilities is 76.6 percent. The average length of stay for all patients discharged from Florida LTACHs between July 2001 and June 2002 was 42.2 days. The 65+ age cohort accounted for approximately 77 percent of the LTACH discharges in Florida for that period. Relevant Demographics of Sarasota County The 2003 population of Sarasota County was 343,966, which was 25.8 percent of the total District 8 population. In 2008, which is the third year of operation for Petitioner's proposed LTACH, the population of Sarasota County is projected to increase by 6.2 percent to 365,439. Over that same period, the population of District 8 as a whole is projected to increase by 10.4 percent. The 65+ age cohort, which is the group most likely to utilize LTACH services and the group that utilizes LTACH services at the highest rate, represents 31.2 percent of Sarasota County's 2003 population and 31.5 percent of the county's projected 2008 population. By contrast, in 2003 the District 8 average for the 65+ age cohort was 26.9 percent and the statewide average was 17.5 percent. Sarasota County has a higher percentage of persons in the 65+ age cohort than do any of the counties that currently have an LTACH facility. Pinellas County, with 22 percent of its population in the 65+ age cohort (and 82 licensed and approved LTACH beds), has the highest rate of the counties with LTACHs. There are four acute care hospitals in Sarasota County, two of which -- Sarasota Memorial Hospital and Doctors Hospital of Sarasota -- are located in the city of Sarasota in close proximity to the proposed location of Petitioner's LTACH. The other two hospitals in Sarasota County -- Bon Secours Venice Hospital and Englewood Community Hospital -- are located in the southern part of the county and are 16 miles and 28 miles, respectively, from the proposed location of Petitioner's LTACH. In the CON application, Petitioner stated that the four hospitals in Sarasota County would "provide a solid base of patients" for the proposed LTACH. The application further stated that patients would also likely come from three hospitals in Charlotte County -- Charlotte Regional Medical Center, Fawcett Memorial Hospital, and Bon Secours St. Joseph Hospital - - and one hospital in DeSoto County -- Desoto Memorial Hospital -- even though the Charlotte County hospitals are almost 40 miles from the proposed site of Petitioner's LTACH and the DeSoto County hospital is more than 40 miles from the proposed site. The record does not reflect how many total acute care beds are in these hospitals, nor does it reflect whether any of the hospitals are trauma centers or whether they have any specialty programs that might impact (either positively or negatively) the potential LTACH patient pool for Petitioner's proposed facility. Approved LTACH in District 8 There are no LTACHs currently operating in Sarasota County or District 8. HealthSouth received a CON in October 2002 to establish a freestanding 40-bed LTACH in Sarasota County, but that facility has not yet opened. HealthSouth is behind schedule in the development of its LTACH. If HealthSouth does not "break ground" on its LTACH by April 2004, its CON will expire; however, as of the date of the hearing, HealthSouth's CON was still valid. The Agency expressed a concern in the SAAR that "the ultimate development of the HealthSouth LTCH [sic] in District 8 is uncertain" based upon legal and financial problems at HealthSouth. However, as of the date of the hearing, the Agency had not received any formal indication from HealthSouth that it is not going forward with the development of its Sarasota County LTACH. HealthSouth did not seek to intervene in this proceeding. Numeric Need for Petitioner’s Proposed LTACH Petitioner has the burden to demonstrate "need" for its proposed LTACH. The Agency does not publish a fixed-need pool for LTACHs, nor is there an Agency rule which provides a specific formula or methodology for determining numeric need for an LTACH.2 HealthSouth's 40 approved, but not yet operational LTACH beds must be factored into the analysis of need for any additional LTACH beds in District 8. Accordingly, it is necessary for Petitioner to demonstrate a numeric need for at least 84 LTACH beds for its application to be granted. The application states that “the primary service area [for Petitioner’s proposed LTACH] is Sarasota County and the broader service area includes portions of Charlotte County and DeSoto County . . . .” This service area encompasses an approximately 40-mile radius around the site of the proposed facility, and includes the eight acute care hospitals referenced above. In contrast to the application’s description of the service area, Petitioner’s expert witness in the area of LTACH development, Greg Sasserman, testified that the “actual” service area for Petitioner’s proposed LTACH would be a 10 to 20-mile radius around the facility. That distance is a more reasonable estimate of the distance that patients would likely travel for LTACH services. In its application, Petitioner attempted to demonstrate numerical need for the proposed facility under two distinct methodologies, one based upon "use rate" and another based upon "length of stay." “Use Rate” Methodology Petitioner’s "use rate" methodology projected the number of LTACH patient days that will likely be generated by Sarasota County residents based upon the utilization rates of LTACH services by the residents of the counties in which LTACH facilities are currently located. The utilization rates for the existing facilities were calculated by age cohort and were shown as a number of patient days per 1,000 persons in each age cohort. Those rates were then applied to the projected population of Sarasota County in 2008 in each age cohort in order to calculate a projected number of patient days that will be generated by Sarasota County residents in that year. Those patient days were then "grossed up” by an occupancy standard of 80 percent and then "grossed up" by an additional 44.5 percent to account for patients coming to the facility from outside of Sarasota County. The utilization rate calculated under this methodology is not a true “statewide” rate. The existing LTACHs are concentrated in only six of the states 67 counties, and more significantly, Petitioner excluded the facilities in Miami-Dade and Pinellas Counties from its calculations because their utilization rates were, according to Petitioner, “misleadingly conservative.” The 44.5 percent out-of-county rate was purportedly derived from the experience of the other LTACHs operating in Florida. However, the record does not include the raw data upon which that rate was calculated, and it does not reflect whether the rate includes the two facilities excluded from the calculation of the “statewide” utilization rate or the distances from which out-of-county patients are drawn to the facilities. Nor can the 44.5 percent rate be squared with the calculations of potential LTACH discharges from the eight area hospitals as part of the “length of stay” methodology (discussed below), which reflect only 24.7 to 26.8 percent of the patients coming from hospitals outside of Sarasota County.3 Petitioner's calculations produced an estimate of 29,654 LTACH patient days generated by Sarasota County residents in 2008, which translated into an average daily census (ADC) of 81 patients and a need for 101 LTACH beds; and an estimate of 53,431 LTACH patient days, which translated into an ADC of 146 patients and a need for 182 LTACH beds when the out-of-county residents were taken into consideration. Use rate methodologies are commonly used by health planners to project need for acute care beds and other types of services. However, in the LTACH context, a use rate methodology is not necessarily a reliable indicator of bed need because the existing LTACHs are not evenly distributed statewide and the utilization rates for the existing LTACHs vary significantly. The unreliability of Petitioner’s “use rate” analysis is further demonstrated by the fact that Petitioner excluded two of the existing facilities in the calculation of its “statewide” utilization rate. If the utilization rates of those facilities were included, the number of patient days and bed need projected by Petitioner would have been lower. “Length of Stay” Methodology Petitioner’s "length of stay" methodology projected bed need based upon an analysis of the discharges from the eight District 8 hospitals identified above. More specifically, the analysis focused on the discharges that Petitioner considered to be “appropriate” LTACH admissions based upon the nature of the patient’s diagnosis and the length of the patient’s stay at the hospital. Open heart surgery DRGs were included in the analysis, and DRGs “for people age 0 to 17, obstetric and gynecological care, newborns, alcohol and drug abuse, rehabilitation and psychoses” were excluded from the analysis. The application also makes various references to LTACH-appropriate diagnoses by Major Diagnostic Category (MDC) and "program area"; however, the specific discharges identified by Petitioner as being potential LTACH patients from the eight hospitals are not broken down by DRG in the application. Petitioner used two approaches to determine whether the patient is an “appropriate” LTACH patient from a length of stay perspective. Both approaches estimate the number of days that patients who otherwise would have remained in and been discharged from an acute care hospital would have likely spent at an LTACH, if one was available The first approach, which was characterized in the application and at hearing as the more “conservative” measure, only considered patients whose length of stay at the acute care hospital was at least 15 days longer than the geometric mean length of stay (GMLOS) for the patient's DRG (hereafter “the GMLOS plus 15 methodology.)” The estimated number of patient days produced by the GMLOS plus 15 methodology is the sum of the patients' actual lengths of stay less the GMLOS, which represents the number of days that the patients would likely stay in the LTACH facility. The second approach, which was characterized in the application as the more “aggressive” measure, considered all patients whose length of stay was more than 15 days (hereafter “the LOS plus 15 methodology”). The estimated number of patient days produced by the LOS plus 15 methodology is the sum of the patients' actual lengths of stay less 15 days, which again reflects the number of days that the patients would likely stay in an LTACH facility. The GMLOS is a statistically-adjusted value for all cases within a DRG that takes into account “outlier” cases,4 transfer cases, and other cases that could skew an arithmetic average length of stay. The GMLOS is calculated by the federal government. The only difference in the two approaches is that the GMLOS plus 15 methodology includes only those patients with considerably longer lengths of stay than expected for their diagnoses (i.e., 15 days in excess of the GMLOS for the applicable DRG), whereas the LOS plus 15 methodology includes all patients with long lengths of stay (i.e., in excess of 15 days) irrespective of their diagnoses. Patients who, because of co-morbidities, otherwise complex medical conditions, or frailties due to age, have lengths of stay in excess of the GMLOS plus 15 days are generally appropriate LTACH patients, particularly if the patient would otherwise remain in the ICU of the acute care hospital. In such circumstances, an LTACH would likely be the more appropriate setting for the patient from both a financial and patient-care standpoint. The GMLOS plus 15 methodology resulted in an estimated 13,263 LTACH patient days, which translates into an ADC of 36.3 patients and a need of 45 LTACH beds based upon an 80 percent occupancy standard. The LOS plus 15 methodology resulted in an estimated 21,753 LTACH patient days, which translates into an ADC of 59.6 patients and a need for 74 LTACH beds based upon an 80 percent occupancy standard. The patient days computed through the GMLOS plus 15 methodology and the LOS plus 15 methodology were characterized in the application and at the hearing as the lower and upper ends, respectively, of the projected LTACH patient days in the area to be served by Petitioner’s proposed LTACH. The mid-point of that range, 17,508 patient days, was then broken out by age cohort and was used to compute “hospital specific” utilization rates by age cohort. Those “hospital-specific” utilization rates were then multiplied by the projected future population of the respective age cohorts in the area to be served by Petitioner’s LTACH – Sarasota County and one-half of the population of Charlotte County – to project the total number of LTACH beds needed in 2008. No adjustment was made for out-of-county admissions because the hospitals included in both of the length-of-stay methodologies already included projected admissions from out-of- county hospitals. The end-result of the mid-point analysis and, hence, the end-result of Petitioner’s “length of stay” methodology was a projected need for 67 LTACH beds in 2008. Under the circumstances of this case, the GMLOS plus 15 methodology provides a more reasonable projection of LTACH patient days than does the LOS plus 15 methodology or the mid- point analysis. Specifically, the LOS plus 15 methodology is based upon the premise that physicians would be more likely to transfer their patients who would otherwise require long hospital lengths of stays to an LTACH “as soon as possible in their treatment regiment when LTAC [sic] beds are available,” but the record is devoid of competent evidence, such as letters or testimony from local physicians, that would provide support for that premise. Both of the “length of stay” methodologies appear to assume a 100 percent capture rate of the LTACH-appropriate patients by Petitioner’s proposed facility. The record is devoid of competent evidence demonstrating the reasonableness of that assumption, either with or without the HealthSouth facility in place. For example, the record does not include any tentative transfer agreements or other documentation that demonstrates a willingness of the local hospitals to transfer patients to Petitioner’s LTACH if it is constructed.5 Furthermore, based upon Mr. Sasserman’s definition of the service area of Petitioner’s proposed LTACH, it was not reasonable to include the patient days generated by discharges from five of the eight hospitals used by Petitioner in its “length of stay” methodologies, since those hospitals are outside of the 10 to 20-mile radius identified by Mr. Sasserman. Finally, there is no basis in the record to conclude that any overstatement of the bed need resulting from the inclusion of hospitals outside of the service area as defined by Mr. Sasserman would be offset by referrals from Manatee Memorial Hospital, which is located in District 5 approximately 10 miles north of the proposed site for Petitioner’s LTACH. The testimony on this point by Mr. Sasserman and Petitioner's Health Planner is pure speculation. Ultimate Findings Regarding Numeric “Need” The bed need projected by Petitioner through its “use rate” methodology is not reliable because of the significant shortcomings in that methodology described above. Of the two measures used by Petitioner as part of its “length of stay” methodology, the GMLOS plus 15 methodology is more reasonable than the LOS plus 15 methodology; however, neither methodology resulted in a projected bed need that is sufficient to account for HealthSouth’s 40 approved beds and Petitioner’s 44 proposed beds.
Recommendation Based upon the foregoing findings of fact and conclusions of law, it is RECOMMENDED that the Agency for Health Care Administration issue a final order denying Petitioner’s application for a Certificate of Need to establish a 44-bed LTACH in Sarasota County. DONE AND ENTERED this 15th day of March, 2004, in Tallahassee, Leon County, Florida. S T. KENT WETHERELL, II 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 15th day of March, 2004.
The Issue Whether Rule 59C-1.038, the acute care bed need rule, is an invalid exercise of delegated legislative authority. Whether this rule challenge should be dismissed as an untimely attack on a published fixed need pool.
Findings Of Fact In August 1994, the Agency for Health Care Administration ("AHCA") published a numeric need of zero for additional acute care beds in AHCA District 9, Subdistrict 5, for southern Palm Beach County. Pursuant to Subsection 408.034(3), Florida Statutes, AHCA is the state agency responsible for establishing, by rule, uniform need methodologies for health services and facilities. In September 1994, NME Hospitals, Inc. d/b/a Delray Community Hospital, Inc. ("Delray") applied for a certificate of need ("CON") to add 24 acute care beds for a total construction cost of $4,608,260. AHCA published its intent to approve the application on January 20, 1995, in Volume 21, No. 3 of the Florida Administrative Weekly. By timely filing a petition, Bethesda Memorial Hospital, Inc. ("Bethesda") challenged AHCA's preliminary decision in DOAH Case No. 95-0730. Bethesda is also located in AHCA District 9, Subdistrict 5. On May 24, 1995, Bethesda also filed the petition in this case challenging Rule 59C-1.038, Florida Administrative Code, the acute care bed need rule. Pursuant to the acute care bed need rule, AHCA's August 1994 notice published its finding that zero additional acute care beds will be needed in the southern Palm Beach County subdistrict by July, 1999. The data, formulas, and calculations used in arriving at the number zero were not published. AHCA and Delray argue that the publication put persons on notice to inquire into the population data, occupancy rates, or the calculations leading to the published need number. An AHCA rule bars a person from seeking, and AHCA from making, any adjustments to the fixed need pool number if the person failed to notify AHCA of errors within ten days of publication. Still another rule defines "fixed need pool" as the " . . . numerical number, as published. " Bethesda is not contesting and, in fact, agrees that the fixed need pool number as published, zero, is correct. Using AHCA's definition of the fixed need pool, Bethesda's challenge is not barred because it failed to notify AHCA of an error in the fixed need pool number within 10 days of publication. Bethesda is challenging as irrational and invalid subsections (5), (6), and (7) of the acute care bed need rule. Subsection (5) directs the local health councils to determine subdistrict bed need consistent with the methodology for determining district bed need. Under that provision, total projected patient days of acute care needed in a district is calculated by adding together the projected patient care days needed in medical/surgical, intensive care, coronary care, obstetric, and pediatric beds. Each of these separate bed need projections is computed, in general, by multiplying projected population in the district for the appropriate age or gender group times a factor which is the product of the statewide discharge rate and the average length of stay for that particular type of care. After the total projected acute care patient days for district residents is computed, the number is adjusted to reflect historical patient flow patterns for acute care services, for out-of-state residents served in the district, for residents of other districts served in the district, and for residents of the district served outside the district. The rule includes specific percentages to apply for each patient flow group for each of the eleven districts. After the total number of beds needed in the district is derived, that number is decreased by the number of existing licensed or approved beds to get the number of additional acute care beds needed in the district, if any. Bethesda is challenging subsections (7)(a), (b), and (c) of the acute care rule, which authorize adjustments to the calculations from subsections (5) and (6) to achieve desired occupancy levels, based on historic utilization of acute care beds in a district. Bethesda is also seeking a determination that subsections 7(d) and (e) are invalid. Each of those subsections of the rule refer to (5)(b), although AHCA's expert witness testified that they should refer to (6)(b). Subsection (7)(d) requires at least 75 percent occupancy in all hospitals in the district before new acute care beds normally are approved, regardless of the net need projected by the formulas. Subsection (7)(e) allows approvals under special circumstances if net need is projected by the formulas and the applicant facility's occupancy rate equals or exceeds 75 percent. Subsection (7)(e), the provision directly related to the Delray application, is as follows: (e) 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 a net need for beds is shown based on the formula described in paragraphs (5)(b), (7)(a), (b), (c), and (8)(a), (b), (c), and provided that the hospital's average occupancy rate for all licensed acute care beds is at or exceeds 75 percent. The determination of the average occupancy rate shall be made based on the average 12 months occupancy rate made available by the local health council two months prior to the beginning of the respective acute care hospital batching cycle. Phillip C. Rond, III, Ph.D., was the primary architect of the rule, beginning in 1981. The rule was initially published in 1982, and adopted in 1983. Constants in the rule formulas, including use rates, average lengths of stay, occupancy standards and patient flow patterns were taken from a 1979 survey of some state hospitals. Because data used for the constants in the formulas was expected to change, subsection (6) also provides, in pertinent part, that: Periodic updating of the statewide discharge rates, average lengths of stay and patient flow factors will be done as data becomes available through the institution of statewide utilization reporting mechanisms. Patient flow factors were updated in March 1984 to reflect a change in the realignment of counties in Districts 5 and 8. No other constants have been updated since the rule was adopted in June 1983. More current data is available. The Hospital Cost Containment Board ("HCCB") began collecting statewide hospital inpatient discharge data in the fourth quarter of 1987, which became available by the fall of 1988. AHCA now collects the data. Using the rule, the projected net need for acute care beds in 1999 in District 9 is 1,442 additional beds. By contrast, with the factors updated by Dr. Rond, the projected net need is a negative 723 or, in other words, District 9 has 723 more acute care beds than it will need in 1999. The updated formulas show a need for a total of 3,676 beds in District 9, which already has 4,399 licensed or approved acute care beds. Since 1983, hospital utilization has declined in both rates of admissions or discharges, and in average lengths of stay. Although the occupancy goals in the rule are 75 to 80 percent, depending on the type of hospital service, the occupancy rate achieved by using the number of beds projected by the rule methodology is 45 to 52 percent. The statewide occupancy rate in acute beds is approximately 50 percent in 49,215 licensed beds. The formulas in the rule show a statewide net need for 6,000 more beds in 1999, but updated constants in the same formulas result in a total statewide need for approximately 36,000 acute care beds in 1999, or 13,000 fewer beds than currently exist. Statewide utilization of acute care hospital beds declined from 1187.2 days per 1000 population in 1983 to 730.5 days per 1,000 in 1993, despite increases in the percentage of the elderly population. By 1987, AHCA's predecessor realized that the need methodology in the rule was grossly overestimating need and inconsistent with its health planning objectives. Subsection (7)(d) was added to the rule to avoid having a published fixed need based on the outdated methodology in subsections (5), (6) and 7(a)- (c). The occupancy data is also, as the 1987 amendment requires, that reported for the most recent 12 months, available 2 months before the scheduled application cycle. In August 1994, AHCA published a numeric need of zero for District 9, Subdistrict 5, rather than 1,442, the calculated net need predicted by the formulas in the rule, because all subdistrict hospital occupancy rates did not equal or exceed 75 percent. Elfie Stamm of AHCA, who is responsible for the publication of fixed need pools, confirmed that the 1987 amendment to the rule was an efficient and cost-effective way to avoid publishing need where there was no actual need. She confirmed Dr. Rond's conclusions that the formulas are no longer valid and produce excessive need numbers, as in projecting a need for 6,000 or 7,000 more acute care beds in the state. She also confirmed that none of the constants in the formula have been updated as required by subsection 6. Ms. Stamm claims that the information needed to update the formulas cannot be obtained easily from any statewide utilization reporting mechanism. One problem, according to Ms. Stamm, is the possibility of including patients in acute care beds with comprehensive rehabilitation, psychiatric, or substance abuse problems, although it is not lawful for acute care providers to place patients with these primary diagnoses in licensed acute care beds and all data bases have some miscoding of diagnoses. She also testified that some factors required in the formulas are not included in HCCB data base. In addition, she testified that AHCA is in the process of filing a notice to repeal the acute care bed need rule. The filing of the notice of repeal, published in Volume 21, Florida Administrative Week, pp. 4179-4180 (6/23/95) was confirmed by Bethesda's Request For Official Recognition, which was filed on July 20, 1995, and is granted. Ms. Stamm also noted that rules for other need-based health services have facility-specific special circumstances provisions, which are not tied to numerical need, otherwise the special circumstances are not really facility- specific. Need rules make no sense, according to Ms. Stamm, without an exception in the absence of a determination of need. Subsection (7)(e) of the acute care rule requires a finding of numeric need and a 75 percent occupancy rate at the applicant facility. Ms. Stamm's records indicate that AHCA's predecessor adopted the facility-specific provisions tied to net need at the same time it adopted the 75 percent average district occupancy standard to overcome the problems with the net need formula. AHCA asserts that the admittedly irrational need methodology when combined with the 1987 amendment achieves a rational result. Because the need methodology always over estimates numeric need, facilities exceeding 75 percent occupancy have an opportunity to demonstrate special circumstances. Daniel Sullivan, Delray's expert, also testified that problems exist in extracting acute care bed from specialty bed utilization data, in hospitals which have both. He also agreed with Ms. Stamm that the 1987 amendment corrects the erroneous projections of the formula to give a rational outcome from the rule as a whole when not all hospitals in a subdistrict equal or exceed 75 percent occupancy and when one hospital, over 75 percent occupancy, attempts to establish a special circumstance, despite the fact that the need methodology itself is always wrong in projecting numeric need. Ms. Stamm testified that one district is approaching 75 percent occupancy in all hospitals. Mr. Sullivan testified that, if and when that occurs, then the formula is intended to, but does not, reflect the number of additional beds needed. An alternative methodology is required to determine bed need. AHCA, with its responsibility for the data base formerly collected by the HCCB, receives discharge data and financial worksheets from every hospital in the state. The claim that AHCA cannot update the formulas because its data may be unreliable is rejected as not credible. The data now available is more reliable than the 1979 data used in developing the rule, which was not collected from a formalized statewide reporting system, but from a sample of hospitals. The claim that AHCA cannot use its data base from mandatory statewide reporting mechanisms to extract the data needed to update the formulas is also rejected. The rule contemplated ". . .the institution of statewide utilization reporting mechanisms." Dr. Rond's work to update the formulas before the final hearing began on May 23, 1995. Dr. Rond used a total of approximately 1.5 million acute care discharges from the AHCA (formerly, HCCB) data base for the 1992 calendar year. At the time of the final hearing, Dr. Rond had not separated days of care for medical/surgical, intensive and coronary care. The data can be taken from hospital financial data, including detailed budget worksheets which are submitted to AHCA. Separate data are anticipated in the formula because the computation of need for the different bed categories is based on different occupancy goals. For medical/surgical and intensive care beds, the goal is 80 percent occupancy, but it is 75 percent for coronary care for persons age 0 to 64. For persons 65 and older, the rule applies a combined occupancy standard of 79.7 percent for all three bed categories, which assumes that approximately 4 percent of the combined days of older patients will be spent in coronary care. Dr. Rond reasonably applied the 79.7 percent occupancy standard to the combined days for persons under 65, in arriving at the total district bed need for 3,676 beds. To check these results and to assume a worse case scenario of all patient days attributable to coronary care beds, for which more beds are needed to maintain a lower occupancy, Dr. Rond worked the formula using 75 percent occupancy as the goal for medical/surgical, intensive and care coronary care beds combined. Although the base number increased by 100, the calculations and adjustments in the rule yielded the same number of total acute care beds needed in the district, 3,676. That reliably confirms that the maximum number of acute care beds needed in District 9 is 3,676 by 1999. AHCA could use its data base to update formulas and achieve rational results in the rule by using the hospital financial data to distinguish coronary care days for patients 0-64 to include in the formula, or by using a rational blended occupancy standard in a rule amending the existing methodology. AHCA demonstrated that the 1987 amendment overrides the exaggerated numeric need number to yield a rational published fixed need pool in the absence of 75 percent occupancy in all acute care beds in a subdistrict. AHCA also demonstrated that because the projected need is always excessive under the formula, hospitals are allowed to demonstrate special circumstances, although it is absurd to include a requirement of numeric need in a provision for special circumstances. AHCA's claim that the excessive need projection is, therefore, irrelevant is rejected. Net need under the rule formula fails to give any rational indication of the number of beds needed when all hospitals in a subdistrict reach 75 percent occupancy.
The Issue This case concerns four Certificate of Need ("CON") applications ("CONs 9891, 9992, 9893, and 9894") that seek to establish long-term acute care hospitals ("LTCHs") in Miami-Dade County (the "County" or "Miami-Dade County"), a part of AHCA District 11 (along with Monroe County). Promise Healthcare of Florida XI, Inc. ("Promise") in CON 9891, Select Specialty Hospital-Dade, Inc. ("Select-Dade") in CON 9892, and Kindred Hospitals East, L.L.C. ("Kindred"), in CON 9894, seek to construct and operate a 60-bed freestanding LTCH in the County. Miami Jewish Home and Hospital for the Aged, Inc. ("MJH"), in CON 9893, seeks to establish a 30-bed hospital within a hospital ("HIH") on its existing campus in the County. In its State Agency Action Report (the "SAAR"), AHCA concluded that all of the need methodologies presented by the applicants were unreliable. Accordingly, AHCA staff recommended denial of the four applications. The recommendation was adopted by the Agency when it issued the SAAR. The Agency maintained throughout the final hearing that all four applications should be denied, although of the four, if any were to be granted, it professed a preference for MJH on the basis, among other reasons, of a more reliable need methodology. Since the hearing the Agency has changed its position with regard to MJH. In its proposed recommended order, AHCA supports approval of MJH's application. MJH and Promise agree with the AHCA that there is need for the 30 LTCH beds proposed by MJH for its HIH and that MJH otherwise meets the criteria for approval of its application. MJH seeks approval of its application only. Likewise, the Agency supports approval of only MJH's application. Promise, on the other hand, contends that there is need for a 60-bed facility as well as MJH's HIH and that between Promise, Select- Dade and Kindred, based on comparative review, its application should be approved along with MJH's application. Although Promise's need methodology supports need for more LTCH beds than would be provided by approval of its application and MJH's, its support for approval is limited to its application and that of MJH. Like Promise's methodology, Select-Dade and Kindred's need methodologies project need for many more beds than would be provided by the 60 beds each of them seek. Unlike Promise, however, neither Select-Dade nor Kindred supports approval of MJH's application. Each proposes its application to be superior to the other applications; each advocates approval of its respective application alone. Given the positions of the parties reflected in their proposed recommended orders, whether there is need for at least an additional 30 LTCH beds in District 11 is not at issue. Rather, the issues are as follows. What is the extent of the need for additional LTCH beds in District 11? If the need is for at least 30 beds but less than 60 beds, does MJH meet the criteria for approval of its application? If the need is for 60 beds or more, what application or applications should be approved depends on what applications meet CON review criteria and on the number of beds needed (60 but less than 90, 90 but less than 120, 120 but less than 150, 150 but less than 180, 180 but less than 210, and 210 or more) and whether there is health- planning basis not to grant an application even if the approval would meet a bed need and all four applicants otherwise meet review criteria. Finally, based on comparative review, what is the order of approval among the applications that meet CON need criteria? Ultimately, the issue in the case is which if any of the four applications should be approved?
Findings Of Fact The Parties "[D]esignated as the state health planning agency for purposes of federal law," Section 408.034(1), Florida Statutes, AHCA is responsible for the administration of the CON program and laws in Florida. See §§ 408.031, Fla. Stat., et seq. As such, it is also designated as "the single state agency to issue, revoke, or deny certificates of need . . . in accordance with present and future federal and state statutes." § 408.034(1), Fla. Stat. Promise Healthcare of Florida XI, Inc. ("Promise") is a wholly-owned subsidiary of Promise Healthcare, Inc. The applicant for CON 9891, Promise proposes the construction of a 60-bed freestanding LTCH to be located in Miami-Dade County, Florida. Select-Dade, the applicant for CON 9892, proposes the construction of a 60-bed freestanding LTCH to be located in Miami-Dade County, Florida. It is a wholly-owned subsidiary of Select Medical Corporation ("SMC"). The largest operator of LTCHs in the country, SMC operates 96 LTCHs in 24 states. The Miami Jewish Home and Hospital for the Aged is an existing not-for-profit provider of comprehensive health and social services in Miami-Dade County. The applicant for CON 9893, MJH proposes the creation of a 30-bed hospital within a hospital (HIH) LTCH by the renovation of a former acute care hospital building on its existing campus in Miami-Dade County, Florida. Kindred is the applicant for CON 9894 and proposes the construction of a 60-bed freestanding LTCH to be located in Miami-Dade County, Florida. Kindred is a wholly-owned subsidiary of Kindred Healthcare, Inc. ("Kindred Healthcare"). Kindred Healthcare operates 85 LTCHs in the country, eight of which are in the State of Florida. One of the eight is in Miami-Dade County. Twenty-three of Kindred Healthcare's LTCHs are operated by Kindred as well as seven of the eight Florida LTCHs. Kindred has also received CON approval for another LTCH in Florida. It is to be located in Palm Beach County in LTCH District 9. The District and its LTCHs Miami-Dade and Monroe Counties comprise AHCA District The population of Monroe County is 80,000 and of Miami-Dade County, 2.4 million. As to be expected from the population's distribution in the District, the vast majority of the District's health services are located in Miami-Dade County. The greater part of the County's population is in the eastern portion of Miami-Dade County, with population densities there 3-4 times higher than in the western portion of the County. But there is little to no space remaining for development in the eastern portion of the County. Miami-Dade County has an urban development boundary that shields the Everglades from development in the western portion of the County. Still, the bulk of population growth that has occurred recently is in the west and that trend is expected to continue. While the growth rate on a percentage basis is higher in the more-recently developed western areas of the County, the great majority of the population is and will continue to be within five miles of the sea coast on the County's eastern edge. At the time of hearing, there were three LTCHs operating in the District with a total of 122 beds: Kindred- Coral Gables, Select-Miami, and Sister Emmanuel. All three are clustered within a radius of six miles of each other in or not far from downtown Miami. The three existing LTCHs in the District are utilized at high occupancy levels. Kindred's 53-bed facility receives most of its referrals from a within a 10 mile radius. It has operated for the 11-year period beginning in 1995 with an occupancy level from a low of 82.08 percent to a high of 92.86 percent. The occupancy levels for 2004 (82.08 percent) and 2005 (84.90 percent) show occupancy recently at a relatively stable level within the range of optimal functional capacity which tends to be between 80 and 85 percent when facilities are equipped with semi-private rooms. With gender and infection issues in a facility with semi-private rooms, admissions to those facilities are usually restricted above 85 percent. Select operates a 40-bed LTCH on one floor of a health care service condominium building in downtown Miami. It began operation in 2003 as part of legislatively-created special Medicaid demonstration project. Its occupancy levels for the two calendar years of 2004 and 2005 were 83.39 percent and 95.10 percent. Sister Emmanuel Hospital for Continuing Care ("Sister Emmanuel") is a 29-bed HIH located at Mercy Hospital in Miami. It became operational in 2004 with an occupancy level of 82.64 percent, and attained an occupancy level of 85.46 percent in 2005. Kindred's Broward County LTCHs Kindred operates two LTCHs in Broward County (outside of District 11); one is in Ft. Lauderdale, the other in Hollywood. From 1995 to 2003, Kindred-Hollywood's occupancy rate ranged from a low of 65.17 percent to a high of 72.73 percent, generally lower than the state-wide occupancy rate. For the same period, Kindred-Ft. Lauderdale's rate was significantly higher, between 83.69 percent and 91.65 percent. Both LTCHs have experienced occupancy rates significantly lower than the state-wide rates in 2004 and 2005. Kindred-Ft. Lauderdale's occupancy in 2004 fell substantially from earlier years to 66.41 percent and then even farther in 2005 to 57.73 percent. Kindred-Hollywood's rates for these two years were also well below the state's at 59.74 percent and 58.04 percent, respectively. Historically used by residents of District 11, the Hollywood facility served 4,292 patients from Miami-Dade County in the eleven year period from 1995 through 2005. For the same period, the Ft. Lauderdale facility served 275 Miami-Dade residents. Kindred assigns its clinical liaisons to hospitals in a territorial manner to minimize competition for referrals between its two facilities in Broward County and Kindred-Coral Gables. LTCHs A "Long-term care hospital" means a general hospital licensed under Chapter 395, which meets the requirements of 42 C.F.R. Section 412.23(e) and seeks exclusion from the acute care Medicare prospective payment system for inpatient hospital services. § 408.032(13), Fla. Stat. (2005), and Fla. Admin. Code R. 59C-1.002(28). Under federal rules, an LTCH must have an average Medicare length of stay (LOS) greater than 25 days. LTCHs typically furnish extended medical and rehabilitation care for patients who are clinically complex and have multiple acute or chronic conditions. Patients appropriate for LTCH services represent a small but discrete sub-set of all patients. They are differentiated from other hospital patients in that, by definition, they have multiple co-morbidities that require concurrent treatment. Patients appropriate for LTCH services tend to be elderly, frail, and medically complex and are usually regarded as catastrophically ill although some are young, typically victims of severe trauma. Approximately 85 percent of LTCH patients qualify for Medicare. Generally, Medicare patients admitted to LTCHs have been transferred from general acute care hospitals and receive a range of services at LTCHs, including cardiac monitoring, ventilator support and wound care. In 2004, statewide, 92 percent of LTCH patients were transferred from short-term acute care hospitals. That figure was 98 percent for District 11 during the same period of time. The single most common factor associated with the use of long-term care hospitals are patients who have pulmonary and respiratory conditions such as tracheotomies, and require the use of ventilators. There are three other general categories of LTCH patients as explained by Dr. Muldoon in his deposition: The second group is wound care where patients who are at the extreme end of complexity in wound care would come to [an] LTCH if their wounds cannot be managed by nurses in skilled nursing facilities or by home health care. The third category would be cardiovascular diseases where patients compromise[d by] injury or illness related to the circulatory system would come [to an LTCH.] And the fourth is the severe end of the rehabilitation group where, in addition to rehabilitation needs, there's a background of multiple medical conditions that also require active management. (Kindred Ex. 8 at 10-11). Effective October 1, 2002, the federal Centers for Medicare and Medicaid Services ("CMS") established a new prospective payment system for long term care hospital providers. Through this system, CMS recognizes the patient population of LTCHs as separate and distinct from the populations treated by short-term acute care hospitals and by other post acute care providers, such as Skilled Nursing Facilities ("SNFs") and Comprehensive Rehabilitation Hospitals ("CMRs"). The implementation by CMS of categories of payment designed specifically for LTCHs, the "LTC-DRG," indicates that CMS and the federal government recognize the differences between general hospitals and LTCHs when it comes to patient population, costs of care, resources consumed by the patients and health care delivery. Under the LTCH reimbursement system, each patient is assigned a Diagnosis Related Group or "DRG" with a corresponding payment rate that is weighted based upon the patient's diagnosis. The LTCH is reimbursed the predetermined payment rate for that DRG, regardless of the costs of care. These rates are higher than what CMS provides for other traditional post-acute care providers. Since the establishment of the prospective pay system for LTCHs, concerns about the high reimbursement rate for LTCHs, as well as about the appropriateness of the patients treated in LTCHs, have been raised by the Medicare Payment Advisory Committee ("MedPAC") and the Centers for Medicaid and Medicare Services. CMS administers the Medicare payment program for LTCHs, as well as the reimbursement programs for acute care hospitals, SNFs, and CMRs. MedPAC's role is to help formulate federal policy on Medicare regarding services provided to Medicare beneficiaries (patients) and the appropriate reimbursement rates to be paid to health care providers. The 2006 MedPAC report reported that LTCHs were making a good margin or profit, and recommended against an annual increase in the Medicare reimbursement rate for the upcoming fiscal year. In 2006, CMS adopted a reimbursement rate rule for LTCHs for 2007 that did not raise the base rate, and made other changes that reflect the ongoing concerns of CMS regarding LTCHs. 42 C.F.R. Part 412, May 12, 2006. In that rule, CMS found that approximately 37 percent of LTCH discharges are paid under the short-stay outliers, raising concerns that inappropriate patients may be being admitted to LTCHs. CMS made other changes to the reimbursement system which, taken as a whole, actually reduced the reimbursement that LTCHs will receive for 2007. Even with the concerns raised by MedPAC and CMS and recent changes in federal fiscal policy related to LTCHs, the distinction between general hospitals and LTCHs and the legitimate place for LTCHs in the continuum of care continues to be recognized by the federal government. One way of looking at recent developments at the federal level was articulated at hearing by Mr. Kornblat. Federal regulatory changes will reduce the reimbursement LTCHs receive when treating short-term patients (short-term outliers). "On the other end of the spectrum, there are patients who stay significantly longer than would be expected on average, long- stay outliers, and the reimbursement for those patients was also modified." Tr. 163. There have been other changes with regard to LTCH patients who require surgery the LTCHs cannot provide and patients with a primary psychiatric diagnosis or a primary rehab diagnosis. Requiring the LTCH to "foot the bill" for surgery that it cannot provide for its patients and the elimination from LTCHs of patients with a primary psychiatric or rehab diagnosis send a strong signal to the LTCH industry specifically and those who interact with it: LTCHs should admit only the medically complex and severely acutely ill patient who can be appropriately treated at an LTCH. Despite recent changes at the federal level and the clear recognition by the federal government that LTCHs have a place in the continuum of health care services, AHCA remains concerned about LTCHs in Florida. AHCA's Concerns Regarding LTCHs In deciding on whether to approve or deny new health care facilities, the Agency is responsible for the "coordinated planning of health care services in the state." § 408.033(3)(a), Fla. Stat. In carrying out this responsibility, AHCA looks to federal rules and reports to assist in making health care planning decisions for the state. Regarding LTCHs, MedPAC has reported, and CMS has noted that, nationwide, there has been a recent, rapid increase in the number of LTCHs: "It [LTCHs] represents a growth industry of the last ten years." Nationwide there has also been a huge increase in Medicare spending for LTCH care from $398 million in 1993 to $3.3 billion in 2004. AHCA has also become concerned about the recent rapid increase in LTCH applications in Florida. From 1997 through 2001 there were 8 LTCHs in the state. Starting in 2002, there was a marked increase in the number of applications for LTCHs and the number of approved LTCHs rose quickly to the current 14 in 2006. In addition, 9 new LTCHs have been approved and are expected to be licensed in the next 1-3 years. When all of the approved hospitals are licensed the number of available beds will rise from 876 to 1,351 (adding the approved 475 beds), over a 50 percent increase in LTCH beds statewide. In addition, AHCA is concerned that the occupancy level of LTCHs over the entire state appears to be falling over the last 11 years. In response to the rise in LTCH applications over the last several years, and given the decrease in occupancy of the current LTCHs, the Agency has consistently voiced concerns about lack of identification of the patients that appropriately comprise the LTCH patient population. Because of a lack of specific data from applicants with regard to the composition and acuity level of LTCH patient populations, AHCA is not convinced that there is a need for additional LTCHs in the state or in District 11. There are several reasons for this concern. First, AHCA believes, like MedPAC, that there may be an overlap between the LTCH patient populations and the population of patients served in other health care settings, such as SNFs and CMRs. Kindred's expert, Dr. Muldoon, noted that length of stay in the general acute care hospital has been shortened over the last few years because there are new more effective medical treatments, and because the "post-acute sector has emerged as the place to carry out the treatment plan that 20 years ago may been provided in its entirety in the short-term hospital." (Kindred Ex. 8 at 23). To AHCA, what patients enter what facilities in this "post-acute sector" is unclear. In the absence of the applicants better identifying the acuity of the LTCH patient population, AHCA has reached the conclusion that there may be other options available to those patients targeted by the LTCH applicants. In support of this view, AHCA presented a chart showing SNFs in District 11 that offer to treat patients who need dialysis, tracheotomy or ventilator care. These conditions are typically treated in LTCHs. In addition, AHCA believes that some long-stay patients can be appropriately served in the short-stay acute care hospitals, rather than requiring LTCH care. The length of stay in 2005 for the typical acute care hospital for most patients is five to six days. (Kindred Ex. 8, Dr. Muldoon Depo, at 23). Some hospital patients, however, are in need of acute care services on a long-term basis, that is, much longer than the average lengths of stay for most patients. Thus, patients who may need LTCH services often have lengths of stay in the acute care hospitals that exceed the typical stay. AHCA believes that these long-stay patients can be as appropriately served in the short stay acute care hospitals as in LTCHs. AHCA'S Denial of the Four Applications and Change of Position with regard to MJH On December 15, 2005, the Agency issued its SAAR after review of the applications. The SAAR recommended denial of all four applications based primarily on the Agency's determination that none had adequately demonstrated need for its proposed LTCH in District 11. In denying the four applications, AHCA relied in part on reports issued the Congress annually by MedPAC that discuss the placement of Medicare patients in appropriate post-acute settings. Appropriate use of long term care hospital services is an underlying concern that we [AHCA] have and had the federal government has as evidenced by their MedPAC reports and the CMS information in its most recent proposed rule on the subject. (Tr. 2486). The June 2004 MedPAC report states the following about LTCHs: Using qualitative and quantitative methods, we find the LTCH's role is to provide post- acute care to a small number of medically complex patients. We also find that the supply of LTCHs is a strong predictor of their use and those acute hospitals and skilled nursing facilities are the principal alternatives to LTCHs. We find that, in general, LTCH patients cost Medicare more than similar patients using alternative settings but that if LTCH care is targeted to patients of the highest severity, the cost is comparable. Given these concerns, AHCA looked to the four applicants to prove need through a needs methodology that provides sufficient information on the patient severity criteria to better define the patients that would mostly likely be appropriate candidates for LTCHs. AHCA found the need methodologies of three of the four applicants (Kindred, Promise, and Select) "incomplete" because they lacked specific information on the severity level of the patients the applicants plans to admit, and therefore they "overstate need." AHCA pointed to a former LTCH provider that did provide detailed useful information on the acuity level of its patients, and the acuity level of its patients in reference to similar patients in SNFs. Other then MJH, the applicants presented approaches to projecting need that are based, in one way or another, on long- stay patients in existing acute care hospitals. In the Agency's view these methods "significantly overstate need." The method creates a "candidate pool" for the future long-term care hospital users. But it does not include enough information on severity of illness of the patients, in AHCA's view, to give a sense of who might be expected to appropriately use the service. Further, the Agency sees no reason to believe that all long-stay patients in acute care short-stay hospitals are appropriate candidates for long-term hospital services. Lastly, AHCA believes that LTCH applicants should develop an "acuity coefficient or an acuity factor," tr. 2627, to be considered as part of an LTCH need methodology. The need methodology employed by MJH differed substantially from the methodologies of the other three applicants. Because it is more conservative and yields a need "approximately a tenth of what the other three propose," tr. 2500, at the time of hearing AHCA was much more comfortable with MJH's need methodology. By the time AHCA filed its PRO, its comfort with MJH's need methodology had solidified and improved to the point that AHCA changed its position with regard to MJH. Describing MJH's "use rate model" as conservative, see Agency for Health Care Administration Proposed Recommended Order, at 24, AHCA proposed the following finding of fact in support of its conclusion that MJH's application be approved: "Miami Jewish Home projected a reasonably reliable bed need using approved, conservative, but detailed and supportable, need methodologies." Id. at 25. MJH MJH, is an existing not-for-profit provider of comprehensive health and social services in Miami-Dade County. As recited in the Omissions Response to CON 9893: [MJH's] mission is to be the premier multi- component, not-for-profit charitable health care system in South Florida, guided by traditional Jewish values, dedicated to effectively and efficiently serving a non- sectarian population of elderly, mentally ill, disabled, and chronically ill people with a broad range of the highest quality institutionally-based, community-based and ambulatory care services. MJH Ex. 1. Originally founded in 1945 to provide residential care for Jewish persons unable to access services elsewhere, MJH is now in its 62nd year of operation. MJH enjoys a good reputation within its community. MJH is located at Northeast Second Avenue and 2nd Street in north-central Miami in one of the most densely populated areas of the County. Known as “Little Haiti,” the surrounding community is primarily low income, and is a federally designated “medically underserved area.” A “safety net” provider of health care services, MJH's SNF is the largest provider of Medicaid skilled nursing services in the State of Florida. MJH assists its patients/residents in filing Medicaid applications, and also assists individuals in applying for Medicaid for community-based services. This same kind of assistance will also be provided to patients of the MJH LTCH. A 2004 study conducted by the Center on Aging at Florida International University identified unmet needs among elders living within the zip codes surrounding MJH. The study notes that the greatest predictors of need for home and community-based services are poverty, disability, living alone, and old age. Several of the zip codes within the MJH PSA were found to have relatively large numbers of at risk elders due to poverty and dramatic community changes. The study has assisted MJH in identifying service gaps within the community, and in focusing its efforts to serve this at-risk population. Following its most recent JCAHO accreditation survey, both MJH’s hospital and SNF received a three-year “accreditation without condition,” which is the highest certification awarded by JCAHO. MJH is a national leader in the provision of comprehensive long-term care services. MJH has been recognized on numerous occasions for its innovative long-term and post- acute care programs. The awards and recognitions include the Gold Seal Award for Excellence in Long Term Care, the "Best Nursing Home" Award from Florida Medical Business and "Decade of Excellence Award" from Florida Health Care Association. An indicator of quality of care, AHCA’s “Gold Seal” designation is especially significant. Of the 780 nursing homes in Florida, only 13, including MJH, have met the criteria to be designated as Gold Seal facilities. MJH operates Florida's only Teaching Nursing Home Program. Medical students, interns, and other health professionals rotate through the service program in the nursing home and hospital on a regular basis. Specifically, MJH serves as a student and resident training site for the University of Miami and Nova Southeastern University Medical Schools, and the Barry University, FIU, and University of Miami nursing schools. The LTCH would enhance these capabilities and give physicians in training additional opportunities. Not only will this enhance their education, but also will contribute to the high quality of care to be provided in the MJH LTCH. MJH has been the site and sponsor of many studies to enhance the delivery of social and health services to elderly and disabled persons. Most recently, MJH was awarded a grant to do research on fall prevention in the nursing home. MJH is committed to continue research on the most effective means of delivering rehabilitative and long-term care services to a growing dependent population. The development of an LTCH at MJH will enhance the opportunities for this research. MJH operates Florida’s first and only PACE Center (Program of All-inclusive Care for the Elderly) located on the main Douglas Gardens campus. The program provides comprehensive care (preventive, primary, acute and long-term) to nursing home eligible seniors with chronic care needs while enabling them to continue to reside in their own home as long as possible. MJH was recently approved by the Governor and Legislature to open a second PACE site, to be located in Hialeah. The proposed 30-bed LTCH will be located on MJH’s Douglas Gardens Campus. The Douglas Gardens Campus is the site of a broad array of health and social services that span the continuum of care. These programs include community outreach services, independent and assisted living facilities, nursing home diversion services, chronic illness services, outpatient health services, acute care hospital services, rehabilitation, post-acute services, Alzheimer’s disease services, pain management, skilled nursing and hospice. LTCH services, however, are not currently available at MJH. Fred Stock, the Chief Operating Officer of MJH is responsible for the day-to-day operation of the MJH nursing home and hospital and has 24 years experience in the administration of long-term care facilities. An example of Mr. Stock’s leadership is that when he came to MJH, its hospice program had management issues. He assessed the situation and then made a management change which has resulted in a successful turnaround of the program. There are now 462 skilled nursing beds licensed and operated by MJH at the Douglas Garden’s Campus. All of these beds are certified by Medicare. Community hospitals have come to rely on these skilled nursing beds as a placement alternative for their sickest and most difficult-to-place, post-acute patients. The discharges of post-acute patients in the SNF at Douglas Gardens more than doubled from 350 in FY 2002 to 769 in FY 2005. Dr. Tanira Ferreira is the Medical Director of the MJH ventilator unit. Dr. Ferreira is board-certified in the specialties of Internal Medicine, Pulmonary Diseases, Critical Care Medicine, and Sleep Disorders. Dr. Ferreira will be the Medical Director of the MJH LTCH. In addition to Dr. Ferreira, MJH has five other pulmonologists on its staff. MJH also employs: a full-time Medical Director (Dr. Michael Silverman); three full-time physicians whose practices are restricted to MJH hospital and SNF patients; and four full-time nurse practitioners whose practices are restricted to residents of the SNF. MJH employs two full-time psychiatrists, two full-time psychologists, and seven full-time Master’s level social workers. The MJH medical staff also includes many specialist physicians such as cardiologists, surgeons, orthopedists, nephrologists and opthamologists, and other specialists are called for a consultation as needed. A number of the MJH patients/residents are non-English speakers. However, many of the MJH employees, including all of its medical staff, are bilingual. Among the languages spoken by MJH staff are Haitian, Spanish, Russian, Yiddish, French, and Portuguese. This multi-language capability greatly enhances patient/resident communication and enhances MJH’s ability to provide supportive services. The proposed project is the development of a 30-bed LTCH in Miami-Dade County. The LTCH will be located in renovated space in an existing facility and will conform to all the physical plant and operating standards for a general hospital in Florida. The estimated project cost is $5,315,672. The first patient is expected to be admitted by July 1, 2007. The LTCH will be considered an HIH under Federal regulations 42 CFR Section 412.22(e). The LTCH will comply with these requirements including a separate governing body, separate chief medical officer, separate medical staff, and chief executive officer. The LTCH will perform the hospital functions required in the Medicare Conditions of Participation set forth at 42 CFR Section 482. In addition, fewer than 25 percent of the admissions to the LTCH will originate from the MJH acute care hospital, and less than 15 percent of the LTCH operating expenses will be through contracted services with any other MJH affiliate, including the acute care hospital. The separate LTCH governing body will be legally responsible for the conduct of the LTCH as an institution and will not be under the control of the MJH acute care hospital. Finally, less than five percent of the annual MJH LTCH admissions will be re-admissions of patients who are referred from the MJH SNF or the MJH hospital. Each referral to the LTCH will be carefully assessed using the InterQual level-of-care criteria to ensure that the most appropriate setting is chosen. MJH is also a member of the ECIN (Extended Care Information Network) system. As a member of this system, MJH is able to make referrals and place patients who may not be appropriate for its own programs. Only those patients who are medically and functionally appropriate for the LTCH will be admitted to the LTCH program. Many patients admitted to the MJH LTCH will have complex medical conditions and/or multiple-system diagnoses in one or more of the following categories: Respiratory disorders care (including mechanical ventilation or tracheostomy care) Surgical wound or skin ulcer care Cardiac Care Renal disease care Cancer care Infectious diseases care Stroke care The patient and family will be the focus of the interdisciplinary care provided by the MJH LTCH. The interdisciplinary care team will include the following disciplines: physicians, nurses, social workers, psychologists, spiritual counselors, respiratory therapists, physical therapists, speech therapists, occupational therapists, pharmacists, and dietitians. MJH uses a collaborative care model that will be replicated in the LTCH and will enhance the effectiveness of the interdisciplinary team. The direct care professionals in the LTCH will maintain an integrated medical record, so that each member of the care team will have ready access to all the information and assessments from the other disciplines. Nursing staff will provide at least nine hours of nursing care per patient per day. Seventy-five percent of the nursing staff hours will be RN and LPN hours. Therapists (respiratory, physical, speech and occupational) will provide at least three hours of care per patient day. The MJH medical staff includes a wide array of specialty consultants that will be available to LTCH patients. The specialties of pulmonology, internal medicine, geriatrics and psychiatry will be available to each patient on a daily basis. A complete listing of all of the medical specialties available to MJH patients was included with its application. The interdisciplinary team will meet at least once per week to assess the care plan for each patient. The care plan will emphasize rehabilitation and education to enable the patient to progress to a less restrictive setting. The care team will help the patient and family learn how to manage disabilities and functional impairments to facilitate community re-entry. Approval of the LTCH will allow the MJH to "round out" the continuum of care it can offer the community by placing patients with clinically complex conditions in the most appropriate care setting possible. This is particularly true of persons who would otherwise have difficulty in accessing LTCH services. MJH has committed to providing a minimum of 4.2 percent of its patient discharges to Medicaid and charity patients. However, Mr. Stock anticipates that the actual percentage will be higher. If approved, MJH has committed to licensing and operating its proposed LTCH. MJH already has a number of the key personnel that will be required to implement its LTCH, including the Medical Director and other senior staff. In addition, MJH has extensive experience gleaned from both its acute care hospital and SNF in caring for very sick patients. In short, MJH has the clinical, administrative, and financial infrastructure that will be required to successfully implement its proposed LTCH. Approval of the MJH LTCH will dramatically reduce the number of persons who are now leaving the MJH PSA to access LTCH services. The hospitals in close proximity to MJH have LTCH use rates that are very low in comparison to other hospitals that are closer to existing LTCHs. Thus, it is likely that there are patients being discharged from the hospitals close to MJH that could benefit from LTCH services, but are not getting them because of access issues or because the existing LTCHs are perceived to be too far away. A number of hospitals located close to MJH are now referring ventilator-dependent patients to MJH, and would also likely refer patients to the MJH LTCH. Because the majority of the infrastructure required is already in place, the MJH HIH can be implemented much more quickly and efficiently than can a new freestanding LTCH. For example, ancillary functions such as billing, accounting, human resources, housekeeping and administration already exist, and the LTCH can be efficiently integrated into those existing operations on campus. MJH will be able to appropriately staff its LTCH through a combination of its current employees and recruitment of new staff as necessary. In addition, MJH will be establishing an in-house pharmacy and laboratory within the next six months, which will also provide services to LTCH patients. On-site radiology services are already available to MJH patients. MJH has an excellent track record of successfully implementing new programs and services. There is no reason to believe that MJH will not succeed in implementing a high quality LTCH if its application is approved. MJH's Ventilator Unit By the time ventilator-dependent and other clinically complex patients are admitted to a nursing home they have often exhausted their 100 days of Medicare coverage, and have converted to Medicaid. Since Medicaid reimbursement is less than the cost of providing such care, most nursing homes are unwilling to admit these types of patients. Thus, it is very difficult to place ventilator patients in SNFs statewide. The problem is further exacerbated in District 11 by the lack of any hospital-based skilled nursing units. With the recent closure of two SNF-based vent units (Claridge House and Greynolds Park) there are now only three SNF-based vent units remaining in District 11. They are located at MJH, Hampton Court (10 beds), and Victoria Nursing Home. MJH instituted a ventilator program in its SNF in early 2004. Many of the patients admitted into the ventilator program fall into the SE3 RUG Code. On July 1, 2005, there were 24 patients in the SE3 RUG code in MJH. Only one other SNF in District 11 has more than four SE3 RUG patients in its census on an average day. Over 60 percent of the Medicare post-acute census at the MJH SNF falls into the RUG categories associated with extensive, special care or clinically complex services. This mix of complex cases is about three times higher than average for District 11 SNFs. Although some of the patients now admitted to the MJH SNF vent unit would qualify for admission to an LTCH, there are also a number of patients who are not admitted because MJH cannot provide the LTCH level of care required. SNF admissions are required to be initiated following a STACH admission. MJH has actively marketed its vent unit to STACHs. Similarly most LTCH admissions come from STACHs and, like MJH’s efforts, LTCHs also market themselves to STACHs. Hospitals providing tertiary services and trauma care will generate the greater number of LTCH referrals, with approximately half of all LTCH patients being transferred from an ICU. The implementation of the MJH ventilator unit required the development of protocols, infrastructure, clinical capabilities and internal resources beyond those found in most SNFs. Dr. Ferreira conducted pre-opening comprehensive staff education. These capabilities will serve as a precursor to the development of the next stage of service delivery at MJH: the LTCH. MJH’s vent unit provides care for trauma victims, and recently received a Department of Health research grant to develop a program for long-term ventilator rehab for victims of trauma. Jackson Memorial Hospital is experiencing difficulty in placing "certain" medically complex patients, who at discharge, have continuing comprehensive medical needs. MJH is the only facility in Dade County that has accepted Medicaid ventilator patients from Jackson. Mt. Sinai Medical Center also has difficulty placing medically complex patients, particularly those requiring ventilator support, wound care, dialysis and/or other acute support services. Mt. Sinai is a major referral source to MJH and supports its LTCH application. MJH has received statewide referrals, including from the Governor's Office and from AHCA, of difficult to place vent patients. Most of these referrals are Medicaid patients. Ten of the MJH vent beds are typically utilized by Medicaid patients. Although MJH would like to accommodate more such referrals, there are financial limitations on the number of Medicaid patients that MJH can accept at one time. Promise Promise owns and operates approximately 718 LTCH beds outside of Florida and employs an estimated 2,000 persons. Promise proposes to develop and LTCH facility in the western portion of the County made up of 59,970 gross square feet, 60 private beds including an 8-bed ICU, and various ancillary and support areas. The projected costs to construct its freestanding LTCH is $11,094,500, with a total project cost of $26,370,885. As a condition of its CON if its application is approved, Promise agrees to provide three percent of projected patient days to Medicaid and charity patients. Select Select-Dade proposes to locate its 60-bed, freestanding LTCH in the western portion of Miami-Dade County. The Agency denied Select-Dade's application because of its failure to prove need. Otherwise, the application meets the CON review criteria and qualifies for comparative review with the other three applicants. Select-Dade proposes to serve the entire District, but it has targeted the entire west central portion of the County that includes Hialeah, Hialeah Gardens, Doral, Sweetwater, Kendall, and portions of unincorporated Miami. This area is west of State Road 826 (the "Palmetto Expressway"), south of the County line with Broward County, north of Killian Parkway and east of the Everglades ("Select's Target Service Area"). To be located west of the Palmetto Expressway, east of the Florida Turnpike, north of Miller Drive and south of State Road 836, the site for the LTCH will be generally in the center of Select's Target Service Area. Approximately 700,000 people (about 30 percent of the County's population) reside within Select-Dade's Target Service Area. This population of the area is expected to grow almost ten percent in the next five years. The rest of the County is expected to grow about five and one-half percent. Kindred Kindred proposes to construct a 60-bed LTCH in the County. It will consist of 30 private rooms, 20 beds in 10 semi-private rooms, and 10 ICU beds. The facility would include the necessary ancillary service, including two operating rooms, a radiology suite, and a pharmacy. Kindred utilizes a screening process before admission of a patient to assure that the patient needs LTCH level care that includes the set of criteria known as InterQual. InterQual categorizes patients according to their severity of illness and the intensity of services they require. Every patient admitted to a Kindred hospital must be capable of improving and the desire to undergo those interventions aimed at improvement. Kindred does not provide hospice or custodial care. In addition, through its reimbursement process, the federal government provides strong disincentives toward LTCH admission of inappropriate patients. Furthermore, every Kindred hospital has a utilization review (UR) plan to assure that patients do not receive unnecessary, unwanted or harmful care. In addition to the UR plan, the patient's condition is frequently reviewed by nursing staff, respiratory staff and by a multi-disciplinary team. Kindred had not selected a location at the time it submitted its application. Kindred anticipates, however, that its facility if approved would be located in the western portion of the County. Stipulated Facts As stated by Kindred in its Proposed Recommended Order, the parties stipulated to the following facts (as well as a few other related to identification of the parties): Each applicant timely filed the appropriate letter of intent, and each such letter contained the information required by AHCA. Each CON application was timely filed with AHCA. Following its initial review, AHCA issued a State Agency Action Report ("SAAR") which indicated its intent to deny each of the applications. Each applicant timely filed the appropriate petition with AHCA, seeking a formal hearing pursuant to Sections 120.569 and 120.57, Fla. Stat. In the CON batch cycle that is the subject of this proceeding, Promise XI proposed to construct a 59,970 square foot building at a total project cost of $26,370,885.00, conditioned upon providing 3 percent of its patient days to Medicaid and charity patients. Select proposes to construct a 62,865 square foot building at a total project cost of $22,304,791.00, conditioned upon providing 2.8 percent of its patient days to Medicaid and charity patients. MJHHA proposes to renovate 17,683 square feet of space at a total project cost of $5,315,672.00, conditioned upon providing 4.2 percent of its patient days to Medicaid and charity patients. Kindred proposes to construct a 69,706 square foot building at a total project cost of $26,538,458.00, conditioned upon providing 2.2 percent of its patient days to Medicaid and charity patients. Long term hospitals meeting the provisions of AHCA Rule 59A-3.065(27), Fla. Admin. Code, are one of the four classes of facilities licensed as Class I hospitals by AHCA. The length of stay in an acute care hospital for most patients is three to five days. Some hospital patients, however, are in need of acute care services on a long- term basis. A long-term basis is 25 to 34 days of additional acute are service after the typical three to five day stay in a short-term hospital. Although some of those patients are "custodial" in nature and not in need of LTCH services, many of these long-term patients are better served in a LTCH than in a traditional acute care hospital. Within the continuum of care, the federal government's Medicare program recognizes LTCHs as distinct providers of services to patients with high levels of acuity. The federal government treats LTCH care as a discrete form of care, and treats the level of service provider by LTCHs as distinct, with its own Medicare payment system of DRGs and case mix reimbursement that provides Medicare payments at rates different from what the Medicare prospective payment system ("PPS") provides for other traditional post-acute care providers. The implementation by the Centers for Medicare and Medicaid Services ("CMS") of categories of payment design specifically for LTCHs, the "LTC-DRG," is a sign of the recognition by CMS and the federal government of the differences between general hospitals and LTCHs when it comes to patient population, costs of care, resources consumed by the patients and health care delivery. Joint Pre-hearing Stipulation at 4, 6-7, 9-10. Applicable Statutory and Rule Criteria The parties stipulated that the review criteria in Subsections (1) through (9) of Section 408.035, Florida Statutes (the "CON Review Criteria Statute"), apply to the applications in this proceeding. Subsection (10) of the CON Review Criteria Statute, relates to the applicant's designation as a Gold Seal Program Nursing facility. Subsection (10) is applicable only "when the applicant is requesting additional nursing home beds at that facility." None of the applicants are making such a request. MJH's designation as a Gold Seal Program is not irrelevant in this proceeding, however, since it substantiates MJH's "record of providing quality of care," a criterion in Subsection (3) of the CON Review Criteria Statute. The Agency does not have a need methodology for LTCHs. Nor has it provided any of the applicants in this proceeding with a policy upon which to determine need for the proposed LTCH beds. The applicants, therefore, are responsible for demonstrating need through a needs assessment methodology of their own. Topics that must be included in the methodology are listed Florida Administrative Code Rule 59C-1.008(2)(e)2., a. through d. Subsection (1) of the CON Review Criteria: Need Not only does AHCA not have an LTCH need methodology in rule or a policy upon which to determine need for the proposed LTCH beds, it did not offer a methodology for consideration at hearing. This is the typical approach AHCA takes in LTCH cases; demonstration of LTCH need through a needs assessment methodology is left to the parties, a responsibility placed upon them in situations of this kind by Florida Administrative Code Rule 59C-1.008(2)(e)2. MJH's Need Methodology Unlike the other three applicants, all of whom used one form or another of STACH long-stay methodologies, MJH utilized a use-rate analysis which projects LTCH utilization forward from District 11's recent history of increased utilization. A use-rate methodology is one of the most commonly used health care methodologies. The MJH use-rate methodology projected need based upon all of District 11. The methodology projected need for 42 LTCH beds in 2008, with that number growing incrementally to 55 beds by 2012. Because statewide LTCH utilization data is not reliable when looking at any particular district, MJH developed a District 11 use-rate, by age cohort, to yield a projection of LTCH beds needed. The use-rate is derived from the number of STACH admissions compared to the number of LTCH admissions, by age cohort. Projected demographic growth by age cohort was applied to determine the number of projected LTCH admissions. The historic average LTCH LOS in District 11 was applied to projected admissions and then divided by 365 to arrive at an ADC. That ADC was then adjusted for an occupancy standard of 85 percent, which is consistent with District 11. A number of states have formally adopted need methodologies that use an approach similar to MJH's in this case. Kindred has used a shortcut method of the use rate model in other states for analyzing proposed LTCHs "when there is not much data to work with." Tr. 1744. The methodology used by MJH was developed by its expert health planner, Jay Cushman. The methodology developed by Mr. Cushman was described by Kindred's health planner as "a couple of steps beyond" Kindred's occasionally-used shortcut method. Kindred's health planner described Mr. Cushman's efforts with regard to the MJH need methodology as "a very nice job." Tr. 1745. Mr. Cushman created a use-rate by examining the relationship between STACH admissions and LTCH admissions. The use-rate actually grows as it is segmented by age group, and thus the growth in the elderly population incrementally increases the utilization rate. MJH’s application demonstrated how LTCH utilization has varied greatly statewide, and how the District 11 market has a significant history of utilizing LTCH services. For planning purposes the history of District 11 is a significant factor, and the MJH methodology is premised upon that history, unlike the other methodologies. MJH demonstrated a strong correlation between STACH and LTCH utilization in District 11, where 98 percent of LTCH admissions are referred from STACHs. MJH also demonstrated that the south and western portions of Miami-Dade have overlapping service areas from the three existing LTCHs, while northeastern Miami-Dade has only one provider with a similar service area, Kindred Hollywood in neighboring District 10. This peculiarity explains why the LTCH out-migration trend is much stronger in northeastern portions of the District. The area most proximate to MJH would enjoy enhanced access to LTCH services, including both geographic and financial access, if its program is approved. In short, as AHCA, now agrees, MJH demonstrated need for its project through a thorough and conservative analysis. All parties agree that the number of LTCH beds yielded by MJH's methodology are indeed needed. Whether more are needed is the point of disagreement. For example, Mr. Balsano plugged the 2003 use rate into MJH's methodology instead of the 2004 used by MJH. Employment of the 2003 use rate in the calculation has the advantage that actual 2004 and 2005 data can serve as a basis of comparison. Mr. Balsano explained the result: "The number of filled beds in 2005 in District 11 would exceed by 33 beds what the use rate approach would project as needed in 2005." Tr. 370. The reason, as Mr. Balsano went on to explain, is that the use-rate changed dramatically between 2002, 2003, and 2004. Thus MJH's methodology, while yielding a number of beds that are surely needed in the District, may yield a number that is understated. This is precisely the opposite problem of the need assessment methodologies of the other three applicants, all of which overstated LTCH bed need in the District. The Need Methodologies of the Other Three Applicants The need methodologies presented by the other applicants vary to some degree. All three, however, are based on STACH long-stay data. Long-stay STACH analyses rely upon a number of assumptions, but fundamentally they project need forward from historic utilization of STACHs. The methodologies used by each of these three applicants identify patients in STACHs whose stays exceeded the geometric mean of length of stay plus fifteen days (the "GMLOS+15 Methodologies"), although the extent of the patients so identified varied depending on the number of DRGs from which the patients were drawn. Each of the proponent’s projects would serve only a relatively small fraction of the District 11 patients purported by the GMLOS+15 Methodologies to be in need of LTCH services. The lowest projected need of the three was produced by Promise: 393 beds in 2010. Promise's methodology is more conservative than that of Kindred and Select. Unlike the latter two, Promise reduced the number of potential projected admissions to be used in its calculation. The reduction, in the amount of 25 percent of the projection of 500 beds, was made because of several factors. Among them were anticipation that MedPAC's suggestions for ensuring that patients were appropriate for LTCH admission, which was expected to reduce the number of LTCH admissions, would be adopted. The methodologies proposed by Kindred and Select-Dade did not include the Promise methodology's reduction potentially posed by the impact of new federal regulation. Kindred's methodology projected need for 509 new LTCH beds in District 11; Select-Dade's methodology projected need for 556 beds. One way of looking at the substantial bed need produced by the GMLOS+15 Methodologies used by Promise, Select and Kindred was expressed by Kindred. As an applicant proposing a new hospital of 60 beds, when its need methodology yielded a need in the District for more than 500 beds, Kindred found the methodology to provide assurance that its project is needed. On the other hand, if the methodology was reliable then the utilization levels of the two Kindred hospitals in Broward County in relative proximity to a populated area of District 11 would have been much higher in 2004 and 2005, given the substantial out-migration to those facilities from District 11. The Kindred and Select methodologies are not reliable. Their flaws were outlined at hearing by Mr. Cushman, MJH's expert health planner who qualified as an expert with a specialization in health care methodology. Mr. Cushman attributed the flaws to Promise's methodology as well but as explained below, Promise's methodology is found to be reliable. Comparison of the projections produced by MJH's use rate methodology with the projections produced by the other three methodologies results in "a tremendous disconnect," tr. 1233, between experiences in District 11 upon which MJH's methodology is based and the GMLOS+15 Methodologies' bed need yield "that are three or four or five times as high as have actually been expressed in the existing system." Id. One reason in Mr. Cushman's view for the disconnect is that the GMLOS+15 Methodologies identify all long-stay patients in STACHs as candidates for LTCH admission when "there are many reasons that patients might stay for a long time in an acute care facility that are not related to their clinical needs." Tr. 1234. This criticism overlooks the limited number of long-stay patients in STACHs used by the Promise methodology but is generally applicable to the Select and Kindred methodologies. Mr. Cushman performed detailed analysis of the patients used by Kindred in its projection to reach conclusions applicable to all three GMLOS+15 Methodologies. Mr. Cushman's analysis, therefore, related to actual patients. They are based on payor mix, discharge status, and case mix. The analysis showed that the GMLOS+15 Methodologies are "disconnected from the fundamental facts on the ground," tr. 1240, in that the methodologies produce tremendous unmet need not reconcilable with actual utilization experience. Some of the gaps based on additional case mix testing were closed by Kindred's expert health planner. The additional Kindred test, however, did not completely close the gap between projected unmet need and actual utilization experience. Mr. Cushman summed up his basis for concluding that the GMLOS+15 Methodologies employed by Kindred, Select-Dade and Promise are unreliable: [W]e have an untested method that's disconnected from actual utilization experience on the ground. And it provides projections of need that are way in excess of what the experience would indicate and way in excess of what the applicants are willing to propose and support [for their projects.] So for those reasons, I considered [the GMLOS+15 method used by Kindred, Select-Dade and Promise] to be an unreliable method for projecting the need for LTCH beds. Tr. 1243-44. The criticism is not completely on point with regard to the Promise methodology as explained below. Furthermore, at hearing, Mr. Balsano made adjustments to the Promise GMLOS+15 Methodology ("Promise's Revised Methodology"). Although not sanctioned by the Agency, the adjustments were ones that made the Agency more comfortable with the numeric need they produced similar to the Agency's comments at hearing about MJH's methodology. For example, if the number of needed beds were reduced by 50 percent (instead of 25 percent as done in Promise's methodology) to account for the effect of federal policies and alternative providers and if an 85 percent occupancy rate were assumed instead of an 80 percent occupancy rate, the result would be reduce the LTCH bed need yielded by Promise's methodology to 200. These adjustments make Promise's Revised Methodology more conservative than Select's and Kindred's. In addition, Promise's methodology commenced with a much fewer number of STACH patients because Promise based on its inquiry into the patient population that is "using LTCHs in Florida right now." Tr. 351. Examination of AHCA's database led to Promise's identification of patients in 169 DRGs currently served in Florida LTCHs. In contrast, Select-Dade and Kindred, used 483 and 390 DRGs respectively. Substantially the same methodology was used by Promise in Promise Healthcare of Florida III, Inc. v. AHCA, Case No. 06-0568CON (DOAH April 10, 2007). The methodology, prior to the 25 percent reduction to take into account the effects of new federal regulations, was described there as: Long-stay discharges were defined using the following criteria: age of patient was 18 years or older; the discharge DRG was consistent with the discharge DRGs from a Florida LTCH; and the ALOS in the acute care hospital was at the GMLOS for the specific DRG plus 15 days or more. Applying these criteria reduced the number of DRGs used and the potential patient pool. Id. at 19 (emphasis supplied.) The methodology in this case produced a number that was then reduced by 25 percent, just as Promise did in its application in this case. The methodology was found by the ALJ to be reliable. If the methodology there were reliable then Promise's Revised Methodology (an even more conservative methodology) must be reliable as well as the numeric need for District 11 LTCH beds it yields: 200. Such a number (200) would support approval of MJH's application and two of the others and denial of the remaining application or denial of MJH's application and approval of the three other applications. Neither of these scenarios should take place. However high a number of beds that might have been projected by a reasonable methodology, no more than two of the applications should be granted when one takes into consideration the ability of the market to absorb new providers all at once. Tr. 518-520. Nonetheless, such a revised methodology would allow approval of MJH and one other of the applicants. Furthermore, there are indications of bed need greater than the need produced by MJH's methodology. Market Conditions, Population and History The large majority of patients admitted to LTCHs are elderly, Medicare beneficiaries. Typically, elderly persons seek health care services close to their homes. This is often because the elderly spouse or other family members of the patient cannot drive to visit the patient. This contributes to the compressed service areas observed in District 11. Historic patient migration patterns show that for STACH services, there is nine percent in-migration to Miami- Dade, and only five percent out-migration from Miami-Dade, a normal balance. Most recent data for LTCH service, however, shows an abnormal balance: three percent in-migration and 22 percent out-migration. The current utilization of existing LTCHs in District 11 and the high out-migration indicates that additional LTCH beds are needed. Notably, of the 400 District 11 residents who accessed LTCH care in Broward County in 2004, 114 (over 25 percent) lived in the 15 zip codes closest to MJH. MJH’s location will allow its LTCH to best impact and reduce out- migration from District 11 for LTCH services. Neither Kindred nor Promise has a location selected, and while Select-Dade has a “target area,” its actual location is unknown. None of the existing LTCHs in District 11 or in District 10 have PSAs that overlap with the area around MJH. For example, the Agency had indicated that there was no need in the case which led to approval of the Sister Emmanuel LTCH at Mercy Hospital. It was licensed in July of 2002, barely half a year after the Select-Miami facility was licensed. Both facilities were operating at or near optimal functional capacity less than two years from licensure without adverse impact to Kindred-Coral Gables. The utilization to capacity of new LTCH beds in the District indicate a repressed demand for LTCH services. The demand for new beds, however, is not limited to the eastern portion of the County. The demand exists in the western portion as well where there are no like and existing facilities. Medicare patients who remain in STACHs in excess of the mean DRG LOS become a financial burden on the facility. The positive impact on them of an LTCH with available beds is an incentive for them to refer LTCH appropriate patients for whom costs of care exceeds reimbursement. There were a total of 1,231 adult discharges from within Select-Dade's targeted service area with LOS of 24 or more days in calendar year 2004. Medical Treatment Trends in Post-Acute Service The number of LTCHs in Florida has increased substantially in recent years. The increase is due, in part to the better treatment the medically complex, catastrophically ill, LTCH appropriate patient will usually receive at an LTCH than in traditional post acute settings (SNFs, HBSNUs, CMR, and home health care). The clinical needs and acuity levels of LTCH- appropriate patients require more intense services from both nursing staff and physicians that are available in an LTCH but not typically available in the other post acute settings. LTCH patients require between eight to 12 nursing hours per day and daily physician visits. CMS reimbursement at the Medicare per diem rate would not enable a SNF to treat a person requiring eight to 12 hours of nursing care per day. CMR units and hospitals are inappropriate for long- term acute care patients who are unable to tolerate the minimum three hours of physical therapy associated with comprehensive medical rehabilitation. The primary focus of an LTCH is to provide continued acute care and treatment. Patients in a CMR are medically stable; the primary focus is on restoration of functional capabilities. Subsection (2): Availability, Quality of Care, Accessibility, Extent of Utilization of Existing Facilities There are 27 acute care hospitals dispersed throughout the County. Only three are LTCHs. The three existing LTCHs, all in the eastern portion of the County, are not as readily accessible to the population located in the western portion as would be an LTCH in the west. Approval of an application that will lead to an LTCH in the western portion of the County will enhance access to LTCH services or as Ms. Greenberg put it hearing, "if only one facility is going to be built, the western part of the county is where that needs to go." Tr. 2101. See discussion re: Subsection (5), below. In confirmation of this opinion, Dr. Gonzalez pointed out several occasions when he was not able to place a patient at one of the existing LTCHs due to family member reluctance to place their loved one in a facility that would force the family to travel a long distance for visits. LTCH appropriate patients are currently remaining in the acute care setting with Palmetto General and Hialeah Hospital among the busiest of the STACHs in the County. Both are within Select-Dade's targeted service area. From 2002 to 2005 the number of LTCH beds in the District increased from 53 to 122. During the same period, the number of patient days increased from 18,825 to 37,993. Recently established LTCH facilities in District 11 have consistently reached high occupancy levels, approaching 90 percent at the time of hearing. From 2001 to 2004, the use rate for LTCH services grew from 3.07 per 1,000 to 6.51 per 1,000. The increase in use rate for those aged 65 and over was even more significant; from 19.32 per 1,000 to 41.67 per 1,000. Kindred's Miami-Dade facility is licensed at 53 beds; of those seven are in private rooms; the facility has 23 semi- private rooms. As far back as 2001, the facility has operated at occupancy rates in excess of 85 percent; in 1998 and 1999 its occupancy rate exceeded 92 percent and 93 percent, respectively. More recently, it has operated at an ADC of 53 patients; 100 percent capacity. Several physicians and case managers provided support to Kindred's application by way of form letters, indicating patients would benefit from transfers to LTCHs and "an ever growing need for (these) services." Kindred's daily census has averaged 50 or more patients since 2004. Unlike an acute care hospital, Kindred has not experienced any seasonal fluctuations in its census, running at or above a reasonable functional capacity throughout the year. Taking various factors into consideration, including the number of semi-private beds, the facility is operating at an efficient occupancy level. Looking ahead five years, the capacity at Kindred's facility cannot be increased in order to absorb more patients. As designed, the facility cannot operate more efficiently than it has at 85 percent occupancy. Select's facility, located in a medical arts building, houses 34 private and six semi-private beds. In 2005, Select's facility operated at an average occupancy of almost 88 percent. Unlike Kindred, Select can add at least seven more beds to its facility by converting offices. As a hospital within a hospital, Sister Emmanuel's 29-bed facility is subject to limits on the percentage of admissions it can receive from "host" Mercy Hospital; even with such restrictions, its 2005 occupancy rate was 84.6 percent. Because of gender mix and infection opportunities, among other reasons, it is difficult to utilize semi-private beds. Only three District facilities offer ventilator care: MJHHA, HMA Hampton Court, and Victoria Nursing Home. Other health care facility settings do not serve as reasonable alternatives to the LTCH services proposed here. In 2004, roughly one quarter of District 11 residents, (nearly 400 patients), requiring LTCH services traveled to District 10 facilities. In 2005 that number fell to 369, or about 22 percent. Although there is a correlation between inpatient acute care services and LTCH services, the out-migration of patients requiring LTCH services indicated above differs markedly from the out-migration numbers generated by acute care patients. The primary north-south road configurations in the county are A1A, U.S. 1 and I-95 on the east and the Palmetto Expressway on the west. The primary east-west road configurations are composed of the Palmetto Expressway extension, S.R. 112; the Airport Expressway feeding into the Miami International Airport area and downtown Miami, S.R. 836 to Florida's Turnpike, and the Don Shula Expressway in the southwest. Assuming no delays, a trip by mass transit, used by the elderly and the poor, from various areas in Miami-Dade to the nearest LTCH outside District 11 (Kindred Hollywood) runs two to four hours one way. These travel times pose a special hardship to the elderly traveling to a facility to receive care or visit loved ones. While improvements in the system are planned over the next five years, they will not measurably change the existing travel times. These factors, along with high occupancy levels in District 11 LTCHs, indicate the demand for LTCH services in the District exceeds the existing bed supply. The three existing LTCHs have recently operated at optimal functional capacity or above it. On December 31, 2005, Select Specialty Hospital-Miami was operating with 95 percent occupancy. Subsection (3): Ability of the Applicant to Provide Quality of Care and the Applicant's Record of Providing Quality of Care As discussed above, MJH has the ability to provide high quality of care to its LTCH patients and an outstanding record of providing quality of care. Select-Dade has the ability to provide quality of care to its LTCH patients and a record providing quality of care. In treating and caring for LTCH patients, Select-Dade will use an interdisciplinary team of physicians, dieticians, respiratory therapists, physical therapists, occupational therapists, speech therapists, nurses, case managers and pharmacists. Each will discipline will play an integral part in assuring the appropriate discharge of the patient in a timely manner. The Joint Commission on Accreditation of Hospital Organizations (JCAHO) has accredited all Select facilities that have been in existence long enough to qualify for JCAHO accreditation. Both Select and Promise use various tools, including Interqual Criteria, to assure patients who need LTCH services are appropriately evaluated for admission. All Promise facilities are accredited by JCAHO. Promise has developed and implemented a company-wide compliance program, as well as pre-admission screening instruments, standards of performance and a code of conduct for its employees. Its record of providing quality of care was shown at hearing with regard to data related to its ventilator program weaning rate and wound healing rates. None of the parties presented evidence or argument that any of the other applicants was unable to provide adequate quality of care. The Agency adopted its statements from the SAAR at pages 43 through 45. The SAAR noted the existence of certain confirmed complaints at the two existing LTCH providers in Florida Select and Kindred. The number of confirmed complaints is relatively few. Kindred, for example, had 12 confirmed complaints with the State Department of Health at its seven facilities during a three-year period, less than one complaint per Kindred hospital every two years. Each applicant satisfies this criterion. Subsection (4): Availability of Resources, Health and Management Personnel, Funds for Capital and Operating Expenditures, Project Accomplishment and Operation The parties stipulated that all applicants have access to health care and management personnel. Select-Dade, Kindred and MJH all have funds for capital and operating expenditures and project accomplishment and operation. In turn, each of these three contends that Promise did not demonstrate the availability of funds for its project. This issue is dealt with below under the part of this order that discusses Subsection (6) of the Statutory CON Review Criteria. Subsection (5): Access Enhancement The applicants stipulated that "each of the applicants' projects will enhance access to LTCH services for residents of the district to some degree." All four applicants get some credit under this subsection because approval of their application will enhance access by meeting need that all of the parties now agree exists. Select-Dade and Promise propose to locate their projects in the western portion of the County. Kindred did not indicate a location. Location of an LTCH in the western portion of the County will enhance geographic access. MJH's location is in an area that has reasonable geographic access to LTCH services. But approval of its application, given the unique nature of its operation, chiefly its charitable mission, will enhance access to charity and Medicaid recipients. Approval of Select-Dade's application will also enhance cultural access to the Latin population in Hialeah. A substandard public transportation system for this population makes traveling to visit hospitalized loved ones an insurmountable task in some situations. Select-Dade has achieved a competent cultural atmosphere in its LTCH opened in the County in 2003. It has in excess of 100 multi-lingual employees, many of whom communicate in Spanish. The staff effectively communicates with patients with a variety of racial, cultural and ethnic backgrounds. Every new LTCH must undergo a qualifying period to establish itself as an LTCH for Medicare reimbursement. Specifically, the average LOS for all Medicare patients must meet or exceed 25 days. During the qualifying period the LTCH is reimbursed by Medicare under the regular STACH PPS, that is paid on a DRG basis as if the patient were in an ordinary general acute care hospital with its lower reimbursement. Upon initiation of their LTCH services, Promise, Kindred and Select all intend to restrict or suppress admissions to ensure longer LOS to meet the Medicare 25 day average LOS requirement, and to “minimize the costs” of obtaining LTCH certification and reimbursement. MJH will not be artificially restricting its LTCH admissions during the initial 6 month Medicare qualification period, even though the cost of providing services during this period will likely exceed the STACH Medicare reimbursement. MJH’s opening without suppressing admissions (as in the case of Sister Emmanuel), will enhance access by patients in need of these services during the initial qualification period. Subsection (6): Immediate and Long-term Financial Feasibility a. Short-Term Financial Feasibility Short-term financial feasibility is the ability of an applicant to fund the project. None of the parties took the position that the MJH project was not financially feasible in the short term. MJH's current assets are equal to current liabilities, a short-term position found by AHCA to be weak but acceptable. The financial performance of MJH, however, has been improving in the past three years. Expansion of existing services, improved utilization of services, and the development of new programs have all contributed to a significant increase in operational revenue and total revenue during that period. MJH has a history of receiving substantial charitable gifts (ranging from $6.2 million to $13.2 million annually during the past three years) and can reasonably expect to receive financial gifts annually of between $4-5 million in the coming years. However, MJH is moving away from reliance on charitable giving, and toward increasing self-sufficiency from operations. Approval of the LTCH will play a major role in achieving that goal. In addition, MJH has total assets, including land and buildings, of approximately $150 million. The cost to implement the proposed MJH LTCH is $5,319,647. The projected cost is extremely conservative in the sense of overestimating any potential contingency costs that could be incurred. MJH has the resources available to fund the project through endowments and investments (currently $41 million) as well as from operating cash flow and cash on hand. Select-Dade has an adequate short-term position and Kindred a good short-term position. None of the parties contest the short-term financial feasibility of either Select-Dade or Kindred. In contrast, both Select-Dade and Kindred contested the short-term financial feasibility of Promise. In accord is MJH's position expressed in its proposed recommended order: "Promise did not demonstrate the availability of funds for its project." Miami Jewish Home & Hospital For the Aged, Inc.'s Proposed Recommended Order, at 37. Promise's case for short-term financial feasibility rests on the historical relationship between the principals of Promise, Sun Capital Healthcare, Inc., and Mr. William Gunlicks of Founding Partners Capital Management Company ("Founding Partners.") The relationship has led to great success financially over many years. For example, through the efforts of Mr. Gunlicks, Sun Capital has generated over $2 billion in receivable financing. Founding Partners is an investment advisor registered with the Security Exchange Commission, the Commodity Futures Trading Commission, the National Futures Association and the State of Florida. As a general partner, it manages two private investment funds: Founding Partners Stable Value Fund and Founding Partners Equity Fund. Founding Partners also manages an International Fund for non-U.S. investors. Its base is composed of approximately 130 individuals with high net worth and access to capital. Founding Partners provided Promise with a "letter of interest" dated October 12, 2005, which indicated its interest in providing the "construction, permanent, and working capital financing for the development of a 60 bed long-term acute care hospital to be located in Dade County, Florida." Promise Ex. 3, Exhibit Promise XI, Gunlicks 4, 6-27-06. The letter makes clear, however, that it is not a commitment to finance the project: "The actual terms and conditions of this loan will be determined at the time of your loan request is approved. Please recognize this letter represents our interest in this project and is not a commitment for financing." Id. Testimony at hearing demonstrated a likelihood that Promise would be able to fund the project should it's application be approved. Mr. Balsano opined that this is sufficient to meet short-term financial feasibility: "[I]t's not required at this point that firm funding be in place. . . . [W]e have an appropriate letter from Mr. Gunlicks' organization that they're interested and willing to fund the project. It kind of goes to the second issue, which is, well, what if there were some issue in that regard? Would this project be financed. And I guess I would just have to say bluntly that in doing regulatory work for the last 20-some years, that if an applicant has a certificate of need for a given service, most lending institutions view that as a validation that the project is needed and can be supported. My experience has been that I have never personally witnessed a project that was approved that could not get financing. Tr. 392. Other expert health planners with considerable experience in the CON regulatory arena conceded that they were not aware of a CON-approved hospital project in the state that could not get financing. Despite the proof of a likelihood that Promise's project would be funded if approved, however, Promise failed to demonstrate as MJH, Select-Dade and Kindred continue to maintain, that funds are, indeed, available to fund the project. In sum, Promise failed to demonstrate the short-term financial feasibility of the project. The projects of MJH, Select-Dade and Kindred are all financially feasible in the short-term. b. Long-Term Financial Feasibility Long-term financial feasibility refers to the ability of a proposed project to generate a positive net revenue or profit at the end of the second full year of operation. MJH’s projected patient volumes are both reasonable and appropriate, given its current position in the community, the services it currently provides, and the need for LTCH services in the community. MJH’s projected payor mix was largely based upon the historical experience of the three existing LTCHs in the District, with the exception of the greater commitment to charity and Medicaid patients. The higher commitment to Medicaid/charity is consistent with MJH’s historical experience and status as a safety net provider. Sister Emmanuel is a 29-bed LTCH located within Mercy Hospital. As a similarly-sized HIH, a not-for-profit provider, and an entity with the same kind of commitment to Medicaid/charity patients, Sister Emmanuel is the best proxy for comparison of the financial projections contained in the MJH application. MJH projected its gross revenues based upon Sister Emmanuel’s general charge structure, adjusted for payor mix and inflated at 4 percent per year. The staffing positions, FTEs and salaries contained on Schedule 6 of each of the applications were stipulated to represent reasonable projections. MJH’s Medicaid net revenues were calculated by determining a specific Medicaid per diem rate using the Dade County operating cost ceiling and 80 percent of the capital costs. Given that many LTCH patients exhaust their allowable days of Medicaid coverage, 70 percent of the revenue associated with MJH’s Medicaid patient days were “written off” in total. Similarly, patient days associated with charity care and bad debt reflected no net revenue. MJH's Medicare net revenues were determined using the specific diagnosis (DRG) of each projected patient. For the first six months of operation it was assumed that MJH would receive the short-stay DRG reimbursement, and in the second 6 months and second year of operation would receive the LTCH DRG payment. Net revenues for the remaining payor categories were based upon the historical contractual adjustments of MJH. MJH’s projected gross and net revenues for its proposed LTCH are conservative, reasonable and achievable. However, if MJH has in fact understated the net revenues that it will actually achieve, the impact will be an improved financial performance and improved likelihood of long-term financial feasibility. MJH’s staffing expense projections were derived from its Schedule 6 projections (which were stipulated to be reasonable) with a 28 percent benefit package added. Non- ancillary expense costs were based upon MJH’s historical costs, while ancillary expenses (lab, pharmacy, medical supplies, etc.) were based upon the Sister Emmanuel proxy. Capitalized project costs, depreciation and amortization were derived from Schedule 1 and the historical experience of MJH, as were the non- operating expenses such as G&A, plant maintenance, utilities, insurance and other non-labor expenses. MJH’s income and expense projections are reasonable and appropriate, and demonstrate the long-term financial feasibility of MJH’s proposed LTCH. John Williamson is an Audit Evaluation and Review Analyst for AHCA. He holds a B.S. in accounting and is a Florida CPA. Mr. Williamson conducted a review of the financial schedules contained in each of the four applications at issue. In conducting his review, Mr. Williamson compared the applicants’ financial projections with the “peer group” of existing Florida LTCHs. With regard to the MJH projections, Mr. Williamson noted: Projected cost per patient day (CPD) of $1,087 in year two is at the group lowest value of $1,087. Projected CPD is considered efficient when compared to the peer group with CPD falling at the lowest level. The apparent reason for costs at this level are the low overhead costs associated with operating a hospital-within- a-hospital. MJH Ex.34, depo Ex. 4, Page 3 of 5. Mr. Williamson further concluded that MJH presented an efficient LTCH project, which is likely to be more cost- effective and efficient than the other three proposals. In its application, Kindred projected a profit of $16,747 at the end of year two of operation. Schedule 8A listed interest expense "as a way of making a sound business decision." Tr. 1458. Interest expense, however, is not really applicable because Kindred funds new projects out of operation cash flows. If the interest expense is omitted, profit before taxes would roughly $1.5 million. Taking taxes into consideration, the profit at the end of year two of operation would be roughly $1 million. Promise's projections the facility will be financially feasible in the long term are contained in its Exhibit 2, Schedules 5, 6, 7 and 8A and related assumptions. The parties agreed the information contained in Promise's Schedule 5, and the supporting assumptions, were reasonable. Schedule 5 indicates Promise projects an occupancy rate in Year 2 of 76.1 percent, based on 16,660 patient days and an ADC of 45.6 patients. To reach projected occupancy rates, Promise would have to capture roughly 15-17 percent of the LTCH market in Year 2. AHCA concluded Promise's project would be financially feasible in the long term. Only Select questioned Promise's projected long term financial feasibility. The attack, evidenced by Select Exhibits 12 and 14, was composed of a numbered of arguments, considered below: The estimated Medicare revenue per patient projected by Promise was high, and among other factors, erroneously assumed Medicare would increase reimbursement by an average of 3 percent per year. In determining a project's long-term financial feasibility, AHCA looks to the facility's second full year of operation, and, assuming reasonable projections, determines if there is a net positive profit. The analysis AHCA uses to determine the reasonableness of an applicant's projections in Schedules 7A and 8A begins with a comparison of those figures against a standardized grouping developed over the years and consistently applied by the agency as a policy. In this instance, the grouping consisted of all LTCHs operating in Florida in 2004; a total of 11 facilities; eight operated by Kindred and three operated by Select. The analysis is based on Revenue Per Patient Day (RPPD). Promise estimated it would generate an average RPPD of $1,492 in Year 2, and a net profit for the same period of $2,521.327. Using the above process, AHCA concluded that Promise's projected net income per patient day appeared reasonable. At the time of hearing, other Promise facilities were receiving an average RPPD higher than $1,400; compared to the projected "somewhat over" $1,500 it would expect to receive in Year 2 of its Miami-Dade facility. Approximately half of the existing Promise facilities (including West Valley and San Antonio) received Medicare RPPDs in excess of $1,500. As opposed to total revenue per patient, revenue on a per patient day is the one figure associated with the expenses generated to treat a patient on a given day. A comparison of net RPPDs projected by Promise with those of other applicants and the state median indicate Promise's revenue projections are reasonable. While Medicare recently opted not to increase the rate of LTCH reimbursement for the 2006-07 fiscal year, it is the first year in four that the program has done so. Compared to Promise's assumption that Medicare reimbursement would increase yearly by 3 percent on average, Select assumed a rate of 2.4 percent. The ALOS projected by Promise was too long. In projecting need, Select projected an ALOS similar to Promise's projection. Compared with the statewide ALOS of 35 days, Select's is about 28 days. This is the result of a combination of managing patients and their acuity. Assuming Promise's ability to manage patients in a manner similar to Select and achieve a like ALOS, Promise would have room available to admit more patients. There is no reason to assume Promise could not attain a similar ALOS with a similar population than that served by Select; others have done so. Like other segments of the health care industry, LTCH providers will manage patient care to the reimbursement received from payors. The CMI projected by Promise was too high. The prospective payment system is based to a great extent on how patients' diagnoses and illnesses are "coded," or identified, because the information is translated into a DRG, which, in turn, translates directly into the amount of reimbursement received. Each DRG has a "weight." By obtaining the DRG weight for each patient treated in a hospital, one can obtain the average weight, which will correspond to the average cost of care for the hospital's patients. The term for this average is Case Mix Index (CMI). Each year Medicare determines the rate it will pay for treatment of patients in LTCHs, adjusted for each market in the U.S. to account for variations in labor costs. Mr. Balsano assumed the new facility would experience an average CMI of 1.55 and that Medicare would reimburse the facility based on existing rates with an annual inflation of 3.0 percent. Mr. Balsano then reduced the estimated Medicare RPPD generated by those assumptions by 15 percent. While Select's expert criticized Promise's projected CMI adjusted reimbursement rate for Medicare patients (approximately $50,000) as to high, Select's own Exhibit 12, p. 8, indicates a projected reimbursement of $41,120.44 based on an average CMI of 1.0. However, at hearing it was verified that Select's Miami facility operated at an average CMI of 1.23. Applying a CMI of 1.23 generates an average projected Medicare reimbursement of $50,618 per patient, a number similar to that projected by Mr. Balsano. Select Ex. 14, pages 9-16, contains data on, among other things, the CMI of 161 DRGs used by Promise's expert. The data was taken from each of the existing LTCHs in Florida. In 2004, the statewide average CMI was 1.231. Also in 2004, four of 11 LTCHs in Florida experienced an average CMI of 1.4 or higher. Other Florida facilities have experienced an average CMI at or above 1.59. Indeed, other Florida facilities have experienced average CMIs and ALOS similar to that of the Select facility. While Promises operates no facility with an average CMI of 1.55, it has several with average CMIs of 1.3 or 1.4. Promise expects Medicare will take future steps to restrict the admission of patients with lower CMIs' the effect being more complex patients will access LTCHs than currently do, increasing the average CMI in LTCHs. Reducing the number of lower acuity patients admitted to LTCHs in future years will likely increase the CMI of those admitted. There is a direct correlation between CMI and ALOS. If, in fact, the CMI experienced by Promise's facility is less than 1.55, it will in turn generate a lower ALOS. Applying the reduction in reimbursement advanced by Promise's witness (15 percent) would in turn reduce the projected CMI in Promise's facility from 1.55 to 1.05. Because reimbursement coincides with acuity and ALOS, a representation that reducing one of the three does not likewise affects the others is not realistic. Whatever the CMI and ALOS for LTCHs will be in the future will be governed to a great extent by the policies established by the federal government. The federal government's reimbursement system will drive the delivery of patient services and the efficiencies the system provides, so that, in fact, the providers of care manage patients to the reimbursement provided. Whether the average CMI at Promise's facility reaches 1.55 in the future is subject to debate; however, it is reasonable that the status quo will not likely continue; thus, regardless of a facility's current CMI, more complex patients will access the facility in the future. Various sensitivity analyses generated to test the reliability of Select's criticisms in this area do not indicate any material change in the projected Medicare reimbursement. The interest rate on the loaned funds was 9 percent, rather than 7 percent. The estimated expenses did not include sufficient funds to pay the following: the necessary ad valorem taxes the required PMATF assessment the premiums to obtain premises insurance physician fees housekeeping expenses in Year 1 Using the same standardized "grouping" analysis, AHCA calculated Promise's projected costs per patient day and found them reasonable. Because the projected increase in ad valorem taxes and the PMATF assessment will not be payable until 2010, it is not necessary to borrow additional funds to meet these obligations. Select's expert concluded that, depending on a number of scenarios, the result of the appropriate calculations would produce a loss to Promise's project of between $624,636 and $902,361 of year 2. Assuming they represented sensitivity analyses which included various assumptions based on criticisms from Select. The impact of Select's suggested adjustments, reduced by overstated costs in Promise's application Schedule 8A, increased Promise's projected Year 2 net income from the initial estimate of $2,521,327 to $2,597.453. Even if the 15 percent reduction previously included in Mr. Balsano's assumptions on Medicare reimbursement were not considered, and assuming a lower CMI consistent with the existing statewide average (1.43 vs. 1.23), or that Promise's experience in District 11 will be similar to Select's, Promise's facility would still be financially feasible. Select's witness conceded that if Promise's facility experienced a lower ALOS, the demand for additional LTCH services is high enough to allow the facility to admit additional patients ("backfill"). While assuming a lower reimbursement due to lower acuity patients admitted to Promise's facility, Select's witness did not similarly assume any reduction in expenses associated with treatment of such lower acuity patients. In reality, if revenues are less than expected a facility reduces expenses to generate profits. Select's witness also conceded that Promise could reduce the management fee to reduce costs and generate a profit. The testimony of Promise's Chairman, Mr. Baronoff, established the company would take measures to reduce expenses to assure the profitability, including reducing the facility's corporate allocation. Such a reduction by itself would reduce expenses by between $1 million and $1.5 million. Reduction in corporate allocation has occurred before to maintain the profitability of a Promise facility. With regard to Select-Dade, its forecasted expenses, as detailed on Schedules 7A and 8A of its application are consistent with Select-Miami's historical experience in Miami. Evaluation of the revenues and expenses detailed in Select-Dade's Schedules 7A and 8A (and drawing comparison with SMC's 96 other hospitals, with particular attention paid to the Select-Miami facility), its profitability after year one indicates that Select-Dade's project will be financially feasible in the long term. In sum, all four applicants demonstrated long-term financial feasibility. Subsection (7): Extent to Which the Proposal Will Foster Competition that Promotes Quality and Cost-effectiveness Competition benefits the market. It stimulates providers to offer more programs and to be more innovative. It benefits quality of care generally. Competition to promote quality and cost-effectiveness is generally driven by the best combination of high quality and fair price. The introduction of a new LTCH providers to the market would press Sister Emmanuel, Kindred-Coral Gables and Select-Miami to focus on quality, responsiveness to patients and would drive innovations. Approval of any of the applications, therefore, as the Agency recognizes, see Agency for Health Care Administration Proposed Recommended Order, at 36, will foster competition that promotes quality and cost-effectiveness. Competition that promotes quality and cost- effectiveness will best be fostered by introduction to the market of a new competitor: either MJH or Promise. Between the two, Promise's application for 60 rather than 30 beds proposed by MJH, if approved, would capture a larger market share and promote more competition. On the other hand, MJH's because of its long-standing status as a well-respected community provider, particularly in the arenas of cost-effectiveness and quality of care, would be very effective in fostering competition that would promote both quality and cost-effectiveness. Kindred and Select dominate LTCH services in Florida with control over 86 percent of the licensed and approved beds: Kindred has eight existing LTCHs and one approved LTCH yet to be licensed; Select has three existing LTCHs and six approved projects in various stages of pre-licensure development. In 2005 the District 11 LTCH market shares were: Kindred-Coral Gables: 42 percent; Select-Miami: 35 percent; and Sister Emmanuel: 23 percent. Approval of Promise would only slightly diminish Select-Miami’s market share and would reduce Sister Emmanuel to a 16 percent share. A Select-Dade approval would give the two Select facilities a combined 54 percent of the market. A Kindred approval would give its two Miami-Dade facilities a combined 57 percent market share. An MJH approval would give it about 16 percent of the market, Sister Emmanuel would decline to 19 percent and Select-Miami and Kindred-Coral Gables would both have market shares above 30 percent. MJH's application is most favored under Subsection (7) of the Statutory Review Criteria. Subsection (8): Costs and Methods of Proposed Construction The parties stipulated to the reasonableness of a number of the project costs identified in Schedule 1, as well as the Schedule 9 project costs. All parties stipulated to the reasonableness of the proposed construction schedule on Schedule 10 of the application. Those additional costs items on Schedule 1 of the respective applications that were not stipulated to were adequately addressed through evidence adduced at final hearing. Given the conceptual-only level of detail required in the schematic drawings submitted as part of a CON application, and based on the evidence, it is concluded that each of the applicants presented a proposed construction design that is reasonable as to cost, method, and construction time. Each applicant demonstrated the reasonableness of its cost and method of construction. Accordingly each gets credit under Subsection (8) of the CON Statutory Review Criteria. But under the subsection, MJH's application is superior to the other three applications. The subsection includes consideration of "the costs and methods of energy provision and the availability of alternative, less costly, or more effective methods of construction." § 408.035(8), Fla. Stat. As an application proposing an HIH rather than a free-standing facility, not only can MJH coordinate its operations with other types of service settings at expected energy savings, its application involves less construction and substantially less cost that the other three applications. Subsection (9): Past and Proposed Provision of Services to Medicaid and Indigent Patients A provider's history of accepting the medically indigent, Medicaid and charity patients, influences patients and referral sources. Success with a provider encourages these patients on their own or through referrals to again seek access at that provider. As a safety net provider, MJH has a history of accepting financially challenged patients, many of whom are medically complex. Its application is superior to the others under Subsection (9) of the Statutory Review Criteria. Promise does not have a history of providing care in Florida. It has a history of providing health care services to Medicaid and the medically indigent at some of its facilities elsewhere in the country. As examples, its facility in Shreveport, Louisiana, provides approximately 7 percent of its care to Medicaid patients and a facility in California provides about 20 percent of its service to Medicaid patients. MJH committed to the highest percentage of patient days to Medicaid: 4.2 percent. Promise proposes a 3.0 percent commitment; Select-Dade and Kindred, 2.8 percent and 2.2 percent, respectively. Select-Dade's proposed condition is structured so as to allow it to include Medicaid days from a patient who later qualifies as a charity patient, thus accruing days toward the condition without expanding the number of patients served. Select-Dade's targeted service area, moreover, has fewer proportionate Medicaid beneficiaries identified (13 percent) as potential LTCH patients than identified by the methodologies used by the applicants (21 percent), indicating that Select's targeted area is generally more affluent than the rest of the County. Kindred does not have a favorable history of providing care to Medicaid and charity patients. For example, during FY 2004, Sister Emmanuel provided 6.1 percent of its services to Medicaid and charity patients. During this same period, Kindred-Coral Gables provided only 1.08 percent of its services to Medicaid and charity patients. Of all four applicants, Kindred proposes the lowest percentage of service to such patients: 2.2 percent. It has not committed to achieving the percentage upon its initiation of services. Its proposed condition and poor history of Medicaid and indigent care merit considerably less weight than the other applicants and reflects poorly on its application in a process that includes comparative review. MJH's proposed condition, although the highest in terms of percentage, is not the highest in terms of patient days because the facility it proposes will have only half as many beds as the facilities proposed by the other three applicants. Nonetheless, the proposal coupled with its past provision of health care services to Medicaid patients and the medically indigent, which is exceptional, makes MJH the superior applicant under Subsection (9) of the Statutory Review Criteria. Subsection (10) Designation as a Gold Seal Program None of the applicants are requesting additional nursing home beds. The subsection is inapplicable to this proceeding.
Recommendation Based on the foregoing Findings of Fact and Conclusion of Law it is RECOMMENDED that the Agency for Health Care Administration issue a final order that: approves Miami Jewish Home and Hospital for the Aged, Inc.'s CON Application No. 9893; approves Select Specialty Hospital-Dade, Inc.'s CON Application No. 9892; denies Promise Healthcare of Florida XI, Inc.'s CON Application No. 9891; and, denies Kindred Hospitals East LLC's CON Application No. 9894. DONE AND ENTERED this 17th day of May, 2007, 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 17th day of May, 2007. COPIES FURNISHED: Dr. Andrew C. Agwunobi, Secretary Agency for Health Care Administration Fort Knox Building III, Suite 3116 2727 Mahan Drive Tallahassee, Florida 32308 Craig H. Smith, General Counsel Agency for Health Care Administration Fort Knox Building III, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308 Richard Shoop, Agency Clerk Agency for Health Care Administration Fort Knox Building III, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308 W. David Watkins, Esquire Karl David Acuff, Esquire Watkins & Associates, P.A. 3051 Highland Oaks Terrace, Suite D Tallahassee, Florida 32317-5828 Sandra E. Allen, Esquire Agency for Health Care Administration Fort Knox Building 3, Mail Stop 3 2727 Mahan Drive Tallahassee, Florida 32308-5403 F. Philip Blank, Esquire Robert Sechen, Esquire Blank & Meenan, P.A. 204 South Monroe Street Tallahassee, Florida 32301 Mark A. Emanuele, Esquire Panza, Maurer & Maynard, P.A. 3600 North Federal Highway, Third Floor Fort Lauderdale, Florida 33308 M. Christopher Bryant, Esquire Oertel, Fernandez, Cole & Bryant, P.A. 301 South Bronough Street, Fifth Floor Tallahassee, Florida 32302-1110
The Issue Whether Promise Healthcare of Florida III, Inc.'s (Promise) Certificate of Need (CON) Application No. 9870 should be approved to establish a 40-bed freestanding Long Term Care Hospital (LTCH) in Agency for Health Care Administration (AHCA or Agency) Service District 3.
Findings Of Fact Parties AHCA. The Agency for Health Care Administration is the state agency authorized to evaluate and render final determinations on CON applications to pursuant to Section 408.034(1), Florida Statutes. Promise. Promise Healthcare of Florida, III, Inc. (Promise) is the applicant in this proceeding. Promise is a newly-formed and wholly-owned subsidiary of Promise Healthcare, Inc. (Promise Healthcare). Promise Healthcare is a Florida corporation headquartered in Boca Raton, Florida. Promise Healthcare was established in July 2003 when it acquired the assets of Camelot Healthcare. Promise Healthcare owns and operates 13 LTCHs in six other states; seven freestanding facilities and six hospitals- within-hospitals. One additional freestanding LTCH is scheduled to begin operation in 2007 in Bossier, Louisiana. Promise Healthcare does not presently operate any facilities, including LTCHs, in Florida. CON applications and preliminary agency action Promise timely filed an appropriate letter of intent, which contained the information requested by AHCA. Promise timely applied for a CON to establish a 40-bed freestanding LTCH in Lake County, one county of the 16 counties in District 3. The project will consist of 47,951 square feet at construction costs of $11,244,400 and a total project cost estimated at $20,901,826. As a condition of approval, Promise agreed to provide two percent of patient days of Medicaid and charity care. During the same batching cycle, Select Specialty and Leesburg Regional filed CON applications to provide LTCH services in District 3. All applications were deemed complete and comparatively reviewed by AHCA. The numbers and assumptions in Schedule 1 of Promise's CON application represent reasonable projections. The information contained in Schedule 2, 3, and 6, and the assumptions relating to Schedule 3 and 6 represent reasonable projections. The information contained in Schedule 4 is not required. The information contained in Schedules 9 and 10 represent reasonable estimates, including the days required to complete the project. The Agency's review of the CON application complied with all statutory and regulatory requirements. The Agency's review of the CON applications resulted in the issuance of a State Agency Action Report (SAAR) on December 16, 2005. The Agency recommended the denial of the three CON applications. Leesburg Regional and Select Specialty requested formal administrative hearings, but dismissed their cases prior to the final hearing. Promise argues there is a need for additional LTCH beds and services in District 3 and AHCA disagrees. AHCA also argues Promise cannot obtain the required funds to build and operate the LTCH. LTCH services The classes of facilities licensed as hospitals by the Agency, "Class I or general hospitals" include long term care hospitals, which are identified as having an average length of patient stay (ALOS) of 25 days for all beds, Section 408.032(13), Florida Statutes, and also comply with 42 C.F.R. Section 412.23(e)(1994). See Fla. Admin. Code R. 59C-1.002(28). Some hospital patients need acute care services on a long-term basis. A long-term basis is 25 to 34 days of additional acute care service after the typical stay in a short- term hospital. Although some of these patients are "custodial" in nature and not in need of LTCH services, many of these long- term patients may be better served in a LTCH than in a traditional short-term acute care hospital. LTCHs typically furnish extended medical and rehabilitation care for patients who are clinically complex and have multiple acute or chronic conditions. Patients appropriate for LTCH services are differentiated from other hospital patients in that, by definition, they have multiple co- morbidities that require concurrent treatment. Patients appropriate for LTCH services tend to be elderly, frail, and are usually regarded as catastrophically ill. Generally, Medicare patients admitted to LTCHs have been transferred from general acute care hospitals and receive a range of services at LTCHs, including cardiac monitoring, ventilator support, and wound care. These patients require daily physician involvement, extensive nursing care, and appropriate respiratory, occupational, speech, and physical therapies, usually accompanied by some type of technologically advanced support. Quite commonly, the technological support includes a ventilator. The level of care provided in an LTCH is generally analogous to that provided in an ICU (Intensive Care Unit) in a short-term acute care hospital. However, the staff at general acute care hospitals has different orientations than staff at LTCHs. The staff at general acute care hospitals is geared toward shorter lengths of stay (five days or less) than are more typical in an LTCH, where extended lengths of stay are more appropriate. An LTCH is distinguished within the health care continuum by the high level of care the patient requires, the interdisciplinary treatment model it follows, and the duration of the patient's hospitalization, which averages 25 days or more for patients requiring complex medical care. 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, skilled nursing facilities (SNFs), skilled nursing units (SNUs), and comprehensive medical rehabilitation facilities (CMRs). The complex medical, nursing, and therapeutic requirements necessary to serve the LTCH patient with a high acuity level are generally beyond the capability of these other post-acute care facilities on a sustained basis, i.e., 25 days or more. Services provided in LTCHs are also 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. Additionally, most nursing homes provide two to three hours of nursing care per patient per day, whereas LTCHs provide on average in excess of seven hours of nursing care per patient day. Families and other caregivers play a critical role regarding the delivery of care to LTCH patients. Many LTCH patients are elderly and are a special population, with special needs. They commonly have to manage multiple problems, including financial difficulties, drug management, transportation logistics, and sometimes fragile mental and physical conditions. Older patients as well as older caregivers also have a more difficult time driving, for example, two hours and over long distances. The federal government recognition of LTCHs Within the continuum of care, the federal government's Medicare program recognizes LTCHs as distinct providers of services to patients with high levels of acuity. The federal government treats LTCH care as a discrete form of care and provides a separate Medicare payment system of diagnostic related groups (DRGs) and case mix reimbursement that provides Medicare payments at rates different from what the Prospective Payment System (PPS) provides for other traditional post-acute providers. Under the LTCH reimbursement system, each patient is assigned a DRG with a corresponding payment rate that is weighted based upon the patient's diagnosis. LTCHs are reimbursed the predetermined payment rate for that DRG, regardless of the costs of care. These rates are higher than what the federal Centers for Medicare and Medicaid Services (CMS) provides for other traditional post-acute care providers. Effective October 1, 2002, CMS established a new prospective payment system for long-term care hospital providers, the "LTC-DRG". CMS recognizes the patient population of LTCHs as separate and distinct from the population treated by short-term acute care hospitals and by other post-acute care providers, as well as costs of care, resources consumed by the patients and health care delivery. Since the establishment of the PPS for LTCHs, concerns about the high reimbursement rate for LTCHs, as well as about the appropriateness of the patients treated in LTCHs, have been raised by the Medicare Payment Advisory Commission (MedPAC) and CMS. CMS administers the Medicare payment program for LTCHs, as well as the reimbursement programs for acute care hospitals, skilled nursing facilities, and comprehensive rehabilitation hospitals. MedPAC's role is to help formulate federal policy on Medicare regarding services provided to Medicare beneficiaries (patients) and the appropriate reimbursement rates to be paid to the health care providers. The 2006 MedPAC report found that LTCHs were making a good margin or profit, and recommended against an annual increase in the Medicare reimbursement rate for the upcoming fiscal year. In 2006, CMS adopted a reimbursement rate rule for LTCHs for 2007 that did not raise the base rate and made other changes that reflect the ongoing concerns of CMS regarding LTCHs. AHCA's concerns regarding long term care hospitals In deciding on whether to approve or deny new health care facilities, the Agency is responsible for the "coordinated planning of health care services in the state." AHCA looks to federal rules and reports to assist in making health care planning decisions for the state. MedPAC has reported, and CMS has noted that, nationwide, there has been a recent, rapid increase in the number of LTCHs. Nationwide there has also been a huge increase in Medicare spending for LTCH care from $398 million in 1993 to $3.3 billion in 2004. Because of what it perceives to be a lack of specific data from applicants with regard to the composition and acuity level of LTCH patient populations, AHCA is not convinced that there is a need for additional LTCHs in the state. AHCA believes there may be an overlap between the LTCH patient populations and the population of patients served in other healthcare settings, such as SNFs and CMR facilities. AHCA also believes some long-term patients can be appropriately served in the short-term acute care hospitals, rather than requiring LTCH care. In the absence of the applicants better identifying the acuity level of the LTCH patient population, AHCA has reached the conclusion that there may be other health care options available to those patients targeted by the LTCH applicants, and there are enough approved and operating LTCHs in each District of the state, including District 3. Applicable statutory and rule criteria The parties stipulated that Subsections 408.035(1)-(9), Florida Statutes, apply in this proceeding. In addition, in the absence of agency policy regarding long-term care hospital beds and services, the criteria under Florida Administrative Code Rule 59C-1.008(2)(e) 2.a.-d., apply and include consideration of the following topics, except where they are not 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. Population, demographics, and dynamics and medical treatment trends and market conditions Promise plans to develop its LTCH in Lake County and to target and serve Leesburg Regional and its sister hospital, The Village's Regional Hospital, and secondarily to serve the Citrus and Hernando Counties, located to the west of Sumter County and in the most southern portion of District 3. Promise plans to offer a full array of LTCH services. The Leesburg Regional Medical Center, Inc. health care facilities Leesburg Regional (owned by Leesburg Regional Medical Center, Inc. (LRMC)) is a 309-bed (including a 32 intensive care unit (ICU)) acute care hospital located in the City of Leesburg, Lake County, Florida (east of Sumter County). It serves primarily Lake and Sumter Counties. Leesburg Regional provides a variety of primary and tertiary care services, including obstetrics and comprehensive rehabilitation. It is the only provider of neurosurgery and open heart services in the area. LRMC also owns a 120-bed skilled nursing facility located one mile from Leesburg Regional providing oncology services. Leesburg Regional is the safety net provider for the community it serves. There is evidence that patients stay longer at Leesburg Regional, i.e., exceeding an ALOS of five days, which affects not only patient care, but causes the facility to incur additional costs. Leesburg Regional's emergency room relies on the availability of beds to place patients. The inability to move patients from its cardiac intensive care and critical care units to more appropriate settings like an LTCH affects the care patients receive. Sixteen of the 32 intensive care beds are located on the third floor of Leesburg Regional and eight of these beds are used exclusively for open heart patients and seven of these eight beds must be available on an almost constant basis because of the large open heart surgery volume. Adding more ICU beds is not likely to resolve the problems experienced at Leesburg Regional. Fourteen ICU beds, including a four-bed dialysis unit, are located on the second floor and will be expanded to 16 beds in March 2007 with the relocation of the dialysis unit. Patients on the second floor face conditions similar to those patients on the third floor ICU. Leesburg Regional performed approximately 865 open heart surgery cases in 2005 and expects to perform 1,000 to 1,250 open heart surgery cases annually during the next five years. Leesburg Regional ranks the third largest, by volume, open heart surgery provider in the state. At the time of hearing, Leesburg Regional ICU was operating at 100 percent occupancy and was expected to continue at that level until April 2007. Leesburg Regional is designated as a "primary stroke center" by the federal government, serving as the first responder for stroke patients in the LRMC service area. Leesburg Regional is also designated as a Medicare disproportionate share hospital, indicative of a large population of Medicare patients. The average age of a patient treated at Leesburg Regional's cardiac unit is between 75 and 85 years. These patients often have co-morbidities, i.e., multiple health problems, such as congestive heart failure, diabetes, obesity, and respiratory issues. Over the past several years, over 16 percent of the patients served by LRMC's hospitals were Medicaid or low income patients. The patient transport process from Leesburg Regional to an existing LTCH is lengthy and potentially labor intensive. It requires at least 24 hours advance notice to the only emergency medical services (EMS) provider. The planned transport is subject to change due to the number of EMS vehicles available and emergencies. Also, the transferring hospital may need to provide, for example, a respiratory therapist to accompany the patient to an LTCH. Patients referred by Leesburg Regional to LTCHs are usually unable to be weaned from ventilators, often require complex wound care, wound vacs, extensive intravenous (IV) therapy, and extensive therapy care. The Village's Regional Hospital (Villages Regional) also owned by LRMC, is a 62-bed acute care facility, including 12 ICU beds, located within the development known as The Villages. The Villages is located in Lake, Sumter, and Marion Counties (north of Leesburg). By 2008, Villages Regional will expand to a 198-bed hospital. The Villages is a large development which currently comprises approximately 40,000 people and is projected to increase to approximately 100,000 when it is fully built out. The Village's limits residents to persons who are at least 55 years old. By comparison, Weston, Florida, in Broward County, is fully built out at 40,000 residents. District 3 District 3, comprised of 13 counties, is the largest geographical AHCA service district in the state with approximately 11,000 square miles. The main population centers in District 3 are Gainesville, Ocala, and Leesburg. Lake and Sumter Counties are projected to be the fastest growing counties by population in District 3. Lake County is the fastest growing county in terms of resident growth about 38,200 new residents are projected between 2005 and 2010 or about one out of every four new residents in District is projected to reside in Lake County. Promise expects a significant portion of LTCH referrals from Lake and Sumter Counties. Between 2005 and 2010, the 65 and older population is predicted to grow by 14.2 percent statewide, but 21 percent in Lake County and 22 percent in Sumter County. The same age group is projected to grow at 16.8 percent on a district-wide basis. T 359-360; PE 2 at 30. The total population is growing at a high rate and the elderly rate is growing at a higher rate. Due to the location of the Ocala National Forest, the projected growth in District 3 is projected to occur along the southeastern portion of District, i.e., the general corridor of Interstate 4, the Florida Turnpike, and US Highways 441. Other than the primary roads of US Highway 441 and 27, the roads are open with few street lights. Except for Alachua County (mainly because of the University of Florida), there is no mass transportation system available to the residents of District 3. The aging of the population limits somewhat the times of day that family members can travel. The road configuration, travel times, lack of public transportation and the age of the patients and families in the Leesburg service area make access to LTCHs in Gainesville and Ocala difficult. Quantifying the need for additional LTCH beds in District 3 Section 408.035(1), Florida Statutes The Agency has not adopted a need methodology for LTCH services. There is no published fixed need pool for LTCHs. Need is determined on a district-wide basis, here District 3. In order to determine whether there is a need for its project, Promise examined the population estimates and the number of acute care beds for District 3, discharge data from area acute care hospitals, and the lengths of stay of the patients treated at those hospitals. (Since the application was filed, the number of long term stay days of care in acute care hospitals in District 3 increased from 63,429 in 2004 to 68,602 in 2005.) Promise performed its analysis on a district-wide basis and also offered an analysis based on its targeted primary service area Lake and Sumter Counties and secondary area, Citrus and Hernando Counties. Promise used the Geometric Mean Length of Stay + 15 (GMLOS + 15) analysis of long stay patients in acute care short stay hospitals in District 3 to demonstrate need for additional LTCH beds in District 3. The Agency has accepted the GMLOS + 15 methodology to show need for an additional LTCH. See generally Select Specialty Hospital - Escambia, Inc. vs. Agency for Health Care Administration, DOAH Case No. 05-0319CON, 2005 Fla. Div. Adm. Hear. LEXIS 1095 (DOAH June 17, 2005; AHCA July 11, 2005); Select Specialty Hospital - Marion, Inc. vs. Agency for Health Care Administration, DOAH Case No. 03-2483CON, 2004 Fla. Div. Adm. Hear. LEXIS 1658 (DOAH July 14, 2004; AHCA Sept. 15, 2004). Promise identified the number of long-stay patients discharged from the District 3 hospitals as a starting point to quantify the number of patients who have used LTCH services in the past. Long-stay discharges were defined using the following criteria: age of patient was 18 years or older; the discharge DRG was consistent with discharge DRGs from a Florida LTCH; and the ALOS in the acute care hospital was at the GMLOS for the specific DRG plus 15 days or more. Applying these criteria reduced the number of DRGs used and the potential patient pool. In 2004, there were 1,667 acute care hospital discharges from District 3 hospitals that met the above criteria. These 1,667 discharges averaged a length of stay of 38 days, significantly higher than the overall ALOS (approximately five days) for patients discharged from short- term acute care hospitals. Applying the weighted average annual growth rate of the population 18 years or older to District 3 2004 LTCH discharges resulted in a total of 1,978 potential acute care discharges in the five-year planning horizon (2005-2010) and specifically for 2010. Promise then reduced the 1,978 discharges by 25 percent to 1,484, anticipating the effect of CMS policies which, according to Mr. Balsano, would restrict admissions to LTCHs and encourage LTCHs to care for the more complex patients or would provide alternatives to patients who would otherwise be admitted to an LTCH. (A high percentage of LTCH patients are Medicare recipients. Approximately 80 to 85 percent of admissions to Promise Healthcare LTCHs are Medicare patients. Promise projects 87.9 percent of its patient days will be Medicare and Medicare HMO.) Promise held the ALOS of 34.5 days constant and then multiplied the ALOS times the projected number of admissions in 2010 resulting in 51,198 LTCH patient days (1,484 x 34.5). Promise assumed that the proposed facility would experience an ALOS of 34.5 days reflecting 2004 LTCH experience in Florida, and would operate at 80 percent occupancy. These assumptions yielded a need for 175 LTCH beds by 2010. However, the net bed need by 2010 is 100 after deducting 75 LTCH beds which are approved and operational in District 3. If the number of potential LTCH patients is reduced by 50 percent rather than 25 percent and the occupancy standard is increased to 85 percent, the methodology yields a need for approximately 47 beds (122 minus 75 beds). On Schedules 7B and 8A for the second year of operation, Promise projected it would provide 11,216 patient days with an average occupancy of 76.8 percent in order to achieve a net profit of $1,048,236 in year two of operation. In order to break even in year two, Promise used a sensitivity analysis and updated data and projected it would need approximately 9,551 patient days at an ALOS of 34.5 or approximately 17 to 19 percent of the total projected patient days for District 3 (56,090). T 475, 489, 820. Promise assumed a loss of approximately 15 percent of patient days from the 11,216 patient days projected. T 450-451. Stated otherwise, Promise needs approximately 276 discharges by year two (2010) to break even. T 472-475, 825. Notwithstanding the above, the Agency is concerned with the type of patients (acuity levels) who are appropriate for LTCH services, suggesting in part that admitted LTCH patients are not always appropriate for that level of care and that LTCH applicants have traditionally overstated the need for additional LTCH beds. The basis of this concern lies in part on information obtained by the Agency from CMS which, as interpreted by the Agency, indicates that 37 percent of patients historically admitted to LTCHs are short-term outliers, and that 29 percent of patients admitted were totally inappropriate for such admissions. In 2002, CMS adopted a rule intended to assure that if an LTCH did not expend funds to treat a patient, CMS would receive the benefit of the lower cost and/or more efficient care. For reimbursement purposes, the rule defined short-term as the admission of any patient who stayed in an LTCH no loner than 5/6ths, or 83 percent, of the ALOS at a particular LTCH. The rule generates a number of short-term outliers equal to approximately 37 percent of all admissions. Mr. Balsano considered this rule in reducing the potential number of admissions at the District 3 LTCHs in the future by 25 percent as well as reducing the projected Medicare reimbursement rates the proposed facility would receive. The Agency also relied on a 2004 study of LTCH admissions which found that 29 percent of the patients were inappropriate for hospital care and that CMS supported the study's findings. However, within the same exhibit, CMS recognized that 17.4 percent of the admissions reviewed in the survey involved payment errors and only 5.9 percent involved admission denials. Payment errors involve billing and coding errors not whether an admission was appropriate. The percentage of admission or payment denial claims for acute care hospitals during 2004 was 4.6 percent, similar to the LTCH percentage. In order to ensure patients are appropriate for LTCH care, Promise has implemented a number of programs including an increased level of patient admission scrutiny. Promise uses a standardized model for its admission criteria known as InterQual. InterQual is a set of measurable, clinical indicators that reflect a patient's need for hospitalization. Rather than being based on diagnosis, InterQual criteria consider the level of illness of the patient and the level of services required. InterQual measures the acuity levels of LTCH patients. Using InterQual criteria provides some assurance that patients will be suitable for LTCH care, i.e., actually need LTCH care because of the severity of the patient's illness. InterQual criteria include discharge screens, which require an LTCH to continuously determine whether a patient can be more appropriately cared for in an alternative setting. Approximately 35 to 60 percent of patient referrals are admitted at Promise Healthcare LTCHs. Promise has used InterQual criteria since the summer of 2004, and at the time of hearing, used the criteria in all of its facilities. (However, Promise did not specifically offer proof of actual results of using InterQual in its LTCHs. T 718.) Thirty-nine of 52 Quality Improvement Organizations (QIO) have adopted InterQual as their criteria to review the appropriateness of LTCH admissions. Notwithstanding Promise's use of InterQual criteria, the Agency is still concerned regarding inappropriate admissions at Promise's proposed facility as well as existing LTCHs. But see AHCA 13. During a 2004 site visit to SemperCare Hospital in Orlando (Select Specialty Hospital - Orlando), AHCA obtained information which identified patients' severity of illness. The Agency viewed this report favorably and suggested that this information is the type of evidence which should be included in a LTCH CON application to establish the segment of patients appropriate for admission to an LTCH. See AHCA 17. During rebuttal, Promise called Dr. Grigonis as a witness. Dr. Grigonis' firm was retained by SemperCare to generate the data and analysis presented to the Agency in AHCA Exhibit 17. Dr. Grigonis was retained by Promise to analyze data similar to AHCA Exhibit 17. T 742-743. According to Dr. Grigonis, the severity of illness of a patient can be determined by using a product developed by 3M Corporation and is part of a system known as APR-DRG (all patient refined (APR)) patient classification system. The DRG is the standard classification system developed by Medicare and others to classify patients into broad diagnostic groups. The APR is a further refinement of the DRG system and is used by CMS and is in use in many acute care hospitals in the country. A patient's severity of illness is broken down into four categories: minor, moderate, major, and extreme and ranked from a low of one to a high of four along these categories. Patients with the highest severity (major and extreme/ three and four) have the highest probability of being admitted into an LTCH. Dr. Grigonis was asked to determine the severity of a potential pool of LTCH patients considering only Medicare patients. First, he used a database of Medicare patients known as MedPAR data and for the period October 1, 2004 through September 30, 2005. Second, filters were applied to the data based upon both clinical and length of stay information. Only patients who stayed the GMLOS plus 15 days for similar patients in that DRG category were considered. This filter was provided by Mr. Balsano. Patients were also eliminated who based on their DRG were not appropriate for an LTCH. Patients who expired were also eliminated. When these stringent filters were applied to the MedPAR data from 2004-2005, the potential patient population was reduced to less than one half of one percent of all Medicare patients discharged from the four acute care hospitals studied, Leesburg Regional, Villages Regional, Florida Hospital Waterman, and Citrus Memorial Hospital,1 which are not LTCHs, serving Lake, Sumter, Citrus, and Hernando Counties within District 3. T 751-753, 794. This exercise was followed to determine the reasonable number of candidates who were appropriate for admission to an LTCH from this geographic area. Once the patients were separated using the filters, Dr. Grigonis applied the same algorithm obtained from the 3-M Corporation to determine the APR/DRG and severity of illness, i.e., determine the proportion of patients falling within each severity level. This is the same methodology he used for the SemperCare study. The data generated by Dr. Grigonis indicated that during October 2004 to September 2005, the four acute care hospitals generated 192 Medicare only filtered patients who would be appropriate for admission to an LTCH (but did not receive LTCH care) and 87.5 percent of the 192 patients were in the major and extreme severity of illness categories - levels three and four. (Patient origin data was not included as part of the data or analysis.) Of the 192 patients, Leesburg Regional, Villages Regional, Citrus Memorial, and Florida Hospital Waterman accounted for 69, 13, 60, and 50 patients, respectively. Leesburg Regional and Villages Regional had a very high proportion of high severity cases. Dr. Grigonis compared the SemperCare study (AHCA 17) with his recent four-hospital study (Promise R 3) and concluded that the population of patients selected for the four-hospital study had an overall higher degree of severity than patients who were typically treated at the SemperCare LTCH. T 760. Dr. Grigonis ultimately concluded that the results from the four-hospital study indicate a group of patients that are suitable for LTCH admissions. T 765. Dr. Grigonis also opined that the range of the total number of patients appearing on Promise Exhibit 29 is consistent with his findings. T 767-777.2 He also stated that "over time there are likely to be higher severity patients in certain areas, and that would also be consistent with the fact that the data we analyzed from the pure Medicare file was a year older." T 768. During its surrebuttal case, in response to Promise Rebuttal Exhibit 3, the Agency reviewed Florida statewide LTCH actual discharge data for calendar year 2005. This exhibit provided the number of discharges/cases and percentages by DRG code of patients who were discharged from an LTCH. AHCA SR 1. The data was not limited to Medicare patients. T 830-849. The Agency compared this data set with Promise Rebuttal Exhibit 3 and concluded that many of the patients treated at the four acute care facilities would not necessarily be candidates for LTCH care. For example, for LTCH statewide discharges, approximately 50 percent of the discharges were from DRGs 475 (respiratory system diagnosis with ventilator support/22.40 percent); 87 (pulmonary edema and respiratory failure/8.55 percent); 271 (skin ulcers/5.97 percent); 416 (septicemia age greater than 17/4.44 percent); 88 (chronic obstructive pulmonary disease/4.26 percent); and 79 (respiratory infections and inflammation greater than age 17 with CC/3.91 percent). AHCA SR 1. For these same DRGs, for the patients discharged from the four acute care facilities, the percentages were: 1.6 percent/475; 1.04 percent/87; 0 percent/271; 6.2 percent/416; 2 percent/88; and 5.2 percent/79. Only three patients were classified within DRG 475. The percentages of patients classified within DRGs 416, 88, and 79 are closer in comparison than the others. The Agency reiterated that the purpose of this surrebuttal exhibit was to show the DRGs of patients actually admitted to an LTCH. T 849. The data used by Dr. Grigonis is different from the data used by the Agency, in part, because the patient pools are different and the point of discharge is different. T 847. The Agency also pointed out that, although there has been a steady increase in the state's population for the past ten years, and an increase in the number of LTCHs starting in 2003, there has been a decline in the utilization of LTCHs. See, e.g., AHCA 13 and 14. (The expansion of LTCHs statewide has been significant. As of 2003, there were 740 licensed LTCH beds in 10 facilities statewide. The number of beds rose to 876 by 2005 in 14 facilities and, as of July 2006, there were 475 additional LTCH beds approved statewide in nine facilities. Id.) Past occupancy numbers assist somewhat in predicating future need for future health care services in general. However, the needs of each service district should be analyzed, not lumped into one category of need because of the wide variation in bed occupancy in the various districts and because the numbers do not necessarily indicate whether access or bed availability is a problem. Despite the Agency's concerns, Promise's methodology is more conservative than those applied in other LTCH CON applications in part because Promise considered the potential impact of future CMS actions which would tend to remove the lower acuity patients from being admitted or considered appropriate for admission to LTCHs. Also, the study performed by Dr. Grigonis, see also endnote 2, is persuasive that there is a pool of potential patients who need LTCH services within Promise's service area. Availability, extent of utilization, accessibility, and quality of care of like services in District 3 - Section 408.035(2), Florida Statutes LTCHs Kindred - Hospital East (Kindred Marion) is the only operational LTCH in District 3 and is located in Ocala, Marion County, Florida, north of Lake and Sumter Counties. The drive time from Leesburg Regional to the Kindred Marion LTCH is approximately one hour to one hour and 15 minutes. (It takes approximately 30 minutes to drive from Leesburg Regional to Villages Regional and thus approximately 30 minutes north to Kindred Marion from Villages Regional.) Kindred Marion is a "hospital within a hospital" (HIH) because it is physically located in Munroe Regional Medical Center (Munroe Regional), an acute care hospital. Kindred Marion has 30 beds in 15 semi-private rooms and one bed in an isolation room. Kindred Marion's short-term hospital Medicare provider number became effective November 30, 2005, and its long-term hospital Medicare provider number became effective June 1, 2006. Notwithstanding its status as an HIH, it is separately licensed.3 From October 2005 through November 2006, Kindred Marion reported 147 admissions to its LTCH, of which 42, 21, and 20 came from Munroe Regional, Shands Hospital, and Ocala Regional, respectively, and no referrals from Village's Regional and two admissions from Leesburg Regional. During the same time period, Kindred Marion received 6.8 and 2.7 percent of its patients from Lake and Sumter Counties, respectively, or about two patients a day from these two counties.4 Since it opened, Kindred Marion's average daily census (ADC) has experienced an upward trend. By October 2006, Kindred Marion's ADC reached as high as 20 to 22 patients, with an occupancy rate of approximately 71 percent. Utilization and inefficiencies may result when a facility is composed solely of semi-private rooms. These include assuring that patients of different sexes are not housed in the same room; that contagious diseases or infections are appropriately considered; the location of additional equipment, such as monitoring devices, for patients with co-morbidities; and that the preferences of patient's families are taken into consideration. In light of some these limitations, unlike facilities with private bed configurations, facilities with semi-private bed configurations have more limited capacity. Kindred Marion's realistic occupancy threshold is approximately 70 to 75 percent. (However, Kindred Marion can add LTCH beds without CON review by notifying the Agency of the addition. Kindred Marion can also become a freestanding LTCH without CON review. It is not certain whether Kindred Marion intends to expand or whether it has the capacity to expand or whether it intends to become a freestanding LTCH.) There is some evidence that Leesburg Regional's personnel have not been successful in placing patients at Kindred Marion, although the attempts were not quantified with any precision. See, e.g., T 224-225, 241-242. One additional LTCH (Select Alachua) has been approved and is under construction by Select in Gainesville, Alachua County, Florida, close to the Shands Hospital System (Shands). It will be a freestanding LTCH with 44 beds and is expected to open in 2008. Select Alachua was approved on the basis that the majority of its patients would be generated by Shands and would serve patients in Alachua County and surrounding counties. AHCA expects Shands will refer the large majority of its patients requiring LTCH services to Select Alachua. See Finding of Fact 119 regarding the ability to expand. The drive time to Select Alachua from Leesburg Regional is approximately one hour and a half to two hours. Kindred Marion and Shands Alachua are closer to Leesburg Regional and the residents of Lake and Sumter Counties than the Kindred Tampa LTCHs. However, given the relationship between Shands and Select Alachua, the evidence is not persuasive that Select Alachua will be a viable alternative for the residents of Lake and Sumter Counties and potential patients from, for example, from Leesburg Regional and Villages Regional, needing LTCH services. CMRs, SNFs, SNUs, and Home Health Agencies The CMR unit in the Leesburg Regional service area has not been available for ventilator patients. Also, CMR units are rarely appropriate for the LTCH patient, in part, because LTCH patients are not able to tolerate the minimum three hours of daily therapy associated with CMR care. The services offered at SNFs and SNUs and by home health agencies in District 3 are not appropriate substitutes for the services offered at an LTCH and needed by typical LTCH patients. LTCH services outside District 3 For calendar year 2004 and prior to the operation of an LTCH in District 3, over half (143 out of 259) of the District 3 resident/patients receiving LTCH services were discharged from the Kindred - North Florida LTCH in Green Cove Springs, Florida. An additional 25 percent of the District 3 resident/patients were discharged from Kindred - Central Tampa and Kindred - Tampa. Kindred Hospital - Bay Area - Tampa (Kindred Tampa), in District 6 to the west of District 3, is an existing LTCH with 73 licensed LTCH beds. Patients at Leesburg Regional requiring LTCH services are transported two hours away to the Kindred Tampa facility. Kindred Tampa terminated its family bus shuttle service in the Leesburg Regional service area so family members (of patients admitted to Kindred Tampa) must provide or find transportation and travel two hours each way to visit patients at Kindred Tampa. Patients are transported by EMS. Kindred also operates another LTCH in District 6 known as Kindred Hospital - Central Tampa with 102 licensed LTCH beds. For calendar years 2002 through 2005, the occupancy levels for Kindred Tampa were 67.50, 65.93, 62.64, and 59.49, respectively. For the same time period, the occupancy levels for the Kindred Hospital - Central Tampa LTCH were 79.42, 70.33, 69.52, and 63.05, respectively. (Both Kindred LTCHs have been operational at least since 1995.) The two Kindred LTCHs in the Tampa area are potential alternatives for the residents in Lake and Sumter Counties, but the driving time to and from these facilities is problematic both for the patients and caregivers and costs are incurred by the transferring facility. For example, a respiratory therapist from Leesburg Regional will often accompany the patients, which keeps the therapists out of the hospital for six to seven hours. T 218. There is one LTCH (opened in 2003) in Orlando known as Select Specialty Hospital - Orlando (formerly SemperCare - Orlando), with 35 LTCH beds and another LTCH under construction in southern Orange County within a few miles of the current LTCH known as Select Specialty Hospital - Orange with 40 approved, but not operational LTCH beds. The approximate drive time from Leesburg Regional to the Orlando LTCH is approximately one hour and 20 minutes and one hour and 50 minutes to the LTCH (Orange County) under construction. For calendar year 2004 and 2005, the occupancy levels at the Select Specialty - Orlando LTCH were 71.28 and 73.37 percent, respectively, compared to the statewide average of 67.14 and 64.70 percent, respectively. Select Specialty - Orlando is associated with the Florida Hospital5 System, whereas the approved Select Specialty Hospital - Orange facility is associated with the Orlando Regional Health System. Both LTCHs were approved in part based on patients residing in the Orlando/Orange County area within District 7 needing LTCH services and on the specific needs of these large hospital systems and the patients they serve.6 T 627, 690, 700-705, 709, 729-731. (Select Specialty - Orlando is an LTCH within the multiple buildings of Florida Hospital in Orlando. T 627). Given the special relationships forged by these Orlando/Orange County LTCHs with existing health care systems, the evidence is not persuasive that they are viable alternatives for the residents of Lake and Sumter Counties or, for that matter, other residents in District 3, except perhaps LTCH eligible patients from Florida Hospital Waterman. Also, Leesburg Regional has tried to place patients with the Select Orlando but bed availability has been a problem. Travel is also problematic. As of July 2006, University Community Hospital, Inc. has been approved to operate a 50-bed LTCH in Pasco or Pinellas County, Florida. T 714. The record is scanty on any details regarding this facility and its proposed service area. The ability of Promise to provide quality of care - Section 408.035(3), Florida Statutes Promise is a new development stage corporation without a track record in Florida. However, Promise demonstrated that it can provide quality of care should its project be approved and that its parent company has a history of providing quality of care. About one-half of Promise Healthcare's facilities are accredited by the Joint Commission of Accreditation of Health Care Organizations (JCAHO). Promise Healthcare expects the remaining facilities will be similarly accredited within the next 18 months. LTCHs must go through a six-month demonstration period (unless extended) when they are treated like an acute care hospital. During this period, they must demonstrate that they are caring for medically complex patients who have an ALOS of more than 25 days. The hospital is reimbursed at the acute care hospital rate. In addition to JCAHO accreditation and the use of InterQual admission measures indicated above, Promise Healthcare utilizes a number of other outcome measurement systems, including JCAHO's ORYX performance measurement system, medication error rate determinations and best practices standards. ORYX is a national clinical outcome database operated by Healthcare Data, Inc., under a contract with JCAHO, which enables providers like Promise to evaluate and compare themselves to others in the industry by reporting indicators, such as infection control, ventilator dependency and weaning, and wound healing. In order to maintain JCAHO accreditation, JCAHO requires LTCH facilities to report nine indicators on a quarterly basis. T 41-48. The availability of resources, including health personnel, management personnel, and funds for capital, and operating expenditures, for project accomplishment and operation-- Section 408.035(4), Florida Statutes The parties agree that Promise has or will be able to recruit or otherwise obtain sufficient resources, including health and management personnel, to accomplish the project. The parties have differing views on the availability of funds for capital and operating expenditures discussed herein. The extent to which the proposed services will enhance access to health care for residents of the service district -- Section 408.035(5), Florida Statutes Based primarily on the experiences of personnel at Leesburg Regional, the need assessment performed by Mr. Balsano, and the study performed by Dr. Grigonis, see also endnote 2, approval of the project is likely to enhance access to LTCH services for the residents of District 3. The immediate and long-term financial feasibility of the proposal -- Section 408.035(6), Florida Statutes Promise is required to prove its project will be financially feasible in the short-term by establishing its ability to fund the project, and in the long-term by establishing a positive net revenue or profit at the end of the second full year of its projected operation. In other words, can Promise obtain financing to fund the project and is the project likely to generate a profit at the end of its second year of operation? AHCA argues that Promise is a development stage company with assets of $60,000 and no results from operations. In addition, AHCA argues that Promise did not provide audited financial statements for its parent company, Promise Healthcare, and that, as a result, AHCA cannot perform a review of Promise's short and long-term position. In essence, AHCA questions Promise's ability to obtain financing necessary to fund this project and associated working capital. Promise noted that it intends to fund the project through debt financing and provided a letter of interest. AHCA does not consider a letter of interest a firm commitment to lend. Promise is a start-up company. Promise included an audited financial statement of itself in its CON application, but not for its parent company. There is nothing unusual about establishing a separate start-up company. At the time of the hearing, Promise Healthcare, the parent company, generated net patient revenues in excess of $200 million. Promise Healthcare is viable and profitable, as indicated in its financial documents and financial history. With the exception of its first year of operation, Promise Healthcare has been profitable, and, in 2005, Promise Healthcare generated positive retained earnings in excess of $10 million. In 2006, Promise Healthcare's profits "are just about the same for 2005." Mr. Leder stated that there were some changes in reimbursement that lowered some of the revenue and also one company was in a start-up mode with a six month loss, which was absorbed, but if considered a start-up, net income for the year would have been approximately $5 million. T 147. Mr. Leder explained that audits are being done now which "hopefully will be completed within the next six or eight weeks." He further explained that when Promise Healthcare purchased other companies, there was a significant amount of financial information unavailable to list on statements which were auditable. Nevertheless, he opined that their financial statements "are fairly accurate" and the balance sheet and income statement have "always been fair and reasonable." T 148. Through the years, Promise Healthcare has been successful in securing financing as needed. Promise Healthcare's "sister" company, Sun Capital, obtained in excess of $250 million in loans through the efforts of Founding Partners Capital Management Company (Founding) and its principal, Mr. Gunlicks. Founding acts as the general partner in managing investment funds. Founding is registered with the U.S. Securities and Exchange Commission, the federal Commodities Futures Trading Commission, and the Florida Division of Securities. Promise Healthcare has obtained approximately $15 million in loans from Founding for two facilities in Nederland (by mortgage) and Bossier City (by construction loan). T 148-149. It is not uncommon for health care companies to rely on letters of intent to finance their facilities at this stage of the CON process. Mr. Gunlicks and his associates have conducted extensive due diligence into Promise Healthcare's plans to expand in Florida. If approved, Mr. Gunlicks and the entities that he controls stand ready, willing, and able to provide the necessary financing for the Promise project. According to AHCA's financial expert, Mr. Fitch, there has not been a CON project which, if approved, was not developed due to lack of financing. If the Agency approves the application, it is reasonable to expect that the project will be financed appropriately. Promise produced credible evidence in this regard. Promise, by and through the testimony of witnesses employed at Leesburg Regional, proved that it had strong support from LRMC. Promise's projected occupancy rates are based on the methodology proposed by Mr. Balsano, including the adjustments contained in the projected number of patients who would likely be admitted to District 3 LTCHs. The projected occupancy rates for years one and two, although presenting a challenge for Promise in today's LTCH/health care climate, are reasonable. The projected Medicare revenues as well as the overall net revenues per patient day included in the application are reasonable. The proposed costs per patient day are reasonable. Overall, the projections that Promise's project will at least break even and potentially generate a profit in excess of $1.2 million at the end of the second year of operations, although challenging, is reasonable. The Agency raised legitimate concerns regarding the financial ability of Promise and its parent. The lack of audited financial statements for the parent is troubling, but not dispositive. The financial ability of Promise and its parent to fund and operate this project presents a credibility issue in this de novo hearing. Based on the totality of the evidence presented, Promise proved by a preponderance of the evidence that the project is likely to be financially feasible in the short-term and long-term. The extent to which the proposal will foster competition that promotes quality and cost-effectiveness -- Section 408.035(7), Florida Statutes Approval of Promise's application would provide competition in the District 3 LTCH market and reduce expensive and time consuming patient transfers. In addition, it is likely to provide efficiencies in various departments of hospitals, such as those operated by LRMC. Access to the proposed Promise project is likely to decompress LRMC's emergency departments and intensive care units, reduce hospital stays, and provide better care for patients. The costs and methods of the proposed construction, including the costs and methods of energy provision and the availability of alternative, less costly, or more effective methods of construction -- Section 408.035(8), Florida Statutes The parties agree that the estimated construction costs of the Promise project are reasonable and that the architectural plans submitted by Promise comply with all statutory and rule requirements. The applicant's past and proposed provision of health care services to Medicaid patients and the medically indigent -- Section 408.035(9), Florida Statutes If awarded a CON, Promise agreed to provide a combined two percent of the facility's total annual patient days to Medicaid and charity patients. At hearing, Promise reiterated its commitment contained in the application. In some states in which Promise operates, the Medicaid program does not provide any benefits for LTCH patients. In its Shreveport facility, the percentage of Medicaid patients averages between five and ten percent. It is even higher in Promise's Phoenix facility. Promise provides care to patients who do not have any reimbursement available-allowing its facilities to do so on a case-by-case basis. Promise's commitment to Medicaid and charity care is accounted for in the application's projections. Promise's commitment compares favorably with the level of similar care provided by existing LTCH facilities in Florida. The applicant's designation as a Gold Seal Program nursing facility pursuant to s.400.235, when the applicant is requesting additional nursing home beds at that facility -- Section 409.035(10), Florida Statutes The parties agree this criterion does not apply.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Agency for Health Care Administration issue a final order granting Promise Healthcare of Florida III, Inc.'s CON Application No. 9870. DONE AND ENTERED this 10th day of April, 2007, in Tallahassee, Leon County, Florida. S CHARLES A. STAMPELOS 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 10th day of April, 2007.
The Issue At issue in this proceeding is whether Petitioner is liable for overpayment of Medicaid claims for the period of January 1, 2000, through December 31, 2001.
Findings Of Fact Based upon the stipulations of the parties and the evidence presented at the hearing, the following relevant Findings of Fact are made: AHCA is the state agency charged with the regulation of the Medicaid program in the State of Florida, and has the authority to perform Medicaid audits and recover overpayments, pursuant to Section 409.913(2), Florida Statutes (2001). Petitioner is a Florida, not-for-profit corporation that was enrolled as a Medicaid provider during the audit period of January 1, 2000, through December 31, 2001. At all times relevant to this proceeding, Petitioner was authorized to provide medical services to Medicaid recipients. The selection of records for and the conduct of the audit was not a matter of controversy between the parties. The records in this case were requested within the five-year window for record retention and agency investigation provided by Section 409.913(8), Florida Statutes (2001). Medicaid pays a per diem rate for inpatient hospital care and treatment. This per diem payment covers all services and items furnished during a 24-hour period. The audit in this case dealt exclusively with inpatient services at Florida Hospital Orlando. Deborah Lynn, a medical health care program analyst for AHCA, reviews the agency's inpatient hospital audits. The agency selects a hospital for audit on a random basis, then selects for review, a random sample of that hospital's patients admitted during the audit period. AHCA sends a demand letter to the hospital, which then sends the relevant patient records to AHCA. Florida Hospital Orlando had to be granted a 30-day extension, but eventually provided, to AHCA, all of the requested patient records. The hospital's records were first provided to a nurse consultant, who reviewed the records and made a suggestion as to the number of days for which Medicaid reimbursement should be denied for each patient. The nurse consultant's suggestions were then calculated into an initial overpayment amount and included in the PAAR that was sent to Petitioner on January 20, 2005. Ms. Lynn acknowledged that the PAAR constituted the first notice to Petitioner that AHCA disputed the length of stay for some patients based on medical necessity. After receipt of the PAAR, Petitioner was given the opportunity to examine the days that AHCA preliminarily questioned and to provide additional information in defense of its Medicaid billings. Petitioner did, in fact, submit additional documentation. The records, the additional documentation, and the nurse consultant's recommendations were then forwarded to a peer reviewer; a physician who uses his or her medical expertise to determine the medical necessity of the services provided. In this case, AHCA employed the services of two peer reviewers. Dr. Laura Machado was the peer reviewer for the inpatient medical cases, and Dr. Rahul Mehra was the peer reviewer for the inpatient psychiatric cases. The peer reviewers prepared reports that offered their opinion as to which days of the patients' stays were medically necessary. Ms. Lynn then calculated the amount of the alleged overpayment and communicated that number to Petitioner in the FAAR on August 20, 2005. General issues regarding patient discharge Ms. Lynn emphasized that the peer reviewer's job is simply to determine medical necessity, not to base coverage decisions on the convenience of the patient or provider. She rejected the suggestion that it was any part of the peer reviewer's task to consider what medical facilities and services are actually available in Orlando, at the time that Florida Hospital Orlando is contemplating the discharge of a Medicaid patient. Ms. Lynn reiterated that hospital services under Medicaid are governed by the Hospital Services Coverage and Limitations Handbook, the January 2001 edition, of which sets forth the following "Service Requirements": Medicaid reimburses for services that are determined medically necessary, do not duplicate another provider's service, and are: individualized, specific, consistent with symptoms or confirmed diagnosis of the illness or injury under treatment, and not in excess of the patient's needs; not experimental or investigational; reflective of the level of services that can be safely furnished, and for which no equally effective and more conservative or less costly treatment is available statewide; furnished in a manner not primarily intended for the convenience of the recipient, the recipient's caretaker, or the provider. The fact that a provider has prescribed, recommended, or approved medical or allied care, goods, or services does not, in itself, make such care, goods or services medically necessary or a covered service. Dr. Ross Edmundson, Florida Hospital Orlando's medical director for health care management, agreed that a clinically stable patient can be sent home or to a skilled nursing facility "in the best of all worlds." However, Dr. Edmundson pointed out that a safe discharge plan is the primary consideration for the treating physician, and that a safe discharge plan may render things "medically appropriate" that might not be found "medically necessary" in purely clinical terms. For example, if the Medicaid patient has no home, and the physician knows full well that the patient will be living under a bridge if he is discharged to the street, then it would be grossly negligent to discharge that patient without a plan to get better care. If the patient requires $500.00 worth of medications for the next month, and has no way of obtaining them, it may be grossly negligent for the physician to send that patient out the hospital door. The InterQual Products Group's "ISD" (Intensity of Service, Severity of Illness Discharge Screens) is a nationally recognized set of utilization guidelines that are used by hospitals, Medicare and Medicaid. Dr. Edmundson believed that these discharge guidelines take into account factors beyond the purely clinical: You have to have a discharge treatment plan. If you have a place to follow up, if you have [an] accepting physician, if you have a plan, if you know that they can make it there in the next three days and they require follow up in three days, that is clinically appropriate. If you know that they're going to live in a box and you have no assurance that this patient is going to make it back for any follow up, then that is not a safe discharge plan, it is not medically appropriate. The criteria really has that built in. There are barriers that may prevent the hospital from discharging the patient to an alternative location. Even if the hospital has met the patient's immediate medical needs and the patient's condition is stable, the patient's financial situation may present an obstacle to placement in a skilled nursing facility or to obtaining home health care. Petitioner has access to "a very limited number" of providers who will give follow up care to Medicaid and self-pay patients. Most nursing homes in central Florida have a set number of Medicaid beds for which they will accept patients. Medicaid will not pay a sub- acute facility or a nursing home any additional money for expensive medications, such as intravenous (IV) antibiotics, which further discourages those facilities from accepting patients upon their discharge from the hospital. Florida Hospital Orlando has severe capacity problems. The hospital has about 1,800 beds on its seven campuses, and these beds are almost always full. It is not unusual for the hospital's daytime census to be above 100 percent, with patients backed up in the emergency room waiting for beds to become available. Given the hospital's capacity problems and the low reimbursement rates of Medicaid, there is no motive for Petitioner to keep Medicaid patients in the hospital any longer than is absolutely necessary. Tammy Rikansrud is Florida Hospital Orlando's director of case management, utilization management, and denial management. According to Ms. Rikansrud, the hospital begins discharge planning within 24 hours of a patient's admission, and immediately begins seeking referrals as soon as its knows the patient will be discharged to a sub-acute facility. A major problem is that many facilities limit the number of beds for Medicaid patients, if they accept Medicaid patients at all. Robert Fleener is Petitioner's director of case management. Based on Mr. Fleener's testimony, the impediments to placing a patient in a nursing home from the hospital include rejection of the patient for "payer constraints," meaning that the Medicaid system does not reimburse the nursing home enough to cover its costs. Petitioner seeks placement for its Medicaid patients throughout the state of Florida and beyond, if necessary. Aside from nursing homes, Petitioner uses Shands Teaching Hospital in Gainesville, and long-term acute care hospitals such as those operated by Kindred Healthcare. Petitioner always assesses the practicality of home care for its patients. Petitioner also seeks to place its patients at outpatient clinics when appropriate, but there are few clinics in Orlando, and they are not required to accept Petitioner's discharged patients. The usual practice with ambulatory patients is to discharge them from the hospital, then have them come back on an outpatient basis for follow up treatment. Petitioner always assesses its patients to achieve discharge to the least restrictive setting. If the patient is ambulatory, the hospital will seek to place the patient in an assisted living facility. If the patient is not ambulatory, the hospital will look to a skilled nursing facility that is able to provide the necessary level of service. As also noted by Dr. Edmundson, a patient's need for expensive medicines can make it difficult to place the patient in a nursing home because of reimbursement problems. According to Mr. Fleener, nursing homes look very closely at accepting homeless patients, because they assume that if they accept a homeless person, they will have that patient for the rest of his life. Petitioner has "unavoidable" inpatient days for its Medicaid patients, where the care could have been provided in a nursing home but the hospital was unable to place the patient, due to lack of beds, patient behavior problems, age, or the cost of clinical care. Stephen William Bailey is the clinic coordinator for Petitioner's department of psychiatry. Petitioner has a 76-bed psychiatric inpatient unit and a medical psychiatric unit for patients with a psychiatric diagnosis and co-morbid medical problems. The hospital must obtain state approval to place a psychiatric patient in a nursing home, assuming it is possible to find a nursing home that is equipped to handle psychiatric issues and is willing to take the patient. It is not uncommon for Petitioner to admit a patient with behavior disturbances from a nursing home, stabilize the patient's medications and treatment, then have the nursing home refuse to take the patient back. Petitioner has no incentive to prolong inpatient stays. Patients are waiting to be admitted on a regular basis, and there are, at times, as many as 20 psychiatric patients in the emergency room waiting for admission. Cara Lee Staples is a social worker at Florida Hospital Orlando. She attends treatment team meetings, assesses patients for discharge needs, researches placement availabilities, and meets with families to plan care after discharge. Based on Ms. Staples' testimony, many of Petitioner's Medicaid patients have chronic mental illnesses and often have no involvement with their families. As soon as a patient is admitted, Ms. Staples seeks to obtain a psychosocial history of the patient, which includes where the patient came from and whether the patient can return upon discharge. If the patient cannot return to, for example, the assisted living facility from which he was admitted to the hospital, then Ms. Staples must attempt to find an appropriate placement for the patient so that he may be safely discharged once he is stabilized. According to Ms. Staples, there is a range of alternative placements she may explore, depending on the patient's circumstances. Those persons who do not need structured care may be placed in a boarding home. Those requiring minimal care may be placed in an assisted living facility or a retirement home. The next level of care would be provided by an extended care facility, which Ms. Staples described as an intermediate facility between an assisted living facility and a skilled nursing facility. Other placement options include substance abuse rehabilitation facilities, halfway houses, 28-day programs, and shelters. The hospital faces some placement problems. Assisted living facilities tend not to accept patients with recent histories of drug abuse. Some assisted living facilities will not accept patients who are incontinent and unable to change themselves. Assisted living facilities without locked units will not accept Alzheimer's patients who tend to wander. Nursing homes will often reject young psychiatric patients in need of skilled nursing services because of their age. Nursing homes are generally reluctant to accept Medicaid patients, particularly those who are homeless, because of the difficulty they will face in placing the patient after the need for skilled nursing services has passed. Nursing homes will decline to accept a patient once they learn he is on a psychiatric unit because they "cannot meet their needs," which Ms. Staples described as a global catch-all phrase for their general desire not to accept psychiatric patients. Acute Care Inpatient Hospital Stays By the time of the hearing, the acute care inpatient hospital stays of 16 patients remained at issue. The findings below are set forth in the order that the patients were listed in AHCA Exhibit 6, the recipient spreadsheet indicating the dates of the patients' stays, the dates denied by the peer reviewers, and the amount of claimed overpayment. Patient #1 R.B. R.B. was admitted on March 30, 2001, and was discharged on April 8, 2001. Peer reviewer Dr. Machado determined that two days, April 6 through 8, should be denied due to lack of medical necessity for continued inpatient care.3 Dr. Machado's peer review report stated that R.B. was a 46-year-old female with scleroderma. She was admitted with chest tightness and found to be in near end-stage renal failure. Standard enzyme testing ruled out a heart attack. Her renal function continued to deteriorate and it became clear she would soon need dialysis. R.B. underwent placement of a Tessio catheter on April 3, and arrangements were begun for her to receive dialysis at an outpatient center in her hometown of Pensacola. Dialysis could not be arranged at the outpatient center until after she underwent dialysis at the hospital on April 4 and 5. Dr. Machado agreed that it would not have been safe to discharge her until adequate arrangements had been made for her outpatient dialysis, but that she was medically stable and ready for discharge by April 5 with an outpatient treatment plan in place. The care R.B. received in the hospital after April 5, including two blood transfusions on April 7, could have been provided on an outpatient basis. Petitioner's expert witness, Dr. Yithak Daniel Haim, testified that R.B. was not discharged as scheduled on April 6, because of changes in her mental status. She had nausea and episodes of confusion. On April 7, her medications were reviewed in light of their propensity to cause confusion in persons with poor kidney function. It was thought that her confusion could be due to OxyContin. Dr. Haim also noted that R.B.'s hemoglobin dropped on April 7, and she required a transfusion. Treating physician Dr. Daniel Tambunan confirmed that R.B. was kept in the hospital after April 6 due to her mental status, which developed into hallucinations on April 7. Dr. Tambunan also noted that R.B. had developed tachycardia on April 5 probably due to low hemaocrit, which was measured at 22.3 on April 7. Dr. Haim stated that a normal hematocrit is between 39 and 45. Dr. Haim agreed with Dr. Machado that dialysis can usually be done on an outpatient basis, but that he decided to keep her in the hospital due to the combination of the tachycardia, low hematocrit necessitating a transfusion, and the need to ascertain whether the hallucinations were caused by medications or by the dialysis. The greater weight of the evidence supports Petitioner's position that April 6 through April 8 should not have been denied. Dr. Machado's opinion was supportable regarding the ability of the patient to receive transfusions and dialysis on an outpatient basis. However, Dr. Machado's testimony ignored R.B.'s hallucinations and the reasonableness of keeping her in the hospital until the treating physician could ascertain their cause. Therefore, AHCA offered no evidence to conflict with the testimony of Dr. Haim that it was medically necessary to keep R.B. in the hospital until her discharge on April 8. Patient #2 F.C. F.C. was admitted on March 5, 2001, and was discharged on March 12, 2001. Peer reviewer Dr. Machado determined that the admission should be denied and a 23-hour observation should be approved.4 Dr. Machado's peer review report stated that F.C. was a 43-year-old female smoker with high blood pressure. She was admitted with chest pain for 3 days and by some notes was chronic over the last year. An electrocardiogram (EKG) did not show ischemia, and serial enzyme tests ruled out a heart attack. A spiral computer tomography (CT) scan was negative for a pulmonary embolism, i.e., a blood clot in the lung. Dr. Machado wrote that F.C.'s vital signs were stable, and that an outpatient work-up of her chest pain would have been appropriate. Instead, she had a stress test on March 7. The results were still "pending," according to a note on March 11. She also underwent work-up of abnormal findings on her abdominal CT/ultrasound, all of which could have been accomplished safely in the outpatient setting. Petitioner's physician expert, Dr. Haim, concurred with Dr. Machado's denial of this admission. The treating physician, Dr. Ashok Khanna, offered plausible reasons for keeping the patient in the hospital, including the fact that F.C. was a drug addict who could not be relied upon to comply with testing required to rule out coronary artery disease on an outpatient basis. However, the greater weight of the evidence supports Dr. Machado's denial of the admission. Patient #3 J.C. J.C. was admitted on July 6, 2000, and was discharged on July 12, 2000. Peer reviewer Dr. Machado determined that five days, July 8 through July 12, should be denied due to lack of medical necessity for continued care on the medical ward. Dr. Machado's peer review report stated that J.C. was a 55-year-old male initially admitted to the psychiatric unit due to depression and suicidal gesture. Psychiatric notes indicate that J.C. complained of three weeks of exertional chest pain and fatigue. On July 6, he was transferred to the medical ward for further evaluation. Serial enzyme tests ruled out a heart attack, and he was hemodynamically stable. He underwent a stress test on July 6, with no chest pain and no EKG changes. Dr. Machado concluded that a transfer back to the psychiatric unit or to home would have been appropriate, with follow-up. She believed it was not medically necessary to keep J.C. on the medical ward while awaiting the nuclear images of the stress test, as he was chest pain free and hemodynamically stable after July 7. Treating physician Dr. Luis Allen testified that he was called in on a psychiatric consultation on J.C. while he was in the medical unit. Dr. Allen found that J.C.'s depression was significant and that he would be in need of inpatient psychiatric treatment. Dr. Allen therefore followed J.C. until he was transferred to the inpatient psychiatric unit. The transfer was delayed while the cardiac workup was completed, so that the physicians could be confident that J.C. was medically stable before his transfer back to the inpatient psychiatric unit. Dr. Allen noted that J.C. had a history of depressive disorder and experiencing feelings of helplessness, hopelessness, that his level of depression was significant, and that he had been admitted for a suicidal gesture. Dr. Allen testified that J.C.'s psychiatric symptoms were too marked to be treated on an outpatient basis. Dr. Allen believed that his stay through July 12, was necessary when the combination of the patient's medical and psychiatric conditions are considered. Petitioner's medical expert, Dr. Haim, disagreed with Dr. Machado that J.C. could have been transferred from the medical ward on July 7, because the results of his cardiac stress test did not come back until July 9. Dr. Haim agreed that J.C. could have then been transferred off the medical ward on July 9. However, there was no bed available on the psychiatric floor on July 9. J.C.'s condition indicated a need for acute psychiatric hospitalization, meaning that he could not be discharged home. Thus, he remained on the medical ward until a psychiatric bed became available. Dr. Haim agreed that this was not strictly a medical reason, but contended that the lack of beds constituted a "reality reason" that justified the full admission. Petitioner's expert psychiatrist, Dr. Alan Berns, agreed with Dr. Allen and Dr. Haim that the full stay was justified. The greater weight of the evidence supports AHCA's denial of the last five days of J.C.'s admission. The record established that J.C. ceased to meet the criteria for inpatient admission in the medical unit on July 7. The reason for keeping J.C. on the medical unit after July 7 was administrative convenience, not medical necessity.5 Patient #4 P.C. P.C. was admitted on July 9, 2000, and was discharged on July 17, 2000. Peer reviewer Dr. Machado determined that six days, July 11 through July 17, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that P.C. was a 64-year-old female with a history of leg cellulitis treated with IV antibiotics. She was admitted through the emergency room with increased swelling of her lower left leg, which raised concerns about deep vein thrombosis (DVT), compartment syndrome, or progression of her cellulitis. P.C. also had chronic anemia, which had worsened. DVT was ruled out and she was continued on IV antibiotics. She had a blood transfusion due to her anemia. P.C. declined inpatient gastrointestinal (GI) work for her anemia, so the hospital planned to schedule the GI work after her discharge. Dr. Machado concluded that the remainder of her hospital stay could have been outpatient. She did have one further transfusion of packed red blood cells (PRBCs) on July 12, but she did not need to remain in the hospital for this, as she continued to refuse further workup and was asymptomatic. Dr. Machado testified that the physician's note on July 10 stated that the patient "looks and feels better, no complaints," and that P.C. was sufficiently improved to be discharged on that date. She could have been continued on IV antibiotics at home through home health care, or could have received the treatment at a skilled nursing facility. Dr. Machado acknowledged that the patient received a blood transfusion on July 12, but stated that this could have been provided on an outpatient basis. Blood could have been drawn after the transfusion for lab testing, and there would have been no need to call P.C. back in unless there was a problem with the labs. Dr. Haim testified that P.C. was diabetic, had kidney problems, peripheral vascular disease, and congestive heart failure. As of July 11, the status of her infection had improved, but she was still on IV antibiotics and her hematocrit had dropped. Because P.C. had several chronic medical problems, it was important to raise her hematocrit. She was given the transfusion on July 12, but her hematocrit continued to drop. Dr. Haim testified that this raised intense concerns as to "where is this blood going to." A CAT scan of her abdomen revealed no internal bleeding. Her release was planned for July 15, but blood testing on that date showed that her kidney function had deteriorated. It was feared that one of her IV antibiotics, Vancomycin, was affecting her kidney function. Dr. Haim concluded that her drop in hemoglobin and kidney function necessitated keeping her in the hospital until July 17. J.C.'s treating physician, Dr. Sidiab Elalaoui, testified that he could not have given the patient transfusions in his office. He disagreed with Dr. Machado that J.C. could have been seen in an outpatient setting as of July 11. Dr. Elalaoui noted that J.C. had been getting IV antibiotics at home, but that her condition nonetheless worsened to the point where she had to be brought to the emergency room on July 9. Dr. Elalaoui stated that J.C. did not have a simple infection. She had a bacteremia, a bacteria that went from the skin into the blood of a patient with diabetes and high blood pressure. Dr. Elalaoui could not be sure if her condition was life- threatening, but confidently stated that it was "severe." The greater weight of the evidence supports Petitioner's position that July 11 through July 17 should not have been denied. Dr. Machado's opinion was supportable regarding the ability of the patient to receive a transfusion and IV antibiotics on an outpatient basis. However, Dr. Elalaoui's testimony as the treating physician, in combination with Dr. Haim's expert testimony, credibly established that, whatever the patient's theoretical ability to receive transfusions and IV antibiotics on an outpatient basis, under the actual circumstances, it was medically necessary to keep J.C. in the hospital through July 17. Patient #5 A.F. A.F. was admitted on November 30, 2000, and was discharged on December 7, 2000. Peer reviewer Dr. Machado determined that four days, December 3 through December 7, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that A.F. was a 32-year-old female admitted with a severe headache and apparent new onset seizures. She underwent a CT scan, magnetic resonance imaging (MRI), and an electroencephalogram (EEG), all of which reported negative. She had undergone a lumbar puncture previously for the severe headaches, which was also negative. Dr. Machado concluded that A.F. could have been discharged with oral medications and outpatient follow-up for treatment of her apparent migraine headaches. Instead, A.F. underwent further testing for low back pain, depression, and substance abuse. Dr. Machado believed that the treating physicians allowed the patient's subjective complaints to outweigh the objective clinical findings, noting that A.F. had been seen sitting up and talking on the phone during the time when she was complaining of a severe headache. Petitioner's expert, Dr. Haim, agreed with Dr. Machado that the days denied were redundant. Thus, it is found that the greater weight of evidence supports AHCA's denial of four days of A.F.'s inpatient stay. Patient #6 C.G. C.G. was admitted on May 3, 2001, and was discharged on May 14, 2001. Peer reviewer Dr. Machado determined that eight days, May 5 through May 14, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that C.G. was a 47-year-old female admitted with increasing chest pain and numbness in her left arm. She was found to have a mass in the middle of her chest and possible early superior vena cava (SVC) syndrome. Serial enzymes and a CT scan respectively ruled out heart attack and pulmonary embolism as the cause of her chest pain. She was started on steroids and pain medication, with satisfactory pain relief. Dr. Machado concluded that the remainder of her stay focused on a workup on the mediastinal mass, which could have been done on an outpatient basis. The steroid and pain medications she was receiving through IV, could have been given orally. None of the progress notes documented sufficient evidence for inpatient workup. In her deposition, Dr. Machado explained that the SVC is the large blood vessel that returns blood from the heart to the upper portion of the body. When there is a mass in that area of the chest, it can compress the SVC and cause swelling from the backing up of the blood. Dr. Machado believed that once immediate life-threatening diagnoses such as heart attack and pulmonary embolism were ruled out, and C.G.'s pain was managed with medication, the matter of determining the nature of the mediastinal mass could have been handled on an outpatient basis. Treating physician Dr. Vajihuddin Khan, an internal medicine specialist at Florida Hospital Orlando, testified that he kept C.G. in the hospital until May 14 to complete the workup and perform all the necessary investigations of the mediastinal mass. Dr. Khan noted that C.G. could not have been released and observed daily on an outpatient basis because of transportation or financial problems. The patient lacked outside support. If she had gone out of the hospital with no place to stay and no friends to support her, the workup might never have been completed. C.G. was ultimately diagnosed with Hodgkins lymphoma. Dr. Haim testified that on May 7 the patient was ordered NPO (nothing by mouth) for a biopsy to be performed the next day. She had a swollen face, neck, and arm, with a large, undiagnosed mass in her chest. The chest pain and shortness of breath had not gone away. Dr. Haim believed it would be irresponsible to discharge C.G. in that condition without a diagnosis. The biopsy was performed on May 8 and showed cancer. The pathology report on May 9 was inconclusive, but her physicians knew it was "something bad," either lymphoma or a small cell carcinoma of long standing. Dr. Haim testified that it was important to differentiate the type of cancer because the treatments would be different for each. Therefore, it was necessary to perform surgery to obtain a larger sample of the mass. Dr. Haim noted that the surgery was not performed prior to C.G.'s discharge on May 14, due to the surgeon's unavailability. Dr. Haim agreed that reimbursement for the dates of May 10 through May 14 were "somewhat questionable" because C.G. was, in essence, sitting in the hospital waiting for a surgeon. The record did not disclose why she was forced to wait for a surgeon. The greater weight of the evidence supports AHCA's denial of the last eight days of C.G.'s admission. The record established that C.G. ceased to meet the criteria for inpatient admission in the medical unit on May 5. Patient # 7 C.G. C.G. was readmitted on May 25, 2001, and was discharged on June 3, 2001. Peer reviewer Dr. Machado determined that the last day should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that C.G. was a 47-year-old female admitted for elective mediastinoscopy to diagnose a mediastinal mass. The procedure was attempted but was unsuccessful due to a large goiter. A thoroactomy was performed that diagnosed lymphoma. C.G. did fine postoperatively but her discharge was delayed after an episode of chest pain, determined to be non-cardiac. When the oncologist saw her on May 30, he felt that she had SVC syndrome and called for an urgent radiation oncology consult. She was started on IV Decadron and improved dramatically by June 2, when she received her first radiation treatment. Dr. Machado saw no reason to keep the patient in the hospital for an extra day, until June 3. Dr. Haim testified that it was reasonable to wait for hours after the radiation treatment to note any improvement or adverse reaction. The greater weight of the evidence supports Dr. Haim's opinion that it was medically necessary to keep C.G. in the hospital through June 3. Patient #8 THE. THE. was admitted on April 3, 2001, and was discharged on April 9, 2001. Peer reviewer Dr. Machado determined that three days, April 6 through April 9, should be denied due to lack of clear documentation showing medical necessity for continued inpatient care. Dr. Machado's peer review report stated that THE. was a 21-year-old pregnant patient admitted with dehydration from hyper emesis gravid arum (the nausea and vomiting commonly called "morning sickness"). She had lost 14 pounds in the week prior to admission. This was the only abnormal vital sign, as she was not hypertensive or tachycardia and her electrolytes were not abnormal. No blood urea nitrogen (BUN) or cretonne was documented, indicating normal kidney function. She had two previous admissions for hyper emesis. She received sufficient IV hydration by April 4 for a 10 pound weight gain and was documented by nursing notes as feeling better and tolerating oral medications. She did have an increase in vomiting on April 5, but by April 6 this was much less and her weight was stable. Most of her anti-emetics were switched from IV to oral, and her IV was locked off for the remainder of her hospital stay. Her vital signs remained stable throughout her hospital stay, and there were no new labs to document continuing dehydration. Dr. Machado testified that the record showed no treatment after April 6 that necessitated a hospital stay. She believed THE. could have been discharged on April 6, with outpatient oral anti-nausea medications in a trial to see how she would do. Treating physician Dr. Andre Jakubowski testified that he saw THE. in his office on April 2, for an obstetrical visit. She was about 12 weeks pregnant and complaining of nausea, vomiting and the inability to keep down fluids. At her March 26, visit, she had weighed 134 pounds. On April 2, she weighed 120 pounds, a loss of 14 pounds. Dr. Jakubowksi immediately admitted T.H. to the hospital. For the first few days of T.H.'s hospitalization, Dr. Jakubowski gave her IV hydration in order to correct her electrolytes. On April 6, the IVs were discontinued and T.H. was placed on oral medications and food. Between April 6 and 7, she lost two and one half pounds. Dr. Jakubowski testified that he could not send T.H. home because she was still vomiting and was generally "not in good shape." He started her on IV hydration again and adjusted her medications. She began to eat and was able to go home on April 9. Dr. Jakubowski testified that it is not within the standards of practice to discharge a patient who has been taken off IV hydration without observing for 24 hours to be sure the patient is taking food and/or liquids orally and keeping them down. On April 8, the IV hydration was stopped for the second time. This time, T.H. was able to keep down some oral foods and liquids. Dr. Jakubowski watched her for 24 hours, then sent her home on April 9. Dr. Haim testified that laboratory testing is not necessary to document continued dehydration when the patient continues to vomit. With a weight loss of 14 pounds in a woman who was 12 weeks pregnant, this was clearly more than ordinary morning sickness. T.H. could not be discharged when she could not keep food on her stomach. The greater weight of the evidence supports Petitioner's position that April 6 through April 9 should not have been denied. Dr. Jakubowski's testimony as the treating physician, in combination with Dr. Haim's expert testimony, credibly established that, although T.H. showed some improvement and was taken off IV fluids on April 6, it was reasonable to wait 24 hours to make sure that she could keep down oral nutrition. When she was unable to do so, the IV hydration was resumed and it was reasonable to keep her in the hospital until she was able to keep some food on her stomach. Under all the circumstances, it was medically necessary to keep T.H. in the hospital through April 9. Patient #9 J.H. J.H. was admitted on July 7, 2001, and was discharged on August 7, 2001. Peer reviewer Dr. Machado determined that 21 days, July 17 through August 7, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that J.H. was a 63-year-old female lung transplant candidate with end-stage pulmonary fibrosis. She was admitted with a cough, low-grade fever, and increased shortness of breath. She also appeared to have a postoperative wound infection from recent vascular surgery in the right groin. A CT scan was negative for pneumonia or pulmonary embolism and her respiratory symptoms stabilized and, fairly quickly, returned to their concededly poor baseline. The infected right groin area was the cause for J.H.'s extended stay. The wound was debrided on July 14, following treatment with IV antibiotics and wound care. The infectious disease consultant agreed on July 16 that home IV antibiotics would be appropriate. However, the patient declined a PICC line (peripherally inserted central catheter, a long–term catheter that is inserted into the arm and threaded into central circulation) due to a past poor experience. She therefore, continued to get IV antibiotics in the hospital until July 26, when she was changed to oral Keflex. Dr. Machado found that the reasons for her continued hospital stay were unclear, except for wound care. On August 3, J.H. left the hospital for a few hours on a day pass. During her stay, she underwent other tests that are required for lung transplant evaluation, but weren't necessary during this hospitalization. In her deposition, Dr. Machado testified that J.H.'s lung condition stabilized, but the groin became the problem. The treating physician initially thought the problem was fluid collection, but the increased white blood cell count indicated an infection. J.H. was given proper wound care, but Dr. Machado could not see anything done in the hospital during the last 21 days of J.H.'s stay that required an inpatient stay. Dr. Machado concluded that IV antibiotics and wound care could have been given in a sub-acute skilled nursing facility or with home health. Treating physician Richard Young Feibelman is board certified in internal medicine and pulmonary medicine. He is a pulmonary physician, and was J.H.'s primary physician during this admission. In his deposition, Dr. Feibelman testified that he had followed this patient for some time prior to this admission. J.H.'s primary underlying problem was severe and progressing idiopathic pulmonary fibrosis, a scarring debilitation of the lungs causing progressive shortness of breath and requiring increasing oxygen. It typically results in death within two to three years of the diagnosis. Dr. Feibelman testified that J.H. had recently been evaluated at the University of Miami for a lung transplant, which she desperately wanted despite the high risk of death associated with this surgery. J.H. had been undergoing workup in Miami, about two weeks before this admission, including a carotid artery angiogram and cerebral angiogram. After the angiogram, she developed a pseudo aneurism or partial false leak of the puncture site from the catheter insertion. She then developed a hematoma in that area, with wound infection and fever that made necessary her admission to the hospital. Dr. Feibelman testified that at the time of admission, it was difficult to tell whether the infection came from her groin, or whether it was a superimposed respiratory infection on top of her underlying pulmonary fibrosis. Dr. Feibelman testified that J.H. was extremely sick, near the end stage of chronic fibrotic lung disease. He stated that this was an important hospitalization, because J.H.'s infection had to be under control to ensure she could make it to Miami, and survive the lung transplant. Before her admission, she was on immunosuppressive therapy, which increases the risk of infection. Dr. Feibelman treated her with IV antibiotics, and he described the wound care as "aggressive," an effort to heal the wound before the lung disease killed her. The wound was debrided, and J.H. was seen by an infectious disease consultant and vascular surgery consultant. There was difficulty with her blood pressure, and an episode of arm and leg numbness that required a neurological consultation to rule out a transient ischemic attack or pulmonary embolus. Dr. Feibelman testified that J.H. was still getting IV antibiotics to almost the end of her hospitalization. She was on high flow oxygen. The pain in her wound was such that she required intravenous morphine to change her dressings. Dr. Feibelman concluded that any additional setback for J.H. would have been fatal. Her disease has a fairly rapid stair- step pattern, in which there is a drop-off, then stabilization, then a further drop-off, then stabilization. Dr. Feibelman stated that J.H. could not afford a further drop-off before her lung transplant. Dr. Feibelman testified that J.H. was allowed to leave the hospital for two hours on a day pass. He stated that this was in all likelihood her last chance to go home, and he thought it was worth letting her go. Dr. Haim disagreed that the last 21 days of J.H.'s stay should be denied, but agreed that the last week or two were debatable, depending on the support system she had at home and the possibility of giving her IV antibiotics at home. Dr. Haim stated that she could have had IV antibiotics at home with a PICC line. However, J.H. had prior poor experiences with PICC lines and told her physicians they were not going to "torture" her again. She would also have needed assistance with her oxygen tanks if she went home. Dr. Haim stated that J.H. needed help with all of her activities of daily living (ADLs), as would any patient requiring six liters of oxygen every day. The greater weight of the evidence supports AHCA's denial of the last 21 days of J.H.'s admission. Dr. Machado correctly observed that the patient could have received IV antibiotics, oxygen, and wound care in a skilled nursing facility. She conceded that it would have been correct to keep the patient in the hospital if no skilled nursing facility was available. Petitioner offered no evidence that it attempted to place J.H. in a skilled nursing facility. Dr. Feibelman's concerns about J.H.'s precarious condition are fully credited, but the record as presented established that C.G. ceased to meet the criteria for inpatient admission in the medical unit on July 16. Patient #10 C.J. C.J. was admitted on March 31, 2001, and was discharged on April 9, 2001. Peer reviewer Dr. Machado determined that six days, April 3 through April 9, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that C.J. was a 67-year-old female admitted with abdominal pain, nausea, and vomiting, two days after a colonoscopy. She also had poorly controlled diabetes. The nausea and vomiting resolved quickly after admission, and she was hemodynamically stable throughout her stay. A CT scan of the abdomen and pelvis was performed, which showed a large pelvic mass (likely the recurrence of a previous cancer) with accompanying hydronephrosis (swelling of the kidney caused by obstruction of urine flow). The gastroenterologist cleared C.J. for discharge on April 2. The remainder of her stay involved consultations with urology and oncology specialists regarding the pelvic mass, which could have been accomplished in the outpatient setting. She did not have a ureteral stent (a surgical device implanted to hold the ureter open so that urine can flow freely from the kidneys to the bladder) placed until April 6. Dr. Machado wrote that it was unclear from the notes what was keeping her in the hospital after April 6. In her deposition, Dr. Machado testified that C.J. should have been discharged on April 3 with arrangements for an outpatient workup of the pelvic mass, which was likely a recurrence of the cancer that C.J. had ten years previously. Dr. Machado testified that the placement of the stent was necessary, either during the hospitalization or as an outpatient, and noted that if it had been an emergency, the stent would have been placed sooner in C.J.'s stay. The stent was actually placed one week after C.J.'s admission. Treating physician Dr. Alan Varraux, a specialist in pulmonary medicine, testified that C.J. was weak and somewhat frail, but underwent a colonoscopy because of gastrointestinal symptoms. The procedure caused her much nausea and vomiting. Her complaints and Dr. Varraux' concerns about dehydration led to her hospitalization on March 31. On admission, she was kept NPO and IV fluids were started. She was allowed to start eating on April 1. A GI specialist saw her on April 2 and performed an upper GI endoscopy and a colonoscopy. A consulting oncologist saw the patient on April 3. Dr. Varraux stated that an X-ray showed hydronephrosis, a blockage of the ureter system causing urine to back up into and dilate the kidney. The stent could not be placed on an outpatient basis because C.J. was a debilitated, immuno-compromised cancer patient who could be killed by a urinary tract infection. She was a high risk patient and needed to be cleared by a urological specialist before discharge. The urologist planned to place the stent on April 6, after which C.J. could be discharged if all went well. Dr. Haim testified that the stent was actually placed on April 7 and that C.J. needed to stay in the hospital for an additional 24-to-48 hours to ensure that she had adequate urine output, and that her fever was going down, and that her subjective feelings were improved. After reviewing the depositions of Dr. Varraux and Dr. Haim, Dr. Machado testified that she saw no reason to change her opinion. Dr. Machado stated that the medical record showed C.J.'s cancer was 10 years prior to this admission. This led Dr. Machado to disagree with Dr. Varraux's assumption that C.J. was immuno-compromised and unable to fight infections normally. Dr. Machado also did not see anything in the medical record to support the concerns about a urinary tract infection. C.J.'s urinalysis was normal on admission. There was glucose in her urine, which was consistent with her diabetes. On April 3, her white blood count was normal, indicating that if there was an infection, it had been treated adequately. No culture or urinalysis was performed on that date. Dr. Machado opined that the patient's low grade fever throughout her stay was not a reason to keep her in the hospital, as evidenced by the fact that she was still running a low grade fever on the day she was discharged. She was treated with antibiotics that can be given orally, but that they chose to give via IV. Dr. Machado could find no notes from the primary treating physician from April 6, until his discharge note on April 9. Dr. Machado found nothing in the medical record to support the view that the stent had to be placed in the hospital. The stent certainly needed to be placed, to allow urine to drain properly from the kidney. However, this is not always an inpatient procedure. Dr. Machado noted that the urologist's post-operative orders were written as outpatient orders. The greater weight of the evidence supports AHCA's denial of the last six days of C.J.'s admission. The record as presented established that C.J. ceased to meet the criteria for inpatient admission in the medical unit on April 3. Patient #11 G.M. AHCA Exhibit 6 indicates that G.M. was admitted on July 6, 2000, and was discharged on July 20, 2000. However, Dr. Machado noted that the records provided by the hospital indicate that this patient was admitted directly to a rehabilitation facility, and was never an acute medical admission. Therefore, Dr. Machado determined that the entire 14-day stay should be denied. Dr. Haim testified that his own abbreviated notes show that the patient came into the hospital, was in respiratory failure, had shortness of breath, coded with cardiac arrest, and was intubated. However, the lack of medical records provided by the hospital to AHCA require that Dr. Machado's denial determination be sustained. Patient #12 N.P. N.P. was admitted on January 8, 2000, and was discharged on January 14, 2000. Peer reviewer Dr. Machado determined that two days, January 13 through January 14, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that N.P. was a 46-year-old male with cardiomyopathy and poorly controlled diabetes. He was admitted with increasing shortness of breath, fever, cough, and chest pain. He was started on antibiotics, diuretics, and respiratory treatments, and his symptoms improved rapidly. Serial enzyme tests ruled out a heart attack and his chest pain was not thought to be cardiac in nature. On January 11, his IV medications were changed to oral, and it was felt he was near ready for discharge. Though he felt better, his oxygen saturations were slow to improve and he was still saturating in the high 80's on room air, which was not his baseline. Dr. Machado found it reasonable to monitor him for one more day in the hospital to see if this would improve before sending him home with oxygen. The saturations stayed the same the following day, and he continued to feel better. Dr. Machado concluded that discharge would have been safe on January 12, with home oxygen and close outpatient follow-up. Dr. Haim testified that N.P. could have probably been discharged on January 13 on oxygen, with outpatient follow-up. He noted that N.P. was started on a new medication on January 12, and that it is reasonable to keep the patient for an additional 24 hours to gauge his response. However, Dr. Haim also noted that the medication was a diuretic, not an antibiotic. The greater weight of the evidence supports AHCA's denial of the last two days of N.P.'s admission. The record, as presented, established that N.P. ceased to meet the criteria for inpatient admission in the medical unit on January 13. Patient #13 T.S. T.S. was admitted on September 18, 2001, and was discharged on October 16, 2001. Peer reviewer Dr. Machado determined that 12 days, October 4 through October 16, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that T.S. was a 55-year-old man with cirrhosis and a history of drug abuse and hypertension. He was admitted for treatment of a non-healing venous ulcer that had failed outpatient therapy. T.S. was also experiencing uncontrolled pain despite oral pain medications. He was started on broad-spectrum antibiotics and underwent debridement of the wound on September 21. Plans were made to discharge him to a skilled nursing facility for several weeks of wound care and IV antibiotics. On September 22, his pain continued to be poorly controlled. On September 25, he had a fever of 101 degrees and diarrhea, which prompted further work- up. Cultures of the wound continued to show a polymicrobial infection that also had a fungal component. Oral Sporanox was added to his medications to address the fungal component. On September 27, a pain management consultation was requested due to his continued uncontrolled pain, mostly during dressing changes. T.S. was placed on a Dilaudid PCA (patient controlled analgesic) pump, which did control his pain. During this time, T.S. had waxing and waning mental status due to the pain medications. Because of T.S.' increasing lethargy, the PCA was discontinued on October 3, and T.S. was thenceforth maintained on oral medications. The patient also had chronic anemia on admission that slowly worsened during his stay. He was found not to be acutely bleeding and was transfused PRBCs with improvement in his hematocrit level. Dr. Machado agreed that it would have been difficult to discharge T.S. to a skilled nursing facility for dressing changes and IV antibiotics if he was unable to tolerate dressing changes without a PCA pump for pain. Therefore, she would allow the hospitalization through October 3, when T.S. was switched to oral pain medications and could have gone to a skilled nursing facility. Treating physician Dr. Pradeep Vangala testified that he saw T.S. in his office, prior to his hospital admission. T.S. came in with what appeared to be cellulitis in his legs, and was treated with oral antibiotics. When the condition failed to respond to treatment, Dr. Vangala admitted T.S. to the hospital. Dr. Vangala stated that T.S. was kept in the hospital after October 4 because his cellulitis had not resolved and the patient was not stable enough to be changed to oral antibiotics. Dr. Vangala testified that it was not a simple decision to send T.S. home with IV antibiotics because of complicating issues. Secondary to his cirrhosis, T.S. had significant edema in most of his body in general, and his legs in particular. Dr. Vangala stated that T.S. required close observation of his skin integrity and his cellulitis, and that his cirrhosis was the cause of the edema. Though T.S. had a lot of excess fluid, most of it was in the subcutaneous tissues rather than the blood vessels. This means that his fluid status had to be closely monitored, because of the danger that he might become intravascularly volume depleted, which could affect renal function. Dr. Vangala stated that the cirrhosis had altered T.S.'s mental status for a significant portion of his stay, and that an acutely confused patient is not a candidate for discharge because he is not able to follow discharge instructions. T.S. also had significant anemia, which meant that his hemoglobin had to be watched. Dr. Haim testified that during all 12 days denied by Dr. Machado, T.S. was still running a fever and had a depressed mental status. He was still receiving IV medications, and had a significant swelling of the abdomen that required drainage. His mental status was abnormal, and physicians were having a very difficult time titrating his pain medications. A neurologist was called in on October 4 because of T.S.'s impaired mental status. The gathering fluids in his body were causing swelling and making it difficult for him to breathe. Dr. Haim testified that the IV antibiotics that T.S. was receiving after October 4 had to be closely supervised, though he conceded that a skilled nursing facility could handle their administration. The greater weight of the evidence supports AHCA's denial of the last twelve days of T.S.'s admission. Dr. Machado's opinion that the medications administered via IV after October 4 could have been given outside of the inpatient hospital setting was uncontradicted. Dr. Vangala's concerns were genuine, but mostly consisted of monitoring actions he wished to perform as a precaution, rather than acute care needs. The record as presented established that T.S. ceased to meet the criteria for inpatient admission in the medical unit on October 4. Patient #14 & 15 (two denials) F.T. F.T. was admitted on March 27, 2000, and was discharged on May 13, 2000. Peer reviewer Dr. Machado determined that: March 27 through April 4, should be approved for treatment of coagulopahty, subdural hematoma, and evaluation of the patient's near-syncopal episode; April 14 through May 7, should be approved for chemotherapy and treatment of neutropenic fever; and that April 5 through April 13 (nine days) and May 8 through May 13, (five days) should be denied because the treatment during those periods could have been administered on an outpatient basis. Dr. Machado's peer review report stated that F.T. was a 49-year-old female with metastatic breast cancer admitted with a coagulopathy and near-syncope (almost fainting). She was found to have new subdural hematomas and a new pathologic fracture of the left femur. Her coagulopathy was reversed and she was evaluated by neurosurgery and radiation oncology. No surgery was recommended. She began palliative radiation therapy to the brain and left femur on March 29. She remained hemodynamically and neurologically stable, and neurosurgery signed off on the case on April 4. Her pain responded well to radiation. Between April 4 and April 14, F.T.'s hospital care involved continued radiation therapy and the biopsy of a left auxiliary lymph node (on April 11) to determine the receptor status of the breast cancer, which would enable the oncologist to decide if chemotherapy would be of benefit. Dr. Machado concluded that this evaluation and the radiation therapy could have been done on an outpatient basis. On April 14, F.T. began chemotherapy over two days and very soon began experiencing fever and neutropenia (an abnormally low level of neutrophils, the white blood cells produced in the bone marrow) and then respiratory distress. She was started on IV antibiotics and IV diuretics. She was also started on a feeding tube due to poor oral food intake. The fevers and neutropenia were resolved by April 25, but she continued to decline, with increased shortness of breath requiring more diuresis to clear fluid from the lungs. It became evident she was deteriorating and her feeding tube was discontinued by May 4. By May 8, comfort measures only were initiated, and evaluation for inpatient hospice care was requested. In her deposition, Dr. Machado testified that between April 4 and April 14, F.T. was receiving radiation as her main treatment, as well as further evaluation to determine whether anything more could be done for her cancer. Everything she received during this period, including the lymph node biopsy, could have been done on an outpatient basis. On April 14, she started chemotherapy, which was reasonable to perform in the hospital. By May 8, the medical chart notes indicate the initiation of "comfort only" measures. At that point, she could have been sent home with hospice care or to a hospice house. Dr. Machado conceded that some patients do receive inpatient hospice care, but she testified that this should have been done in a hospice bed, not an acute care medical bed. Dr. Haim testified that during the period of April 5 through April 13 F.T. was receiving radiation to the fractured femur and awaiting a lymph node biopsy. The initial pathology report did not have sufficient material for receptor studies, so a surgical biopsy would need to be performed. On April 5, the hospital social worker was awaiting orders to transfer F.T. back to the nursing home from which she had been admitted. However, F.T. had an episode of nausea and vomiting. Dr. Haim stated that nausea and vomiting in a patient who has metastasis is extremely serious because it could indicate more brain swelling or bleeding in the brain. She was started on IV Decadron, a steroid given to combat nausea and vomiting in chemotherapy patients. Dr. Raul Castillo, F.T.'s oncologist, met with a pathologist on April 7 to discuss the need for an open biopsy of F.T. On April 8, F.T. was receiving radiation and was started on a new chemotherapy drug, IV Aredia. She was monitored closely for side effects. Pain management was a persistent problem. On April 10, she was given the open biopsy under a local anesthetic. On April 13, an orthopedist saw her and ordered a specially fitted brace, because she had difficulty sitting due to her spinal problems. Dr. Haim concluded that, because of all the treatments F.T. was undergoing, including IV chemotherapy, it was mandatory for her to stay inpatient from April 5 through April 13. She could not get the chemotherapy in a non-acute care facility. The hospital was the only place she could get IV Aredia, the brace hadn't arrived until April 13, and she had multiple problems that required monitoring, including advanced cancer and bleeding in the brain. Dr. Haim did not believe that a skilled nursing facility had the capacity to handle F.T. As to F.T.'s second stay, Dr. Haim testified that from May 8 to May 13 F.T. was very weak. Her abdomen was markedly distended, which could have meant that her bowels weren't working well. The abdomen was X-rayed. Her platelet count was dangerously low. An oncology note dated May 9 stated that she was a full code (meaning that all resuscitative efforts must be attempted), by her own choice. Her full code status forced the doctors to plan the performance of tests on her abdomen. On May 10, F.T. voluntarily changed her instructions to DNR (do not resuscitate). Comfort measures were instituted and hospice was consulted. However, when the hospice nurse arrived, F.T. was out having an ultrasound preparatory to having the abdominal fluid drained. The hospice decided not to see her, because she was getting a procedure. The hospice nurse never saw the patient on that day. F.T. was admitted to hospice on May 13. The oncologist, Dr. Castillo, testified that his medical group first saw F.T. in March 2000. F.T. had been diagnosed with breast cancer in 1988, and treated in Puerto Rico, with a left mastectomy and chemotherapy. After she completed the chemotherapy, she had radiation. She indicated that in December 1999, she was told she had metastatic cancer in her bones. In 2000, she developed pain over her hips and legs. She had a fracture over her left femur and a prosthesis over her left leg. She had been taking Coumadin, and Dr. Castillo became involved when F.T. presented with bleeding secondary to Coumadin toxicity. Dr. Castillo described this as a very complex and emotional case in which a few strands of information had to be pieced together to determine the best case management. F.T.'s case history was incomplete because she spoke only Spanish, creating a big language barrier with most of the hospital staff. Dr. Castillo speaks Spanish, and was able to get a "full but scattered history" from F.T. F.T. had metastatic disease. Dr. Castillo testified that it is extremely important to determine if the patient has an estrogen receptor or hormonally positive tumor. Patients who are hormonally sensitive have a much higher probability of responding positively to therapy. To make things more difficult, F.T. developed a subdural hematoma, for which the medical team had to correct her coagulation. Dr. Castillo testified that they felt uncomfortable discharging F.T. while treatment planning was underway. Because of her previous exposure to chemotherapy and radiation, F.T. was at high risk for complications such as the sepsis that eventuated. Dr. Castillo emphasized that this was a complex case, and that the treatment team lacked all the information necessary to make rapid and clear decisions. One event followed another, and the team concentrated on trying to catch up and get the patient somewhat stable. Dr. Castillo stated that a problem with treating a patient this sick on an outpatient basis is the lack of supervision by a specialist. Such a patient will not have access to a site where she is going to get one-to-one care from the oncological standpoint. Dr. Castillo agreed that radiation is commonly done on an outpatient basis, but he noted that this was a patient who had bleeding on the brain and was getting radiation to the brain. If she was in a skilled nursing facility and had a subdural hematoma, she would probably have died on her way to the hospital. Dr. Castillo stated this his group's philosophy is to discharge a patient when they consider the patient stable, and not to leave a patient in the hospital for a mere workup. However, this was a patient who could become a neurosurgical emergency case at any moment. Dr. Castillo concluded that he would not have done anything different in the management of this patient. The greater weight of the evidence supports Petitioner's position that April 5 through April 13 and May 8, through 13 should not have been denied. Dr. Castillo's testimony as the treating oncologist, in combination with Dr. Haim's expert testimony, credibly established that F.T.'s condition was so precarious that her entire inpatient stay was medically necessary. Patient #16 J.Y. J.Y. was admitted on March 3, 2001, and was discharged on March 21, 2001. Peer reviewer Dr. Machado determined that five days, March 16 through March 21, should be denied due to lack of medical necessity for continued inpatient care. Dr. Machado's peer review report stated that T.S. was a 39-year-old female admitted due to swallowing a dental appliance that had become lodged in her esophagus. She underwent an endoscopy on March 3, and the foreign body was removed with some difficulty. Her esophagus was perforated due to the foreign body. She began having fever and pain. A CT scan on March 6 showed extensive edema and air tracking compatible with esophageal perforation. She was kept NPO, hyperalimentation (feeding tube) was started, and IV antibiotics were continued. Clinically, she improved and conservative treatment was continued, as the patient wished to avoid surgery. Dr. Machado believed that it was still prudent to monitor and treat the patient in the hospital, due to the possibility of serious complications from this type of injury. A barium swallow was performed on March 12, which showed that a leak persisted in the esophagus. Because she was clinically so much better, it was decided to repeat the CT scan of the neck on March 14 to check for improvement. The scan showed that soft tissue gas and swelling had decreased considerably. By March 16, her IV antibiotics were discontinued and it was decided to give her a trial of fluids by mouth. She was hemodynamically stable and afebrile. Dr. Machado concluded that she should have been discharged with home health and hyperalimentation, with an outpatient swallowing study and close follow-up, rather than waiting in the hospital until March 19. Dr. Haim testified that the March 12 barium swallow results led to J.Y.'s being kept on no food by mouth and total parenteral nutrition (TPN, another term for a feeding tube). On March 16, J.Y. was noted to be clinically stable, but her liver function tests were noted to be high. Dr. Haim noted that the TPN itself could be causing the liver problems, so there was a GI consult. The gastroenterologist recommended a change of antibiotics as a possible solution to the increase in her liver enzymes. Contrary to Dr. Machado's statement, no trial of fluids by mouth was done on March 16. As of March 17, J.Y.'s orders were still nothing by mouth and TPN, and her liver enzymes continued to rise. On March 18, there were no major interventions and J.Y. was scheduled for a swallow study. Her liver enzymes were noted to be decreasing for the first time. Her swallow study results were pending on March 19. Also on March 19, a pulmonary note indicated phlebitis in J.Y.'s arm, in the area of the IV feeding. An order to replace the PICC line was written. On March 20, J.Y. was started on an oral diet and her PICC line was replaced. The gastroenterologist recommended a liquid diet for several weeks. On March 21, the patient was discharged home on IV TPN and a liquid diet. Dr. Haim concluded that the acute care setting was required for the denied days. Treating surgeon Dr. Stephen Huber testified J.Y. came in having swallowed her partial plate, which had become lodged in her esophagus. The emergency room physicians could not get it out, and so J.Y. was taken to surgery. Dr. Huber kept her NPO because he was afraid she might have torn her esophagus. He placed her on IV antibiotics and ordered a swallow study, which revealed a small leak. Dr. Huber called in an infectious disease specialist to manage J.Y.'s antibiotics because he was worried about contamination from the leak spreading into her neck. She was started on IV feedings and the medical team watched for an abscess to develop in her neck. A few days later, another swallow study was performed, which indicated the leak was smaller but still persistent. J.Y. was kept NPO and kept on IV antibiotics. A third swallow study showed the leak had resolved, and she was started on regular food the next day. Once she was cleared by all her specialists, she was discharged from the hospital. Dr. Huber testified that he kept J.Y. in the hospital after March 16 mainly to watch her. Even after the last swallow study, there was still a small leak in the esophagus. J.Y. had not eaten for a couple of weeks, but she was getting better clinically, and Dr. Huber decided to feed her. He started with clear liquids, then advanced her diet slowly over the next few days. Dr. Huber testified that J.Y. had to be watched for fevers and neck swelling caused by her eating, and that he could not evaluate her progress if she was at home. J.Y. did not speak English, and there would have been difficulty monitoring her condition if she were not under direct observation. Seeing her on an outpatient basis might prove harmful if she developed an abscess in her neck, or sepsis. J.Y. remained on IV antibiotics until she was discharged. The greater weight of the evidence supports AHCA's denial of the last five days of J.Y.'s admission. Dr. Huber's concerns were genuine, but mostly consisted of monitoring actions he wished to perform as a precaution, rather than acute care needs. The record as presented established that J.Y. ceased to meet the criteria for inpatient admission in the medical unit on March 16. Psychiatric Inpatient Hospital Stays By the time of the hearing, the psychiatric inpatient hospital stays of 13 patients remained at issue. The findings below are set forth in the order that the patients were listed in AHCA Exhibit 6. Patient #1 H.A. H.A. was admitted on February 20, 2001, and was discharged on February 26, 2001. Peer reviewer Dr. Rahul Mehra determined that three days, February 23 through February 26, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that H.A. was a 57-year-old female with a diagnosis of schizophrenia. She was admitted, involuntarily, under the Baker Act on February 20, for reasons unclear in the medical record. As of February 23, the patient was not actively suicidal, homicidal, manic, or in complicated withdrawal. The patient's psychotic symptoms may have persisted, but these symptoms appeared to be baseline. As of February 23, she did not need 24-hour nursing care. She was discharged on the same dose of anti-psychotic medication she was taking upon admission. Dr. Mehra concluded that, as of February 23, outpatient was the appropriate level of care. Generally the Medicaid standards for whether a patient requires an inpatient psychiatric bed are as follows: whether the patient is actively suicidal or homicidal; whether the patient is so acutely psychotic that her ability to care for herself is impaired; whether the patient is physically aggressive or manic; and whether the patient is having a "complicated withdrawal" from alcohol or drugs that might cause a seizure or other acute health problem. Dr. Mehra testified that it was appropriate to admit H.A. because of her family's concerns that she might have a handgun and was threatening self-harm. At admission, she was having some psychotic symptoms, displaying disorganized thoughts. However, as of February 23, she was no longer actively suicidal or homicidal, which was the reason she was admitted. She did have psychiatric symptoms, such as visual and auditory hallucinations, but these seemed to be her baseline level of functioning. No changes were made to her medications. Dr. Mehra's recommendation would have been to transfer H.A. to an outpatient setting or a nursing home. Petitioner's psychiatric expert, Dr. Alan S. Berns, testified that H.A. was admitted through the emergency room for increasing auditory hallucinations and religious preoccupation. She had a history of non-compliance with her outpatient treatment, mood swings, and unpredictable impulse control. She denied hallucinations, but was noted to talk to herself as a religious preoccupation. The diagnostic impression was of acute exacerbation of chronic undifferentiated schizophrenia, and rule out schizoaffective disorder. The psychosocial note on admission stated that the patient is talking to God and dead relatives. The emergency room nursing notes stated that the patient wanted to hurt an unidentified person with a handgun. She was labile, crying and laughing. H.A. had been admitted to Lifestream, a mental health center in Lake County, three times since November 2000. She had a history of "cheeking" her medications. She required assistance with her ADLs and ate poorly. The social worker reported that she did not attend group therapy sessions, and that she heard God talking to her all the time. By February 24, H.A. showed an improved mood and affect, with no overt agitation. She was observed talking to herself and appeared to be responding to internal stimuli. She also demonstrated some looseness of association. On February 25, she denied auditory or visual hallucinations and any suicidal or homicidal ideations. However, she remained seclusive, with pressured speech and a depressed, blunted affect. She refused to participate in groups. She was discharged on February 26 with improved mood and affect, no evidence of delusions, and denied hallucinations and suicidal or homicidal ideations. She was diagnosed with schizoaffective disorder and discharged with prescribed Seroquel and Paxil, the same medications she was taking on admission. Dr. Berns concluded that H.A. warranted another day or two past the February 23 discharge authorized by Dr. Mehra. The nursing notes from February 23 show the patient alert and oriented times three (time, place and person), and indicate that she was cooperative, pleasant, and denied suicidal ideation. However, H.A. also stated that God talks to her and she sees the Holy Spirit. On February 24, she was still exhibiting some looseness of association, indicating that her thinking was not organized, which could in turn affect her ability to perform her ADLs. At this point, she did not appear a danger to herself in terms of intentionally inflicting harm. Dr. Mehra agreed that the symptoms cited by Dr. Berns, such as looseness of association, loose thoughts and disorganization, can be indicative of the need for a longer inpatient stay. However, loosening of associations is a common finding in a patient with schizophrenia, which is a lifelong disorder. Looseness of association in a schizophrenic patient does not, in and of itself, invoke the Medicaid guidelines that the patient is acutely and gravely psychotic. Dr. Mehra reasoned that if her acute condition had been such a great concern, then her antipsychotic medication could have been increased to effect a change in the observed loosening of associations. Her subtherapeutic dosage was never changed during her inpatient stay. Dr. Mehra concluded that just having loosening of associations is not sufficient, under the Medicaid guidelines, to continue an inpatient level of care. The greater weight of the evidence supports AHCA's denial of the last three days of H.A.'s admission. The record as presented established that H.A. ceased to meet the criteria for inpatient admission in the psychiatric unit on February 23. Patient #2 T.E. T.E. was admitted on March 19, 2001, and was discharged on March 27, 2001. Peer reviewer Dr. Mehra determined in his report that three days, March 25 through March 27, should be denied due to lack of medical necessity for continued inpatient care. However, in his deposition, Dr. Mehra testified that he now agreed with the hospital that the entire stay should be approved.6 Patient #3 S.G. S.G. was admitted on October 25, 2001, and was discharged on November 1, 2001. Peer reviewer Dr. Mehra determined in his report that four days, October 28, through November 1, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that S.G. was a 42-year-old male with diagnosis of schizophrenia admitted, involuntarily under the Baker Act from Orange County Jail, for psychotic symptoms and homelessness. The patient had a previous admission to Florida Hospital Orlando's psychiatric unit, having been discharged on October 8, 2001. Dr. Mehra found that as of October 28, the patient was not actively suicidal, homicidal, manic, or in complicated withdrawal. The patient's psychotic symptoms were still present but improved. The patient was cooperative, directable, and interacting with some peers. His delusions appeared to be chronic. He did not require seclusion or restraints and was not a management problem on the unit. His vital signs, appetite, mood and sleep were stable. Placement became an issue during his inpatient stay. Dr. Mehra concluded that the appropriate level of care as of October 28, was outpatient, with closely supervised living arrangements. In his deposition, Dr. Mehra testified that S.G. was admitted with auditory hallucinations and delusions, reaching the level of psychotic behavior. During his stay, S.G.'s speech and thought became more organized. His psychotic symptoms became less intrusive. By October 28, he was not a danger to himself or others and was ready to move into a sub-acute setting. Dr. Mehra conceded that S.G. was probably still delusional on October 28, but noted that his delusions were probably chronic and at this time were not interfering with his ability to perform his ADLs. Treating physician Dr. Rex A. Birkmire testified that S.G. was initially very psychotic, delusional, and disorganized. He had not been taking his prescribed medications. S.G. thought that one of the his nurses was the Queen of England. He heard voices and had a religious preoccupation about Satan, aliens and dragons. Staff at the jail believed S.G. needed a higher level of care, and therefore had him admitted to the hospital under the Baker Act. Dr. Birkmire testified that as late as October 31, S.G. was still so psychotic, he thought the medication Artane was a "gasoline pill." His conversation continued to be irrelevant and rambling. He said that "people see the smell but they don't see me." On October 31, S.G. was so disorganized that he could not identify the medications he would need to stay stable, and hospital staff felt he could not maintain his basic ADLs. Dr. Birkmire noted that by October 28, S.G. was "passively compliant" with his medications, meaning that he would take them when the nurses gave them to him. Dr. Birkmire stated that S.G. could have been managed in a skilled nursing facility with a 24-hour nursing staff. Dr. Berns testified that S.G.'s prior admission on October 8 raised questions as to the adequacy of his prior treatment, his compliance upon discharge, and his stress level during the interval between admissions. The notes for the current admission stated that S.G. was readmitted due to medication noncompliance. The admission note stated that the patient was psychotic and disorganized, with jumbled thoughts, and had ideas of reference as to the television, i.e., that it was sending him special messages. S.G. was reported to be hyper-religious, and carried a Bible. He had a history of hearing voices and was diagnosed with chronic undifferentiated schizophrenia. A note from Dr. Luis Allen stated that on October 28 the patient was still "very loose," meaning his thoughts were disorganized and psychotic. Hospital staff reported that S.G. remained delusional, and there was concern from the social worker that he might be responding to internal stimuli. On October 29, a note reported that S.G. was psychotic and manic, though starting to make more sense. On October 30, the notes stated that S.G. was still rambling and tangential, and that his ADLs were not good. Staff was concerned that S.G. was not committed to taking medication as an outpatient. Dr. Birkmire recommended the decanoate form of antipsychotic medicine, a long-acting intramuscularly administered form. Dr. Berns stated that some of these medications can be injected such that a dose can last from two- to-four weeks, which can improve patient compliance. A November 2 note from an advanced registered nurse practitioner stated that the patient seemed confused when given discharge instructions to follow up at Lakeside Alternatives. Dr. Berns testified that such confusion can be a red flag that the patient is not ready for discharge. Dr. Berns stated that S.G.'s chronic schizophrenic condition could deteriorate if he were discharged without being well stabilized and not committed to following through with his medications. Dr. Berns concluded that the length of stay was appropriate, and that he might have kept S.G. in the hospital even longer if he appeared confused on the day of discharge. In response, Dr. Mehra testified that the psychotic symptoms, including delusions that his nurse was the Queen of England, did not mean that S.G. must remain in the hospital. Dr. Mehra stated that the note in the chart that the patient believed Artane was a "gasoline pill" was not necessarily a delusion, but could have been an uneducated patient's way of saying that the pill peps him up. Dr. Mehra argued that if the medical concern was persistent psychotic symptoms so severe that he needed hospitalization, then the medical team needed to make dosage adjustments. However, no such adjustments were made after October 26. Dr. Mehra's opinion remained that S.G. did not meet Medicaid guidelines as of October 28. The patient was cooperative on the unit, not a management problem. He could live outside the hospital and still have delusions that someone was the Queen of England. Dr. Birkmire testified that S.G. was rambling, disorganized, and thought the nurse was Queen even after October 28, but that does not necessarily mean he should be in the hospital. Dr. Mehra pointed out that plenty of people walking the street have schizophrenia, are psychotic, and ramble. Dr. Mehra stated that one possible reason S.G. was kept in the hospital was concern as to where he would go upon discharge, because he came from jail and was homeless. Dr. Mehra testified that Medicaid does not cover the period of time when someone is needing placement. Dr. Mehra agreed that S.G. showed psychotic symptoms, but stated that the psychotic symptoms should affect the patient's ability to function in order to justify inpatient treatment. This patient was taking his medications, eating, participating in activities on the unit, and was directable. He did not require any means of seclusion or physical restraints and did not demonstrate aggressive behavior. Dr. Mehra agreed with Dr. Berns that it is a concern any time a patient is readmitted, because it speaks to the chronic nature of schizophrenia and psychoses, and how the symptoms persist over a period of time. "Loose" symptoms probably continued until the day he left. Dr. Mehra also agreed with Dr. Berns that schizophrenics can stabilize and have their thoughts become more organized, depending on the patient and his response to medications. One patient can be loosely organized and live on the street, and another may regain full control of his thoughts. However, Dr. Mehra saw no reason to amend his original opinion. A patient with loose thinking, who is psychotic and disorganized, may need hospitalization, if he is not taking his medicine, not eating, not sleeping appropriately, or is being aggressive. Otherwise, those symptoms may be as good as things are going to get for this patient, given that he has been in jail, has had frequent inpatient hospitalizations, and is homeless. Again, Dr. Mehra noted that S.G. remained in the hospital for several days with no changes to his medications. The greater weight of the evidence supports AHCA's denial of the last four days of S.G.'s admission. The record as presented established that S.G. ceased to meet the criteria for inpatient admission in the psychiatric unit on October 28. Patient #4 C.M. C.M. was admitted on March 2, 2001, and was discharged on March 10, 2001. Peer reviewer Dr. Mehra determined in his report that four days, March 6 through March 10, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that C.M. was a 16-year-old male admitted, involuntarily, under the Baker Act from the Orange County jail, where he was banging his head on the wall. Hospital records indicated concerns by hospital staff that C.M. was malingering in an effort to avoid his pending legal woes. Dr. Mehra found that, as of March 6, C.M. was not actively suicidal, homicidal, grossly psychotic, manic, or in complicated withdrawal. Dr. Mehra concluded that the patient should have been discharged to the juvenile detention center, with psychiatric consultation. In his deposition, Dr. Mehra testified that the hospital had a range of diagnoses for this patient, from "malingering," meaning that he was intentionally inventing his symptoms, to a concern about schizophrenia. Dr. Mehra's review of the hospital records led him to conclude that C.M. was malingering, based primarily on C.M.'s statement to his grandmother that he would get himself placed in the psychiatric unit anytime he went to jail. The record stated that C.M. said the devil was telling him to kill himself, but the physician and staff all thought C.M. was malingering. C.M.'s statement about killing himself led to no increase in precautions, and his medications were not increased until the next day. By the end of C.M.'s hospitalization, the physician was talking about tapering the boy completely off of Respiradol, an anti- psychotic, and was convinced that C.M. was feigning his symptoms. Attending physician Scott D. Farmer7 contended that C.M. remained very dangerous on March 6, because he was still complaining of command hallucinations. "Command hallucinations" cause the patient to believe there are voices telling him to act in a dangerous way, and are recognized as a "unique risk factor" justifying inpatient care. On March 6, C.M. was hearing the voice of his grandfather reassuring him, but he was also hearing the voice of the devil telling him to kill himself. Dr. Farmer testified that patients have been known to kill themselves when they have persisting command hallucinations, and this was a patient who bangs his head against a brick wall. This was an indication that his medications had not been properly adjusted, and that they could not be so adjusted on an outpatient basis. Dr. Farmer's opinion was that it was "ludicrous" to think this patient could be placed in a more complex environment and get better. To discharge C.M. on March 6, would have constituted "abandonment." Dr. Farmer contended that it is a "glib assumption" to say that C.M. was faking his illness, and it is not within the spirit of psychiatry to prejudge that a patient is falsifying his expressed distress. The tradition in medicine is to compassionately adjust medication to remedy the symptom complex, which in this case pointed toward schizophrenia. Dr. Farmer stated that the faking allegation is "a reflection of the lowest form of psychiatric practice. It is a departure from the Hippocratic oath to do no harm. It is an assumption that you can climb inside of somebody else's head and then make conclusions that are a distinct departure from what the patient is saying." Dr. Farmer pointed out that C.M. had been treated at least once for a prior suicide attempt. He also pointed out that a "first break" psychotic episode is the best opportunity for treatment to have a favorable impact in the case of a patient with command hallucinations. Subsequent episodes require more aggressive treatment and higher doses of medications. C.M. was being treated with antipsychotic and antidepressant medications. On March 6, he was taking Wellbutrin, an antidepressant that has the lowest likelihood of triggering manic-type symptoms, and Risperdal. His medications were increased on March 6 and March 7. On March 8, C.M. was still responding to internal stimuli, carrying on a conversation with an internal voice. Dr. Farmer agreed that C.M. was stabilized by March 9 and should have been discharged on that date rather than on March 10. Dr. Berns noted the suspicions of malingering, but also considered that jail staff could not handle C.M., that he appeared to be in imminent danger of harming himself, and he had been treated for at least one suicide attempt in the past. These factors raised concerns as to how much of C.M.'s behavior was malingering and how much indicated genuine illness. Dr. Berns was influenced by the fact that C.M. requested an increase in his dosage of Risperdal, which is not a medication that can be abused or used for intoxication. Dr. Berns acknowledged that C.M. lost some credibility with his statement that he would continue getting Baker Acted if incarcerated. He also acknowledged that C.M.'s age and impulsiveness made it harder to determine the extent of his malingering, but that there was undoubtedly some malingering present in this case. Dr. Berns concluded, as did Dr. Farmer, that C.M. could have been discharged a day or two earlier than March 10. Dr. Mehra replied that it was the treating physician, Dr. Birkmire, who concluded that C.M. was malingering. The auditory hallucinations on March 6 were not sufficient to keep him in the hospital where the treating physician and the medical team believed he was making up the symptoms. Nothing in the testimony of Dr. Farmer or Dr. Berns caused Dr. Mehra to change his opinion. Both doctors referred to this patient's having a diagnosis of schizophrenia. Dr. Mehra called this a "serious and unusual diagnosis" for a 16-year-old, similar to a diagnosis of cancer in that the patient will have to live with it for the rest of his life. Dr. Mehra would expect that such a diagnosis would have led the treatment team to meet with C.M.'s family to offer the appropriate treatment planning and education regarding schizophrenia, but the record indicated that no such meeting occurred. Dr. Mehra believed that there was cause to admit C.M. for evaluation, because he was only 16 years old. Even if it turned out he was malingering, it was prudent to admit him for four days to evaluate him. The greater weight of the evidence supports AHCA's denial of the last four days of C.M.'s admission. The record as presented established that C.M. ceased to meet the criteria for inpatient admission in the psychiatric unit on March 6. Patient #5 L.M. L.M. was admitted on May 16, 2001, and was discharged on May 22, 2001. Peer reviewer Dr. Mehra determined in his report that three days, May 20 through May 22, should be denied due to lack of medical necessity for continued inpatient care. Florida Hospital Orlando did not contest Dr. Mehra's denial of three days for this admission. Patient #6 H.P. H.P. was admitted on March 7, 2001, and was discharged on March 14, 2001. Peer reviewer Dr. Mehra determined in his report that four days, March 10 through March 14, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that H.P. was a 34-year-old female admitted, involuntarily, under the Baker Act for "suicidal ideation and auditory hallucinations." As of March 10, H.P. was not actively suicidal, homicidal, psychotic, manic, or in complicated withdrawal. Her sleep, vital signs, and appetite were stable. Dr. Mehra concluded that the patient no longer needed 24-hour psychiatric nursing care and could have gone back to the skilled nursing facility on March 10. Outpatient was the appropriate level of care. In his deposition, Dr. Mehra testified that H.P.'s improvement was such that she could have been discharged on March 10. There was no deterioration in her condition after March 10. She denied suicidal or homicidal ideations, hallucinations, and delusions throughout the day. Dr. Mehra noted that H.P. was HIV-positive and obese, and would therefore chronically be at risk for suicidal ideation. She had been hospitalized many times for suicidal ideation and auditory hallucinations. H.P. claimed to have jumped from a five-story building when she was 18 years old. Dr. Mehra did not think H.P. was schizophrenic, though her attending physician was concerned about major depression with psychotic features. Attending physician Dr. Luis Allen testified that H.P. was admitted from a skilled nursing facility. She had had multiple psychiatric hospitalizations, and on this admission was presenting with psychotic symptoms, hearing voices and having suicidal thoughts. Dr. Allen conceded that there was one day during her stay when H.P. reported not having suicidal thoughts, but he added that these thoughts resumed the next day. Given that H.P.'s history made her a higher risk for suicide, Dr. Allen felt that he had to ensure she was stable psychiatrically before she could return to the skilled nursing facility. Dr. Berns testified that the March 7 admission note indicated that H.P. had a history of depressive disorder and multiple psychiatric admissions to Florida Hospital Orlando. H.P. reported insomnia and auditory hallucinations, which were mostly command and derogatory hallucinations, voices calling the patient "stupid" and a "dummy" and saying that she should kill herself. H.P. had a history of several suicide attempts and had been taking Risperdal, Prozac, and Remeron. She reported suicidal thoughts, but no plan, and was alert times three. The admitting diagnosis was major depressive disorder, recurrent with psychotic features, and rule out mood disorder secondary to medical condition with depressive-like features. Dr. Berns agreed that the March 10 notes showed that H.P. was depressed with blunted affect, and that she denied suicidal ideation. On March 11, the notes indicated that H.P.'s mood was improved, that she slept better during the night, that her suicidal thoughts were significantly decreased, and that she was compliant with her medications and reported no auditory hallucinations or delusions. However, the attending physician continued to note that she was depressed and hopeless, and the social worker reported that H.P. discussed having no desire to live any more and be a burden to her children. On March 13, her behavior was improved, she had a very good appetite, and she had no hallucinations, delusions, or suicidal thoughts. On March 14, she denied suicidal ideation and hallucinations and was discharged back to the skilled nursing facility. Dr. Berns testified that H.P.'s stay was necessary, and disagreed that she could have been discharged to a skilled nursing facility on March 10. She had a previous suicide attempt, and she had suicidal thoughts and heard voices telling her to harm herself. Dr. Berns agreed that many people express thoughts such as those H.P. expressed to the social worker on March 11, but Dr. Berns pointed out that many people do not also have previous attempts or voices telling them to kill themselves. H.P. was showing improvement by March 11, but her symptoms were still present. Dr. Berns did not agree with Dr. Mehra that suicidal ideation, without the means to carry out a plan, is never sufficient to keep a patient in the hospital. Dr. Berns stated that if the patient is having thoughts of suicide and staff is documenting that the patient is helpless and hopeless, the suicide risk may be sufficient to hospitalize the patient, particularly where there have been previous hospitalizations and suicide attempts. After reviewing the testimony of Dr. Allen and Dr. Berns, Dr. Mehra maintained his opinion that H.P. should have been discharged on March 10. Dr. Mehra pointed out that no physician saw the patient on March 10, probably because it was a Saturday.8 Regardless of the day of the week, if the patient's condition is so acute that it is necessary to hospitalize her with a risk factor of attempted suicide, then she should be seen by a doctor. The unit notes for March 10 indicate she was showing no evidence of psychoses or suicidal ideation. Dr. Mehra noted that even H.P.'s mental status exam at the time of admission showed no active suicidal plan. If the patient is sick enough to be in the hospital, then she should have been seen by a physician. Dr. Mehra concluded that there appeared no need for H.P. to be seen by a doctor on March 10. The greater weight of the evidence supports AHCA's denial of the last four days of H.P.'s admission. The record as presented established that H.P. ceased to meet the criteria for inpatient admission in the psychiatric unit on March 10. Patient #7 J.R. J.R. was admitted on March 2, 2001, and was discharged on March 15, 2001. Peer reviewer Dr. Mehra determined in his report that ten days, March 5 through March 15, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that J.R. was a 46-year-old female admitted from a skilled nursing facility under the Baker Act for being "suicidal." As of March 5, the patient was not actively suicidal, homicidal, psychotic, manic, or in complicated withdrawal. She needed intensive outpatient treatment to address her depressive symptoms, eating issues, and possible addiction to narcotics. J.R. refused to see a chronic pain specialist while in the hospital. Dr. Mehra's report also raised a "serious quality of care concern" in the fact that the patient was immediately referred for electroconvulsive therapy ("ECT") treatment based on "unclear, poorly documented reasons." The physician's decision to use ECT was based purely on the patient's report, without documentation, of failed past treatments. The patient reported that she had not had psychotherapy in years. In his deposition, Dr. Mehra testified that the admitting concern was that the patient was suicidal in her skilled nursing facility, and Dr. Allen had her Baker Acted into the hospital. Dr. Farmer performed the psychiatric evaluation on admission and stated that the patient was not actively suicidal, not psychotic, and was angry with Dr. Allen for Baker Acting her. Dr. Farmer diagnosed J.R. with major depression and anorexia nervosa. Dr. Mehra concluded that J.R. should have been discharged on March 3, because the initial admission evaluation did not establish medical necessity for an inpatient psychiatric hospitalization. Dr. Mehra testified that it was difficult to determine why the patient was at this level of care because Dr. Farmer found that J.R. was not suicidal, actively suicidal with a plan, or psychotic, had no form of auditory or visual hallucinations, and was not manic or involved in a complicated withdrawal. Dr. Mehra stated that, while he did not come out and say that J.R. should not have been Baker Acted, Dr. Farmer did repeatedly note the patient's anger at being Baker Acted, which is highly unusual in a psychiatric evaluation report. Dr. Mehra also observed that the Baker Act documents were not dated, and, thus, there was no way of saying they were completed on the day of J.R.'s admission. A Baker Act is an involuntary commitment of a patient, and it involves a patient's rights. Dr. Mehra stated that because the papers must be completed within a specified period of time of having face-to- face contact with the patient, they must note the date and time. The failure to fill out the papers completely causes a concern about the appropriateness of the patient's admission. Dr. Mehra believed that the indication for ECT was not clear. There are specific criteria to initiate a patient on ECT, and the medical record here did not support it. One of the ECT criteria is that the patient must have failed a minimum of three antidepressants at adequate doses for an adequate length of time. The only documentation in the record was the patient's own report that medications had not worked. There was no objective data in the record regarding her medication history. Dr. Mehra found it very significant that J.R. had no prior psychiatric hospitalizations, especially in the context of her being given ECT. There was not adequate medical confirmation that she had failed previous antidepressant therapy. Dr. Mehra noted that J.R. had a history of a cervical spinal fusion and issues of lumbar back pain, which should have raised concerns about inducing a grand mal seizure by way of ECT. Dr. Mehra saw no MRI or CAT scan of the brain, which is usually done prior to the administration of ECT in order to rule out a mass in the brain. Dr. Allen testified that J.R. had a history of recurrent depression with psychotic features and an eating disorder. During her initial visit to the skilled nursing facility, she was found to be anxious, depressed, and experiencing some suicidal thoughts, and was referred for inpatient treatment. She had experienced significant weight loss and had issues of untreated depression. ECT was initiated and performed three times a week. Dr. Allen conceded that ECT may sometimes be done on an outpatient basis, but stated that J.R.'s history of psychiatric symptoms and the low level of support she had in the community necessitated inpatient placement. She was at a nursing home and would have had to be transported at 5:30 every morning for the treatment. There was no transportation available for her to come in as an outpatient. Dr. Allen also felt that she needed to remain inpatient because of the confusion and disorientation that she was developing with each treatment. Dr. Allen stated that the primary reason for keeping J.R. as an inpatient was to give the ECT treatment. He decided to complete the course of ECT treatment and discharge her back to the nursing home. Dr. Farmer was the second opinion doctor who actually performed the ECT treatment. Dr. Allen stated that ECT is usually reserved for patients who are considered treatment intolerant or "refractory" to treatment, with a history of failing different trials of medications or having developed side effects, or patients who had a very high risk of suicide. With J.R., it was not clear how much the eating disorder was playing into her depression, but Dr. Allen believed that her inner functions were clearly deteriorating, as evidenced by the fact that she was in a nursing home at age 45. The ECT was to address her primary mood symptoms and appetite. At the nursing home, she was only eating percent of her meals. She was eating 50 percent of her meals when she left the hospital, and she continued to show improvement at the nursing home. Dr. Berns testified that J.R. showed a history of cervical spinal cord injury. J.R. was a nurse, and an aggressive patient had caused her injuries when she worked in the emergency room. She had a history of anorexia, depression, and alcohol abuse. She was agitated in the emergency room during admission. J.R. claimed she had had trials on all available antidepressants, which were only partially helpful or failed. She had insomnia and took Klonopin for restless legs. On admission, she was also taking OxyContin, Wellbutrin, Flomax, Trazedone, and Zofran. Her mental status examination indicated lethargy and monotone speech. She was depressed, helpless, and hopeless, but denied suicidal ideation. J.R. stated that she had an overdose of medications at age 16. On March 5, she was withdrawn, depressed, and complained of anergia (lack of energy) and anhedonia (inability to experience pleasure). She also complained of dizziness. She showed the same symptoms on March 6, hopeless and helpless but denying suicidal thoughts. She again complained of feeling weak and dizzy, and she had low blood pressure, 73 over 53. Because of concerns that her medications may have been the cause of her medical complaints, the treatment team decided to withhold all psychiatric medicines and initiate ECT. J.R.'s first ECT treatment was on March 7. On that date, she was depressed, withdrawn, had anergia, but no suicidal thoughts. Her dose of Klonopin was lowered. On March 8, her mood was depressed and she showed anergia, anhedonia, and a variable appetite. A trial of a new anti-depressant, Remeron, was commenced, and J.R. was given an Ambien sleeping pill at bedtime. Her blood pressure was still low, 70 over 50, and the treatment team decided to withhold the OxyContin. J.R. refused to see the psychiatrist and stayed secluded in her room. By the second ECT treatment on March 9, J.R.'s mood was improving and her blood pressure was up to 90 over 57. ECT was scheduled for three days during the next week. On March 10, J.R. complained of depression and suicidal thoughts, and stayed alone in her room most of the day. On March 11, she again stayed secluded in her room, depressed and with flat affect. On March 12, she slept fairly well and ate 75 percent of her meals. Her mood was improving and suicidal thoughts decreased. She had ECT in the morning then rested in bed most of the day. She was depressed and anxious, with poor insight and judgment. On March 13, her mood was improving and she denied suicidal ideation. J.R. was more goal oriented and showed less psycho- motor retardation. She was scheduled for discharge on March 14, after her ECT treatment. She had the treatment, but her discharge was placed on hold because the skilled nursing facility did not want to accept her. Dr. Berns was not sure why the facility did not want to take J.R. back, unless they considered her a problem patient or didn't want to handle a depressed patient. Such problems can hold up discharge. Dr. Berns testified that it would not be acceptable to discharge this patient to the street, and that the length of stay was medically necessary. Dr. Mehra countered that feeling depressed, helpless and hopeless is not enough to justify an inpatient admission. On J.R.'s mental status exam upon admission, Dr. Farmer documented no suicidal ideation, no psychotic symptoms, and patient anger at being Baker Acted. She was given ECT for reasons that Dr. Mehra thought were not very well documented in the medical record. Dr. Mehra stated that it is a complex question as to whether giving her the ECT treatment is reason enough for an inpatient admission. J.R. was taking a heavy narcotic medication, OxyContin, which can make one depressed, withdrawn, and isolative. Dr. Mehra could find in the medical record no real theory as to why J.R. was still taking these medications. For reasons unclear to Dr. Mehra, there seemed to be an immediate desire to give her ECT. Dr. Mehra believed that March 2, should be authorized just to see what was going on with her, given that she had been Baker Acted by a physician. However, as far as the record indicated, J.R. had no history of inpatient psychiatric treatments. Dr. Mehra found it very unusual that ECT treatments would be given on a patient's first inpatient stay. ECT is routinely performed on an outpatient basis. Dr. Mehra acknowledged that where patient compliance is a problem, it may be proper to keep the patient in the hospital. However, this did not seem to be the case with J.R. One of the cornerstones of the decision process leading to ECT is a documented failure of past antidepressant therapy, and Dr. Mehra found no such documentation in the medical record. Lack of support in the community or transportation problems are not reason enough, standing alone, to keep someone in the hospital. The greater weight of the evidence supports AHCA's denial of the last ten days of J.R.'s admission. The record as presented established that J.R. ceased to meet the criteria for inpatient admission in the psychiatric unit on March 3. Patient #8 J.R. J.R. was admitted on March 16, 2001, and was discharged on March 27, 2001. Peer reviewer Dr. Mehra determined in his report that six days, March 21 through March 27, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that J.R. was a 24-year-old male with a diagnosis of schizoaffective disorder, who was admitted for symptoms of agitated, psychotic behavior. As of March 21, the patient was not actively suicidal, homicidal, manic, or in complicated withdrawal. J.R. may have still been psychotic, but the symptoms had improved. The patient's vital signs, sleep, and appetite had stabilized. He no longer needed 24-hour psychiatric nursing care. Dr. Mehra concluded that sub-acute treatment was the appropriate level of care as of March 21. In his deposition, Dr. Mehra testified that J.R. met discharge criteria on March 21, and could have gone into a sub- acute setting such as a skilled nursing facility or group home. Dr. Mehra could not recall whether there was a problem with bed availability in the skilled nursing facility, but added that after a patient meets discharge criteria, it is not Medicaid's responsibility to pay for a longer hospital stay while the patient awaits placement. Once the schizophrenic's acute crisis is resolved, he no longer meets the criteria for medical necessity. Dr. Birkmire was the treating physician, and his notes indicated compliance and improvement on March 20, and 21. J.R. was having some religious preoccupation, which was probably a baseline issue for him. Religious preoccupations in people with schizophrenia are sometimes chronic and never go away. Dr. Mehra testified that being psychotic and having auditory hallucinations do not alone establish grounds for remaining inpatient. Such symptoms are consistent with a diagnosis of schizophrenia, which is chronic. Dr. Mehra stated that J.R. would probably display these symptoms no matter how long he was kept in the hospital. Dr. Mehra testified that there is a vast difference between J.R.'s having an auditory hallucination and his having a command auditory hallucination to harm himself or someone else. Treating physician Dr. Birkmire testified that J.R. was very disorganized on admission, with a grandiose religious preoccupation that indicated he was in a manic stage. J.R. was also sexually inappropriate with some of the other patients and staff. On March 21, he was inappropriate, grandiose, and sexually preoccupied. On March 22, he was still very psychotic, hearing the voice of Britney Spears, with whom he had delusions of being married. He had sexually explicit conversations with other patients and staff. On March 23, he was severely agitated, requiring staff intervention. He was illogical and bizarre, talking to himself in the hallway, reporting that he heard voices and remaining delusional. He was hearing the voice of the devil and having paranoid thoughts about people around him. On March 24, he was unchanged, still delusional and still hearing voices, though they were becoming less intense and he was becoming less manic. Dr. Berns testified that schizoaffective disorder is an illness with symptoms of psychosis in the absence of symptoms of a mood disorder such as mania or depression. It is a chronic mental illness. Dr. Berns agreed with Dr. Mehra that the presence of a chronic illness is not grounds for hospitalization. It is only when the condition becomes acute, where the patient presents a danger to himself or others, that an inpatient psychiatric hospital may be the option. J.R.'s admission note stated that he was previously hospitalized in January 2001, for agitation, bizarre delusions, and concerns about violent behavior. He stated dead people were talking to him. On March 17, the psychiatrist noted marked auditory hallucinations, grandiosity, paranoid delusions, and tangential thought processes. The plan was to keep him on Risperdal, an antipsychotic, and Lithium, which is a mood stabilizing, anti-manic medication. Dr. Berns testified that the usual practice with these medications is to start with a low dose and build it up slowly and gradually. If the patient is in the hospital, the physician can be more aggressive because he can closely monitor blood work and vital signs. Lithium takes seven to ten days to build up to a therapeutic level. The medical notes from March 22 showed that J.R. remained psychotic, had auditory hallucinations, had delusional thoughts regarding Britney Spears and Judy Garland, was responding to internal stimuli, and was sexually preoccupied. On March 23, he was agitated and illogical with bizarre ideation. On March 24, he continued to report auditory hallucinations and was labile and agitated about his upcoming discharge. On March 25, he was still having auditory hallucinations, but less of the manic behavior. On March 26, there was some improvement in his mood and his auditory hallucinations were resolving, but he was still having problems in a group situation. Dr. Berns noted that on March 17, J.R. tried to kiss a nurse, then called her a "bitch with an attitude." There were concerns about his impulse control and potential for committing a sexual offense if released before he was fully stabilized. Dr. Berns agreed with the length of hospitalization, because J.R. had shown poor impulse control and sexually inappropriate behavior on admission, had been admitted two months earlier, and there were concerns about psychosis and violent behavior. The greater weight of the evidence supports AHCA's denial of the last six days of J.R.'s admission. The record as presented established that J.R. ceased to meet the criteria for inpatient admission in the psychiatric unit on March 3. Patient #9 N.R. N.R. was admitted on March 4, 2001, and was discharged on March 16, 2001. Peer reviewer Dr. Mehra determined in his report that three days, March 13 through March 16, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that N.R. was a 21-year-old female admitted, involuntarily, under the Baker Act for psychotic agitation and delusions. The authorization was based on the patient's presenting symptoms, diagnosis of intrauterine pregnancy, and a positive test for syphilis. As of March 13, the patient was not actively suicidal, homicidal, manic, or in complicated withdrawal. Her vital signs, sleep, and appetite were stable. She was taking prescribed medications and following unit rules. Her psychosis had decreased and she was not an immediate danger to herself or others. She appeared to have support from her mother. Dr. Mehra concluded that outpatient was the appropriate level of care. In his deposition, Dr. Mehra testified that the physician's note of March 13 stated that N.R. had no recent episodes of bizarre behavior, and no episodes of agitation or aggressive behavior. The physician progress notes for March 14 say the same thing. N.R.'s initial evaluation was for a psychotic disorder, and she was diagnosed with schizophrenia, a chronic condition. Dr. Mehra noted that the hospital did not obtain N.R.'s previous psychiatric history and that her mother could have been contacted about N.R.'s medical records. The treating physician, Dr. Allen, testified that N.R. was psychotic and grandiose on admission, and was noted at the jail to be head-banging, smearing feces, and playing in the toilet. She was pregnant with an unknown gestational age, and had a positive Rapid Plasma Reagin ("RPR") test for syphilis. She needed a lumbar puncture to determine if she had some form of neurosyphilis or another disorder that could influence her psychiatric behavior. Dr. Allen testified that she lacked the social network for the lumbar puncture to be done on an outpatient basis. He conceded that on March 13 she was compliant with her medication, but stated that she was still disorganized. The initial RPR was performed on March 4, but it took an additional ten days for the lumbar puncture to be successfully performed. One puncture was performed on March 10, but the specimen was not good, so another puncture was performed on March 14. Dr. Allen testified that the lumbar puncture requires the patient to remain very quiet in a hunched position as the needle is going through her back. An agitated, restless patient could cause problems. N.R. needed to be stabilized before the puncture could be performed. N.R. was found positive for syphilis and was discharged to her mother's home rather than to the jail. A visiting nurse went to the home to give the treatments. Dr. Allen explained that N.R. was not released to her mother between lumbar punctures because the mother had a history of depression, according to the history provided by N.R., which Dr. Allen conceded may not have been accurate. Dr. Allen also noted that N.R.'s mother was very difficult to contact. Dr. Berns testified that N.R. was psychotic on admission. At the jail, she was stripping off her clothing and hearing voices telling her that her husband was messing around with other women. She was hitting her head against the wall to get rid of the voices. Dr. Berns concluded that the length of stay was medically necessary and reasonable. N.R. was a pregnant female from the jail, psychotic, with self-destructive behavior. Her physician wanted to be extra careful in view of N.R.'s being pregnant. She was placed on antipsychotic medication. N.R. was eventually going back to the jail, and they wanted to stabilize her condition as much as possible, because the jail is a very stressful place to be. Dr. Berns agreed that, as of March 13, N.R.'s psychosis had decreased and she was not an immediate danger to herself or others. The greater weight of the evidence supports AHCA's denial of the last three days of N.R.'s admission. The record as presented established that N.R. ceased to meet the criteria for inpatient admission in the psychiatric unit on March 13. Patient #10 R.S. R.S. was admitted on May 19, 2001, and was discharged on May 30, 2001. Peer reviewer Dr. Mehra determined in his report that nine days, May 22 through May 30, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that R.S. was a 38-year-old female diagnosed with schizophrenia who was admitted, involuntarily, under the Baker Act for worsening of her psychotic symptoms. As of May 22, the patient was not actively suicidal, homicidal, manic, or in complicated withdrawal. The patient was still psychotic but the symptoms were not worsening and appeared to be at baseline, given that they remained unchanged even at discharge. After May 22, the record disclosed no evidence that R.S. required seclusion or emergency treatment orders to force her to take medications. She voluntarily took medications and followed unit rules, and no longer needed 24-hour nursing care. Dr. Mehra concluded that sub-acute treatment was the appropriate level of care as of May 22. Dr. Mehra testified that his main concern with the length of R.S.'s stay was that she persisted with active psychotic symptoms even up to the day of discharge. There was not much difference in her symptoms between May 22 and May 30. After May 22, she received no medication on an "as needed" ("PRN") basis and required no seclusion or restraints. Despite her other noted symptoms, if she was not having command hallucinations telling her to hurt herself, she could have been treated as an outpatient. Dr. Mehra agreed that if she expressed a desire to hurt herself after May 22, then any such days should be authorized. Dr. Mehra did not have the benefit of R.S.'s past medical records and stated this made it difficult to determine if R.S. was a patient whose baseline level of functioning is so low that she could never care for herself. He agreed that the attending physician had the advantage of having seen the patient in person. Attending physician Dr. Birkmire testified that R.S. had just been discharged from a medical psychiatric unit but had to be readmitted on a Baker Act because she was not taking her medications and was psychotic again, hearing and responding to voices and unable to communicate in a meaningful manner to staff. On May 22, she was noted as unchanged from admission, still psychotic, disorganized, hearing voices, depressed, and oriented only to person and place, not date and time. She was secluded in her room, not going to any treatment groups. Dr. Birkmire testified that his greatest concern was that R.S. was so psychotic and disorganized that she would probably not take her medications and would not be not able to care for herself. On May 23, she remained regressed, bizarre and psychotic. She was "dirty and careless" in her ADLs, and was complaining of suicidal command ideations, voices telling her to hurt herself by taking an overdose. On May 24, she was showing mild signs of improvement and was a little less reclusive and bizarre. She was still hearing voices and claimed they were telling her to be a friend to everybody, but she also admitted to suicidal ideations. On May 25, R.S. was still psychotic with suicidal ideations, disorganized, and paranoid. By this time, the voices were telling her to do more good things than bad, but she needed more time to stabilize on her medications. She was still dirty and careless in her appearance, and depressed with a flat affect. On May 26, she showed further mild improvement, was less paranoid, and reported that the voices were less intense. She remained in her room most of the time, and her ADLs were still careless. On May 27, R.S. was now oriented to person, place and time. The treatment team still thought she would stop taking medications on release, and made, further, two more medication changes. On May 28, the hallucinations had resolved. She was more logical and organized and less paranoid, though her ADLs were still poor. By May 29, she was safe to go home, but her appearance was still disheveled, she had poor concentration and hygiene, still heard voices, and was depressed and anxious. She went to group therapy, but was distracted by auditory hallucinations. Dr. Birkmire believed that May 29 or May 30, would be an appropriate discharge date. Dr. Berns testified that R.S.'s diagnosis was schizophrenia. He acknowledged that R.S.'s suicidal ideations on May 22 did not indicate a plan. However, in the hospital setting, a patient may not always reveal her plans for fear of prolonging the hospitalization. On May 23, she was having command auditory hallucinations to hurt herself with a plan to overdose. Dr. Berns found this very serious, because patients have been known to hoard medications in the hospital in order to take an overdose. After reciting the same day-by-day review conducted by Dr. Birkmire, Dr. Berns concluded that he concurred with the length of stay. Noting that she was still psychotic, mumbling and hallucinatory at the time of discharge, Dr. Berns testified that he did not necessarily agree that R.S. should have been released even on May 30. In response, Dr. Mehra testified that he agreed with many of the concerns expressed by Dr. Birkmire, but did not see them evidenced in the medical record. Dr. Birkmire testified that as of May 22, R.S. was so psychotic and disorganized that she would probably not take her medications after discharge. Dr. Mehra agreed this is grounds for keeping someone in the hospital, but did not see this concern noted in the medical record. Dr. Mehra also found nothing in the medical record to indicate she was having command auditory hallucinations to hurt herself with a plan to overdose. Dr. Mehra stated that the record did show that on May 23 she was regressed, bizarre and psychotic, but he noted that those symptoms were also present at the time of R.S.'s discharge. However, the psychiatry unit patient notes do indicate that R.S. told a student nurse on May 23, that "she was having suicidal ideations and a plan to overdose." The other notations cited by Dr. Birkmire, with the exception of the May 22 notation regarding staff's suspicions regarding R.S. medication compliance, were all found in the psychiatry unit patient notes. It must be concluded that Dr. Mehra simply overlooked this section of the record, and that if he had seen that the record supported Dr. Birkmire's concerns, Dr. Mehra would have authorized the full stay for R.S. The greater weight of the evidence supports Petitioner's position that May 22 through May 30, should not have been denied. Dr. Mehra's reason for denying these days was not that he disagreed with Dr. Birkmire's concerns regarding the patient, but that he could not find those concerns reflected in the record. In fact, the record supported Dr. Birkmire's concerns and rendered R.S.'s entire inpatient stay medically necessary. Patient #11 D.T. D.T. was admitted on May 9, 2001, and was discharged on May 21, 2001. Peer reviewer Dr. Mehra determined in his report that nine days, May 12, through May 21, should be denied due to lack of medical necessity for continued inpatient care.9 Dr. Mehra's peer review report stated that R.S. was a 14-year-old female with an extensive psychiatric history who was voluntarily admitted from a community mental health center for mood swings, violent tendencies, and a report of auditory hallucinations. As of May 12, the patient was not actively suicidal, homicidal, grossly psychotic, manic, or in complicated withdrawal. The patient was taking her medications, following unit rules, and participating in activities. The patient's sleep, vital signs, and appetite were stable. Dr. Mehra concluded that she did not require 24-hour nursing care as of May 12 and that outpatient was the appropriate level of care. Dr. Mehra testified that D.T. was admitted for mood swings, a history of violence, and some personality issues. She was diagnosed as bipolar, though Dr. Mehra was not clear as to what features led to that diagnosis. She was also diagnosed with post traumatic stress disorder. Further, she was diagnosed under Axis II of the DSM-IV, which includes personality disorders. Dr. Mehra found this significant because personality disorder symptoms are treated differently than bipolar disorder. Dr. Mehra found no indication in the record that D.T.'s history records were ordered or reviewed for purposes of continuity of care and current treatment. She was admitted and served on the inpatient unit, and her medications were continued: such as Wellbutrin, Topamax, and Risperdal. She was started on Geodan, an antipsychotic, which was then changed to Seroquel, and then back to Risperdal. Dr. Mehra concluded that D.T. met discharge criteria as of May 12. Nothing remarkable happened between May 12, and the date of her discharge. She continued to have some difficulties on the unit, but nothing that warranted inpatient care. She could have been sent to outpatient and returned to foster care. Given her documented history, return to a stable group home or foster home would be appropriate to help her engage with her symptoms. For someone with a personality condition to be on an inpatient psychiatric unit can worsen the symptoms. Dr. Mehra believed that D.T. needed an environment with a lot less stimulation and less potential for her to become agitated and act out. Dr. Mehra acknowledged that the records do show serious medication side effects on May 15, such as akathisia, but he stated that people have these symptoms frequently as outpatients. Dr. Mehra was curious as to why D.T. was on three different antipsychotic drugs. Three different psychotropic medications is a concern because there is insufficient clinical and medical data to use them in children when the diagnosis is not clear. They have potential long-lasting side effects, such as tardive dyskinesia, where the patient develops permanent tic- like movements of the lips, mouth and jaw. Cogentin is a medication for side effects from antipsychotics, and its use caused Dr. Mehra to question whether research was done as to whether this child had been on such medications before. Dr. Mehra stated that such research is essential, especially when the patient starts showing side effects. Treating physician Dr. Scott Farmer testified that all of the denied dates represented necessary periods of care to stabilize D.T. and make her discharge safe. On May 15, she was still experiencing severe mood swings and dissociative symptoms, which Dr. Farmer described as "a watershed between normal and neurotic." In a spectrum moving toward psychosis, dissociative thinking has features of both psychotic and neurotic thinking. Dr. Farmer explained that if a physician has a patient in his office who is disassociating, the physician must watch the patient until the patient has demonstrated several hours of improved functioning. If a patient drifts into a dissociative state during psychotherapy and becomes agitated, the patient requires hospitalization. Dr. Farmer stated that D.T. could not have been released on May 15. She was still requiring Haldol due to episodes of anger, and due to her inability to recognize people who are caring for her and distinguish them from threats. Her agitation was so extreme that Dr. Farmer had to change the dose of Haldol. This was complicated by the fact she was having side effects of the antipsychotic medication. Akathisia is an acute dystonia, a side effect of these powerful medications. It is a restlessness, an inability to sit still. The patient wants to stretch her legs and flex her muscles to relieve tension that feels like an unrelenting, very slowly developing cramp. Dr. Farmer testified that akathisia is not as dramatic as other side effects because it looks like the disease itself: the patient is restless, can't sit still, and wants to walk around. There is a ramping up in the intensity of treatment for akathisia, culminating in Propranolol, which itself causes 40 percent of users to have a new onset of major depression within a year. Dr. Farmer stated that D.T. was so resistant to taking medications that at times intramuscular medications were required. Geodon was the medicine initially chosen to treat D.T.'s psychotic features because it has the least likelihood of causing weight gain. By May 15, Dr. Farmer had deemed it a failure and was in the process of replacing it with Seroquel. On May 16, D.T. remained actively psychotic with visual and auditory hallucinations, side effects of the medication. Dr. Farmer testified that the dosages of medicine that could possibly make the hallucinations go away, had the side effect of incapacitating her. D.T. could not sit in group and match her mood and comments to the group process. She was too lethargic to function. Dr. Farmer believed that as of May 16, D.T. could not be anywhere but an acute care setting, with nurses and physicians monitoring her response to medications. On May 17, D.T. reported seeing what she described as flashing lights. She moved out of her lethargy into accelerated speech. She was irritable and paranoid, different from the day before when she appeared overdrugged and lethargic. On May 18, she had a severe reaction to the Seroquel, active symptoms of delusional, confused and agitated behavior. Seroquel was discontinued and a new antipsychotic, Restoril, was introduced. On that day, D.T. was noted to be crying and hallucinating. She saw a man in her room and held so fast to the idea of being in danger that she required additional medication, Ativan PRN, to make her relax. On May 19, her Risperdal had to be further adjusted because she was overly sedated. She was disheveled, easily agitated, and still required PRN medications in addition to her standard medicines. On May 20, she remained labile, easily upset and crying. Dr. Farmer concluded, by stating, that this case involved a very complex juggling of medications to get control of auditory hallucinations with other medications striving to compensate for side effects. Dr. Berns testified that D.T. had an extensive psychiatric history and was admitted for mood swings, violent tendencies, and report of auditory hallucinations. The admissions note stated that D.T. was depressed and angry, "ready to kill everybody." She was having problems with flashbacks regarding her history of fights with her father, and was fearful that her mood changes and lability would cause her to lose her foster placement. She had been in foster care since January 2001. D.T. had a history of arrests for fighting, breaking and entering, grand theft auto, and battery. She had a decrease in appetite with a four-pound-weight-loss in the past week. She said that she felt paranoid a lot, and she overreacted to intrusions into her physical space. She heard voices with command features telling her to cut her arms instead of battering her father, and admitted to some prior plans of killing her father. The May 15 notes showed severe mood swings and unspecified dissociative symptoms. She received Haldol for anger episodes. Dr. Farmer discontinued Geodon and began Seroquel, another antipsychotic also used in the treatment of bipolar disorder. She was also given Cogentin intramuscularly, because she had tremors and akathasia. On May 16, D.T. was anxious with sleep disturbance, undescribed auditory and visual hallucinations, and said she was lethargic. On May 17, she was reporting flashbacks about "angel trumpet," which may have been a psychedelic drug. She was observed to have accelerated speech, irritable, perseverant and loquacious. Dr. Farmer raised the level of the Seroquel. D.T. was incoherent from midnight to 6:30 a.m., with auditory and visual hallucinations. She was seeing people, carrying on conversations, making and unmaking her bed, trying to open a window, and mumbling. At times she was manic, hyper-verbal, crying and laughing. She said she was high on her medications. The Cogentin was discontinued. They raised the Seroquel and put her on intramuscular injection of Ativan. On May 18, she had a severe reaction to Seroquel. She was delusional, confused and agitated, but showed no aggression. Dr. Farmer stopped the Seroquel and started Risperdal. She hallucinated seeing a man in her room. On May 19, she was anxious and irritable, having non-command hallucinations. Her Risperdal dosage was increased. On May 20, she was upset and emotional. She requested Haldol and Ativan to calm herself down. On May 21, she was organized, with no flight of ideas or loose associations, and was discharged. Dr. Berns noted that Dr. Mehra found that by May 12, D.T.'s sleep, vital signs and appetite were stable, that she was following unit rules of participating in activities, and that she was not actively suicidal, homicidal, psychotic, manic or in withdrawal. Dr. Berns disagreed with this assessment. On May 16, D.T. still had active psychotic symptoms. On May 17, she was paranoid and irritable, carrying on conversations with unseen people. On May 18, she had visual hallucinations. On May 19, she was disheveled and easily agitated. On May 20, she was labile, and very easily upset. Dr. Berns agreed with Dr. Farmer that the entire stay was medically necessary. In response, Dr. Mehra testified that he was aware of Dr. Farmer's statement that on May 15 D.T. was still experiencing severe mood swings and disassociative symptoms, "which are a watershed between normal and neurotic." Dr. Mehra did not know what that means in terms of the issues in this case, because disassociation is not sufficient to warrant keeping her in the hospital. Visual and auditory hallucinations, in and of themselves, are not sufficient to keep her in the hospital unless they are command hallucinations. Dr. Mehra pointed out that the treating physician's own discharge summary stated that D.T. has personality issues of concern. "Personality issue" means that a lot of the symptoms do not necessarily indicate a major, Axis I diagnosis such as schizophrenia or major depression, but are more about the patient's character and how she relates to people. Dr. Farmer noted on May 16 that, due to side effects of medication, D.T. was lethargic and could not function. Dr. Mehra stated that this might be sufficient to keep her in the hospital, though, again, the treating physician must keep the treatment options in mind and distinguish between someone with a personality disorder who is experiencing hallucinatory symptoms and someone who is schizophrenic. Dr. Mehra was concerned that the physicians were injecting this 14-year-old child with potent anti-psychotic medications and that she was having an adverse reaction. He was further concerned that she was not having much of a response to the medications, which Dr. Mehra found would not be unusual if the diagnosis were inaccurate. Based on the documentation in the record, Dr. Mehra could not be sure that D.T. had a true psychotic disorder that would respond to anti-psychotic medications. Because D.T. was in the custody of the Department of Children and Family Services, Dr. Mehra believed there was "a great likelihood" that she had been physically and/or sexually abused. Many of her symptoms would be easier to understand in the context of past abuse rather than as a diagnosis of schizophrenia. Dr. Mehra found the record confusing as to the rationale for this hospitalization. In the admission mental status exam, the physician documented no well formulated plan for the patient to harm herself. No psychotic symptoms were noted in the admission mental status exam. However, because the patient was presenting with symptoms such as auditory hallucinations, Dr. Mehra authorized and approved a three-day evaluation period. Dr. Mehra stated that he would authorize less time for an adult, but with a child it is important to take sufficient time to obtain a good history. What confused Dr. Mehra was that the medical record showed no clear documentation of collateral information regarding D.T.'s past to understand why she may be disassociating or having mood swings. The greater weight of the evidence supports AHCA's denial of the last six days of D.T.'s admission. The record as presented established that D.T. ceased to meet the criteria for inpatient admission in the psychiatric unit on May 12. However, as stated in note 10, AHCA's recovery is limited to May 14 through May 21. Patient #12 S.T. S.T. was admitted on March 7, 2001, and was discharged on March 17, 2001. Peer reviewer Dr. Mehra determined in his report that eight days, March 9 through March 17, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that S.T. was a 34-year-old female admitted, involuntarily, under the Baker Act for violent, agitated, self-harming behavior in a community services van. The patient had a history of moderate mental retardation, cerebral palsy, seizure disorder, and abnormal EEGs. The patient was prescribed Haldol PRN, but required none on March 9, 10, or 11. As of March 10, the patient was not actively suicidal, homicidal, grossly psychotic, or manic. Later in her hospitalization, the patient did require Haldol/Ativan PRN on several days, though no adjustments were made to her routine antipsychotic doses. Dr. Mehra noted that the only adjustment made to S.T.'s psychotropic dosing was on March 8, when her Depakote was increased. Dr. Mehra also noted that the patient was allowed to sign a voluntary consent for treatment, when her legal guardian should have been involved in the consent process. Dr. Mehra found in the record no informed consent for psychotropic medications. Dr. Mehra found that the diagnostic studies performed on S.T. could have been done on an outpatient basis. Dr. Mehra concluded that, as of March 10, outpatient treatment with a return to the group home setting was the appropriate level of care. Dr. Mehra testified that S.T. was admitted through the emergency room under the Baker Act when she became violent and agitated in a van operated by Spectrum Community Services, the group home where S.T. lived. Her admitting diagnosis was mental retardation in the moderate range, as well as organic mood disorder. Dr. Mehra found no physician's notes in the record for March 8 or 9, which indicated that S.T. must not have many problems. The nurse's notes for March 9 indicated that S.T. was sleeping through the night. The unit notes from three different shifts on March 9 indicated that her behavior was under control and that she denied hallucinations. All of the above findings in the record, coupled with a lack of physician notes, led Dr. Mehra to conclude that S.T. should be discharged. Dr. Mehra stated that everything relative to making a medical decision must be documented in the record. If something is not in the record, then a peer reviewer must assume it did not happen. In this connection, Dr. Mehra noted there was no indication in the record that the hospital contacted S.T.'s group home for a treatment history and status before initiating invasive diagnostic procedures. On March 8, S.T. had a fall on the unit, hitting and cutting her head on the cinderblock wall. Dr. Mehra stated that a patient with cerebral palsy and mental retardation, who is having acute psychological problems and has fallen, is a grave concern, yet he could find no record that she was seen by a physician on March 9. He did find a March 9, note calling for a consultation with Dr. Henry Comiter regarding S.T.'s fall, but no actual physician's visit on that date. The record indicated that S.T. was placed in restraints on March 13 and engaged in threatening behavior on March 14. However, Dr. Mehra noted there had been no such incidents on March 11 through 12. He testified that, if a patient with these underlying medical and psychiatric conditions is kept long enough, she will probably act out. The incident on March 13, alone, was not enough to keep her without getting a legal guardian involved to continue her voluntary legal status in the hospital. This was a great concern to Dr. Mehra because the hospital appeared to be relying on a voluntary consent form signed by S.T. on March 10. S.T. was mentally retarded and possibly incompetent to admit herself to the hospital. Treating physician Dr. Farmer described S.T. as a 34- year-old mild-to-moderately retarded woman living in a group home. The incident that led to her admission was her deteriorated impulse control, agitation, and aggression directed toward the van and toward the staff and other peers during a van ride on an outing. She struck at her peers and the van driver, bit herself on the hands and arms, and stated that she wanted to hurt herself. S.T. was already scheduled to have an outpatient neurology consultation in late March with Dr. Comiter, out of concern for a seizure disorder that was not adequately managed. Because of her agitation, S.T. was sent to the psychiatric unit, which began the process of adjusting her medications. The adjustments were ongoing on March 8. She was seen by Dr. Comiter on March 8, as indicated by a consultation note in the record.10 Dr. Comiter ordered an EEG and CT scan of the head. The CT brain scan was scheduled for March 9, but S.T. was too agitated to undergo the procedure. On March 10, S.T. remained agitated. She refused a shower and was generally careless with regard to her ADLs. On March 11, Dr. Farmer reduced S.T.'s dosage of Ativan in order to calm her and make her more manageable for the CT scan. On March 12, she was less agitated, but not calm enough for the CT and EEG to be completed. She was too agitated to go for the CT study, but too fatigued from the medications to be functional on the psychiatric unit. The reduction in her Ativan dosage did enable her to respond more promptly to questions. On March 13, staff attempted to transport S.T. in the van for her EEG, but she began swinging her arms and had to be placed in seclusion and restraints. On March 14, Dr. Farmer characterized S.T. as naïve regarding her manic grandiosity, unable to recognize that her reactions are disproportionate to the circumstances. Her vocal volume was threatening and her intrusiveness was with ominous import, but she believed she was justified in her reactions. She was paranoid and misreading the likelihood of danger and pain, and so was attempting to intimidate people away from her. By this time, Dr. Farmer believed that she was reacting well to the adjusted medications. However, when the medications got her to the point at which she was not threatening others, she began having balance problems and falling again. On March 15, the medications had slowed S.T.'s psychomotor skills, and she was not assaultive. She was taking her medicines by mouth. S.T. was beginning to return to baseline and Dr. Farmer began considering discharging her back to the group home. She was denying any suicidal or homicidal ideation and denying hallucinations. She was able to be redirected from biting her hand, which was a continuing problem for S.T. She was able to participate in group therapy without disruption, though her ADLs remained careless. On March 16, she remained restless, distractable, impulsive, and aggressive, though the hospital did manage to complete the EEG on that date. The EEG showed no evidence of a seizure disorder. On March 17, S.T. was discharged back to the group home. Dr. Farmer concluded that it was in the best interest of S.T.'s care to keep her through March 17. It was best to accomplish all the needed adjustments to her anticonvulsant medications on an inpatient basis, especially since there were no EEG results until March 16. Those results could have required further adjustments, and Dr. Farmer believed that relying on the group home to make the changes in her medications might not work and could result in her readmission. Dr. Berns agreed with the length of stay because the attending physicians were not only trying to make sure that S.T. was no longer suicidal, they were trying to decrease her agitation and aggression while completing important diagnostic tests. Dr. Berns thought that authorizing her stay only through March 10 would be premature. The greater weight of the evidence supports Petitioner's position that March 10 through March 17, should not have been denied. Dr. Mehra's opinion in this instance was at least partly based on a misreading of the record, i.e., that S.T. was not seen by a physician on March 8 or 9. Dr. Mehra's concerns regarding S.T.'s consent to treatment are serious, but cannot be resolved on this record and do not appear relevant to the question of the medical necessity of S.T.'s hospital stay. Dr. Farmer's testimony as the treating psychiatrist credibly established that S.T.'s entire inpatient stay was medically necessary. Patient #13 W.W. W.W. was admitted on June 17, 2001, and was discharged on June 28, 2001. Peer reviewer Dr. Mehra determined in his report that five days, June 23 through June 28, should be denied due to lack of medical necessity for continued inpatient care. Dr. Mehra's peer review report stated that W.W. was a 43-year-old female admitted to the psychiatric unit after an intentional overdose of psychotropic medications. Dr. Mehra wrote that his authorization was based on continued documentation of the patient's having command auditory hallucinations to hurt herself. As of June 23, the patient was not actively suicidal, homicidal, psychotic, manic, or in complicated withdrawal. Her sleep, appetite, and vital signs were stable. She was not a management problem on the unit. Dr. Mehra concluded that, as of June 23, the patient no longer required 24-hour nursing care and that outpatient treatment was the appropriate level of care. In his deposition, Dr. Mehra testified that W.W. was on the medical floor, then transferred to the psychiatric floor after two days. On discharge, her Axis I diagnosis was alcohol abuse, history of cocaine abuse, and rule-out schizoaffective disorder. The hospital's discharge summary stated that the reason for admission was severe depression, no psychotic features, and having suicidal thoughts. She was integrated into the milieu of the inpatient psychiatric unit and prescribed antidepressant medication. Dr. Mehra believed W.W. should have been discharged on June 23. He stated that, even on the psychiatric unit's admission mental status exam, dated June 17, Dr. Allen noted that W.W. was slightly more cooperative than on the previous day's consult, meaning that Dr. Allen probably saw her on the medical floor after her overdose. Dr. Allen noted there was no active suicidal ideation with a plan. Dr. Mehra testified that this admission psychiatric exam, standing alone, would indicate that W.W. did not need to be admitted at all. However, the totality of her presentation and history showed numerous overdoses on psychotropic medications. Based on her history and the mental status exam showing her mood was still depressed, her admission through June 22 was approved. Attending physician Dr. Birkmire testified that, in the weeks before the overdose, W.W. described becoming increasingly depressed with feelings of helplessness and hopelessness. Her history showed at least eight or nine other psychiatric hospitalizations. She indicated a history of being sexually abused by her father at age 12. Her admitting diagnosis was schizoaffective disorder, depressed, subtype provisional, depressive disorder not otherwise specified, alcohol abuse and history of cocaine abuse. W.W.'s global assessment of functioning ("GAF") upon admission was 35. Dr. Birkmire testified that a GAF below 50 indicates that a patient should be in a residential program, at least, and that a GAF below 40-to-45 indicates the patient should be on an inpatient unit. On June 23, the medical notes show she was still somewhat confused. Her mood had improved but she still had suicidal ideations. The hospital was holding her in part to see if the blood test for syphilis was negative. She felt better but complained about mood swings. On June 24, a Sunday, there were no notes. Dr. Birkmire explained that the physicians who take rounds on the weekends are required to see each patient on either Saturday or Sunday, but not both days. On June 25, W.W. was less psychotic. Her auditory hallucinations were present but decreased, and she was taking her medications. On June 26, she continued to have auditory hallucinations. She was disorganized, paranoid, and isolating herself. She reported fear in being released because she might make another suicide attempt. She was given Ativan PRN to treat reported anxiety. On June 27, her mood was euthymic (normal, neither depressed nor highly elevated in mood). She showed no psychotic symptoms, denied suicidal ideations, and felt safe for discharge. However, hospital staff said that W.W. still seemed to be responding to voices. Dr. Birkmire stated that he had reservations about releasing W.W. on June 27 because she seemed to be telling the staff she was doing better than she really was. The voices had played a role in her suicide attempt, and, thus, the knowledge that she was still hearing them would be a strong factor in deciding to keep W.W. hospitalized. In Dr. Birkmire's opinion, W.W.'s problems could not be addressed in a skilled nursing facility. Dr. Birkmire testified that, with the exception of psychiatrists who perform peer reviews and medical authorizations, there is not one psychiatrist in the country who would say that a patient should not be in an acute care setting unless she has a definite plan for suicide or the means to complete the plan. Dr. Birkmire stated that no doctor is going to risk his license in that fashion. Whole books are written on how to perform a suicide assessment, and the assessment is based on much more than what the patient tells the physician. It is based on the patient's history, her degree of hopelessness, degree of disorganization, degrees of psychosis, and her access to the means of doing the suicide. Dr. Birkmire stated that one of the best predictors of suicidality is past attempts. Dr. Berns testified that W.W.'s history of multiple psychiatric hospitalizations indicated a probability of chronic illness. He stated that the number of prior hospitalizations automatically raises the question of past suicide attempts, and noted that her history indicated three attempts prior to this one. She was diagnosed with "depressive disorder" but not otherwise specified, as well as alcohol and cocaine abuse. She took an overdose of Haldol, Cogentin, and Sinequan. Dr. Berns explained that Sinequan is an older antidepressant, a tricyclic, with which a higher number of suicides occur. Thus, an overdose of this medication is more serious than overdoses of other medicines. There were concerns about neurosyphilis because she had a positive RPR. Her physicians were concerned that inadequate treatment for this condition would complicate her psychiatric course as well as cause physical complications leading to dementia and death. Dr. Berns concluded that W.W.'s stay was medically necessary. Her physicians were trying to stabilize her condition and treat a gradual illness that can become fatal. In response, Dr. Mehra testified that the evaluation for neurosyphilis on June 23 and 24, was not, standing alone, a ground for keeping the patient in the hospital. Dr. Mehra also stated that the suicidal thoughts, hallucinations, and psychotic disorganized paranoia are not grounds to keep the patient in the hospital, unless she was actively suicidal with a plan and unless her psychoses were causing imminent danger to herself or others, or she was aggressive or noncompliant. W.W. was taking her medications, eating, and sleeping and was not requiring seclusion or restraints on the unit, which meant that she was, at some level, functioning. Dr. Mehra was adamant that Medicaid and InterQual guidelines require more than a suicidal ideation; they require a plan. The greater weight of the evidence supports Petitioner's position that June 23 through June 28, should not have been denied. Dr. Birkmire's testimony as the treating psychiatrist, as well as Dr. Berns' expert testimony, credibly established that W.W.'s entire inpatient stay was medically necessary. Summary of Findings At the time of the hearing, AHCA sought from Petitioner overpayments in the amount of $198,582.54 for 29 patients who stayed at Florida Hospital Orlando between January 1, 2000, through December 31, 2001. The findings of fact above upheld AHCA's denial of days for the following: Acute Care Inpatient Hospital patients 2, 3, 5, 6, 9, 10, 11, 12, 13, and 16; and Psychiatric Inpatient Hospital patients 1, 3, 4, 5, 6, 7, 8, 9, and 11. The findings of fact above found that the greater weight of the evidence supported Petitioner's position that AHCA should not have denied the days for the following: Acute Care Inpatient Hospital patients 1 (2 days x reimbursement rate of $1,168.38 = $2,336.76), 4 (6 days x $1,206.42 = $7,238.52), 7 (1 day at $1,168.38), 8 (3 days x $1,168.38 = $3,505.14), 14 (9 days x $919.27 = $8,273.43), and 15 (5 days x $919.27 = $4,596.35); and Psychiatric Inpatient Hospital patients 10 (9 days x $1,168.38 = $10,515.42), 12 (8 days x $1,168.38 = $9,347.04), and 13 (5 days x $1,168.38 = $5,841.90). The total dollar figure for days that should not have been denied is $52,822.94, reducing the total overpayment due AHCA from Petitioner to $145,759.60.
Recommendation Based on the foregoing Findings of Facts and Conclusions of Law, it is RECOMMENDED that: Respondent, Agency for Health Care Administration, enter a final order revising its Final Agency Audit Report as directed herein. DONE AND ENTERED this 5th day of March, 2007, in Tallahassee, Leon County, Florida. S LAWRENCE P. STEVENSON 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 5th day of March, 2007.
The Issue Whether Certificate of Need Application No. 8614, filed by Vencor Hospitals South, Inc., meets, on balance, the applicable statutory and rule criteria. Whether the Agency for Health Care Administration relied upon an unpromulgated and invalid rule in preliminarily denying CON Application No. 8614.
Findings Of Fact Vencor Hospital South, Inc. (Vencor), is the applicant for certificate of need (CON) No. 8614 to establish a 60-bed long term care hospital in Fort Myers, Lee County, Florida. The Agency for Health Care Administration (AHCA), the state agency authorized to administer the CON program in Florida, preliminarily denied Vencor's CON application. On January 10, 1997, AHCA issued its decision in the form of a State Agency Action Report (SAAR) indicating, as it also did in its Proposed Recommended Order, that the Vencor application was denied primarily due to a lack of need for a long term care hospital in District 8, which includes Lee County. Vencor is a wholly-owned subsidiary of Vencor, Inc., a publicly traded corporation, founded in 1985 by a respiratory/physical therapist to provide care to catastrophically ill, ventilator-dependent patients. Initially, the corporation served patients in acute care hospitals, but subsequently purchased and converted free-standing facilities. In 1995, Vencor merged with Hillhaven, which operated 311 nursing homes. Currently, Vencor, its parent, and related corporations operate 60 long term care hospitals, 311 nursing homes, and 40 assisted living facilities in approximately 46 states. In Florida, Vencor operates five long term care hospitals, located in Tampa, St. Petersburg, North Florida (Green Cove Springs), Coral Gables, and Fort Lauderdale. Pursuant to the Joint Prehearing Stipulation, filed on October 2, 1997, the parties agreed that: On August 26, 1996, Vencor submitted to AHCA a letter of intent to file a Certificate of Need Application seeking approval for the construction of a 60-bed long term care hospital to be located in Fort Myers, AHCA Health Planning District 8; Vencor's letter of intent and board resolution meet requirements of Sections 408.037(4) and 408.039(2)(c), Florida Statutes, and Rule 59C-1.008(1), Florida Administrative Code, and were timely filed with both AHCA and the local health council, and notice was properly published; Vencor submitted to AHCA its initial Certificate of Need Application (CON Action No. 8614) for the proposed project on September 25, 1996, and submitted its Omissions Response on November 11, 1996; Vencor's Certificate of Need Application contains all of the minimum content items required in Section 408.037, Florida Statutes; Both Vencor's initial CON Application and its Omissions Response were timely filed with AHCA and the local health council. During the hearing, the parties also stipulated that Vencor's Schedule 2 is complete and accurate. In 1994, AHCA adopted rules defining long term care and long term care hospitals. Rule 59C-1.002(29), Florida Administrative Code, provides that: "Long term care hospital" means a hospital licensed under Chapter 395, Part 1, F.S., which meets the requirements of Part 412, Subpart B, paragraph 412.23(e), [C]ode of Federal Regulations (1994), and seeks exclusion from the Medicare prospective payment system for inpatient hospital services. Other rules distinguishing long term care include those related to conversions of beds and facilities from one type of health care to another. AHCA, the parties stipulated, has no rule establishing a uniform numeric need methodology for long term care beds and, therefore, no fixed need pool applicable to the review of Vencor's CON application. Numeric Need In the absence of any AHCA methodology or need publication, Vencor is required to devise its own methodology to demonstrate need. Rule 59C-1.008(e) provides in pertinent part: If no agency policy exists, the 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. Vencor used a numeric need analysis which is identical to that prepared by the same health planner, in 1995, for St. Petersburg Health Care Management, Inc. (St. Petersburg). The St. Petersburg project proposed that Vencor would manage the facility. Unlike the current proposal for new construction, St. Petersburg was a conversion of an existing but closed facility. AHCA accepted that analysis and issued CON 8213 to St. Petersburg. The methodology constitutes a use rate analysis, which calculates the use rate of a health service among the general population and applies that to the projected future population of the district. The use rate analysis is the methodology adopted in most of AHCA's numeric need rules. W. Eugene Nelson, the consultant health planner for Vencor, derived a historic utilization rate from the four districts in Florida in which Vencor operates long term care hospitals. That rate, 19.7 patient days per 1000 population, when applied to the projected population of District 8 in the year 2000, yields an average daily census of 64 patients. Mr. Nelson also compared the demographics of the seven counties of District 8 to the rest of the state, noting in particular the sizable, coastal population centers and the significant concentration of elderly, the population group which is disproportionately served in long term care hospitals. The proposed service area is all of District 8. By demonstrating the numeric need for 64 beds and the absence of any existing long term care beds in District 8, Vencor established the numeric need for its proposed 60-bed long term care hospital. See Final Order in DOAH Case No. 97-4419RU. Statutory Review Criteria Additional criteria for evaluating CON applications are listed in Subsections 408.035(1) and (2), Florida Statutes, and the rules which implement that statute. (1)(a) need in relation to state and district health plans. The 1993 State Health Plan, which predates the establishment of long term care rules, contains no specific preferences for evaluating CON applications for long term care hospitals. The applicable local plan is the District 8 1996-1997 Certificate of Need Allocation Factors Report, approved on September 9, 1996. The District 8 plan, like the State Health Plan, contains no mention of long term care hospitals. In the SAAR, AHCA applied the District 8 and state health plan criteria for acute care hospital beds to the review of Vencor's application for long term care beds, although agency rules define the two as different. The acute care hospital criteria are inapplicable to the review of this application for CON 8614 and, therefore, there are no applicable state or district health plan criteria for long term care. (1)(b) availability, quality of care, efficiency, appropriateness, accessibility, extent of utilization and adequacy of like and existing services in the district; and (1)(d) availability and adequacy of alternative health care facilities in the district. Currently, there are no long term care hospitals in District 8. The closest long term care hospitals are in Tampa, St. Petersburg, and Fort Lauderdale, all over 100 miles from Fort Myers. In the SAAR, approving the St. Petersburg facility, two long term care hospitals in Tampa were discussed as alternatives. By contract, the SAAR preliminarily denying Vencor's application lists as alternatives CMR facilities, nursing homes which accept Medicare patients, and hospital based skilled nursing units. AHCA examined the quantity of beds available in other health care categories in reliance on certain findings in the publication titled Subacute Care: Policy Synthesis And Market Area Analysis, a report submitted to the Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, on November 1, 1995, by Levin-VHI, Inc. ("the Lewin Report"). The Lewin Report notes the similarities between the type of care provided in long term care, CMR and acute care hospitals, and in hospital-based subacute care units, and subacute care beds in community nursing homes. The Lewin Report also acknowledges that "subacute care" is not well-defined. AHCA has not adopted the Lewin Report by rule, nor has it repealed its rules defining long term care as a separate and district health care category. For the reasons set forth in the Final Order issued simultaneously with this Recommended Order, AHCA may not rely on the Lewin Report to create a presumption that other categories are "like and existing" alternatives to long term care, or to consider services outside District 8 as available alternatives. Additionally, Vencor presented substantial evidence to distinguish its patients from those served in other types of beds. The narrow range of diagnostic related groups or DRGs served at Vencor includes patients with more medically complex multiple system failures than those in CMR beds. With an average length of stay of 60 beds, Vencor's patients are typically too sick to withstand three hours of therapy a day, which AHCA acknowledged as the federal criteria for CMR admissions. Vencor also distinguished its patients, who require 7 1/2 to 8 hours of nursing care a day, as compared to 2 1/2 to 3 hours a day in nursing homes. Similarly, the average length of stay in nursing home subacute units is less than 41 days. The DRG classifications which account for 80 percent of Vencor's admissions represent only 7 percent of admissions to hospital based skilled nursing units, and 10 to 11 percent of admissions to nursing home subacute care units. Vencor also presented the uncontroverted testimony of Katherine Nixon, a clinical case manager whose duties include discharge planning for open heart surgery for patients at Columbia-Southwest Regional Medical Center (Columbia-Southwest), an acute care hospital in Fort Myers. Ms. Nixon's experience is that 80 percent of open heart surgery patients are discharged home, while 20 percent require additional inpatient care. Although Columbia-Southwest has a twenty-bed skilled nursing unit with two beds for ventilator-dependent patients, those beds are limited to patients expected to be weaned within a week. Finally, Vencor presented results which are preliminary and subject to peer review from its APACHE (Acute Physiology, Age, and Chronic Health Evaluation) Study. Ultimately, Vencor expects the study to more clearly distinguish its patient population. In summary, Vencor demonstrated that a substantial majority of patients it proposes to serve are not served in alternative facilities, including CMR hospitals, hospital-based skilled nursing units, or subacute units in community nursing homes. Expert medical testimony established the inappropriateness of keeping patients who require long term care in intensive or other acute care beds, although that occurs in District 8 when patients refuse to agree to admissions too distant from their homes. (1)(c) ability and record of providing quality of care. The parties stipulated that Vencor's application complies with the requirement of Subsection 408.035(1)(c). (1)(e) probable economics of joint or shared resources; (1)(g) need for research and educational facilities; and (1)(j) needs of health maintenance organizations. The parties stipulated that the review criteria in Subsection 408.035(1)(e), (g) and (j) are not at issue. (f) need in the district for special equipment and services not reasonably and economically accessible in adjoining areas. Based on the experiences of Katherine Nixon, it is not reasonable for long term care patients to access services outside District 8. Ms. Nixon also testified that patients are financially at a disadvantage if placed in a hospital skilled nursing unit rather than a long term care hospital. If a patient is not weaned as quickly as expected, Medicare reimbursement after twenty days decreases to 80 percent. In addition, the days in the hospital skilled nursing unit are included in the 100 day Medicare limit for post-acute hospitalization rehabilitation. By contrast, long term care hospitalization preserves the patient's ability under Medicare to have further rehabilitation services if needed after a subsequent transfer to a nursing home. (h) resources and funds, including personnel to accomplish project. Prior to the hearing, the parties stipulated that Vencor has sufficient funds to accomplish the project, and properly documented its source of funds in Schedule 3 of the CON application. Vencor has a commitment for $10 million to fund this project of approximately $8.5 million. At the hearing, AHCA also agreed with Vencor that the staffing and salary schedule, Schedule 6, is reasonable. (i) immediate and long term financial feasibility of the proposal. Vencor has the resources to establish the project and to fund short term operating losses. Vencor also reasonably projected that revenues will exceed expenses in the second year of operation. Therefore, Vencor demonstrated the short and long term financial feasibility of its proposal. needs of entities serving residents outside the district. Vencor is not proposing that any substantial portion of it services will benefit anyone outside District 8. probable impact on costs of providing health services; effects of competition. There is no evidence of an adverse impact on health care costs. There is preliminary data from the APACHE study which tends to indicate the long term care costs are lower than acute care costs. No adverse effects of competition are shown and AHCA did not dispute the fact that Vencor's proposal is supported by acute care hospitals in District 8. costs and methods of proposed construction; and (2)((a)-(c) less costly alternatives to proposed capital expenditure. The prehearing stipulation includes agreement that the design is reasonable, and that proposed construction costs are below the median in that area. past and proposed service to Medicaid patients and the medically indigent. Vencor has a history of providing Medicaid and indigent care in the absence of any legal requirements to do so. The conditions proposed of 3 percent of total patient days Medicaid and 2 percent for indigent/charity patients proposed by Vencor are identical to those AHCA accepted in issuing CON 8213 to St. Petersburg Health Care Management, Inc. Vencor's proposed commitment is reasonable and appropriate, considering AHCA's past acceptance and the fact that the vast majority of long term care patients are older and covered by Medicare. services which promote a continuum of care in a multilevel health care system. While Vencor's services are needed due to a gap in the continuum of care which exists in the district, it has not shown that it will be a part of a multilevel system in District 8. (2)(d) that patients will experience serious problems obtaining the inpatient care proposed. Patients experience and will continue to experience serious problems in obtaining long term care in District 8 in the absence of the project proposed by Vencor. Based on the overwhelming evidence of need, and the ability of the applicant to establish and operate a high quality program with no adverse impacts on other health care providers, Vencor meets the criteria for issuance of CON 8614.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Agency for Health Care Administration issue CON 8614 to Vencor Hospitals South, Inc., to construct a 60-bed long term care hospital in Fort Myers, Lee County, District 8. DONE AND ENTERED this 3rd day of March, 1998, 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 Filed with the Clerk of the Division of Administrative Hearings this 3rd day of March, 1998. COPIES FURNISHED: Sam Power, Agency Clerk Agency for Health Care Administration Fort Knox Building 3 2727 Mahan Drive, Suite 3431 Tallahassee, Florida 32308-5403 Paul J. Martin, General Counsel Agency for Health Care Administration Fort Knox Building 3 2727 Mahan Drive, Suite 3431 Tallahassee, Florida 32308-5403 Kim A. Kellum, Esquire Agency for Health Care Administration Fort Knox Building 3 2727 Mahan Drive, Suite 3431 Tallahassee, Florida 32308-5403 R. Terry Rigsby, Esquire Geoffrey D. Smith, Esquire Blank, Rigsby & Meenan, P.A. 204 South Monroe Street Tallahassee, Florida 32301
The Issue The issue is whether HRMC's Certificate of Need Application No. 9881 to establish an 84-bed acute care hospital in the Viera area, Brevard County, Florida, AHCA Subdistrict 7-1, should be approved.
Findings Of Fact THE PARTIES The Agency AHCA is the single state agency responsible for administering the certificate of need ("CON") program, and is authorized to evaluate and make final determinations on CON applications pursuant to the Health Facilities and Services Development Act, Sections 408.031-408.045, Florida Statutes. HRMC and Affiliates The applicant is HRMC, a not-for-profit Florida corporation, holding tax exempt status pursuant to Section 501(c)(3) of the Internal Revenue Code. HRMC is a subsidiary of Health First, Inc. ("Health First"), an Internal Revenue Code Section 501(c)(3) holding company having no shareholders. It is managed by a 13-member board of directors, 11 of whom are outside directors representing a cross section of community leaders. As a 501(c)(3) organization, Health First pays no dividends to shareholders. It has no shareholders so it pays no dividends. Its earnings are reinvested in the community to support existing health programs, to offer new or expanded services, and to provide new medical facilities in areas of need. Health First's overall mission in Brevard County is to enhance the health status of the citizens by providing direct health care services, education, outreach, and prevention. For the first six years of its operation, Health First was headquartered in Viera, but in 2003 it relocated to larger corporate offices in South Rockledge. Through its subsidiaries, Health First operates three hospitals in Brevard County: Holmes Regional Medical Center ("Holmes") in Melbourne, Palm Bay Community Hospital ("PBCH") in Palm Bay, and Cape Canaveral Hospital ("CCH") in Cocoa Beach. All three hospitals are Joint Commission on Accreditation of Healthcare Organizations (JCAHO) accredited. HRMC is the license holder for Holmes and PBCH. Holmes is a 514-bed acute care hospital. It has 504 acute care and 10 Level II neonatal intensive care unit beds. It also provides adult open-heart surgery services, is a designated Level II Trauma Center, and is the only trauma center serving Brevard and Indian River Counties. Holmes sits on approximately 18 acres of land. Health First has acquired an extensive amount of property surrounding the hospital that is available for future growth and development. PBCH, a 60-bed acute care hospital, is located south of Holmes. It opened in 1992, sits on 30-40 acres of land about a mile from an I-95 interchange, and offers neither obstetrical nor tertiary services. CCH is a 150-bed acute care hospital. Health First Health Plans, Inc., is a health maintenance organization and subsidiary of Health First, operating in Brevard County. Wuesthoff Wuesthoff Health System, founded in 1941, is a not- for-profit health care system. In addition to its two acute care hospitals, Wuesthoff Health System owns and operates a nursing home, an assisted living facility, a hospice, a home health agency, a diagnostic imaging center, and other health care related services. Wuesthoff Health System also operates a foundation whose role is to identify and raise capital to support the organization's mission. Wuesthoff-Rockledge is a not-for-profit 267-bed acute care hospital comprised of 240 acute care beds, 17 psychiatric beds, and 10 Level II Neonatal Intensive Care Unit beds. Wuesthoff-Rockledge also provides tertiary care services, including adult open-heart surgery and angioplasty. It is currently constructing a 24-bed Intensive Care Unit expansion that is due to come online in October 2006, bringing the licensed bed complement of that facility to 291. Wuesthoff-Melbourne, which opened in December 2002 as a 65-bed acute care hospital, is located approximately five miles north of Holmes. It is a not-for-profit hospital with a current acute care bed complement of 115. Both Wuesthoff hospitals are JCAHO accredited. HRMC's CON APPLICATION The letter of intent and CON application at issue were timely and properly filed, and satisfy the requirements of Section 408.039, Florida Statutes. The signatures on Schedules C and D-1 of the application are authentic. The application satisfies the content requirements of Section 408.037, Florida Statutes. HRMC has sufficient availability of funds for capital and operating expenditures for the proposed project in accordance with Subsection 408.035(4), Florida Statutes. The proposed project demonstrates immediate financial feasibility in that HRMC has sufficient availability of funds for capital and operating expenses for the proposed project in accordance with Subsection 408.035(6), Florida Statutes. The criterion set forth in Subsection 408.035(10), Florida Statutes, is not applicable to this project. The projected staffing patterns (i.e., the numbers of full time equivalents ["FTEs"] needed by position) and the projected salaries in Schedule 6A of the CON application filed by HRMC are reasonable with regard to the census levels projected in the application. HRMC's CON application adequately documents the reasonableness of the costs (including equipment), methods of construction, and energy provision for its project as proposed in its application, in accordance with Subsection 408.035(8), Florida Statutes. Further, the architectural drawings submitted by HRMC with its application satisfy applicable code requirements. Schedules 1, 2, 3, 9, and 12 of the application are accurate. The time intervals contained in the project completion forecast depicted in Schedule 10 of the HRMC application are reasonable. This proceeding involves the third application by HRMC for this proposed hospital. The first application, CON No. 9759, was preliminarily denied by AHCA. Following an extensive final hearing, a Recommended and Final Order of denial were entered. Official recognition was taken of that in Holmes I. The second application, CON No. 9836, was filed in the first batching cycle of 2005. The head of the CON program, Jeff Gregg, recommended, and AHCA preliminarily denied, that application. The denial was initially challenged by HRMC, but the Petition was later dismissed voluntarily. By the time HRMC filed its second CON application for this project, the CON law had been amended to allow existing hospitals to increase their bed capacity at existing facilities without a CON. The third Application, CON No. 9881, was filed in the second batching cycle of 2005, and was followed by a preliminary decision in December of that year. Following staff review of the application at issue here, Mr. Gregg went to then-AHCA Secretary Alan Levine without a recommendation to either approve or deny the application. At that brief meeting (five minutes or so), Secretary Levine seemed interested in approving the application and appeared to be aware of the conditions proposed by the applicant. Wuesthoff attempted to tie the meeting with Secretary Levine, and his desire to approve the HRMC application, to Health First's support of a Medicaid reform bill then pending before the Florida Legislature that Secretary Levine strongly supported. While it is not uncommon for hospitals to communicate their views on particular CON proposals to AHCA officials, including the secretary, Wuesthoff alleges that the timing of this meeting, along with the fact that AHCA had recently denied the same CON application filed by HRMC, makes this a coincidence that cannot be ignored. Moreover, it is not uncommon for CON applicants to apply more than once in separate batching cycles for AHCA approval of a proposed hospital in a particular area. The circumstances relating to the need for a hospital in a particular area change over time due to numerous factors such as population increases, high utilization of existing providers, and the needs of underserved segments of the population. For example, Wuesthoff-Melbourne filed three CON applications with AHCA before it was finally approved. HRMC's application proposes to establish an 84-bed acute care hospital in the unincorporated area of Viera in Brevard County. AHCA reviews each application filed in a separate batching cycle independent of the reviews it conducted on previous applications filed for the same project. On December 16, 2005, AHCA issued its State Agency action report ("SAAR"), summarizing its findings and conclusions based upon its review of HRMC's application, and recommended approval of CON No. 9881. AHCA reaffirmed its support of HRMC's CON application through the final hearing testimony of Jeff Gregg. The Proposed Site Viera is an unincorporated area in south-central Brevard County being developed by The Viera Company ("TVC"), a for-profit land development company. TVC is a wholly-owned subsidiary of A. Duda & Sons, Inc. The initial Viera Development of Regional Impact ("DRI") contained approximately 3,200 acres east of I-95. Its residential area was planned for 4,200 units, and is currently 98 percent built out. The DRI development order was later amended to add 6,000 acres west of I-95, bringing the size of the DRI to over 9,000 acres. The current DRI entitlements include approximately 19,000 residential units, 2.9 million square feet of commercial space, and 3.7 million square feet of office space. The DRI also includes a 7,500 seat stadium and spring training facility for the Washington Nationals, a Major League baseball team. Six public schools are planned for the area west of I-95 as part of the DRI. Three elementary schools and a high school have already opened. A 107,000-square-foot Veterans' Administration clinic is located on the Viera DRI, just north of the proposed Viera Medical Center ("VMC") site. Authority to construct a 150-bed hospital within the confines of the Viera DRI has already been approved as part of the Master Development Plan. Viera has experienced tremendous growth in the past 10 years. As of June 1, 2006, over 7,200 homes were built. Nearly one million square feet of commercial space and another million square feet of office space have been completed. Demand for home-sites in the Viera area has been strong. According to testimony, people have camped overnight to wait in line to purchase certain properties. A. Duda & Sons owns another 11,500 acres adjacent to Viera, and it has applied to add that land to the current DRI. The proposed addition to the DRI would bring the residential entitlements to almost 30,000 units, the commercial entitlements to over 3.4 million square feet, and the office entitlements to over 3.5 million square feet. By 2010, the greatest growth in population will occur in the 45-64 age cohort, followed by the 65+ age cohort--the age cohort with the highest utilization of acute care inpatient services. At build out, Viera will have approximately 30,000 residential units and a population expected to exceed 40,000. The Proposed Viera Medical Center Health First began to seriously evaluate the opportunity to establish a new hospital in the Viera area in 2002. Based upon the significant growth in the area, Health First purchased 50 acres from TVC, located just west of I-95, at the intersection of Wickham Road and Lake Andrew Drive. VMC will initially offer a full array of acute care services, not including obstetric services, but including a larger than average emergency department. It will occupy 20 acres of the 50-acre site and will be part of an integrated health care campus (Viera Health Park) that will offer a continuum of services to the community, emphasizing health, wellness, and outpatient care, while also meeting the needs of the more acute patients. The remaining 30 acres will be used for related health care functions that are not CON reviewable. Health First also has an option to purchase additional land at the same location, if needed for future expansion. The purchase price allotted to the 20-acre portion of the site is approximately $3.4 million, which is reasonable. In determining the scope of the initial services, HRMC analyzed the population, demographics, service area, and needs of the community. The land purchase contract contained an exclusivity provision, not uncommon for large commercial projects such as this. The provision prevents the sale by TVC of any property to be used to construct another acute care hospital in Viera. TVC had a similar contractual provision with Wuesthoff relating to property it purchased in Viera in the 1990s. Prior to the hearing, the exclusivity provision in HRMC's contract covering hospital beds and services was amended to apply to a period of seven years from the opening of VMC. This reduced term of exclusivity is considered reasonable when examining the circumstances in this case. All planning and zoning approvals to develop the site as a health care center are already in place. Viera is the central seat of Brevard County government. The Moore Justice Center and the school board headquarters are located a short distance from the proposed site. The site is located within the Viera DRI. As a condition of the DRI, the developer was required to spend $8 million to make road improvements throughout the county to address traffic congestion. VMC will share management and administrative support services with HRMC rather than duplicating those services. Health First's goal is to operate all of its hospitals in a similar fashion to share overhead expenses, consolidating where practical to provide a more efficient operating model. The proposed total project cost for VMC is approximately $105 million. This cost was not challenged and is reasonable. Outside the proposed project, Health First anticipates developing the proposed site in several phases. Phase I will consist of a medical office building and a large health and fitness center. Health First has already been awarded a disturbance permit for this phase, allowing it to move sod and dirt on the site. Phase II will consist of the proposed hospital (VMC) and, potentially, additional office space. Health First has set a system-wide goal and adopted a formal policy to convert its inpatient facilities to private rooms. Accordingly, all of the beds at VMC will be in private rooms. This is consistent with the recent trend for hospitals to be designed with all private rooms. The American Institute of Architects ("AIA") recently adopted a draft guideline requiring 100 percent private rooms, with semi-private only by exception. The AIA guidelines form the basis for hospital design regulations in most states, including Florida. Private rooms are important not just as a means of enhancing patient comfort, but also to protect patient privacy, to achieve operational efficiencies, and to provide an optimal clinical environment. They provide a quieter, more restful environment, lessen the risk of infection, and better protect the patient's confidential clinical information under the Health Insurance Portability and Accountability Act ("HIPAA"). Mr. Jeff Leitner, Wuesthoff's Director of Plant Operations, acknowledged that HIPAA now imposes stringent privacy requirements on hospitals. The standard of care has evolved over the years to where patients now prefer private rooms. Due to changing hospital environments, the Health First Board of Directors is troubled that patients at their most sick and most vulnerable time are required to share a room with a stranger and the stranger's family. Wuesthoff agrees that private rooms should be offered to patients. It implemented a local marketing campaign to promote the fact that all of the rooms at its Melbourne facility are private. Both of Wuesthoff's architect witnesses acknowledged that private rooms are now the standard in the industry. The evidence strongly demonstrates that private rooms are no longer just an amenity; they offer substantial benefits to patients. Prior to undertaking this project, HRMC considered a number of alternatives to ensure that VMC was the best option. One alternative was to expand Holmes. This option was dismissed for several reasons: the cost of expansion; the traffic congestion that would result from further enlargement of that facility; and the need to decompress the Holmes campus rather than expand its patient volume. Holmes is a regional tertiary referral center, with a comprehensive open-heart program, a Level II trauma center, helicopter medical transport, and a Level II Neonatal Intensive Care Unit (NICU). It also provides comprehensive high-level cancer services. In its most recent fiscal year, Holmes received approximately 1,200 trauma visits, and almost 70,000 total visits to its emergency department ("ED")--almost twice that of other hospitals in the county. Approximately 3,300 births were recorded at Holmes during that same period. Holmes is a very large, busy, and congested hospital. Occupancy tends to remain above 80 percent and has, on occasion, exceeded 100 percent. For example, on the Thursday night before the final hearing, 15 patients admitted to the hospital had to wait on gurneys in the ED hallways before rooms became available. VMC will help decompress Holmes' high occupancy. Holmes is already seeking to address its high occupancy concerns by adding a new patient tower that is called the "North Expansion." This additional space will allow Holmes to consolidate its cardiac services into a new heart center. The North Expansion will also include a new ED and trauma center. The North Expansion will include 108 progressive care beds to provide care for cardiac patients. "Shelled-in" space is available to add 36 more progressive care beds in the future. Holmes does not intend to increase its licensed bed capacity with the new beds in the North Expansion, but will convert semi-private rooms in the older part of its facility to private rooms, keeping the total number of licensed beds the same. This will allow it to move to its all private beds platform and relieve congestion in the hospital. Given Holmes' persistent high utilization rates and the strong regulatory preference (from the new AIA guideline) for all private rooms, this approach is reasonable. Wuesthoff argues that increasing the number of licensed beds at Holmes is preferable to approving VMC. Moreover, it argues, the North Expansion and the footprint of the hospital will support as many as 144 licensed beds. Finally, states Wuesthoff, Holmes has significant land holdings adjacent to the existing facility. Each of these reasons, states Wuesthoff, supports Holmes' ability to add the additional beds it requires at the existing location rather than expanding into Viera. Holmes' campus is only 18 acres, which is relatively small for a large regional hospital. The facility has been expanded in piecemeal fashion over the years. That approach is not the optimal way to go. Even after completion of the North Expansion, over 50 percent of the existing facility is more than 25 years old. The useful life for a modern hospital is 25 to 30 years. Because of the hospital's age, many of the spaces used to support patient care are undersized by current standards, particularly for the level of care provided. Further increasing Holmes' licensed beds will strain its already overburdened support services. Holmes admittedly has acquired a number of parcels adjacent to its campus, but will use these to ease its chronic parking shortage. Holmes currently provides a substantial amount of employee parking at an off-site location. It shuttles employees back and forth by bus. Wuesthoff's efforts to show that these adjacent properties could be used to expand Holmes' existing facility were not persuasive. Any new construction must comply with current development regulations. No credible evidence was produced to demonstrate that the major development proposed for Holmes by Wuesthoff could be undertaken consistent with storm water, parking, and other local zoning and development regulations. Expanding PBCH beyond the currently planned expansion from 60 to 100 beds was also considered by HRMC as an alternative to developing a new hospital in Viera. That facility, however, is located at the far southern end of Brevard County, which is too distant to adequately and efficiently serve the needs of the Viera residents. Adding beds at CCH is not an option, given its land- locked location and the access difficulties concerning its existing beds, which are located about 17 miles from Viera. Because the actual facility is owned by a special taxing district and is leased to Health First, its beds cannot be transferred to Viera. In preparing and filing its CON application in this case, HRMC chose to begin its analysis regarding the size and scope of its services anew. HRMC cites several major factors in support of its project: a) a large population base, significant population growth, and a high percentage of elderly residents; b) improving access without an adverse impact on other providers; c) enhancing quality of care; d) enhancing access to cost- effective, quality care for all residents of the service area, including the uninsured and Medicaid patients; and overwhelming community support for the project. HRMC also proposed the following conditions for approval: a) To demonstrate its commitment to addressing the health care needs of the greater Brevard County community, Health First agrees to provide not less than 19.5 percent of all inpatient admissions and births across the entire Health First System for patients in "safety net" categories, including Medicaid, Medicaid HMO, self-pay, uninsured, KidCare, and charity. By the year 2010, VMC's first full year of operation, these categories are projected to represent approximately 10,000 annual admissions and $380 million in hospital charges. This commitment will remain in effect for no less than five years; Health First will continue to provide a majority of local support for the Brevard Health Alliance, a federally qualified health clinic whose mission is to provide care to the poor and uninsured. This commitment will represent a minimum of $4 million commencing in 2005 and continuing through 2007; Recognizing that Health First must work together with other social service organizations in its community, Health First commits to provide support in the form of cash, goods, and services not less than annually for the next 10 years. This support represents a total commitment of $1 million to support organizations such as the American Cancer Society, Brevard Alzheimer's Association, Habitat for Humanity, United Way, and many more health and social service organizations; Health First has created a Special Needs Plan (SNP) for low-income seniors, which provides enriched health and drug benefits for poor citizens. This program is administered by Health First Health Plans and is approved by the Centers for Medicare and Medicaid Services (CMS). Their goal is to ensure that all low income seniors in the community have access to essential medical services through this program. Health First commits to provide this essential community service for no less than four years; HRMC, the tertiary flagship hospital of Health First and the CON applicant, commits to continue to operate the region's only Level II Trauma Center and air ambulance for those patients requiring the highest level of care, through at least 2015; Mindful of the need to promote tomorrow's health care work force, HRMC commits to provide half of the instructional costs associated with the expansion of the nursing program at Brevard Community College. HRMC will provide $2 million by 2014 for a grand total of over $3 million after matching funds from the Florida Legislature; and Finally, responding to concerns surrounding the continued growth and success of existing hospital facilities in the community, HRMC agrees not to seek to expand the number of licensed acute care beds at VMC for its first five years of operation, unless the average occupancy rate of VMC, based on inpatient utilization of all licensed acute care beds, exceeds 80 percent for at least a 12-month period. AHCA's Review of HRMC's Application AHCA reviews CON applications under the criteria set forth in Section 408.035, Florida Statutes. AHCA does not give equal weight to each criterion. It has a standard procedure it follows each time it reviews a CON application, which was followed by AHCA in its review and consideration of CON No. 9881. Each CON application is reviewed by a staff analyst assigned to the project (assisted by the Agency accountant and architect) who makes a report of his or her analysis. This is called the State Agency Action Report ("SAAR"). The draft SAAR is then reviewed by the analyst's supervisor, who edits and forwards the SAAR to the CON Bureau Chief. The chief typically confers with the deputy secretary of AHCA. A final draft SAAR is prepared, and a meeting is scheduled with the secretary of AHCA to review the proposed decisions and obtain the secretary's input. In 2004, the Florida Legislature significantly amended the CON law. Bed additions to hospitals were deregulated, and the requirement to publish a fixed need pool was eliminated. Existing providers now have opportunities they never had before to make their own decisions. AHCA no longer considers individual beds to be the building blocks of a hospital proposal. Rather, it looks more holistically at a proposal for a new hospital. Each CON applicant is now encouraged to frame the issues they deem most important. In approving HRMC's application, AHCA determined that a number of key circumstances significantly differed from the first CON application. First, this application was considered after the 2004 changes to the CON application were fully implemented. Second, significant population growth had occurred in Viera during the two years since HRMC's first application. In fact, the time this application was filed, the projected population growth in Viera exceeded what had earlier been projected for this period in HRMC's first application. Further, the overall occupancy rate of acute care beds within the county continued to increase, despite the opening of Wuesthoff-Melbourne in late 2002, and the net addition of 94 acute care beds in the county between 1999 and 2004. Admissions, average length of stay, and patient days are all higher, and the average daily census grew faster than the beds that have been added. The proposed opening date of the VMC facility is nearly two years later than the first proposed date, which accounts not only for two more years of population growth, but two more years that Wuesthoff-Melbourne will have been open. Wuesthoff-Melbourne now will have seven years of operations, and ample opportunity to become fully established before the Viera facility opens. HRMC made adjustments in response to previous AHCA criticisms relating to financing and ability to staff the proposed facility, which AHCA considered significant. Finally, HRMC significantly changed the proposed conditions to the project, from what was a typical Medicaid and charity care utilization proposal, to a much broader, health system-wide commitment. APPLICATION OF STATUTORY CRITERIA Subsection 408.035(1), Florida Statutes: The need for the health care facilities and health services being proposed. Service Area The primary service area ("PSA") for VMC consists of Zip Codes 32904, 32934, 32935, 32940, and 32955. The proposed PSA contains one zip code not contained in HRMC's first application--32904. This change was made to increase the proposed PSA from 69 percent of total hospital discharges to 75 percent of total discharges which follows a more traditional definition of a PSA. HRMC's health care planning expert, Dr. Finarelli, also re-examined the area, which revealed that the market shares of the existing hospitals had changed over time. As a result, VMC's market share was increased in some zip codes and reduced in others. The fact that VMC's market share increased in Zip Code 32904 from 8 percent to 10 percent caused that zip code to be shifted from the secondary service area to the PSA. Even if, as Wuesthoff asserts, the 32904 Zip Code is in the secondary service area rather than the PSA, this would result in only 100 fewer patient admissions per year, an insignificant amount. Wuesthoff asserted that VMC's proposed PSA is too large, yet it approximates the size and same number of zip codes of the PSA proposed by Wuesthoff in the CON application for its Melbourne facility. Wuesthoff also argued that the proposed VMC PSA includes two existing hospitals. While true, the two existing facilities are located about 10 miles from the proposed site. They are on the extreme outer fringes of the PSA. 2. Population Growth Population is a fundamental aspect of virtually every proposal for a new hospital. The proposed project will be located in Zip Code 32940, which is projected to grow by more than 12,000 persons between 2005 and 2012. It is easily the zip code with the largest projected growth in Brevard County. Two other zip codes in the PSA, 32955 and 32904, are projected as the second and third fastest growing zip codes in the county between 2005 and 2012. The projected population growth in the proposed PSA compares favorably with the population growth for nine previous hospital projects approved by AHCA. It falls within the mid- range of those other projects. AHCA recognized that the VMC proposal is in a rapidly growing suburban community. Brevard County has experienced explosive growth in population, construction, and services. Area physicians testified that their practices have grown significantly in the past five years, requiring additional hiring to handle the increase. The population of the VMC PSA alone is projected to increase by almost 30,000 people between the filing of the application in 2005 and 2012, VMC's projected third year of operation. This increase represents half of the entire county's growth during that time period. Such market conditions fit well within the type of conditions that have led AHCA to approve other CONs for new hospitals. The projected population in the greater Viera community and the absolute number of new people, as well as the annual growth rates, support the proposed hospital. Just as population growth drives the need for new schools, roads, and utilities, the health care infrastructure must be increased to meet the rise. The healthcare infrastructure is a key component of the Viera DRI. Sound health planning creates capacity where it is needed. Hospital discharges involving residents of the VMC PSA are projected to grow by 41 percent between 2004 and 2012, reflecting strong growth in the PSA. Physicians who practice in Brevard County and support the proposed Viera project overwhelmingly agreed that the area population is growing rapidly. That population growth results in an increase in patients seeking health care services. The average age of the PSA's population is also increasing as older people move into the PSA. The population in the PSA over 65 is projected to grow three times as fast as the rest of the county. This is significant because people over 65 utilize hospital services at the greatest level. Wuesthoff claims that the tremendous increase in population in Viera does not create a need for a hospital there. In July 2003, however, Mr. Emil Miller, Wuesthoff's CEO, expressly acknowledged in a letter to The Viera Company the significant growth of Viera, and sought to purchase land in Viera upon which to build a hospital, specifically recognizing that such a hospital was necessary to "meet the health care needs of that growing population." He sought, at that time, the exclusive right to be the only hospital in Viera. This letter was written after the opening of Wuesthoff-Melbourne. From a health planning perspective, new capacity should be created where population growth is occurring, as opposed to expanding existing hospitals that are site- constrained, short on space, poorly configured, aged, and far removed from the area experiencing high population growth. HRMC has demonstrated that the tremendous population growth in Viera requires the creation of new capacity. 3. Need Analysis AHCA no longer has a numeric methodology for calculating need for short-term acute care beds or services. Therefore, HRMC presented its own analysis of need for a new acute care hospital in this case. Although no longer in rule form, AHCA considers a five-year planning horizon for short-term acute care facilities to be appropriate. The amended CON law allows hospital providers to add beds at existing facilities without prior CON approval, but that fact does not control whether a hospital should be built in Viera. If that were the case, AHCA would never approve another hospital because existing hospitals would always argue that they can meet need over time by adding beds at existing facilities--a plainly inappropriate way to meet future health care needs. Instead, new capacity should be created where population growth is occurring. HRMC sufficiently demonstrated short term need. HRMC also demonstrated a long term need. An underlying premise of HRMC's application is that sound long- range planning requires consideration of where the needs are today and where they are projected to be in the future. Dr. Finarelli testified as an expert on the need for VMC. His role was twofold: (1) to determine whether there is a population-based need for a new hospital in Viera; and (2) to judge the impact such a hospital would have on existing hospitals in south and central Brevard County. Dr. Finarelli prepared a need analysis for this project. He examined the past five years of utilization and noted strong and steady growth in the sub-district. He then projected future growth to 2012, the third year of operation for VMC. He projected a steady growth in demand, and he determined that the acute care bed need would extend beyond the capacity available and planned to be added in Brevard County. In the last five years, a significant number of acute care beds have been added to the county's bed inventory: Holmes added 76 beds, Wuesthoff-Rockledge added 22 beds, and Wuesthoff- Melbourne opened with 65 beds in late 2002 and now has 115 beds. Wuesthoff-Rockledge will open a 24-bed ICU within the next year. PBCH is planning a 40-bed expansion, bringing its total to 100 beds. PBCH will also shell-in an additional 40 beds for future expansion at that campus. Even if these additional beds come online earlier than anticipated, the effect on the proposed VMC will be minimal because PBCH is located more than 20 miles from Viera. Mr. Mark Richardson, Wuesthoff's health planner, acknowledged that PBCH does not really serve the residents of Viera. Although these planned expansions will help meet the increased demand for acute care services, they are not a long- term solution for future growth or the demand for acute care services in Brevard County. The health care industry generally recognizes an optimum occupancy level of 75 percent. Annual occupancy rates at this level mean that a hospital sometimes runs in excess of 90 percent at peak times. When this is combined with the fact that the midday census typically runs 10 percent higher than the reported midnight census, the conclusion is that hospitals with an optimum occupancy level of 75 percent actually run at greater than 100 percent occupancy at times. Table 21 of HRMC's application demonstrates that if the proposed VMC is denied, Holmes' occupancy rates will escalate from 83 percent in 2010 to 86 percent in 2012; PBCH's occupancy rates will increase from 77 percent to 80 percent in 2012; and Wuesthoff-Rockledge's occupancy rates will increase from 74 percent to 77 percent in 2012, even after the currently planned bed increase at PBCH and Wuesthoff-Rockledge. Dr. Finarelli noted that the two fastest growing segments of the population in VMC's PSA are the over 65, and 45 to 64 age groups, who demand the most acute care. The over- 65-age demographic is two percent higher than the county as a whole, and nearly three percent higher than the statewide average. Demand for inpatient services in Brevard County has increased by approximately seven percent per year from 2000 to 2005. In the Viera PSA alone, however, the demand for inpatient services has grown by 40 percent between 2000 and 2004, twice the 19 percent growth for the rest of the county. Dr. Finarelli determined that the VMC project is appropriately sized and is expected to be well utilized. He based this upon the fact that the VMC PSA alone had 4,000 more discharges from 2000 to 2004. Total discharges in Brevard County during that time period increased by 10,000 during that same period. The rate of increase in discharges for the VMC PSA was twice that of the county as a whole. The VMC PSA is expected to have 20,400 discharges by 2012, compared with 14,400 in 2004, a substantial increase. Dr. Finarelli's assumption that VMC will capture 20 percent of the PSA's market is conservative and reasonable. In preparing his need analysis, Dr. Finarelli analyzed the appropriate change in age-specific discharge rates in Central and South Brevard County. Decisions about how to project future use rates is a market-based analysis that varies from market to market. Historical trends and comparative use rates must be examined to determine the most appropriate growth factor to use. Although historical use rates had increased an average of four percent per year from 2000 to 2004, Dr. Finarelli chose to use a more conservative admission rate increase of approximately 1.5 percent per year from 2005 and 2010, less than half the most recent historical experience. Mr. Richardson, Wuesthoff's health planner, acknowledged that he would employ a forecast model in basically the same form as Dr. Finarelli's. However, he differed as to the appropriate change in use rates and how the VMC market share should be allocated. Dr. Finarelli assumed a 1.5 percent annual increase in use rate through 2010, and a flat rate thereafter in his analysis. Mr. Richardson argued that the 1.5 percent annual increase was too optimistic, but he acknowledged that the average increase in use rate is one to two percent. When undertaking his analysis, Mr. Richardson testified that he had not examined the District VII Health Council's quarterly reports, whose numbers differed from its annual reports. When shown the quarterly figures, he acknowledged that utilization, on average, had trended upward. Using the revised data, the growth rate in admissions from 2002 to 2003 was 6.8 percent, with a six percent increase in patient days for the same time period. Mr. Richardson agreed that no authoritative literature exists stating that use rates should be kept flat for this type of analysis. He acknowledged he had previously used a variable use rate trend when he prepared the CON application for Wuesthoff's Melbourne facility. The use rate analysis employed by Dr. Finarelli is more persuasive. Another material factor that must be considered in this analysis is that between 2003 and 2005, all Brevard County hospitals experienced a significant increase in average length of stay ("ALOS") of a quarter day or more. This caused the average daily census to increase steadily even as admission growth slowed to some extent. The most likely explanation is that patient acuity increased in the county. As a result of the upward trend in ALOS, Dr. Finarelli used an estimated 4.27 days as the projected length of stay in this case. This figure is consistent with Wuesthoff-Melbourne's current 4.29-day ALOS and PBCH's projected 4.27-day ALOS in 2010. Wuesthoff does not oppose the 4.27-day ALOS projection. The projection is reasonable in this case. Dr. Finarelli projected growth in demand for adult medical-surgical and total acute care, using both admissions and average daily census ("ADC"). According to his calculations, residents of the VMC PSA will account for a cumulative increase in ADC of 75 patients by 2012, while ADC will increase by 188 patients in the county overall. The county-wide increase translates to a need for an additional 250 beds in Brevard County by 2012. Table 15 of the CON application shows hospital utilization of just above the 75 percent optimum level by the end of the third year of operation. Even using a flat admission rate, as proposed by Wuesthoff, the average annual occupancy would still be 74 percent by 2012. Dr. Finarelli's analysis took into account the current addition of 24 ICU beds at Wuesthoff-Rockledge, the shelled-in space for 19 beds at Wuesthoff-Melbourne, and the planned 40-bed expansion at PBCH. Even Mr. Richardson acknowledged the need for more than 126 additional beds by 2010. In allocating market share to VMC, Dr. Finarelli relied on his experience that patients are more likely to shift from one hospital to another within the same system than they are to change systems altogether. For that reason, for the four zip codes of the PSA that are in South Brevard County, Dr. Finarelli assumed that 75 percent of the patients admitted to VMC would be patients who would otherwise use a Health First facility, while 25 percent of the patients admitted would otherwise use a Wuesthoff facility. Mr. Richardson agreed with Dr. Finarelli's volume forecast, but opined that the patients should be allocated 50/50 between Wuesthoff and Holmes. It is more persuasive that patients will keep their physicians and remain within a known health system than they are to change health care providers. The utilization and market shares projected for VMC are reasonable and attainable. VMC's application projects that it will capture the majority of its patients from HRMC, but it will also capture some patients from Wuesthoff-Rockledge and Wuesthoff-Melbourne. In spite of this, Dr. Finarelli projects that the Wuesthoff hospitals will gain in excess of 5,000 admissions (a 34 percent increase) from 2004 to 2012. Admissions at Health First would increase by 29 percent during the same period. In addition to projected in-patient admissions, VMC is expected to have more than 26,000 ED admissions by 2012. This projection is reasonable and attainable. The evidence in this case demonstrates that the most appropriate response to address the demonstrated bed need is through the construction of a new hospital in Viera. The population growth in Viera will more than adequately support a new facility, and the new hospital will foster enhanced access for the community. Wuesthoff raised questions as to whether HRMC's application meets the traditional "not normal" circumstances required in the absence of a numerical bed need. Since the repeal of the numeric bed need methodology, however, the concept of "not normal" does not apply, as each applicant may present its own argument for need. HRMC has done so here. 4. Overwhelming Community Support VMC enjoys strong support from the local community. It received hundreds of letters of support from a broad cross- section of the community, including health care and law enforcement officials, physicians, business and civic leaders, and members of the general public. HRMC's application also included petitions containing over 10,000 resident signatures supporting a new hospital for Viera. It developed a website devoted to this project. Prior to the hearing, over 5,000 e-mails were received by the website in support of the new hospital. Samples of these were included in HRMC's application. 5. Disaster Preparedness During the planning process for the proposed hospital, HRMC sought input from a number of county officials as to how the hospital could be designed to improve the county's capability to respond to a large-scale emergency or catastrophic event. HRMC met with Mr. Robert Lay, Director of the county's Emergency Management Office, and several representatives of local law enforcement agencies. The proposed hospital design incorporates a number of their suggestions and has received strong support from local officials. As director of the Brevard County Emergency Management Office, Mr. Lay is responsible for developing emergency management plans to ensure public safety in the event of an emergency or disaster. He considers the site of the proposed VMC to be a significant benefit because of its inland location and proximity to I-95. The hospital would be further inland than any existing hospital in Brevard County, making it less susceptible to damage from hurricanes and other severe storms. Moreover, its close proximity to I-95 provides easy access to hospital services in the event of a hurricane or mass casualty. The proposed VMC would exceed minimum hurricane standards, allowing it to withstand higher wind speeds and projectile impacts of a hurricane or other major storm. The proposed hospital would offer additional "surge capacity" in Brevard County, that is, the ability to expand treatment facilities to accommodate a large influx of patients in a short period of time. Existing hospitals in Brevard County do not have adequate surge capacity. The proposed hospital would be located near the county health department which would provide additional support for the public clinic located there. The hospital could potentially be used to provide Disaster Medical Assistance Team ("DMAT") support. This is a mobile medical team that can provide emergency medical services in the event of a disaster. The proposed hospital would include facilities specifically designed to support a DMAT in the event of a disaster or other catastrophe. This includes a site that is stabilized and equipped with all necessary utility connections, including potable water, power, data, and sanitary sewer. The proposed facility plans to include isolation rooms, negative pressure beds, and special air handling equipment that may be used in responding to a bioterrorism event. Expert meteorologists testified for both HRMC and Wuesthoff at hearing concerning the potential for hurricanes striking Brevard County. Their testimony differed as to the probability of a catastrophic storm hitting the county, based upon historical precedent and the laws of probability. Clearly, CCH is particularly vulnerable to storms because of its location on a causeway nearly surrounded by water. The remaining hospitals in Brevard County, while slightly more vulnerable to storm damage than the proposed VMC due to their locations, appear to be safe from all but the most catastrophic storms. While building the proposed hospital to the latest Florida Building Code requirements will make it more hurricane resistant than the existing hospitals in Brevard County, this fact alone does not render the existing hospitals in the county unsafe or not suitable to withstand the infrequent storms projected to strike Brevard County. The proposed hospital design includes other elements to assist in the event of a disaster. These include multiple- site entrances that can be controlled and used for specific emergency operations; enlarged ambulance access and discharge areas to facilitate efficient flow of multiple emergency vehicles; a ground helipad designed to handle military aircraft; and a large area near the ED that could be used for decontamination purposes. The proposed hospital will improve Brevard County's capability to respond to emergency or mass casualty events. While this finding, in and of itself, does not justify approval of the CON application, it reflects positively in HRMC's favor. Subsection 408.035(2), Florida Statutes: The availability, quality of care, accessibility, and extent of utilization of existing health care facilities and health services in the service district of the applicant. The parties stipulated that the Health First and Wuesthoff hospitals are all JCAHO accredited. The evidence at hearing demonstrates that good quality care is provided at the Health First and Wuesthoff hospitals. Holmes is at or near capacity at times. When full, patients have to be cared for in hallways until appropriate rooms are available. The County Emergency Office has been aware of various occasions when hospitals must be called to determine space availability. Many of the county's hospitals, especially in the winter months, experience extremely high occupancy and limited availability of space for handling emergency patients. A number of area physicians testified regarding the frequent difficulty in getting patients admitted to Holmes due to overcrowding. The proposed VMC offers an opportunity to provide additional capacity for emergency and acute care patients in Brevard County. Continued high occupancy rates can lead to a deterioration of quality of care if patients are forced to wait for available beds. The evidence concerning demand supports the existing hospitals as well as the proposed facility in Viera. HRMC sufficiently demonstrated the availability, quality of care, accessibility, and extent of utilization of existing health care facilities in the district. Subsection 408.035(3), Florida Statutes: The ability of the applicant to provide quality of care and the applicant's record of providing quality of care. AHCA is well aware of HRMC's excellent quality of care record. The quality of care offered by HRMC is excellent. The general consensus of those testifying at hearing was that HRMC has an excellent reputation and provides excellent quality care. A number of Wuesthoff's witnesses acknowledged that Health First has a good reputation for providing quality care. Only one physician testified, on behalf of Wuesthoff, that he believed the care offered at Holmes did not meet his standards. All three Health First hospitals are JCAHO accredited. Both Holmes and PBCH have received Magnet Designation, the highest award given by the American Nurses credentialing Center ("ANCC") for excellence in nursing. Less than two percent of hospitals in the country receive such designation, and no other Brevard County hospitals are so designated. In awarding Magnet Designation to a hospital, the ANCC takes into consideration a number of factors, including nursing management, best practices, quality of care, and the supportive environment for nurses to grow both personally and professionally. Holmes was recently ranked the number one cardiac hospital in Florida. In the past year, Holmes was recognized by the Florida Health Care Coalition for high quality service. Health Grades, a national consumer provider website, recognized Holmes as being in the top five to ten percent rankings in cardiac, vascular, pulmonary, and orthopedic services during the past year. Health First has instituted a hospitalist program at its three existing hospitals. This program provides for inpatient specialty care. Health First has made significant strides in the past several years towards its electronic medical model, which helps the system achieve superior quality in patient safety. All clinical applications are now accessible on the Web to allow physicians access to a patient's records from an off-site location. Health First operates an e-ICU system to improve clinical care management. The e-ICU program covers all six units of critical care patients at the Health First hospitals. Health First has implemented a PACS (picture archive communication system) to provide a digital environment for radiology management. Each of these programs will be implemented at VMC. Holmes has been recognized the past two years as one of the "100 Most Wired" hospitals in the country. The evidence was not persuasive that approving VMC would exacerbate any nursing or physician shortages in Brevard County, thereby having a negative impact on quality of care in the county. HRMC clearly demonstrated its ability to provide quality care. Subsection 408.035(4), Florida Statutes: The availability of resources, including health personnel, management personnel, and funds for capital and operating expenditures, for project accomplishment, and operation The parties stipulated that HRMC has sufficient funds available for capital and operating expenditures for the proposed project. The parties further stipulated that the proposed staffing patterns and the projected salaries contained in Schedule 6A are reasonable for the census levels projected in the application. AHCA, as part of its review, determined that HRMC's application met this criterion. HRMC is confident it can recruit and hire the necessary staff to operate the proposed hospital. HRMC has made the development of leadership and management programs a priority. It has focused heavily on recruitment, especially of nurses. Health First proactively approaches staffing. It had a successful recruiting campaign in the past year and was able to recruit and hire "one hundred nurses in one hundred days." Health First is committed to working with educational organizations to enlarge area nursing programs. With the addition of two new nursing programs in Brevard County, the capacity within the county itself to produce qualified nursing graduates has grown. The most recent data from the Florida Hospital Association shows that both nursing turnover and nursing vacancies have decreased from their reported high in 2001. Human resources for Health First hospitals is a corporate function, and services such as recruitment and retention are provided consistently throughout the health system. In addition to staffing, HRMC and Health First have the ability to provide management resources to establish and operate the proposed hospital. The proposed hospital will benefit from the shared resources and efficiencies gained by its affiliation with Holmes and Health First. Corporate functions, such as human resources, finance, marketing, and information technology are provided throughout the organization. Health First operates on a matrix management system under which one person serves as the matrix director for all three current labs, as well as physical therapy, pharmacy, environmental services, and health information management. Although a shortage of nurses and specialty physicians exists in the county, as well as in the rest of the country, the situation is improving. The shortage has improved over the last three to four years. Today, Health First is able to staff its hospitals with competent, skilled nurses. HRMC and Health First demonstrated that they are able to successfully recruit competent staff for certain specialty staff positions, such as radiology technicians. Health First recently recruited more than 15 new physicians in Brevard County, including specialists in neurology, neurosurgery, gynecology, cardiology, internal medicine, family practice, general surgery, and gastroenterology. HRMC's recruiting success complements the more than 180 physicians with offices in the Viera/Suntree area. The number of physicians in this area continues to grow. Moreover, Wuesthoff is able to attract competent, qualified nurses who meet the community standards in terms of education and experience. HRMC was initially concerned that the opening of the Wuesthoff-Melbourne hospital in late 2002 would have an adverse impact on its ability to retain and recruit necessary staff. The impact Health First has experienced, however, has been slight. Also, Wuesthoff publicly claimed that it had no problems in staffing its new Melbourne facility. Wuesthoff-Melbourne opened in December 2002, and staffed its entire facility. The hospital was able to meet the quality standards for the community. Wuesthoff-Melbourne has continued to improve the quality and availability of services. Dr. Williams, whose group practice has an exclusive provider contract with Wuesthoff to provide emergency physicians, testified that Wuesthoff-Melbourne has a full complement of ED physicians and that the quality of care provided by Wuesthoff's EDs is acceptable and within the standard of care for the community. Within months of opening, Wuesthoff-Melbourne had over 300 physicians on staff. Additionally, Wuesthoff-Melbourne has continued to attract better specialist coverage over the past three years. Disapproving the VMC project will not favorably affect the staffing challenges in the county. The core staff needed to care for patients at the new hospital would be needed to staff increased bed capacities at any of the existing Brevard County hospitals. ED call coverage tends to be a problem in Brevard County. The evidence was not persuasive, however, the problems experienced in Brevard County are as significant as they are in other parts of the state, or that approving VMC would materially exacerbate the problem. The evidence did not demonstrate that physicians would be on call at more than one hospital if VMC were approved. The evidence did not prove that physicians would give up their privileges at other Brevard County hospitals to accept privileges at VMC in such significant numbers that other hospitals would be unable to staff their EDs. The VMC project has drawn considerable support (over 78 letters) from physicians in the community. Many of these physicians are interested in joining the VMC medical staff. HRMC has demonstrated that it has ample resources available and funds for capital and operating expenditures to accomplish its proposed project. Subsection 408.035(5), Florida Statutes: The extent to which the proposed services will enhance access to health care for residents of the service district The extent to which the proposed services will enhance access to health care for residents is a significant criterion when considering a proposal to establish a new hospital. The VMC project demonstrates compliance with this factor as it emphasizes access to emergency services, as well as basic hospital services, in a rapidly-growing suburban market. AHCA has favored projects similar to what HRMC proposes, recognizing a national trend to approve projects that propose smaller, suburban hospitals rather than adding beds at existing urban hospitals. Such applications emphasize access to emergency care and basic hospital services. In terms of traffic, the proposed site is approximately 10 miles from the nearest Wuesthoff hospital, and 15 miles from the nearest Health First hospital. As the population continues to grow, traffic access will diminish. Wuesthoff presented a travel time study analyzing the time it takes to travel from the proposed Viera hospital location to Wuesthoff-Rockledge and Wuesthoff-Melbourne during morning and afternoon peak traffic hours. The report concluded that a driver could reach either Wuesthoff-Rockledge or Wuesthoff-Melbourne from the propose Viera site within 20 minutes during both the morning and afternoon peak hours. The results of the study are not entirely persuasive for several reasons. The study claimed to use three test runs for each of the routes examined, yet the I-95 route result was based upon one run only. Next, the peak times used for the traffic runs were defined as 7:00 a.m. to 9:00 a.m. and 4:00 p.m. to 6:00 p.m. The actual runs, however, were made almost exactly at the beginning of the peak hours as so defined (all between 7:00 and 7:10 a.m. for the morning runs, and between 4:00 and 4:07 p.m. for the afternoon runs). This study failed to take into account any possible variations during both the morning and afternoon rush hours. Moreover, Dr. Finton, a physician who lives in Suntree (near Viera) testified that the same runs take him 30-45 minutes to reach any of the three hospitals, Wuesthoff-Rockledge, Wuesthoff-Melbourne, or Holmes. The travel time study was incomplete and, therefore, unpersuasive that the existing hospitals can be reached within 20 minutes during peak travel times. Diversion The Brevard County EMS system controls which hospital can go on diversion, and works based upon the hospital's good faith assumption that it will act in the patient's best interest. In 2005, the EMS policy on diversion for south Brevard County changed. If any one of the three existing hospitals in that part of the county can take patients, the diversion request is granted. The factors that typically trigger a hospital to diversion status are a sudden influx of critical patients or the lack of available beds in the ED. When a hospital goes on diversion, its emergency room is, in effect, closed. This strains and compromises the delivery of medical services. In 2005, diversion status was granted 53 times. In 2006, approximately 40 diversion requests had been granted as of the date of hearing. Although the number of diversion requests in 2006 was down from the 2005 level, the number of hours the hospitals remained on diversion status increased. In some instances, Holmes' diversion request has been denied because PBCH was at capacity and could not take more patients. A week before final hearing in this case, Holmes was operating at capacity with as many as 36 patients in the ED waiting for beds, and was forced to go on diversion status. Even when a hospital is denied diversion status, the EMS system feels the impact because patients taken to the ED cannot be accepted due to lack of a bed. This forces the paramedic to wait with the patient for an hour or more, which takes the emergency vehicle and the paramedic out of service. Approval of VMC would improve emergency medical care for the residents of Brevard County. Adding more beds in Viera would decompress the high census problems at Holmes and PBCH. The EMS transport times in Midwest Brevard County would decrease, especially in the areas served by the Viera and Suntree stations. The Viera project will enhance access to a rapidly growing community as recognized by the testimony of Brevard County health officials and health care providers. 2. Financial Access HRMC's proposed CON conditions will improve financial access for low income residents of Brevard County. This access is assured through Health First's commitment to approval of VMC on the conditions it proposed. Wuesthoff acknowledged that patients making up the payor class addressed by HRMC's 19.5 percent condition, including Medicaid, Medicaid HMO, self-pay, the uninsured, KidCare, and charity care could be deemed "safety net" patients. Wuesthoff has no system-wide condition concerning safety net care, and Wuesthoff-Rockledge has no hospital-wide condition regarding Medicaid or charity care. HRMC demonstrated that its proposed services will enhance access to health care for residents of the service district. Subsection 408.035(6), Florida Statutes: The immediate and long-term financial feasibility of the proposal Immediate or short-term financial feasibility The parties stipulated that the proposed project demonstrates immediate financial feasibility, since HRMC has sufficient funds for capital and operating expenses available for the proposed project. 2. Long-tern financial feasibility AHCA tests the reasonableness of an applicant's financial projection by comparing them with their peer group experience. In this case, AHCA concluded that HRMC's financial ratios were within the reasonable range. Generally, a CON project is financially feasible if it will at least break even in the second year of operation. However, a project is not financially feasible in the long-term if it continues to show a loss in the second year of operation, unless it is nearing break-even and can demonstrate that it will break even within a reasonable period of time. New hospitals are not developed for a two- to three-year projection. The life of a hospital is measured in decades, not years. The net operating revenue projected on Schedule 7A of the HRMC application, the starting point for the net income/loss projected on Schedule 8A, is reasonable. Holmes projects it will generate a net loss of just over $1 million in its first year of operation, but will generate net profits of $5.6 million and $11.7 million in its second and third years of operation. On a stand-alone basis, the proposed VMC project is financially feasible in the long term. When viewed on a consolidated basis with the other two HRMC hospitals, Holmes and PBCH, as well as on a health system-wide basis (Health First), VMC is also feasible in the long term. Wuesthoff argued that HRMC would be better served by investing its money in some high yield instrument rather than building a new hospital. Since the hospitals involved in this proceeding are not-for-profit corporations, and involved in the health care business, maximizing profits and returns on investment are not of primary concern. Wuesthoff's health planner, Rick Knapp, acknowledged that the primary mission of both Wuesthoff and Health First is to provide medical facilities and services in their communities. The manner in which HRMC accounted for the management fee line item in the VMC pro forma is consistent with how it is treated for PBCH, and is reasonable. The criticisms leveled by Wuesthoff against HRMC's application concerning financial feasibility focused on alleged differences between the Wuesthoff and Health First case mix adjusted prices; the outpatient competition from Phase I of Viera; the "lost" investment opportunity for the funds spent in developing VMC; and the effect the opening of VMC will have by stealing patients from Wuesthoff. The evidence supported HRMC's contention that its case mix adjusted prices were well within, if not below, the parameters for other counties of similar size. Wuesthoff did not quantify the alleged impact the outpatient services in Viera would have on VMC's outpatient revenues. The project is financially feasible with the funds being spent for the delivery of health care services, not for seeking potential "lost" income on investments, which is clearly not a goal of Health First. No provider is guaranteed a particular share of the market. Health First and Wuesthoff must compete with all providers for their patients. Dr. Eisenstadt's challenge to VMC's financial assumptions is not persuasive. He sought to compare the facilities' financial feasibility based upon data from Holmes. The percent to gross revenue, the percent payor mix, the gross charges per case, the gross charges per patient day, and the case mix index are all different, and not comparable. HRMC demonstrated that the proposed VMC project is financially feasible in both the short and long term. Subsection 408.035(7), Florida Statutes: The extent to which the proposal will foster competition that promotes quality and cost-effectiveness Generally, competition for hospital services benefits consumers because it leads to lower prices, creates incentives for hospitals to lower costs, and leads to improved quality of care. The proposed Viera facility will add bed capacity which will permit increased supply to serve the growing demand for health care services. The proposal will add not only beds, but the proposed site is in an area in which demand is demonstrating particularly strong growth. The Viera facility is well-positioned for convenient or lower cost access for patients requiring hospitalization. AHCA believes that a CON applicant can overcome the potential adverse impact associated with a proposed new hospital project on an existing provider by offering to extend access to lower income persons. HRMC and Health First have agreed to do this through the conditions set forth in the application. Table 13 of HRMC's application provided a market share analysis to show what share the new hospital is expected to capture in each of the five zip codes in the PSA by the third year of operation. HRMC projects that it will take three years for it to achieve its market share of just under 21 percent of the PSA. These projections are reasonable and accurate. Table 16 of the application indicates sufficient demand and population growth in the VMC PSA to support the proposed Viera hospital as well as the existing providers. Figure 18 of the application represents the change in adult medical/surgical discharges by hospital between 2004 and 2012, assuming VMC is opened. It demonstrates that, with the exception of Holmes, every other hospital in Brevard County will see a significant increase in adult medical/surgical discharges over that eight-year period, even if VMC is built. As stated previously, the decline in discharges at Holmes is a desirable effect of building the new hospital because it will allow Holmes to decompress its currently over-utilized state. The decision to establish Wuesthoff-Melbourne was a centerpiece in the Wuesthoff Health Systems strategy to be more attractive to managed care providers by having a presence in south Brevard County. Wuesthoff's CEO acknowledged that it is now competitive with Holmes from a pricing perspective. No managed care organization has refused to negotiate with Wuesthoff since it opened its Melbourne facility in late 2002. Wuesthoff hospitals now have managed care contracts with four of the five major managed care organizations. It does not have a managed care contract with Health First Health Plans (HFHP). Wuesthoff has never attempted to contract with HFHP. Candidly, Wuesthoff conceded that it would not be in its best interest to attempt to contract with a competitor (Health First) and therefore promote the competitor's growth. Wuesthoff-Melbourne has attempted in the past to expand its current bed capacity. In 2004, it filed a CON application to expand to 134 licensed beds, a 19-bed addition. Although the 2004 amendments to the CON law eliminated the need to request regulatory approval to add beds, and despite having the space for the additional beds shelled-in, Wuesthoff has yet to expand beyond 155 beds at its Melbourne facility. Wuesthoff argues that its Melbourne campus needs a more complete array of services, including open-heart surgery and Level II NICU, in order to better compete with Holmes. Wuesthoff-Melbourne, however, has not committed to expanding its bed capacity at any time over the next 10 years, nor does it have a current written plan for the timing of any future expansion. Further, Wuesthoff-Rockledge has made no public plans to expand its facility at any particular time in the future. Approval of HRMC's application is pro-competitive. Dr. Lynk's testimony in this regard is persuasive. Dr. Lynk, an industrial organization economist, has testified in CON matters for over 20 years and employs generally accepted data and economic regression methods and variables that are generally accepted in the field of industrial economics. Based upon Dr. Lynk's testimony, it is found that HRMC's prices are within the observed range of hospital prices for inpatient services in competitive Florida markets, and are at competitive levels, as measured by reliable objective data and generally accepted econometric methods. Wuesthoff's prices, on average, are currently lower than HRMC's. This comparison, however, does not determine whether the prices are competitive, because in a comparison of two hospitals, one must necessarily be lower than the other. A comparison of pricing alone does not lead to the conclusion that one hospital is pricing below, at, or above competitive levels. The price difference between HRMC and Wuesthoff hospitals is within the observed range of price differences between hospital providers observed in a large number of other Florida markets with multiple competing hospitals. HRMC provided extensive data to support this finding. Wuesthoff's claim that HRMC's prices are above competitive levels is not supported by the evidence. Construction of the new Viera facility will expand capacity in a market that is experiencing growing demand, particularly in the PSA. Because it will expand bed capacity in Brevard County, especially in the fastest growing area of the county, the new facility will competitively benefit Brevard County residents in need of hospital services. Dr. David Eisenstadt testified on certain competition issues in this case. Dr. Eisenstadt is an experienced and respected economist. However, the opinions he offered concerning the anti-competitive effects of approving HRMC's CON application are not persuasive. Dr. Eisenstadt's testified that the Viera project will harm hospital competition in southern Brevard County if VMC prevents Wuesthoff-Melbourne from expanding to 200 beds from its current size. Further, he opines, the new hospital will increase hospital costs to the managed care industry in Brevard County by approximately $189 million in 2004 dollars. Dr. Eisenstadt's opinion in this regard is not found to be persuasive. Dr. Eisenstadt's opinion on the competitive impact of the new Viera hospital was based upon his regression analysis. In conducting this analysis, Dr. Eisenstadt relies upon his measure of hospital "dominance" in a particular market. The dominance measure used by Dr. Eisenstadt is new and not generally accepted in the economic community. He concludes that managed care prices in Brevard County will decrease by a total of $189 million if the Wuesthoff-Melbourne hospital increases in size from 115 to 200 beds. Of course, Wuesthoff-Melbourne has neither attempted to increase its bed complement to 200 nor to produce evidence of a written long-term plan setting forth a timetable for increasing the hospital's bed capacity. Dr. Eisenstadt's dominance variable regression technique is not found to be persuasive for several reasons. First, it has not been shown to be generally accepted by the economic community. He has never used the particular theory in the past nor can he point to another expert who has employed the theory. The technique he has chosen to use in this case has not been critically reviewed by the community of economists and is, therefore, not persuasive. Dr. Eisenstadt's cost savings analysis is driven entirely by bed numbers. He testified that so long as Wuesthoff-Melbourne increased in size from 115 to 200 beds, while Holmes remained the same size, HRMC's dominance would drop along with the prices leading to his estimated cost savings. In short, a change in number of beds drives a change in dominance. This theory is not borne out when examining all hospitals in Florida. Such an examination reveals no link between bed numbers and a particular hospital's dominance in a market. Dr. Eisenstadt's dominance measure is not consistent with testimony from other managed care representatives about what makes a hospital dominant. The evidence supports that it is not just the size of the hospital, but the availability of complex services that makes one hospital dominant over another. In this case, Holmes provides a far more complex mix of services than Wuesthoff-Melbourne, which should support a more dominant position for Holmes based upon the services offered, not the relative size of the two facilities. Yet, a comparison of the supposed "dominance" of hospitals throughout Florida, as measured by Dr. Eisenstadt, with the acuity of patients (as measured by case mix), shows no correlation. Dr. Eisenstadt's methodology did not accurately predict actual pricing behavior by Holmes when it was applied historically. Dr. Eisenstadt's model predicts that as Wuesthoff-Melbourne grows compared to HRMC, HRMC's prices should decrease. However, Dr. Eisenstadt conceded that the data from December 2002 (when Wuesthoff opened its new 65-bed campus in Melbourne and Holmes added no new beds) through December 2004 showed that Holmes' prices did not decrease relative to Wuesthoff's. This directly contradicts Dr. Eisenstadt's methodology. Dr. Eisenstadt's measure of dominance in his regression study classifies a number of hospitals commonly regarded as prominent and desirable facilities (and therefore expected to have substantial market power), such as the University of Florida Shands teaching hospital, as not being "dominant." Conversely, he classifies several relatively small hospitals as "dominant," such as St. Cloud Hospital, when they do not possess the size and characteristics of traditionally strong hospitals in the marketplace. These classification aberrations, when viewed in the context of a methodology that has not been generally accepted in the economics community renders the reliability of this regression analysis suspect. Dr. Eisenstadt did not testify that building the Viera facility would prevent expansion of Wuesthoff-Melbourne, nor could he testify that Wuesthoff-Melbourne would be expanded at any specific time in the future. His dominance-based pricing analysis also assumed that neither Holmes nor PBCH added beds. His estimate of an increase of $189 million in managed care costs to the market is based upon two unsubstantiated assumptions: 1) that Wuesthoff-Melbourne would not and could not expand if Viera is built; and 2) that Wuesthoff-Melbourne would expand to 200 beds if VMC is not built. The evidence at hearing supports neither of these assumptions. Dr. Eisenstadt also testified that building VMC will be anti-competitive, regardless of whether Wuesthoff-Melbourne's expansion is affected by VMC's construction, because he concludes that VMC's charges will be higher than Wuesthoff's. The result, he opines, will be that VMC will cost managed care organizations and employers $75 million more than if the patients were treated at a Wuesthoff facility. His opinion in this regard is also not persuasive. Dr. Eisenstadt assumes, for purposes of this opinion, managed care organizations cannot direct their patients to less costly hospitals if they perceive prices to be above competitive levels. This is not so as evidenced by the fact that the Harris Corporation, one of Brevard County's largest employers, is self- insured, and it steers its plan members to Wuesthoff because of Wuesthoff's lower prices. Interestingly enough, the Harris Corporation supports the Viera facility proposed here which argues against a perception that VMC will not be competitively priced. The substantial growth in Brevard County, especially in the VMC PSA, supports the establishment of a new hospital. Based upon this fact, and the fact that the new facility proposed by HRMC will not be anti-competitive, the proposed facility will not have a material adverse impact on any existing provider. Subsection 408.035(8), Florida Statutes: The costs and methods of the proposed construction, including the costs and methods of energy provision and the availability of alternative, less costly, or more effective methods of construction The parties stipulated that HRMC's application demonstrated the reasonableness of the costs (including equipment), methods of construction, and energy provision for the project as proposed. The architectural drawings submitted by HRMC with its application satisfy all applicable code requirements. The parties also stipulated that the time intervals contained in the project completion forecast depicted in Schedule 10 of the application are reasonable and require no further proof. Subsection 408.035(9), Florida Statutes: The applicant's past and proposed provision of health care services to Medicaid patients and the medically indigent HRMC provides its fair share of health care services to Medicaid patients and the medically indigent population of Brevard County. Holmes provides the greatest percentage of charity care as well as the largest volume in terms of total dollars of charity care in the subdistrict. It is also the largest provider of Medicaid services in the subdistrict. HRMC is the largest provider of safety net services in the county. It is larger than all other providers combined. Health First and its member hospitals serve everyone, regardless of insurance or ability to pay. It is fully committed to meeting the needs of the citizens of Brevard County, and does so without receiving tax dollars. In the early 1990s, HRMC established HOPE, a Health Outreach, Prevention, and Education project to serve the health care needs of Brevard County. Since that time, HRMC and Health First have created a series of fixed-site medical clinics to serve the medically needy, in addition to providing a mobile health care delivery clinic that travels to various sites within the county. HRMC has a strong record of providing health care services to Medicaid patients and the medically indigent. Subsection 408.035(10), Florida Statutes: The applicant's designation as a Gold Seal Program nursing facility pursuant to §400.235, Fla. Stat., when the applicant is requesting additional nursing home beds The parties stipulated that this criterion is not applicable. CONFORMANCE WITH RULE CRITERIA When no Agency methodology exists to determine need, an applicant is responsible for demonstrating need through a needs assessment methodology which includes: 1) population, demographics, and dynamics; 2) availability, utilization, and quality of like services in the district, subdistrict, or both; 3) medical treatment trends; and 4) market conditions. Fla. Admin. Code R. 59C-1.008(2)(e). HRMC's application conforms with the preferences contained in Florida Administrative Code Rules 59C-1.008(2)(e) and 59C-1.030. To the extent population growth and utilization outpace hospital expansion, the result creates an access issue. Access to inpatient hospital services affects not only convenience, but also matters of life and death. Access is improved by the distribution of health care facilities where the population needs them most, not by concentrating services in one or two areas. The proposed Viera site is in a rapidly growing suburban community, where the projected population growth in the immediate service area is greater than in the planning area as a whole. Additionally, for the VMC PSA, the elderly population and its growth rate are both projected to increase in the future. The HRMC proposal addresses issues of financial access, based upon HRMC and Health First's system-wide condition of 19.5 percent of admissions for Medicaid and charity care patients. Such a substantial commitment to the underserved mitigates Wuesthoff's contention that approval of VMC will force it to take on a greater share of the indigent patient care. AHCA has a CON condition review process in place which requires an annual report as to the provider's performance in meeting its CON conditions, authorizing it by statute to impose a fine of up to $1,000 per day, and up to $365,000 per year, for any violations. AHCA has utilized its enforcement power in the past, imposing fines of as much as $1.5 million in cases of multi-year violations of CON conditions. IMPACT OF VMC ON WUESTHOFF VMC is projected to capture approximately 30 percent of its patient volume from patients who would otherwise seek care at one of the two Wuesthoff hospitals. At the same time, however, based upon the projected growth in Brevard County, especially in the Viera/Suntree area, inpatient admissions to Wuesthoff are expected to grow by more than 30 percent between 2004 and 2012. Therefore, any impact on Wuesthoff from the new VMC will be short term. While some competitive effect will occur due to the establishment of the new facility in Viera, it will not rise to the level of a substantial adverse impact since the income levels of both Wuesthoff-Rockledge and Wuesthoff- Melbourne are reasonably expected to be at higher levels in 2012 than they were in 2006. According to Dr. Finarelli's projections, Health First, even with VMC coming online, will actually lose market share in certain zip codes as the utilization of Wuesthoff- Melbourne increases. Based upon projections, at a system-wide level, the market shares of Health First and Wuesthoff should remain at the same level in 2012 as they were in 2004, 55 percent and 40 percent, respectively. Wuesthoff argues that approval of VMC would have a significant adverse impact on its two hospitals by jeopardizing its ability to meet bond covenants and by inhibiting its ability to implement its expansion projects. In order to quantify the magnitude of the adverse impact, Wuesthoff offered a sensitivity analysis based in part upon an internal template that Wuesthoff had been using for three to four years. A sensitivity analysis is a type of financial analysis. Normally, revenues and expenses would be projected separately in a financial analysis. In this case, Mr. John Van Gorp, Wuesthoff's expert in health planning and financial analysis, made some assumptions that are contrary to the evidence presented at hearing. First, the growth utilization projections for Wuesthoff for the period 2010 to 2012 are the same for each scenario, with or without VMC. Second, Mr. Van Gorp assumed an annual three percent growth in admissions for the internal projections provided for planning purposes at Wuesthoff-Rockledge, yet assumed only a four percent annual increase combined for growth in admission and price increases. Third, Mr. Van Gorp failed to forecast a specific increase in managed care reimbursement rates for the period 2007 to 2012. Next, his analysis accounts for $1.75 million in additional expenses for hypothetical construction projects of $35 million in 2010, but showed no increasing revenues resulting from these expansion projects at either Wuesthoff facility. Finally, he assumes $35 to $40 million of capital expansion projects at a time when its borrowing capacity, as reflected in its own calculations, shows its borrowing ceiling to be below $35 million. The Van Gorp sensitivity analysis is done at too broad a level to predict detailed information such as cash on hand projected out to 2012. The analysis is too general to predict how the specific bond covenants will or will not be met five to six years in the future. Further, Mr. Van Gorp relies heavily on projections made by the same Wuesthoff financial analysts who projected in two previous Wuesthoff-Melbourne CON applications that the project would be financially feasible within two years. AHCA found both those projects not to be financially feasible and denied them. The sensitivity analysis presented by Wuesthoff has too many weaknesses to serve as a basis for denying HRMC's CON application. Armand Balsano, HRMC's expert in health care finance and planning, prepared an analysis of the impact the approval of VMC would have on both Wuesthoff hospitals. Assuming a 2010 opening date for VMC, by 2012 both Wuesthoff-Rockledge and Wuesthoff-Melbourne are projected to be at their pre-VMC admission levels. Whenever a new hospital enters the market, some effect on existing providers will occur. In this case, the entry of VMC will not create a long-term adverse impact on the Wuesthoff Health System, and will not jeopardize its long-term financial feasibility. Since 2000, Wuesthoff-Rockledge has a net worth of $67.5 million. Wuesthoff-Rockledge is the main economic engine of the Wuesthoff Health system, and continues to show strong growth and improvement in net revenues and cash flows, demonstrating a positive operating income. It experienced more than $24 million in cash flow in its most recent fiscal year. In the most recent fiscal year, Wuesthoff-Melbourne generated a cash flow of approximately $2.5 million, a significant improvement over the previous two years. Furthermore, the operations at Wuesthoff-Melbourne have continued to improve in the current fiscal year, reflecting a positive EBIDA (earnings before interest, depreciation, and amortization) of $6.7 million, evidence of a strong financial improvement. Overall, Wuesthoff-Melbourne has made strong financial headway, despite being a new hospital. Although not in line with its projections in previous CON applications, Mr. Fayer testified that it typically takes four to five years for a new hospital to generate a positive operating income. Wuesthoff-Melbourne is in its fourth year of operation and is nearly there. Mr. Fayer acknowledged that Wuesthoff Health System projects to further improve its financial viability in the next three years. Interestingly, Wuesthoff has developed a master plan to add 40 beds at a cost of more than $80 million at the same time stating that it can barely maintain its bond covenants. Wuesthoff has already spent more than $100,000 on developing expansion plans for its Rockledge campus. If built as now proposed in its master plan, Wuesthoff-Rockledge would expand to 410 beds at a cost of at least $123 million, not including equipment. Since the opening of Wuesthoff-Melbourne, Mr. Fayer has not been turned down by any managed care organization he sought to contract with on behalf of Wuesthoff. He has not contacted Health First Health Plan to determine whether that organization would contract with Wuesthoff. Since its opening in December 2002, Wuesthoff- Melbourne has met or exceeded its projected patient volumes. Wuesthoff-Melbourne cannot argue that it has not yet made a profit due to not meeting its patient volumes. Patient volumes continue to rise at the hospital and it is moving toward a positive bottom line. Wuesthoff-Rockledge has enjoyed an excess of revenues over expenses in every year between 1998 and 2005, except in 1999, that year, Wuesthoff lost $11 million, at least a portion of which was due to mismanagement, lack of overall leadership within the system, and volume variances. Wuesthoff took the necessary and appropriate steps to correct the management issues and quickly returned to a positive bottom line. The adverse impact analysis prepared by Mr. Knapp, based upon patient volume information received from Mr. Richardson, did not present an accurate picture of the alleged adverse impact of VMC on the existing Wuesthoff hospitals. It, therefore, is not persuasive.
Recommendation Based upon the Findings of Fact and Conclusions of Law, it is RECOMMENDED that the Agency issue a final order approving HRMC's CON Application No. 9881. DONE AND ENTERED this 26th day of January, 2007, in Tallahassee, Leon County, Florida. S ROBERT S. COHEN 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 26th day of January, 2007. COPIES FURNISHED: Michael J. Glazer, Esquire E. Dylan Rivers, Esquire Martin B. Sipple, Esquire Ausley & McMullen 227 South Calhoun Street Post Office Box 391 Tallahassee, Florida 32302 R. Terry Rigsby, Esquire Daniel C. Brown, Esquire W.Douglas Hall, Esquire Carlton, Fields, P.A. Post Office Drawer 190 Tallahassee, Florida 32302 Karin M. Byrne, Esquire Agency for Health Care Administration 2727 Mahan Drive, Building 3 Mail Station 3 Tallahassee, Florida 32308 William G. Kopit, Esquire Lee Calligaro, Esquire Patricia M. Wagner, Esquire Epstein, Becker and Green, P.C. 1227 Twenty Fifth Street, Northwest Washington, District of Columbia 20037 Jerome W. Hoffman, Esquire Holland & Knight, LLP Post Office Drawer 810 Tallahassee, Florida 32302 David E. Mathias, Esquire Health First, Incorporated 6450 United States Highway 1 Rockledge, Florida 32955 Richard Shoop, Agency Clerk Agency for Health Care Administration 2727 Mahan Drive, Mail Station 3 Tallahassee, Florida 32308 Craig H. Smith, General Counsel Agency for Health Care Administration Fort Knox Building, Suite 3431 2727 Mahan Drive Tallahassee, Florida 32308 Andrew C. Agwunobi, M.D., Secretary Agency for Health Care Administration Fort Knox Building, Suite 3116 2727 Mahan Drive Tallahassee, Florida 32308
The Issue The issue is whether the Agency for Health Care Administration should approve Petitioner’s application for a Certificate of Need to establish a 54-bed freestanding long-term care hospital in Escambia County.
Findings Of Fact Parties Select-Escambia is a subsidiary of Select Medical Corporation (Select), which has been in the business of operating LTCHs since the 1980’s. Select currently operates 99 LTCHs in 27 states, including three in Florida. Select’s Florida LTCHs are located in Orlando, Miami, and Panama City. The Orlando and Panama City LTCHs were formerly operated by SemperCare, Inc. (SemperCare), which Select acquired in January 2005. Three other Select LTCHs –- in Tallahassee, Orlando, and Alachua County -- have been approved by the Agency, but are not yet operational. The Tallahassee LTCH, which was also formerly a SemperCare facility, was originally projected to open in 2006, but that date is no longer certain. The Agency is the state agency responsible for administering the CON program and for licensing LTCHs and other health care facilities. Application Submittal and Review and Preliminary Agency Action In the second batching cycle of 2004 for hospital beds and facilities, Select-Escambia filed with the Agency an application for a CON to establish a 54-bed freestanding LTCH in Escambia County. There were no co-batched applications comparatively reviewed by the Agency with Select-Escambia's application, CON 9800. Select-Escambia’s application was complete, and it satisfied the applicable submittal requirements in the statutes and the Agency's rules. The Agency’s review of Select-Escambia’s application complied with the applicable statutory and rule requirements. The Agency’s review culminated in a SAAR issued on December 10, 2004. The SAAR recommended denial of CON 9800, primarily based upon Select-Escambia’s failure to demonstrate to the Agency’s satisfaction that there is a need for the proposed Escambia County LTCH. The determination in the SAAR that Select-Escambia failed to adequately demonstrate need for its proposed LTCH was largely based upon a 2004 report by MedPAC, which is an organization that advises Congress on issues related to Medicare. The MedPAC report concluded that LTCH patients need to be better defined so as to ensure that the patients treated at LTCHs are of the highest severity and cannot be more cost- effectively treated in other care settings. The Agency formally published notice of its intent to deny CON 9800 in the Florida Administrative Weekly, and Select- Escambia thereafter timely filed a petition challenging the Agency’s denial of its application. The Agency reaffirmed its opposition to Select- Escambia’s application at the hearing through the testimony of Jeffrey Gregg, the bureau chief over the Agency’s CON program. LTCHs Generally An LTCH is defined by statute and Agency rule as “a hospital licensed under chapter 395 which meets the requirements of 42 C.F.R. s. 412.23(e) and seeks exclusion from the Medicare prospective payment system for inpatient hospital services.” LTCHs provide extended medical and rehabilitative care to patients with multiple, chronic, and/or clinically complex acute medical conditions. They serve a patient population whose average length of stay (ALOS) exceeds 25 days. There are two types of LTCHs: hospital-within-a- hospital (HIH) and freestanding. Both types are accepted in the industry, and both types are found in Florida and nationwide. HIH LTCHs are located in the same building or on the same campus as a traditional acute care hospital, which is referred to as the “host hospital.” HIH LTCHs contract with the host hospital for ancillary services such as laboratory and radiology services. HIH LTCHs get the vast majority of their admissions from the host hospital, whereas freestanding LTCHs tend to get their admissions from a number of different hospitals. LTCHs fit into the continuum of care between traditional acute care hospitals and traditional post-acute care facilities such as nursing homes, skilled nursing facilities (SNFs), hospital-based skilled nursing units (SNUs), and comprehensive medical rehabilitation (CMR) facilities. LTCHs are designed to serve patients that would otherwise have to be maintained in a traditional acute care hospital (often in the ICU) where the reimbursement rates may be insufficient to cover the costs associated with a lengthy stay, or be moved to a traditional post-acute care facility where the patient may not receive the level of care needed. Patients with co-morbidities, complex medical conditions, or frailties due to age are typically appropriate LTCH patients, particularly if the patient would otherwise remain in the ICU of a traditional acute care hospital. For such patients, an LTCH is likely the most appropriate setting from both a financial and patient-care standpoint. There is a distinct population of patients who, because of the complexity or severity of their medical condition, are best served in an LTCH. However, there is an overlap between the population of patients that can be served in an LTCH and the population of patients that could also be well- served in the ICU of an acute care hospital or a traditional post acute care setting with ventilator capability. Indeed, as noted in the MedPAC report, “[i]n the absence of LTCHs, clinically similar patients are principally treated in acute hospitals or in freestanding SNFs that are equipped to handle patients requiring a high level of care.” Because of the overlap in patients, it is important for LTCHs to adopt detailed admission criteria to ensure that the LTCH (rather than a SNF, SNU, or CMR) is the most appropriate care setting for the patient. InterQual, which is a private organization that establishes standards for quality of care for a variety of health care settings, has developed model admission criteria for LTCHs. The Interqual criteria are designed to ensure that the LTCH is the most appropriate care setting for the patient, and they are referenced in the MedPAC report as an example of the type of admission criteria that LTCHs should adopt to ensure that they are not treating patients that should be treated in another setting. Mr. Gregg and Karen Rivera, the supervisor of the CON program, acknowledged in their deposition testimony that an LTCH’s use of the InterQual criteria would, at least to some degree, address the Agency’s concern that LTCHs might be serving patients that should be served in a more traditional, less- intensive (and/or less-costly), post-acute care setting. Select utilizes the InterQual criteria as part of its admission process at its existing LTCHs, and it intends to utilize those criteria at its proposed Escambia County LTCH. Specifically, Select’s nurses screen patients prior to admission and, again, shortly after admission to ensure they are LTCH- appropriate patients. Additionally, Select’s nurses and care teams periodically evaluate each patient to ensure that the LTCH is still the most appropriate care setting for the patient and to determine whether the patient is ready for discharge, either to a traditional post-acute care setting or to home. Select also utilizes a third-party organization to review and assess the patient-outcomes achieved at each of its LTCHs. This is a quality assurance/improvement tool because it allows Select to compare and “benchmark” the performance of its LTCHs against each other and against other LTCHs nationwide and it helps to identify functions or services that need improvement. LTCH services are most highly utilized by persons in the 65 and older (65+) age cohort because those persons are more likely to have complex and/or co-morbid medical conditions that require long-term acute care. In calendar year 2003, for example, approximately 77 percent of LTCH patients in Florida were in the 65+ age cohort and approximately 51 percent were in the 75 and older (75+) age cohort. The typical LTCH patient is still in need of considerable acute care, but a traditional acute care hospital may no longer be the most appropriate or lowest cost setting for that care. The vast majority of LTCH admissions are patients transferred directly from a traditional acute care hospital. It is not uncommon for an LTCH patient to be transferred on life support from a critical care unit or ICU after the patient has been diagnosed and stabilized. Nursing homes, SNFs, SNUs, CMR facilities, and home health care are not appropriate for the typical LTCH patient because the patient's acuity level and medical/therapeutic needs are higher than those generally treated in those settings. Indeed, unlike traditional post-acute care settings, which typically do not admit patients who still require acute care, the core patient-group served by LTCHs are patients who require considerable acute care through daily physician visits and intensive nursing care in excess of eight hours of direct patient care per day. LTCH patients are often discharged to a traditional post-acute care facility such as a nursing home, SNF, CMR facility, or home health care. Thus, those facilities cannot be considered as "substitutes" for LTCHs, even though there is some overlap between the services provided to lower acuity LTCH patients and higher acuity patients in those traditional post- acute care facilities. The family of a patient in an LTCH is generally encouraged to be more involved in the patient’s care than it would be if the patient was in the ICU of a traditional acute care hospital. For example, the visiting hours at LTCHs are typically more liberal than the visiting hours of the ICU at a traditional acute care hospital. Medicare reimbursements are the primary source of revenue for LTCHs because, on average, 75 to 85 percent of LTCH patients are covered by Medicare. In this case, Select-Escambia projected that approximately 77 percent of the patient days at its proposed Escambia County LTCH would be generated by Medicare patients. In 2002, the federal government adopted a Medicare prospective payment system (PPS) specifically for LTCHs. That system recognizes the LTCH patient population as being distinct from the patient populations treated by traditional acute care hospitals and post-acute care facilities such as nursing homes, SNFs, SNUs, and CMR facilities, even though there may be some overlap between the patient populations served by LTCHs and those other types of facilities. Under the LTCH PPS, services are reimbursed by Medicare at a predetermined rate that is weighted based upon the patient's diagnosis and acuity, regardless of the cost of care. This reimbursement system is similar to, but uses Diagnosis Related Groups (DRGs) that are different than the DRGs used in the PPS for traditional acute care hospitals. The Medicare reimbursement rates for services to long- stay patients in an LTCH are generally higher than the reimbursement rates for the same services to long-stay patients at a traditional acute care hospital. As a result, there is a financial incentive for hospitals to transfer their long-stay patients to an LTCH. In August 2004, the federal regulations governing Medicare reimbursements for LTCHs were substantially amended. One significant change in the regulations is that the number of admissions that an HIH LTCH can receive from its host hospital and still qualify for reimbursement under the LTCH PPS is generally capped at 25 percent. The effect of that change is that new HIH LTCHs will not be viable in most instances. LTCHs in Florida At the time CON 9800 was filed, there were 12 LTCHs operating in Florida with a total of 799 licensed beds. There were an additional four approved but not yet licensed LTCHs, including the three Select facilities referenced above. There are no licensed or approved LTCHs in District 1, which consists of Escambia, Santa Rosa, Okaloosa, and Walton Counties. There is at least one licensed or approved LTCH in each health planning district, except for Districts 1 and 9.2 The closest Florida LTCH to Escambia County is the former SemperCare (now Select) facility in Panama City, which is in District 2. That facility, which opened in early 2003, is a 30-bed HIH LTCH, and is approximately 100 miles and a two-hour drive from Pensacola. There is or soon will be an LTCH in Mobile, Alabama, which is approximately 60 miles from Pensacola. There was no evidence presented regarding the type, size, utilization, or quality of care at that facility. The existing Florida LTCHs are well-utilized. According to the SAAR, the overall occupancy rate for the Florida LTCH beds was approximately 68 percent in 2003, and several of the facilities had occupancy rates in excess of 80 percent. The newer facilities -– Select’s Miami LTCH, which opened in December 2002, and the former SemperCare (now Select) LTCH in Orlando, which opened in June 2003 -- had considerably lower occupancy rates, which as discussed in the Select-Marion Recommended Order (page 23), is to be expected. If the beds and patient days for those facilities are excluded from the calculation in the SAAR, the overall occupancy rate for the Florida LTCH beds in 2003 would have been slightly above 71 percent. The existing Florida LTCHs receive a majority of their admissions from the county in which they are located, which is consistent with the comment in the MedPAC Report that proximity to an LTCH “quadruples the likelihood that a [patient] will use a long-term care hospital.” Florida LTCHs served patients in 174 of the 527 DRGs in calendar year 2003, but 50 of the DRGs accounted for 91 percent of the cases and 93 percent of the patient days. By far, the most commonly treated DRG is No. 475, which is “respiratory system diagnosis with ventilator support.” Select-Escambia’s Proposed LTCH Select-Escambia’s proposed LTCH will be a 54-bed freestanding facility in 54,090 square feet of new construction. The precise location of the proposed LTCH is not yet known. However, Select-Escambia conditioned approval of its CON application on the facility being located in Escambia County, and the application states that the facility will be located "proximate to the area acute care hospitals." The service area for the proposed LTCH is Escambia County and a 40-mile radius around Pensacola. The service area extends into Alabama on the west and into Santa Rosa and Okaloosa Counties on the east. It excludes Walton County. The service area is reasonable based upon the facts discussed in Part D(2)(a) below, particularly the concentration of the population and the acute care beds in Escambia County, the large elderly population in Escambia County, and the large in-migration to (and small out-migration from) Escambia County for acute care services. The bed complement at the proposed LTCH will be 35 private rooms (five of which are ICU-level), 8 semi-private rooms, and three isolation rooms (one of which is ICU-level). The facility will also include a surgical suite, a gym for physical and occupational therapy, a pharmacy, and laboratory and x-ray facilities. The total project cost is approximately $17.1 million. That cost will be funded by Select from its net cash flow from operations and through borrowings from Select’s bank. The services at the proposed LTCH will include the same “core” services found at other Select LTCHs. Those services are the treatment of pulmonary and ventilator patients, neuro-trauma and stroke patients, medically complex patients, and wound care. Select-Escambia has not negotiated patient transfer agreements with any of the area hospitals, but the CON application does include letters of support from Sacred Heart Hospital-Pensacola in Escambia County and North Okaloosa Medical Center in Okaloosa County. It is not unusual for patient transfer agreements not to have been negotiated at the CON-stage of the development of a new LTCH. The proposed LTCH was projected to open approximately two years after approval of the CON, or in November 2006. That date has been delayed as a result of this proceeding, but the two-year construction period is reasonable. The need projections in the application focus on the first two years of the facility’s operation, 2007 and 2008, as do the utilization and financial projections. Select-Escambia projects that its proposed LTCH will have 8,819 patient days in its first year of operation, and 14,054 patient days in its second year of operation. Those patient days equate to utilization rates of 45 percent in the first year and 71 percent in the second year. Those projections are reasonable and attainable. Select-Escambia projects that its proposed LTCH will generate a net loss of approximately $2.18 million in the first year of operation, and a net profit of approximately $1.19 million. Those projections are reasonable and attainable based upon the utilization projected. In addition to the letter of support from the two hospitals referenced above, the CON application includes letters of support from physicians, local politicians and businesses, the operator of rehabilitation clinics in Pensacola, and the medical director of several nursing homes in Pensacola. The letters of support attest to the general unavailability of LTCH services in Escambia County and, as discussed below, several of the letters specifically state that the traditional post-acute care settings in the area are inadequate for patients in need of long-term acute care. Statutory and Rule Criteria The statutory criteria applicable to the review of Select-Escambia’s application are in the 2004 version of Section 408.035, Florida Statutes.3 The Agency’s rules do not contain any specific criteria relating to LTCHs. The general criteria in Florida Administrative Code Rule 59C-1.008(2)(e)2. are applicable because the Agency does not publish a fixed need pool or a need methodology for LTCHs. That rule requires the applicant to demonstrate that there is a need for its proposed facility or service. Stipulated Criteria The parties’ Joint Pre-hearing Stipulation includes the following stipulations relating to the statutory criteria4: With respect to compliance with Section 408.035(3), Florida Statutes, it is agreed that Select-Escambia has the ability to provide quality programs based on the description of their programs in their CON application and based on the operational facilities of the applicant and/or of the applicant's parent facilities which are JCAHO certified. With respect to compliance with Section 408.035(4), Florida Statutes, it is agreed that Select-Escambia has the ability to provide the necessary resources including health personnel, management personnel and funds for capital operating expenditures, for project accomplishment and operation. With respect to compliance with Section 408.035(6), Florida Statutes, it is agreed that the immediate financial feasibility of the Select-Escambia project is not in dispute. It is further agreed by all parties that the long term financial feasibility of Select-Escambia is not in dispute. The parties agree that, if the projected levels are realized (i.e., need) with respect to compliance there is no disputed issue with respect to compliance with Section 408.035(7), Florida Statutes, in that the project will foster competition that promotes quality and cost effectiveness. The parties agree there are no disputed issues with respect to compliance with Section 408.035(8), Florida Statutes, which relates to an applicant's proposed costs and methods of proposed construction for the type of project proposed. The parties agree there is no disputed issues with respect to compliance with 408.035(9), Florida Statutes, as it relates to Medicaid patients in that Select's Medicaid provision (conditions - Schedule C) exceeds the state average. Section 408.035(10), Florida Statutes, is not at issue with respect to a review of the CON application filed by Select-Escambia. In light of those stipulations, the only statutory criteria still at issue are those relating to “need” –- Section 408.035(1),5 (2), and (5), Florida Statutes -- and the charity care component of Section 408.035(9), Florida Statutes. The issue of “need” was identified as the dispositive issue in this case. Mr. Gregg acknowledged in his testimony at the hearing and in his deposition that other than the issue of “need” there is no basis to deny Select-Escambia’s application. Criteria Related to “Need” The statutory criteria in Section 408.035(1), (2), and (5), Florida Statutes –- i.e., need for the proposed service; availability, quality of care, accessibility, and extent of utilization of the service in the district; and the extent to which the proposed service will enhance access in the district - encompass essentially the same factors that are enumerated in Florida Administrative Code Rule 59C-1.008(2)(e)2. Mr. Gregg testified at the hearing that where there is no LTCH in a district (as is the case in District 1), the Agency presumes that there is some amount of need for LTCH services in the district. However, Select-Escambia has the burden to demonstrate the extent of that need. Demographic, Market, etc. Factors Showing Need Each of the four counties in District 1 is relatively long and narrow. The counties extend from the Gulf of Mexico to the south and the Florida-Alabama line to the north. Escambia County is the westernmost county in District 1, and Walton County is the easternmost county in the district. Santa Rosa County is immediately to the east of Escambia County, and Okaloosa County is between Santa Rosa and Walton Counties. A 40-mile radius around Pensacola, which is the largest city in Escambia County, encompasses all of Santa Rosa County and almost all of Okaloosa County. Although much of Walton County is outside of that radius, it (and all of District 1) is within an hour and a half drive of Pensacola. Walton County is bordered on the east by Washington and Bay Counties, which are in District 2. Panama City, which currently has an LTCH, is in southern Bay County. District 1 had a population of 670,283 in July 2004, with approximately 45.6 percent of that population located in Escambia County. Approximately 13.4 percent of the July 2004 population in District 1 was in the 65+ age cohort, and 5.98 percent of that population was in the 75+ age cohort. Those percentages were lower than the statewide averages of 17.8 percent in the 65+ age cohort and nine percent in the 75+ age cohort. The population of District 1 and the percentages of the population in the 65+ and 75+ age cohorts are almost the same as the population and percentages in District 2, which has one operational (Panama City) and one approved (Tallahassee) LTCH. The population of District 1 is projected to grow approximately 6.91 percent to 716,585 by July 2009, which is five-year planning horizon applicable to this case. The five-year growth rate in District 1 is lower than the 7.93 percent rate that the state as a whole is projected to grow over the same period. However, the projected five-year growth rate in the 65+ and 75+ age cohorts, which most heavily utilize LTCH services, are higher than the statewide growth rates in those age cohorts. Specifically, the 75+ age cohort in District 1 is projected to grow 13.85 percent by July 2009, which is a higher percentage than any other health planning district in the state and nearly twice the statewide rate of 6.33 percent. The 65+ age cohort in District 1 is projected to grow 11.36 percent by July 2009, which is higher than the 9.94 percent statewide rate and higher than all but three of the other health planning districts. Walton County is projected to grow at a higher rate, both as a whole and in the 65+ and 75+ age cohorts, over the applicable five-year planning horizon than any of the other counties in District 1. The higher growth rate is due in large part to the fact that Walton County is considerably smaller than the other District 1 counties. From a raw population perspective, there will be considerably more growth in Escambia and Santa Rosa Counties than in Walton County over the applicable five-year planning horizon. The population of Walton County is expected to increase by only 7,400 persons over that period, while the population of Escambia and Santa Rosa Counties are expected to increase by almost 27,000 persons. As of December 2003, there were approximately 1,800 acute care beds in District 1 at 11 hospitals. For calendar year 2003, the district-wide average occupancy of those beds was 52.4 percent. The three largest hospitals in District 1 are located in Escambia County. Those hospitals -- Baptist Hospital, Sacred Heart Hospital-Pensacola, and West Florida Regional Medical Center -- are all similar in size and account for approximately 1,135 (or 62.6 percent) of the acute care beds in District 1. Sacred Heart Hospital-Pensacola provided a letter of support for Select-Escambia's proposed LTCH, as did two hospitals in Okaloosa County (i.e., Sacred Heart Hospital of the Emerald Coast and North Okaloosa Medical Center). The data presented in the CON application (at pages 000118 to 000121) shows that between 62.4 and 68.4 percent of the “long-stay patients” in District 1 were in the three Escambia County hospitals; that those hospitals had a relatively high (28.8 to 31.6 percent) in-migration rate of long-stay patients from outside of Escambia County; and that there is very little (1.3 to 3.6 percent) out-migration of Escambia County long-stay patients to other District 1 hospitals. Only one District 1 resident was admitted to a Florida LTCH in calendar year 2003, which is a strong indication that LTCH services are not reasonably accessible to District 1 residents even with the establishment of the Panama City LTCH in early 2003. The Panama City LTCH, which is approximately 100 miles from Pensacola, is too far away from Escambia County to be a reasonable alternative for residents of that county. The same is true for the other counties in District 1, except for Walton County which is geographically closer to Panama City than it is to Pensacola. The Panama City LTCH was not expected to serve District 1. According to the SAAR that recommended approval of that LTCH, the facility was projected to get 60 percent of its admissions from its host hospital, Bay Medical Center, and only two of the potential LTCH referrals were projected to come from a District 1 hospital. Those referrals were projected to come from Santa Rosa Medical Center in Santa Rose County, and none of the referrals to the Panama City LTCH were projected to come from Escambia County. Those projections are consistent with the experience of the Panama City LTCH since it opened in early 2003. Only five or six patients from Escambia County have been referred to the Panama City LTCH, and none have chosen to be admitted to the facility. There are no LTCHs or “like services” in District 1 because, as more fully discussed in Part C(1) above, the traditional post-acute care settings such as SNFs, CMRs, and hospital-based SNUs are not substitutes for LTCHs. The data presented in the CON application shows that in calendar year 2003 there were 500 patients treated in District 1 hospitals with LTCH-appropriate DRGs who were in the hospital for a collective 13,942 days beyond the geometric mean length of stay (GMLOS),6 which corresponds to an average of 27.9 days beyond the GMLOS. It is reasonable to expect that that those patients would have been discharged to a post-acute care setting if they no longer needed acute care, and because there were available CMR, SNU, and SNF beds in the district,7 it is reasonable to infer that the patients were still in need of long-term acute care and/or that the available post-acute care facilities did not offer the requisite level of intensive care. This inference is corroborated by the letters of support from local physicians that were included in the CON application. For example, the October 7, 2003, letter to Mr. Gregg from Dr. Donna Jacobi states that: Our skilled nursing facilities and subacute units have had difficulty in managing complex, more unstable patients One facility was equipped and staffed for ventilator patients when it opened; now that ward is for routine SNF care. Our rehabilitation institute is not the place for these patients either – they may be too ill for three hours of therapy daily. Currently some of these patients remain in acute care much longer than necessary and are subjected to iatrogenic [sic] risks, depression, and possible further decline in functional status while becoming more medically stable. Others bounce back and forth between nursing home and hospital, and a few leave our area of the state to find care elsewhere – far from their family and friends who are very important to their recovery. A LTACH [sic] would provide the opportunity for them to remain here in a supportive environment.[8] Letters of support such as Dr. Jacobi’s and those quoted in Endnote 8, with detailed information about the inability to place patients in existing facilities, are the type that the Mr. Gregg identified in Select-Marion (page 60, endnote 5) as being the most useful to the Agency in “validating” the applicant’s numeric need projections. In sum, the demographic and market conditions described above, coupled with the letters of support from local physicians and two of the acute care hospitals in District 1, support the establishment of an LTCH in the district, and more specifically, in Escambia County. Quantification of the Need / Numeric Need Select-Escambia presented two different methodologies in its application to quantify the need for LTCH beds in District 1. The methodologies are similar, but not identical to the methodology recently accepted by the Agency in Select- Marion.9 The methodologies presented in the application each define the potential patients for Select-Escambia’s proposed LTCH as the “long-stay patients” in the existing District 1 acute care hospitals with “LTCH-appropriate DRGs.” That approach is reasonable from a health planning perspective because, as discussed in Part C(1) above, an LTCH is likely the most appropriate setting for such patients from a financial and patient-care standpoint. The methodologies differ in their definition of what constitutes a “long-stay patient,” but they both use the GMLOS as the starting point, which is reasonable from a health planning perspective. Both methodologies define the “LTCH-appropriate DRGs” as the 50 DRGs that are most commonly treated in the existing Florida LTCHs. The focus on the “top 50” DRGs was reasonable from a health planning perspective because those DRGs account for more than 91 percent of the cases and 93 percent of the patient days at the existing Florida LTCHs. GMLOS+15 Methodology The first methodology presented in the application –- “the GMLOS+15 methodology” –- identified all of the patients treated in the District 1 hospitals with LTCH-appropriate DRGs whose length of stay was at least 15 days longer than the GMLOS for the DRG. A similar definition of long-stay patients was accepted by the Agency in Select-Marion. There were a total of 500 potential LTCH patients identified through Select-Escambia’s GMLOS+15 methodology. According to the data included in the CON application (at page 000120), 30 of those patients were Walton County residents and 55 resided outside of District 1. Select-Escambia calculated a total of 19,409 potential LTCH patient days that would be generated by the 500 identified long-stay patients, which equates to an average daily census (ADC) of 53. According to Select-Escambia's health planner (Transcript, at 131), the 19,409 patient-days included all of the days in the patient’s hospital stay as potential LTCH patient days, and not just that portion of the stay that exceeded the GMLOS. The inclusion of all of the days in the patient’s hospital stay as potential LTCH patient days is not reasonable because the vast majority of LTCH patients are transferred from an acute care hospital at some point during the patient’s hospital stay, typically at or after the GMLOS. The effect of including all of the days in the patient’s hospital stay as potential LTCH patient days rather than just the days after the GMLOS is an overstatement of the potential LTCH patient days and the ADC calculated under the GMLOS+15 methodology in Select-Escambia’s application. If only the days beyond the GMLOS were included (as was done in Select-Marion), the result would be 13,941 potential LTCH patient days. If the 875 days attributable to Walton County residents and the 1,596 days attributable to non-District 1 residents were excluded (see Exhibit P2, at 000121), then the total would be 11,471 potential LTCH patient days. The ADC of 53 calculated by Select-Marion under the GMLOS+15 methodology is not reliable because it was based upon the 19,409 patient days. Using the 13,941 or 11,471 patient days referenced above would result in an ADC of 38.2 or 31.4, respectively. Based upon an 80 occupancy standard, those ADCs would translate into a projected need for 40 to 48 LTCH beds in District 1. If a 75 percent occupancy standard was used, the projected LTCH bed need would be 42 to 51 beds. The lower numbers in each of those ranges reflect the exclusion of the patient days attributable to Walton County residents and non- District 1 residents; the higher numbers in those ranges reflect the inclusion of those residents. An 80 percent occupancy standard was accepted by the Agency in Select-Marion and was also used by Select in Select- Sarasota. As stated in the Recommended Order in Select-Marion (at page 37), the 80 percent occupancy standard “better reflects the lower bed turn-over by LTCH patients than does the 75 percent occupancy standard typically applied to traditional, ‘short-term’ acute care hospitals.” GMLOS+7 Methodology The second methodology presented in the application - – “the GMLOS+7 methodology” –- uses a broader definition to identify the potential LTCH patients in District 1. It includes all of the patients with LTCH-appropriate DRGs who were treated in the District 1 hospitals and whose lengths of stay were at least seven days longer than the GMLOS. The broader definition of long-stay patients in the GMLOS+7 methodology resulted in 1,498 potential LTCH patients (see Exhibit P2, at 000117 (Table 1-16(b)), 000120), as compared to the 500 potential LTCH patients identified through the GMLOS+15 methodology. The Agency did not expressly take issue with the broader definition used in the GMLOS+7 methodology to identify the potential LTCH patients, and it cannot be said based upon the record evidence in this case that the definition is inherently unreasonable. In calculating the potential LTCH patient days under the GMLOS+7 methodology, Select-Escambia only included the days that the patient stayed in the hospital beyond the GMLOS, which are referred to in the application as “excess days.” See Transcript, at 132. A similar approach was used in the methodology accepted by the Agency in Select-Marion. The following table, which is derived from the data in Table 1-16(a) in the CON application, summarizes the number of excess days generated by patients in the District 1 hospitals based upon the patient’s county of residence: Escambia County 11,434 Okaloosa County 5,634 Santa Rosa County 3,194 Subtotal: District 1 Residents except for Walton County 20,262 Walton County 1,410 Subtotal: All District 1 residents 21,672 Outside of District 1 2,340 Total 24,012 Select-Escambia then converted the excess days into “forecasted LTCH cases” by dividing the most conservative figure –- the 20,262 days, which excluded Walton County residents and non-District 1 residents -- by the 33.6 ALOS at Select’s existing freestanding LTCHs. The result –- 603 cases –- was then inflated based upon the projected growth rate in District 1 to determine the number of forecasted LTCH cases in 2007 and 2008, which were projected to be the first two years of operation for Select-Escambia’s proposed LTCH. The forecasted cases were then converted into “forecasted LTCH days” by multiplying the number of cases by the same 33.6 ALOS. The conversion of the excess days into forecasted LTCH cases and then back into forecasted LTCH days based upon a 33.6 ALOS is not reasonable because, according to the CON application,10 the initial calculation of the excess days is intended to reflect the number of days that patients would likely spend in the LTCH rather than the short-term acute care hospitals in District 1 if an LTCH was available in the area. The ALOS experienced by Select at its other facilities is irrelevant to that issue. The effect of the conversion step in Select- Escambia’s GMLOS+7 methodology is an overstatment of the forecasted LTCH patient days, as can be seen through a comparison of the data in Tables 1-16(a) and 1-16(b) in the CON application. Table 1-16(b) shows the number of cases associated with the excess days calculated in Table 1-16(a). The 1,498 total cases identified on Table 1-16(b) correlate to the 24,012 total excess days identified on Table 1-16(a). As a result, there is an average of only 16.03 excess days per case. Stated another way, the long-stay patients identified through the GMLOS+7 methodology are staying in the hospital an average of 16.03 days longer than the GMLOS. It is those 16.03 days/case that make up the potential LTCH patient days, but the conversion described above appears to assume that those same patients would stay in Select-Escambia’s proposed LTCH for 33.6 days. There is no logic or reason to that assumption, and as a result, the patient days, ADC, and bed need reflected in Table 1-17 of the application are not reliable. The most reliable projection of bed need that can be calculated based upon the data presented in connection with the GMLOS+7 methodology is derived from the Excess Table 1-16(a), to wit: Bed Need Days ADC (at 80%) Escambia only 11,434 31.3 40 District 1 excluding Walton and non-District 1 20,262 55.5 70 District 1 including Walton; excluding non- District 1 21,672 59.4 75 Accordingly, the GMLOS+7 methodology projects a need for 70 to 75 LTCH beds, depending upon whether Walton County residents are included in the calculation, with 40 of the beds attributable to the excess days generated by Escambia County residents alone. Ultimate Findings Regarding Numeric Need Using the most conservative figures produced by the respective need methodologies presented in the application, there is a need for between 40 (see Finding of Fact 107) and 70 (see Findings of Fact 119 and 120) LTCH beds in District 1. It is reasonable to expect that the “actual” bed need is towards the mid-point of that range -- 55 beds -- because Select-Escambia’s proposed LTCH will likely get some of the potential LTCH admissions from Walton County, as well as some of the potential LTCH admissions from outside of District 1; because as many as seven percent of the facility's patient days will be attributable to patients whose diagnoses are not within the “top 50” DRGs used in the methodologies to identify the potential LTCH patients; and because the methodologies and the fiqures reflected in the preceeding paragraphs do not take into account the growth in admissions and patient days between 2003 (the period used in the methodologies) and 2007 (when Select- Escambia's proposed LTCH is projected to open) that is expected as the population of District 1 grows, particularly in the 65+ and 75+ age cohorts. Accordingly, the preponderance of the evidence establishes that there is a numeric need for the 54 LTCH beds proposed by Select-Escambia. Other Disputed Criteria Section 408.035(9), Florida Statutes, requires consideration of the “applicant's past and proposed provision of health care services to Medicaid patients and the medically indigent.” The statutory reference to “the medically indigent” encompasses what are typically referred to as charity patients. Select-Escambia conditioned the approval of its CON application on the provision of two percent of the patient days at its proposed LTCH to Medicaid patients and 0.8 percent of the patient days to charity patients. It was stipulated that Select-Escambia’s commitment to Medicaid patients exceeds the statewide average for LTCHs, which according to the SAAR is 1.24 percent of patient days. Select-Escambia’s commitment to charity patients is slightly lower than the statewide average for LTCHs, which is 0.94 percent of patient days.11 When viewed collectively, Select-Escambia’s commitment to Medicaid and charity patients -- 2.8 percent of patient days -- exceeds the statewide average for LTCHs of 2.18 percent of patient days. The commitments to Medicaid and charity patients in Select-Escambia’s CON application were based upon Select’s experience at its other LTCHs, and they are reasonable and attainable in District 1. The fact that Select-Escambia’s commitment to charity patients is slightly lower than the statewide average for LTCHs is not significant under the circumstances of this case. Indeed, Mr. Gregg conceded at the hearing that it is not an independent basis to deny Select-Escambia’s application, and that the Agency will accept Select-Escambia’s proposed charity commitment of 0.8 percent of patient days if the CON is ultimately approved.
Recommendation Based upon the foregoing findings of fact and conclusions of law, it is RECOMMENDED that the Agency issue a final order approving Select-Escambia’s application, CON 9800. DONE AND ENTERED this 17th day of June, 2005, in Tallahassee, Leon County, Florida. S T. KENT WETHERELL, II 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 17th day of June, 2005.
The Issue Whether the application of Petitioner Naples Community Hospital, Inc. for a Certificate of Need to add a total of 35 beds to Naples Community Hospital and North Collier Community Hospital should be approved based on peak seasonal demand for acute care beds in the relevant subdistrict.
Findings Of Fact Naples Community Hospital, Inc., ("NCH") holds the license for and operates Naples Community Hospital ("Naples"), a 331 bed not-for-profit acute care hospital, and North Collier Community Hospital ("North Collier"), a 50 bed acute care hospital. NCH also operates a 22 bed comprehensive rehabilitation facility and a 23 bed psychiatric facility. NCH is owned by Community Health Care, Inc., "(CHC"). Both Naples and North Collier are located within Agency for Health Care Administration ("ACHA") district 8 and are the only hospitals within subdistrict 2 of the district. Naples is located in central Collier County. North Collier is (as the name implies) located in northern Collier County approximately 2-3 miles from the county line. NCH's primary service area is Collier County from which approximately 85-90 percent of its patients come, with a secondary service area extending north into Lee County. Neither Naples nor North Collier are teaching hospitals as defined by Section 407.002(27), Florida Statutes (1991). NCH is not proposing a joint venture in this CON application. NCH has a record of providing health care services to Medicaid patients and the medically indigent. NCH proposes to provide health care services to Medicaid patients and the medically indigent. Neither Naples nor North Collier are currently designated by the Office of Medicaid as disproportionate share providers. NCH has the funds for capital and initial operating expenditures for the project. NCH has sufficient financial resources to construct and equip the proposed project. The costs and methods of the proposed construction are reasonable. The Agency for Health Care Administration ("AHCA") is the state agency charged with responsibility for administering the Certificate of Need program. Southwest Florida Regional Medical Center ("Southwest") is a 400 bed for-profit acute care hospital located in Fort Myers, Lee County. Lee County is adjacent to and north of Collier County. Southwest is owned by Columbia Hospital Corporation ("Columbia"), which also owns Gulf Coast Hospital in Fort Myers, and two additional hospitals in AHCA District 8. Southwest's primary service area is Lee County. Although Southwest asserts that it would be negatively impacted by the addition of acute care beds at NCH, the greater weight of the credible evidence fails to support the assertion. The primary market services areas of NCH and Southwest are essentially distinct. However, the facilities are located in such proximity as to indicate that secondary service areas overlap and that, at least during peak winter season periods, approval of the NCH application could potentially impact Southwest's operations. Southwest has standing to participate in this proceeding. Southwest offered evidence to establish that it would be substantially affected by approval of the NCH application. The NCH length-of-stay identified in the Southwest documents is inaccurate and under-reports actual length-of-stay statistics. The documentation also includes demographic information from a zip code (33912) which contributes an insignificant portion of NCH patients, and relies on only two years of data in support of the assertion that utilization in the NCH service area is declining. Southwest's chief operating officer testified that he considers Gulf Coast Hospital, another Columbia-owned facility, to offer more competition to Southwest that does NCH. Further, a physician must have admitting privileges at a hospital before she can admit patients to the facility. Of the physicians holding admitting privileges at Southwest, only two, both cardiologists, also have admitting privileges at NCH. Contrary to Southwest, NCH does not have an open heart surgery program. Accordingly, at least as to physician-admitted patients, approval of the NCH application would likely have little impact. On August 26, 1991, NCH submitted to AHCA a letter of intent indicating that NCH would file a Certificate of Need ("CON") application in the September 26, 1991 batching cycle for the addition of 35 acute care beds to the Naples and North Collier facilities. The letter of intent did not specify how the additional beds would be divided between the two facilities. The determination of the number of beds for which NCH would apply was solely based on the fact that the applicant had 35 observation beds which could be readily converted to acute care beds. The observation beds NCH proposes to convert are equipped identically to the acute care beds at NCH and are currently staffed. The costs involved in such conversion are minimal and relatively insignificant. Included with the letter of intent was a certified corporate resolution which states that on July 24, 1991, the NCH Board of Trustees authorized the filing of an application for the additional beds, authorized NCH to incur related expenses, stated that NCH would accomplish the proposed project within time and budget allowances set forth in the application, and that NCH would license and operate the facility. By certification executed August 7, 1991, the NCH secretary certified that the resolution was enacted at the July 24, 1991 board meeting and that the resolution did not contravene the NCH articles of incorporation or bylaws. Article X, Sections 10.1 and 10.1.3 of the NCH bylaws provides that no CON application shall be legally effective without the written approval of CHC. On September 26, 1991, NCH filed an application for CON No. 6797 proposing to add 31 acute care beds to Naples and 4 acute care beds to North Collier. The CON application included a copy of the NCH board resolution and certification which had been previously submitted with the letter of intent as well as the appropriate filing fee. NCH published appropriate public notice of the application's filing. As of the date of the CON application's filing, CHC had not issued written approval of the CON application prior to the action of the NCH Board of Directors and the filing of the letter of intent or the application. On October 2, 1992, four days prior to the administrative hearing in this case, the board of CHC ratified the actions of NCH as to the application for CON at issue in this case. The CHC board has previously ratified actions of the NCH in such fashion. There is uncontroverted testimony that the CHC board was aware of the NCH application and that no reservation was expressed by any CHC board member regarding the CON application. Although NCH's filing of the CON application without appropriate authorization from its parent company appears to be in violation of the NCH bylaws, such does not violate the rules of the AHCA. There is no evidence that the AHCA requested written authorization from the CHC board. After review of the application, the AHCA identified certain deficiencies in the application and notified NCH, which apparently rectified the deficiencies. The AHCA deemed the application complete on November 8, 1991. As required by statute, NCH included a list of capital projects as part of the CON application. The list of capital projects attached to the application was incomplete. The capital projects list failed to identify approximate expenditures of $370,000 to construct a patio enclosure, $750,000 to install an interim sprinkler system, $110,000 to construct emergency room triage space, and $125,000 to complete electrical system renovations. At hearing, witnesses for NCH attempted to clarify the omissions from the capital projects list. The witnesses claimed that such omitted projects were actually included within projects which were identified on the list. When identifying the listed projects within which the omitted projects were supposedly included, the witnesses testified inconsistently. For example, one witness testified that the patio project was included in the emergency room expansion project listed in the application. Another witness claimed that the patio enclosure was included in an equipment purchase category. Based on the testimony, it is more likely that the patio enclosure was neither a part of an emergency room expansion nor equipment purchase, but was a separate construction project which was omitted from the CON application. Similarly inconsistent explanations were offered for the other projects which were omitted from the capital projects list. The testimony was not credible. The capital projects omitted from the list do not affect the ability of NCH to implement the CON sought in this proceeding. The parties stipulated to the fact the NCH has sufficient financial resources to construct and equip the proposed project. As part of the CON application, NCH was required to submit a pro forma income statement for the time period during which the bed additions would take place. The application failed to include a pro forma statement for the appropriate time period. Based on the stipulation of the parties that the costs and methods of the proposed construction are reasonable, and that NCH has adequate resources to fund the project, the failure to include the relevant pro forma is immaterial. Pursuant to applicable methodology, the AHCA calculates numeric acute care bed need projections for each subdistrict's specific planning period. Accordingly, the AHCA calculated the need for additional acute care beds in district 8, subdistrict 2 for the July, 1996 planning horizon. The results of the calculation are published by the agency. The unchallenged, published fixed need pool for the planning horizon at issue in this proceeding indicated that there was no numeric need for additional acute care beds in district 8, subdistrict 2, Collier County, Florida, pursuant to the numeric need methodology under Rule 59C-1.038 Florida Administrative Code. The CON application filed by NCH is based on the peak seasonal demand experienced by hospitals in the area during the winter months, due to part-time residents. NCH asserts that the utilization of acute care beds during the winter months (January through April) results in occupancy levels in excess of 75 percent and justifies the addition of acute care beds, notwithstanding the numerical need determination. Approval of the CON application is not justified by the facts in this case. The AHCA's acute care bed need methodology accounts for high seasonal demand in certain subdistricts in a manner which provides that facilities have bed space adequate to accommodate peak demand. The calculation which requires that the average annual occupancy level exceed 75 percent reflects AHCA consideration of occupancy levels which rise and fall with seasonal population shifts. The applicant has not challenged the methodology employed by the AHCA in projecting need. Peak seasonal acute care bed demand may justify approval of a CON application seeking additional beds if the lack of available beds poses a credible threat of potentially negative impact on patient outcomes. The peak seasonal demand experienced by NCH has not adversely affected patient care and there is insufficient evidence to establish that, at this time, such peak demand poses a credible threat of potential negative impact on patient outcomes in the foreseeable future. There is no dispute regarding the existing quality of care at Naples, North Collier, Southwest or any other acute care hospital in district 8. The parties stipulated that NCH has the ability to provide quality of care and a record of providing quality of care. In this case, the applicant is seeking to convert existing beds from a classification of "observation" to "acute care". The observation beds NCH proposes to convert are equipped identically to the acute care beds at NCH. Approval of the CON application would result in no net increase in the number of licensed beds. NCH offered anecdotal evidence suggesting that delays in transferring patients from the Naples emergency room to acute care beds (a "logjam") was caused by peak seasonal occupancy rates. There was no evidence offered as to the situation at the North Collier emergency room. The anecdotal evidence is insufficient to establish that "logjams" (if they occur at all) are related to an inadequate number of beds identified as "acute care" at NCH facilities. There are other factors which can result in delays in moving patients from emergency rooms to acute care beds, including facility discharge patterns, delays in obtaining medical test results and staffing practices. NCH asserted at hearing that physicians who refer patients to NCH facilities will not refer such patients to other facilities. The evidence fails to establish that such physician practice is reasonable or provides justification for approval of CON applications under "not normal" circumstances and further fails to establish that conditions at NCH are such as to result in physicians attempting to locate other facilities in which to admit patients. The rule governing approval of acute care beds provides that, prior to such approval, the annual occupancy rate for acute care beds in the subdistrict or for the specific provider, must exceed 75 percent. This requirement has not been met. Applicable statutes require that, in considering applications for CON's, the AHCA consider accessibility of existing providers. The AHCA- established standard provides that acute care bed accessibility requirements are met when at least 90 percent of the residents in an urban subdistrict are within a 30 minute automobile trip to such facilities. At least 90 percent of Naples residents are presently within a 30 minute travel time to NCH acute care beds. The number of acute care beds in the subdistrict substantially exceed the demand for such beds. Additional beds would result in inefficient utilization of existing beds, would further increase the current oversupply of beds, would delay the time at which need for additional beds may be determined and, as such, would prevent competing facilities from applying for and receiving approval for such beds. The financial feasibility projections set forth in the CON application rely on assumptions as to need and utilization projections which are not supported by the greater weight of the evidence and are not credited. Accordingly, the evidence fails to establish that the addition of 35 acute care beds to NCH facilities is financially feasible in the long term or that the income projections set forth in the CON application are reasonable. As to projections related to staffing requirements and costs, the beds are existing and are currently staffed on a daily, shift-by-shift basis, based on patient census and acuity of illness. There is reason to believe that the staffing patterns will remain fairly constant and accordingly the projections, based on historical data, are reasonable. Generally stated, where there is no numeric or "not normal" need for the proposed addition of 35 acute care beds in the relevant subdistrict, it could be predicted that the addition of acute care beds would exacerbate the oversupply of available beds and could cause a slight reduction in the occupancy levels experienced by other providers. In this case, the market service areas are sufficiently distinct as to suggest that such would not necessarily be the result. However, based on the lack of need justifying approval of the CON application under any existing circumstances, it is unnecessary to address in detail the impact on existing providers. The state and district health plans identify a number of preferences which should be considered in determining whether a CON application should be approved. The plans suggest that such preferences are to be considered when competing CON applications are reviewed. In this case there is no competing application and the applicability of the preferences is unclear. However, in any event, application of the preferences to this proposal fail to support approval of the application.
Recommendation RECOMMENDED that a Final Order be entered DENYING the application of Naples Community Hospital, Inc., for Certificate of Need 6797. DONE and RECOMMENDED this 19th day of March, 1993 in Tallahassee, Florida. WILLIAM F. QUATTLEBAUM 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 19th day of March, 1993. APPENDIX TO RECOMMENDED ORDER, CASE NO. 92-1510 To comply with the requirements of Section 120.59(2), Florida Statutes, the following constitute rulings on proposed findings of facts submitted by the parties. Petitioner The Petitioner's proposed findings of fact are accepted as modified and incorporated in the Recommended Order except as follows: 3-4, 6-8, 16-20, 29-36, 38, 41, 44, 47, 49-61, 80, 88, 95-96, 100, 104, 108, 117-119, 122-125, 127, 134-138. Rejected as unnecessary. 15. Rejected as irrelevant. Peak seasonal demand is accounted for by the numeric need determination methodology. There is no credible evidence which supports a calculation of three years of four month winter occupancy to reach a 12 month average occupancy rate. 21-27, 37, 42-43, 62-64, 66, 97, 99, 101-103, 105-107, 109, 120-121, 126. Rejected as not supported by the greater weight of credible and persuasive evidence. 28. Rejected as not supported by the greater weight of credible and persuasive evidence and contrary to the stipulation filed by the parties. Rejected as not supported by greater weight of credible and persuasive evidence which fails to establish that the transfer of patients from emergency room to acute care beds is delayed due to numerical availability of beds. Rejected as not supported by greater weight of credible and persuasive evidence which fails to establish that the alleged lack of acute care beds is based on insufficient number of total beds as opposed to other factors which affect bed availability. Rejected as immaterial and contrary to the greater weight of the evidence Rejected as immaterial and contrary to the greater weight of the evidence which fails to establish reasonableness of considering only a four month period under "not normal" circumstances where the period and the peak seasonal demand are included within the averages utilized to project bed need. 86. Rejected as cumulative. 114. Rejected as unsupported hearsay. Respondent/Intervenor The Respondent and Intervenor filed a joint proposed recommended order. The proposed order's findings of fact are accepted as modified and incorporated in the Recommended Order except as follows: 6, 45, 51, 53, 59-67, 69-70, 94-113. Rejected as unnecessary. 16. Rejected as to use of term "false", conclusion of law. 58. Rejected as not clearly supported by credible evidence. 71-93, 114-124. Rejected as cumulative. COPIES FURNISHED: Douglas M. Cook, Director Agency for Health Care Administration 2727 Mahan Drive Tallahassee, Florida 32308 Sam Power, Agency Clerk Agency for Health Care Administration The Atrium, Suite 301 325 John Knox Road Tallahassee, Florida 32303 Harold D. Lewis, Esquire Agency for Health Care Administration The Atrium, Suite 301 325 John Knox Road Tallahassee, Florida 32303 W. David Watkins, Esquire Oertel, Hoffman, Fernandez, & Cole Post Office Box 6507 Tallahassee, Florida 32314-6507 Edward G. Labrador, Esquire Thomas Cooper, Esquire Agency for Health Care Administration 2727 Mahan Drive Tallahassee, Florida 32308 John D.C. Newton, II, Esquire Aurell, Radey, Hinkle, Thomas & Beranek Monroe Park Tower, Suite 1000 101 North Monroe Street Post Office Drawer 11307 Tallahassee, Florida 32302
