This is the accessible text file for GAO report number GAO-04-709 entitled 'Comprehensive Outpatient Rehabilitation Facilities: High Medicare Payments in Florida Raise Program Integrity Concerns' which was released on September 13, 2004. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Chairman, Committee on Finance, U.S. Senate: United States Government Accountability Office: GAO: August 2004: Comprehensive Outpatient Rehabilitation Facilities: High Medicare Payments in Florida Raise Program Integrity Concerns: GAO-04-709: GAO Highlights: Highlights of GAO-04-709, a report to the Chairman, Committee on Finance, U.S. Senate: Why GAO Did This Study: Comprehensive Outpatient Rehabilitation Facilities (CORF) are highly concentrated in Florida. These facilities, which provide physical therapy, occupational therapy, speech-language pathology services, and other related services, have been promoted as lucrative business opportunities for investors. Aware of such promotions, you raised concerns about whether Medicare could be vulnerable to overbilling for CORF services. In this report, focusing our review on Florida, we (1) compared Medicare’s outpatient therapy payments to CORFs in 2002 with its payments that year to other facility-based outpatient therapy providers and (2) assessed the program’s effectiveness in ensuring that payments to CORFs complied with Medicare rules. What GAO Found: In Florida, CORFs were by far the most expensive type of outpatient therapy provider in the Medicare program in 2002. Per-patient payments to CORFs for therapy services were 2 to 3 times higher than payments to other types of facility-based therapy providers. Higher therapy payments were largely due to the higher volume of services—more visits or more intensive therapy per visit—delivered to CORF patients. This pattern of relatively high CORF payments was evident in each of the eight metropolitan statistical areas (MSA) of the state where nearly all Florida CORFs operated and the vast majority of CORF patients were treated. A consistent pattern of high payments and service levels was also evident for patients in each of the diagnosis categories most commonly treated by CORFs. Differences in patient characteristics—age, sex, disability, and prior inpatient hospitalization—did not explain the higher payments that Florida CORFs received compared to other types of outpatient therapy providers. Steps taken by Medicare’s claims administration contractor for Florida have not been sufficient to mitigate the risk of improper billing by CORFs. After examining state and national trends in payments to CORFs in 1999, the contractor increased its scrutiny of CORF claims to ensure that Medicare payments made to CORFs were appropriate. It found widespread billing irregularities in Florida CORF claims, including high rates of medically unnecessary therapy services. Since late 2001, the contractor has intensified its review of claims from new CORF providers and required medical documentation to support certain CORF services considered at high risk for billing errors. It has also required that supporting medical records documentation be submitted with all CORF claims for about 650 beneficiaries who had previously been identified as receiving medically unnecessary services. The contractor’s analysis of 2002 claims data for this limited group of beneficiaries suggests that, as a result of these oversight efforts, Florida CORFs billed Medicare for substantially fewer therapy services than in previous years. However, our analysis of all CORF therapy claims for that year indicates that the contractor’s program safeguards were not completely effective in controlling per-patient payments to CORFs statewide. With oversight focused on a small fraction of CORF patients, CORF facilities continued to provide high levels of services to beneficiaries whose claims were not targeted by the contractor’s intensified reviews. What GAO Recommends: GAO recommends that CMS direct the Florida contractor to medically review a larger number of CORF claims. While CMS agreed with our findings, it noted that the contractor is already taking appropriate steps to monitor CORF claims. However, given that CORFs continued to bill significantly more per beneficiary than other outpatient therapy providers under the current level of scrutiny, we maintain that enlarging the number of CORF claims reviewed would promote compliance in this vulnerable area. www.gao.gov/cgi-bin/getrpt?GAO-04-709. To view the full product, including the scope and methodology, click on the link above. For more information, contact Leslie G. Aronovitz at (312) 220-7600. [End of section] Contents: Letter: Results in Brief: Background: Florida CORFs Received Higher Average Therapy Payments, Despite Treating Similar Patients: Actions by the Florida Contractor Were Not Sufficient to Ensure Appropriate Payments to CORFs: Conclusions: Recommendation: Agency Comments and Our Evaluation: Appendix I: Objectives, Scope, and Methodology: Appendix II: Comments from the Centers for Medicare & Medicaid Services: Appendix III: GAO Contact and Staff Acknowledgments: GAO Contact: Acknowledgments: Tables: Table 1: Therapy Payments and Units of Service Per Patient by Type of Provider, Florida, 2002: Table 2: Therapy Payments Per Patient by Type of Provider in Selected Florida MSAs, 2002: Table 3: Therapy Payments Per Patient by Provider Type for Selected Diagnosis Categories, Florida, 2002: Table 4: Units of Therapy Service Per Patient by Provider Type for Selected Diagnosis Categories, Florida, 2002: Table 5: Medicare Payments to Selected Florida CORFs for All Types of Services, 2002: Abbreviations: CMS: Centers for Medicare & Medicaid Services: CORF: Comprehensive Outpatient Rehabilitation Facility: CWF: Common Working File: HHS OIG: Department of Health and Human Services Office of Inspector General: MSA: metropolitan statistical area: NCH: National Claims History File OPD: outpatient department: PIP-DCG: Principal Inpatient Diagnostic Cost Group: SNF: skilled nursing facility: United States Government Accountability Office: Washington, DC 20548: August 12, 2004: The Honorable Charles E. Grassley: Chairman: Committee on Finance: United States Senate: Dear Mr. Chairman: Outpatient therapy services are a covered benefit under Medicare--the federal program that finances health services for approximately 40 million elderly and disabled individuals. Each year, about 9 percent of Medicare beneficiaries use outpatient therapy services--defined by the Medicare program as physical therapy, occupational therapy, and speech- language pathology services--to improve mobility and functioning.[Footnote 1] To qualify for coverage of outpatient therapy services under Medicare, beneficiaries must be referred by a physician, have a written treatment plan that is reviewed periodically by a physician, and need therapy for rehabilitation rather than maintenance purposes.[Footnote 2] Several types of facility-based providers offer outpatient therapy services, including outpatient departments (OPD) at hospitals and skilled nursing facilities (SNF), and rehabilitation agencies. These providers deliver services in ambulatory settings such as clinics and community hospital outpatient departments. In many states, therapy services are also available through Comprehensive Outpatient Rehabilitation Facilities (CORF). In 1980, Congress recognized CORFs as potential Medicare participating providers to allow beneficiaries access to both physician and therapy services in one stand-alone outpatient facility.[Footnote 3] CORFs are different from other types of Medicare-certified outpatient therapy providers in that, in addition to physical therapy, regulations require that they offer psychological or social services and the services of a physician who specializes in rehabilitation medicine.[Footnote 4] They are also unique in their authority to provide a variety of nontherapy services- -such as respiratory treatment or nursing care--as medically necessary in the context of a patient's rehabilitation therapy treatment plan.[Footnote 5] In general, services must be provided on the CORF premises at a single, fixed location. However, physical therapy, occupational therapy, and speech-language pathology services may be provided in places other than the CORF's main location, such as in a patient's home. Back disorders, arthritis, soft tissue injuries (such as joint sprains and strains), and neurologic disorders (such as concussion) are common conditions treated at CORFs. In recent years, CORF marketing consultants have actively promoted the establishment of CORFs as lucrative business opportunities for investors. For example, one consultant's marketing materials stated that "every new CORF office is expected to pre-tax net at least $400,000 to $500,000 after a start-up period. . . With or without any medical background, you can own a small medical facility." Aware of such promotions, you raised concerns about whether Medicare could be vulnerable to overbilling for CORF services. In fact, in 2000, the Department of Health and Human Services Office of Inspector General (HHS OIG) reported a high level of improper billing by outpatient therapy providers in several states.[Footnote 6] In this report, we (1) compared Medicare's outpatient therapy payments to CORFs in 2002 with its payments that year to other facility-based outpatient therapy providers and (2) assessed the program's effectiveness in ensuring that payments to CORFs complied with Medicare rules. As agreed with your staff, we focused our review on Florida providers and Medicare's Florida claims administration contractor because, with nearly 200 CORFs in operation at the end of 2002, that state had one-third of the nation's CORFs and far more of these facilities than any other state. To address these issues, we analyzed Medicare claims data for services provided in 2002 (the most current data available) by CORFs, rehabilitation agencies, hospital OPDs, and SNF OPDs. We also interviewed officials at the Centers for Medicare & Medicaid Services (CMS)--the federal agency that oversees the Medicare program--the Florida contractor responsible for processing and paying Medicare's CORF claims, federal law enforcement agencies, and therapy industry experts. In addition, we reviewed relevant investigative reports by the HHS OIG and the Florida claims administration contractor on the improper billing activities of some CORFs. (For a detailed description of our methodology and procedures we followed for evaluating the reliability of the data we used, see app. I.) This work was performed from May 2003 through July 2004 in accordance with generally accepted government auditing standards. Results in Brief: In Florida, CORFs were by far the most expensive type of outpatient therapy provider in the Medicare program in 2002. Per-patient payments to CORFs for therapy services were 2 to 3 times higher than payments to other types of facility-based therapy providers. Higher therapy payments were largely due to the higher volume of services--more visits or more intensive therapy per visit--delivered to CORF patients. This pattern of relatively high CORF payments was evident in each of the eight metropolitan statistical areas (MSA) of the state where nearly all Florida CORFs operated and the vast majority of CORF patients were treated. A consistent pattern of high payments and service levels was also evident for patients in each of the diagnosis categories most commonly treated by CORFs. Differences in patient characteristics--age, sex, disability, and prior inpatient hospitalization--did not explain the higher payments that Florida CORFs received compared to other outpatient therapy provider types. Steps taken by Medicare's claims administration contractor for Florida have not been sufficient to mitigate the risk of improper billing by CORFs. After examining state and national trends in payments to CORFs in 1999, the contractor increased its scrutiny of CORF claims to ensure that Medicare payments made to CORFs were appropriate. It found widespread billing irregularities in Florida CORF claims, including high rates of medically unnecessary therapy services. Since late 2001, the contractor has intensified its review of claims from new CORF providers and required medical documentation to support certain CORF services considered at high risk for billing errors. It has also required that supporting medical records documentation be submitted with all CORF claims for about 650 beneficiaries who had previously been identified as receiving medically unnecessary services. The contractor's analysis of 2002 claims data for this limited group of beneficiaries suggests that, as a result of these oversight efforts, Florida CORFs billed Medicare for substantially fewer therapy services than in previous years. However, our analysis of all CORF therapy claims for that year indicates that the contractor's program safeguards were not completely effective in controlling per-patient payments to CORFs statewide. With oversight focused on a small fraction of CORF patients, CORF facilities continued to provide high levels of services to beneficiaries whose claims were not targeted by the contractor's intensified reviews. We recommend that CMS direct the Florida claims administration contractor to medically review a larger number of CORF claims. In commenting on a draft of this report, CMS agreed with our findings but noted that contractors have limited resources for medical review. The agency also stated that the Florida claims administration contractor is already taking appropriate steps to address concerns about CORF billing. We recognize that contractors can achieve efficiencies by targeting their medical review activities on areas where the financial risk to Medicare is greatest. However, the impact of medical review comes, in part, from the sentinel effect of consistently applying medical review to providers' claims. Given that Florida CORFs continued to bill significantly more per beneficiary than other outpatient therapy providers even after the contractor took steps to examine some claims, we maintain that the Florida contractor could enhance compliance in this area of program vulnerability by enlarging the number of CORF claims reviewed. Background: Medicare Coverage Rules for Outpatient Therapy: All outpatient therapy providers are subject to Medicare part B payment and coverage rules.[Footnote 7] Payment amounts for each type of outpatient therapy service are based on the part B physician fee schedule.[Footnote 8] In 2000, Medicare paid approximately $2.1 billion for all outpatient therapy services, of which $87.1 million was paid to CORFs. To meet Medicare reimbursement requirements, outpatient therapy services must be: * appropriate for the patient's condition, * expected to improve the patient's condition, * reasonable in amount, frequency, and duration, * furnished by a skilled professional, * provided with a physician available on call to furnish emergency medical care, and: * part of a written treatment program that is reviewed periodically by a physician. CMS relies on its claims administration contractors to monitor provider compliance with program requirements. Contractors regularly examine claims data to identify billing patterns by specific providers or for particular services that are substantially different from the norm. Claims submitted by these groups of providers--or for specific services--are then selected for additional scrutiny. Whether such reviews occur prior to payment (prepayment reviews) or after claims have been paid (postpayment reviews), the provider is generally required to submit patient records to support the medical necessity of the services billed. This routine oversight may lead to additional claim reviews or provider education about Medicare coverage or billing issues. Florida CORF Industry: With 567 facilities nationwide at the end of 2002, the CORF industry is relatively small. Although CORFs operated in 41 states at the end of 2002, the industry is highly concentrated in Florida, where 191 (one- third) of all Medicare-certified CORFs are located. By contrast, the state with the second largest number of CORFs at the end of 2002 was Texas, with 53 CORFs. The number of CORF facilities in Florida grew about 30 percent during 2002 and the industry is now largely composed of relatively new, for- profit providers. The CORF industry in Florida continued to grow in 2003, reaching 220 facilities by year's end, of which 96 percent were for profit. The growth in Florida CORFs came after a period of substantial turnover among CORF owners (many closures and new entrants).[Footnote 9] From 1999 to 2002, Medicare payments to Florida CORFs rose substantially and far outpaced growth in the number of beneficiaries that used CORFs. The number of Medicare beneficiaries receiving services from CORFs grew 13 percent, increasing from 33,653 in 1999 to 38,024 in 2002. However, during the same time period, Medicare expenditures for services billed by CORFs rose significantly, with total payments increasing 61 percent, from $48.1 million to $77.4 million. Half of all Florida CORFs received an annual payment of $91,693 or more from Medicare in 1999; by 2002, the median annual payment more than doubled to $187,680.[Footnote 10] Although CORFs were added to the Medicare program to offer beneficiaries a wide range of nontherapy services at the same location where they receive therapy, most Florida CORFs do not provide these types of services. For those that do, only a small proportion of Medicare payments are accounted for by these services. In 2002, 98 percent of Medicare payments to Florida CORFs went to furnish physical and occupational therapy or speech-language pathology services. The mix of services reimbursed by Medicare was very different in 1999, when such therapy accounted for 68 percent of all payments, and the remainder paid for nontherapy services, such as pulmonary treatments and psychiatric care.[Footnote 11] In recent years, payments to Florida CORFs have increasingly shifted toward those made for patients with back and musculoskeletal conditions. Most notably, patients who presented with back disorders accounted for 16 percent of all Medicare payments to Florida CORFs in 1999 and 29 percent of payments in 2002. In addition, payments for treating patients diagnosed with soft tissue injuries increased from 8 percent of Florida CORF payments in 1999 to 24 percent in 2002. One diagnosis group for which there was a notable decrease in the proportion of Medicare payments was pulmonary disorders, which fell from 30 percent of all payments in 1999 to 2 percent in 2002. In 2002, most of the 191 CORFs in Florida were small, with the median CORF in the state treating 150 beneficiaries. CORFs accounted for 15 percent of all Florida Medicare beneficiaries who received outpatient therapy from facility-based providers that year, and 30 percent of Medicare's payments for outpatient therapy services to Florida facility-based providers. In a few areas, however, CORFs represented a substantial share of the outpatient therapy market, particularly in south Florida. For example, CORFs were the predominate providers of outpatient therapy services in Miami, with 53 percent of all facility- based outpatient therapy patients, and treated 29 percent of patients who received outpatient therapy from facility-based providers in nearby Fort Lauderdale. Florida CORFs Received Higher Average Therapy Payments, Despite Treating Similar Patients: In 2002, Medicare's therapy payments per patient to Florida CORFs were several times higher than therapy payments made to other facility-based outpatient therapy providers in the state. This billing pattern was evident in each of the eight Florida MSAs that accounted for the majority of Medicare CORF facilities and patients. Differences in prior hospitalization diagnoses and patient demographic information did not explain the disparities in per-patient therapy payments. Average Therapy Payments to CORFs Substantially Exceeded Payments to Other Facility-Based Providers: Our analysis of claims payment data showed that per-patient therapy payments to Florida CORFs were about twice as high as therapy payments to rehabilitation agencies and SNF outpatient departments, and more than 3 times higher than therapy payments to hospital outpatient departments. (See table 1.) Specifically, at $2,327 per patient, therapy payments for CORF patients were 3.1 times higher than the per- patient payment of $756 for those treated by outpatient hospital-based therapists. Table 1: Therapy Payments and Units of Service Per Patient by Type of Provider, Florida, 2002: Payments per patient; CORFs: $2,327; Hospital OPDs: $756; Rehabilitation agencies: $1,094; SNF OPDs: $1,167. Units of service per patient; CORFs: 108; Hospital OPDs: 37; Rehabilitation agencies: 59; SNF OPDs: 53. Source: GAO analysis of CMS claims data. Note: Table provides average payments and units for therapy services only. [End of table] Higher therapy payments for Medicare patients treated at CORFs were largely due to the greater number of services that CORF patients received.[Footnote 12] As shown in table 1, on average, CORF patients received 108 units of therapy compared with 37 to 59 units of outpatient therapy, on average, at the other types of outpatient providers. Typically, a unit of therapy service represents about 15 minutes of treatment with a physical therapist, occupational therapist, or speech-language pathologist. The pattern of relatively high payments to CORFs was evident in all of the localities where CORFs were concentrated. In 8 of the 14 MSAs in Florida that had CORFs in 2002, CORF payments per patient were higher than payments to all other types of facility-based outpatient therapy providers. These MSAs together accounted for 86 percent of all Florida CORF beneficiaries and 90 percent of the state's CORF facilities. In these localities, per-patient payments to CORFs ranged from 1.2 to 7.4 times higher than payments to the provider type with the next highest payment amount.[Footnote 13] For example, in Fort Lauderdale, the 2002 average CORF therapy payment was $2,900--more than twice the average payment of $1,249 made for beneficiaries treated by rehabilitation agencies. (See table 2.) Table 2: Therapy Payments Per Patient by Type of Provider in Selected Florida MSAs, 2002: Dollars: MSA: Fort Lauderdale; CORFs: $2,900; Hospital OPDs: $916; Rehabilitation agencies: $1,249; SNF OPDs: $889. MSA: Fort Myers-Cape Coral; CORFs: $1,729; Hospital OPDs: $775; Rehabilitation agencies: $1,084; SNF OPDs: $911. MSA: Miami; CORFs: $2,686; Hospital OPDs: $998; Rehabilitation agencies: $1,914; SNF OPDs: $2,025. MSA: Naples; CORFs: $1,986; Hospital OPDs: $859; Rehabilitation agencies: $1,317; SNF OPDs: $856. MSA: Orlando; CORFs: $3,394; Hospital OPDs: $609; Rehabilitation agencies: $1,037; SNF OPDs: $1,266. MSA: Panama City; CORFs: $6,050; Hospital OPDs: $816; Rehabilitation agencies: $793; SNF OPDs: N/A[A]. MSA: Tampa-St. Petersburg-Clearwater; CORFs: $1,495; Hospital OPDs: $801; Rehabilitation agencies: $1,131; SNF OPDs: $1,278. MSA: West Palm Beach-Boca Raton; CORFs: $2,169; Hospital OPDs: $771; Rehabilitation agencies: $1,092; SNF OPDs: $1,091. Statewide; CORFs: $2,327; Hospital OPDs: $756; Rehabilitation agencies: $1,094; SNF OPDs: $1,167. Source: GAO analysis of CMS claims data. Note: Table provides average payments for therapy services only. [A] We found no 2002 outpatient therapy claims for Medicare beneficiaries treated by SNFs in this MSA. [End of table] Patient Characteristics Did Not Explain Higher Average Therapy Payments to Florida CORFs: Some factors that could account for differences in therapy payment amounts--patient diagnosis and indicators of patient health care needs- -did not explain the higher payments that some Florida CORFs received compared with other types of facility-based outpatient therapy providers. Average Therapy Payment by Diagnosis: We found that CORFs received higher per-patient therapy payments than other facility-based providers for patients in each of the four leading diagnosis categories treated at CORFs. For patients with neurologic disorders, arthritis, soft tissue injuries, and back disorders, payments to CORFs were 66 percent to 159 percent higher than payments to rehabilitation agencies and SNF OPDs and higher yet than payments to hospital OPDs.[Footnote 14] (See table 3.) Patients treated for back disorders made up the largest share of Florida CORF patients, at 25 percent. For patients with this diagnosis, average payments to CORFs-- at $1,734--were twice as high as the average payment of $867 made to rehabilitation agencies--the next highest paid provider type. Table 3: Therapy Payments Per Patient by Provider Type for Selected Diagnosis Categories, Florida, 2002: Dollars: Diagnosis category: Neurologic disorders[A]; CORFs: $2,676; Hospital OPDs: $545; Rehabilitation agencies: $1,311; SNF OPDs: $1,032. Diagnosis category: Arthritis[A]; CORFs: $2,168; Hospital OPDs: $679; Rehabilitation agencies: $979; SNF OPDs: $1,029. Diagnosis category: Soft tissue injuries[A]; CORFs: $1,835; Hospital OPDs: $625; Rehabilitation agencies: $929; SNF OPDs: $1,105. Diagnosis category: Back disorders[A]; CORFs: $1,734; Hospital OPDs: $532; Rehabilitation agencies: $867; SNF OPDs: $743. All diagnosis categories; CORFs: $2,327; Hospital OPDs: $756; Rehabilitation agencies: $1,094; SNF OPDs: $1,167. Source: GAO analysis of CMS claims data. Note: Table provides average payments for therapy services only. [A] These four diagnosis categories accounted for 74 percent of all Medicare beneficiaries receiving therapy services exclusively from Florida CORFs in 2002. [End of table] The higher therapy payments to CORFs were driven by the higher volume of therapy services that CORFs provided to their Medicare patients, compared with the volume of services other facility-based outpatient therapy providers furnished to patients in the same diagnosis group. As shown in table 4, for all four leading diagnosis categories, CORF Medicare patients received far more units of therapy, on average, than Medicare patients treated by other outpatient therapy providers. [Footnote 15] Differences across provider types were particularly pronounced for Medicare patients with arthritis. CORFs furnished an average of 100 units of therapy to beneficiaries treated for arthritis. In contrast, non-CORF outpatient therapy providers delivered an average of 33 to 53 units of therapy to Medicare arthritis patients. Table 4: Units of Therapy Service Per Patient by Provider Type for Selected Diagnosis Categories, Florida, 2002: Diagnosis category: Neurologic disorders[A]; CORFs: 115; Hospital OPDs: 28; Rehabilitation agencies: 86; SNF OPDs: 45. Diagnosis category: Arthritis[A]; CORFs: 100; Hospital OPDs: 33; Rehabilitation agencies: 53; SNF OPDs: 49. Diagnosis category: Soft tissue injuries[A]; CORFs: 87; Hospital OPDs: 32; Rehabilitation agencies: 48; SNF OPDs: 54. Diagnosis category: Back disorders[A]; CORFs: 84; Hospital OPDs: 27; Rehabilitation agencies: 49; SNF OPDs: 38. All diagnosis categories; CORFs: 108; Hospital OPDs: 37; Rehabilitation agencies: 59; SNF OPDs: 53. Source: GAO analysis of CMS claims data. Note: Table provides average units for therapy services only. [A] These four diagnosis categories accounted for 74 percent of all Medicare beneficiaries receiving services exclusively from Florida CORFs in 2002. [End of table] Patient Demographics and Prior-Year Hospitalizations: Differences in patient demographic characteristics and prior-year hospital diagnoses--factors that could indicate variation in patient health care needs--did not explain most of the wide disparities in therapy payments per patient across settings.[Footnote 16] When we considered differences in patient age, sex, disability, Medicaid enrollment, and 2001 inpatient hospital diagnoses across provider types, the data showed that patients served by CORFs could be expected to use slightly more health care services than patients treated by other facility-based therapy providers.[Footnote 17] However, we found that, after controlling for these patient differences, average payments for CORF patients remained 2 to 3 times greater than for those treated by other provider types.[Footnote 18] Consistent with this finding, therapy industry representatives we spoke with--including those representing CORFs--reported that, in the aggregate, CORF patients were not more clinically complex or in need of more extensive care than patients treated by other outpatient therapy providers. They told us that patients are referred to different types of outpatient therapy providers based on availability and convenience rather than on their relative care needs. One private consultant to CORFs and other outpatient provider groups noted that there are no criteria to identify and direct patients to a particular setting for outpatient care, and that physicians generally refer patients to therapy providers with whom they have a relationship. Actions by the Florida Contractor Were Not Sufficient to Ensure Appropriate Payments to CORFs: Despite the Florida contractor's increased scrutiny of CORF claims, our analysis of Florida CORFs' 2002 billing patterns suggests that some providers received inappropriate payments that year. In late 2001, after finding widespread billing irregularities among CORF claims, the Florida claims administration contractor implemented new strategies for reviewing claims that were maintained throughout 2002. Although these strategies were successful at ensuring appropriate claims payments for a limited number of beneficiaries, our analysis of 2002 CORF claims found that many CORFs continued to receive very high per-patient payments. 2001 Investigation by Florida Claims Contractor Revealed Pattern of Inappropriate CORF Billing: In 2001, the Medicare claims administration contractor for Florida reviewed about 2,500 claims submitted by CORFs and other facility-based outpatient therapy providers for services provided from January 1999 through February 2001.[Footnote 19] Among these claims, the contractor found widespread billing for medically unnecessary therapy services. These were therapy services related to maintaining rather than improving a patient's functioning, as required by Medicare reimbursement requirements for covering outpatient therapy. Reviews also found claims for the same beneficiary, made by more than one CORF, sometimes on the same day.[Footnote 20] The unlikelihood that a patient would receive treatment from more than one CORF provider when each one was equipped to provide the patient's full range of needed services caused the contractor to investigate further. After interviewing a sample of beneficiaries treated by multiple CORFs, the contractor found that some of the facilities treating these beneficiaries had common owners. It reported that the common ownership was significant, suggesting efforts by the owners to distribute billings for a patient's services across several providers. The contractor stated that this would allow the CORFs' owners to avoid the scrutiny of the Medicare contractor, which typically screens claims aggregated by facility rather than by beneficiary. After conducting additional reviews of a sample of paid claims from these CORFs, it found that 82 percent of payments made were inappropriate, largely due to questions about medical necessity. As a result, the contractor required these CORFs to repay Medicare approximately 1 million dollars and referred some of the CORFs to CMS and the HHS OIG for further investigation.[Footnote 21] In late 2001, the Florida claims administration contractor implemented additional claim review strategies targeting CORF claims. For any new CORF, the contractor began reviewing for medical necessity, prior to payment, about 30 of the first claims submitted. The contractor also began reviewing all therapy claims submitted on behalf of about 650 beneficiaries identified as having high levels of therapy use from multiple CORFs and other facility-based outpatient therapy providers during the 2001 investigation. CORFs and other providers submitting therapy claims for these beneficiaries had to supply documentation of medical necessity before claims were paid. The contractor also conducted prepayment reviews for specific therapy services determined to be at high risk for inappropriate payments, regardless of the beneficiary receiving services.[Footnote 22] The contractor maintained these intensified claim documentation and review requirements throughout 2002. The contractor indicated that the oversight measures put in place for specific beneficiaries were effective at improving the appropriateness of claims payments for therapy services made for those beneficiaries. Specifically, the contractor reported that Florida CORFs billed Medicare $12.1 million for this group in 2000, $10.2 million in 2001, and $7.3 million during 2002. In addition, the contractor denied an increasing percentage of the amount billed each year--46 percent in 2001, and 53 percent in 2002--based on its medical records reviews.[Footnote 23] Florida CORFs' High Medicare Payments Continued After Intensified Claim Reviews: While the contractor succeeded in ensuring that payments to CORFs for this limited group of beneficiaries met Medicare rules, our own analysis of CORF claims submitted in 2002 found several indications that billing irregularities continued. The indicators included a high rate of beneficiaries who received services from multiple CORFs, some CORFs that did not provide any therapy services, and many facilities with very high per-patient payments. Our analysis of 2002 Florida CORF claims by facility showed that the Florida claims administration contractor's efforts to ensure appropriate CORF payments were not completely effective. We found that 11 percent of the beneficiaries who received CORF services in Florida were treated by more than one CORF facility during the year. While Medicare rules do not prohibit beneficiaries from receiving services from multiple providers in a single year, this occurs much more frequently among Florida CORFs than among CORFs in other states. Specifically, in the five other states with the greatest numbers of CORFs at the end of 2001 (Alabama, California, Kentucky, Pennsylvania, and Texas), fewer than 4 percent of beneficiaries received services from more than one CORF during 2002, and in most of these states, the rate was 1 percent or less. Although many CORFs treated a few patients who received services from multiple providers during 2002, a small group of Florida CORFs had very high rates of "shared" patients that year--suggesting that some CORFs may have continued to operate in the patterns first detected by the Florida contractor during its 2001 review. Of the CORFs operating in Florida in 2002, 32 facilities shared more than half of their patients with other CORF providers. At four CORFs, more than 75 percent of the beneficiaries were treated by multiple CORF providers during the year. Staff from the Florida contractor told us that these patterns of therapy use--receiving services from multiple providers during the same time period--complicate their ability to monitor appropriate use of therapy services. Contractor staff routinely analyze claims data to evaluate appropriate levels of service use and identify trends that may suggest excessive use. However, these analyses are normally conducted on claims data aggregated by CORF provider, not aggregated per beneficiary. When beneficiaries receive outpatient therapy services from multiple providers, traditional methods of oversight are less likely to detect high levels of service use and payments. Our review of 2002 Florida claims data also showed that some CORFs were not complying with Medicare program rules about furnishing required services. Although CORFs are permitted to provide nontherapy services, they must be delivered as part of a beneficiary's overall therapy plan of care. However, three Florida CORFs received payments exclusively for nontherapy services--such as pulmonary treatment and oxygen saturation tests--in 2002.[Footnote 24] Four additional providers billed Medicare primarily for nontherapy services, with therapy care accounting for less than 10 percent of their annual Medicare payments. In addition, we found that a number of the CORFs identified during the Florida contractor's 2001 investigation continued to have very high average payments for all services provided in 2002.[Footnote 25] As shown in table 5, several of these facilities were among 21 CORFs with per-patient payments that exceeded the statewide CORF average by more than 50 percent. Among this group of high-cost facilities, the per- patient payment in 2002 ranged from $3,099 to $6,080, substantially above the average payment of $2,036 across all Florida CORFs.[Footnote 26] Table 5: Medicare Payments to Selected Florida CORFs for All Types of Services, 2002: CORF: A; Total Medicare payments: $6,080; Number of beneficiaries treated: 1; Average payment per beneficiary: $6,080. CORF: B; Total Medicare payments: $400,355; Number of beneficiaries treated: 70; Average payment per beneficiary: $5,719. CORF: C; Total Medicare payments: $346,893; Number of beneficiaries treated: 69; Average payment per beneficiary: $5,027. CORF: D; Total Medicare payments: $1,260,324; Number of beneficiaries treated: 255; Average payment per beneficiary: $4,942. CORF: E; Total Medicare payments: $1,648,748; Number of beneficiaries treated: 337; Average payment per beneficiary: $4,892. CORF: F; Total Medicare payments: $879,629; Number of beneficiaries treated: 193; Average payment per beneficiary: $4,558. CORF: G; Total Medicare payments: $1,709,567; Number of beneficiaries treated: 390; Average payment per beneficiary: $4,384. CORF: H; Total Medicare payments: $740,252; Number of beneficiaries treated: 169; Average payment per beneficiary: $4,380. CORF: I; Total Medicare payments: $154,780; Number of beneficiaries treated: 36; Average payment per beneficiary: $4,299. CORF: J; Total Medicare payments: $914,198; Number of beneficiaries treated: 241; Average payment per beneficiary: $3,793. CORF: K[A]; Total Medicare payments: $6,066,325; Number of beneficiaries treated: 1,600; Average payment per beneficiary: $3,791. CORF: L; Total Medicare payments: $705,499; Number of beneficiaries treated: 189; Average payment per beneficiary: $3,733. CORF: M; Total Medicare payments: $309,082; Number of beneficiaries treated: 88; Average payment per beneficiary: $3,512. CORF: N; Total Medicare payments: $631,009; Number of beneficiaries treated: 184; Average payment per beneficiary: $3,429. CORF: O[A]; Total Medicare payments: $398,526; Number of beneficiaries treated: 117; Average payment per beneficiary: $3,406. CORF: P; Total Medicare payments: $544,950; Number of beneficiaries treated: 163; Average payment per beneficiary: $3,343. CORF: Q; Total Medicare payments: $1,467,888; Number of beneficiaries treated: 447; Average payment per beneficiary: $3,284. CORF: R[A]; Total Medicare payments: $5,387,964; Number of beneficiaries treated: 1,723; Average payment per beneficiary: $3,127. CORF: S[A]; Total Medicare payments: $2,415,257; Number of beneficiaries treated: 775; Average payment per beneficiary: $3,116. CORF: T; Total Medicare payments: $632,862; Number of beneficiaries treated: 204; Average payment per beneficiary: $3,102. CORF: U[A]; Total Medicare payments: $1,152,947; Number of beneficiaries treated: 372; Average payment per beneficiary: $3,099. CORF: All others; Total Medicare payments: $49,656,466; Number of beneficiaries treated: 30,401; Average payment per beneficiary: $1,633. Total; Total Medicare payments: $77,429,600; Number of beneficiaries treated: 38,024; Average payment per beneficiary: $2,036. Source: GAO analysis of CMS claims data. Note: This analysis includes all Medicare payments to Florida CORFs for therapy services (physical therapy, occupational therapy, and speech- language pathology services) and other types of CORF services. In 2002, 98 percent of Medicare payments to Florida CORFs were for therapy services. We included all beneficiaries in this analysis, regardless of their total therapy payments for the year and duration of Medicare fee- for-service enrollment. [A] These facilities were among those identified by the Florida claims administration contractor during its 2001 investigation as having high levels of medically unnecessary services and questionable billing practices. [End of table] These relatively high 2002 payments suggested that Florida CORFs responded to the contractor's targeted medical reviews selectively by reducing the services provided to the small number of patients whose claims were under scrutiny. Other patients, outside the scope of the contractor's criteria for medical review, continued to receive high levels of services. The contractor continues to rely on the medical review criteria originally established in late 2001. However, contractor staff reported ongoing concerns about the extent to which CORFs bill for services that may not meet the program's requirements for payment. In particular, they cited the practice of delivering therapy services over relatively long periods of time that only maintain, rather than improve, a patient's functional status.[Footnote 27] Conclusions: Sizeable disparities between Medicare therapy payments per patient to Florida CORFs and other facility-based outpatient therapy providers in 2002--with no clear indication of differences in patient needs--raise questions about the appropriateness of CORF billing practices. After finding high rates of medically unnecessary therapy services to CORFs, CMS's claims administration contractor for Florida took steps to ensure appropriate claim payments for a small, targeted group of CORF patients. Despite its limited success, billing irregularities continued among some CORFS and many CORFs continued to receive relatively high payments the following year. This suggests that the contractor's efforts were too limited in scope to be effective with all CORF providers. Recommendation: To ensure that Medicare only pays for medically necessary care as outlined in program rules, CMS should direct the Florida claims administration contractor to medically review a larger number of CORF claims. Agency Comments and Our Evaluation: CMS officials reviewed a draft of this report and agreed with its findings. Specifically, the agency noted that "disproportionately high payments made to CORFs indicate a need for medical review of these providers." The agency also pointed out that, given the high volume of claims submitted by providers, contractors must allocate their limited resources for medical review in such a way as to maximize returns. Furthermore, CMS stated that the Florida claims administration contractor is already taking appropriate steps to address concerns about CORF billing and is prepared to take additional steps if necessary. We recognize that contractors can achieve efficiencies by targeting their medical review activities at providers or services that place the Medicare trust funds at the greatest risk. However, the impact of medical review comes, in part, from the sentinel effect of consistently applying medical review to providers' claims. Thus, while we support the contractor's focus on new CORF providers, we continue to believe that enlarging the number of CORF claims reviewed would promote compliance with medical necessity requirements. Given that Florida CORFs continued to bill significantly more per beneficiary than other outpatient therapy providers even after the contractor took steps to examine some claims, compliance could be enhanced by aggressively addressing this vulnerability. CMS's comments appear in appendix II. As agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution of it until 30 days from its issue date. At that time, we will send copies of this report to the Administrator of CMS and to other interested parties. In addition, this report will be available at no charge on GAO's Web site at http://www.gao.gov. We will also make copies available to others upon request. If you or your staff have any questions about this report, please call me at (312) 220-7600. Another contact and key contributors are listed in appendix III. Sincerely yours, Signed by: Leslie G. Aronovitz, Director, Health Care--Program Administration and Integrity Issues: [End of section] Appendix I: Objectives, Scope, and Methodology: In this report we (1) compared Medicare's outpatient therapy payments to CORFs in 2002 with its payments that year to other facility-based outpatient therapy providers and (2) assessed the program's effectiveness in ensuring that payments to CORFs complied with Medicare rules. As agreed with the requester's staff, we limited the scope of our review to facility-based outpatient therapy providers and beneficiaries in Florida. Florida accounted for one-third of all CORF facilities at the end of 2002. Our primary data source was CMS's National Claims History (NCH) 100% Nearline File. The NCH file contains all institutional and noninstitutional claims from the Common Working File (CWF)--the system that CMS uses to process and pay Medicare claims through its contractors across the country. We also reviewed data from CMS's Medicare Provider of Service Files, which contain descriptive information on CORF facility characteristics, such as location, type of ownership, and the date of each provider's initial program certification. Finally, we interviewed representatives of CMS's central and regional offices, the Florida claims administration contractor, federal law enforcement agencies, and the therapy industry. To describe the Florida CORF industry and operations, we gathered Medicare claims data from CMS's NCH File for the years 1999 through 2002. In addition to reviewing trends in total Medicare payments to CORFs, we examined changes in the patient case mix by identifying the primary diagnoses listed on claims for beneficiaries treated by CORFs. We also obtained descriptive information on CORFs' characteristics from the Provider of Service Files for 1999 through 2003. This work was performed from May 2003 through July 2004 in accordance with generally accepted government auditing standards. Comparison of Medicare Outpatient Therapy Payments by Provider Type: In this analysis, we compared Medicare therapy payments to four types of facility-based outpatient therapy providers: CORFs, rehabilitation agencies, hospital OPDs, and SNF OPDs. Although CORFs are authorized to offer a wide range of services, we limited our comparison to a common set of therapy services: physical therapy services, occupational therapy services, and speech-language pathology services. To compare Medicare's therapy payments to Florida CORFs with therapy payments to other types of facility-based outpatient rehabilitation therapy providers, we examined 2002 Medicare beneficiary claims data from the NCH File.[Footnote 28] We used the NCH file to identify all beneficiaries who resided in Florida and received outpatient therapy services from in-state providers during 2002. By limiting our review to beneficiaries who were enrolled in part B for all 12 months of the year, we excluded those in managed care and those with less than a full year of fee-for-service coverage. Using beneficiary identification numbers, we aggregated each beneficiary's total outpatient therapy claims from all provider types. We summed the annual number of therapy units billed for each beneficiary as well as the annual line-item payment amounts.[Footnote 29] This allowed us to assign each beneficiary to a provider comparison group. To compare Medicare expenditures for similar patients, we assigned each beneficiary to a diagnosis category based on the primary diagnoses listed in their outpatient therapy claims for the year.[Footnote 30] Our diagnosis groups included: * stroke, * spinal cord injury, * neurologic disorders, * hip fractures, * back disorders, * amputation, * cardiovascular disorders--circulatory, * cardiovascular disorders--pulmonary, * rehabilitation for unspecified conditions, * arthritis, * soft tissue/musculoskeletal injuries, * ortho-surgical, * multiple diagnoses[Footnote 31] and: * other. To consider differences in payment by provider type at the substate level, we compared annual per-patient payments for CORFs and other outpatient facility providers in each of Florida's 20 metropolitan statistical areas. Variation in treatment patterns and payments (for the same diagnosis category) across provider types may suggest that one type of provider treats a patient population with greater needs for service. To consider patient differences, we applied CMS's Principal Inpatient Diagnostic Cost Group (PIP-DCG) model.[Footnote 32] By comparing patients' use of hospital services and inpatient diagnoses (in the calendar year prior to the year they received therapy) and demographic information such as age, sex, disability, and Medicaid enrollment, the PIP-DCG model allowed a comparison of anticipated patient care needs across provider types. We used the PIP-DCG score developed for each beneficiary in combination with the 2002 therapy payment data to conduct an analysis of covariance. Evaluation of the Florida Contractor's Efforts to Ensure Appropriate CORF Claim Payments: To review strategies used by the Florida claims administration contractor to ensure proper CORF payments, we interviewed representatives of CMS's central and regional offices and representatives from the contractor. The contractor provided us with the results of its 2001 investigation of Florida CORFs and its subsequent reports on CORF billing patterns. In addition, we interviewed federal law enforcement agencies involved in investigations of Florida CORF facilities. To assess the effectiveness of the contractor's oversight strategies, we reviewed information developed by the contractor on changes in CORF billing practices. We also analyzed 2002 claims data for CORF services to identify any CORFs with disproportionately high Medicare payments. This analysis included payment data for all claims--for both therapy and nontherapy services. In contrast to our comparison of per-patient payments by provider type, in this analysis we included all beneficiaries, regardless of their total annual therapy payments and duration of Medicare fee-for-service enrollment. Assessment of Data Reliability: We did not independently verify the reliability of CMS's Medicare claims data. However, we determined that CMS's Medicare claims data were sufficiently reliable for the purposes of this engagement. CMS operates a Quality Assurance System designed to ensure the accuracy of its Medicare NCH and CWF data files. Specifically, the agency has procedures in place to (1) ensure that files have been transmitted properly and completely, (2) check the functioning of contractor claims edits, and (3) sample claims from the files that exhibit unusual or inconsistent coding practices (indicating that data elements may be unreliable). In addition, we consulted with CMS's technical staff as necessary to ensure the accuracy and relevance of the data elements used in our analysis. We also screened the files and excluded claims that were denied, claims superseded by an adjustment claim, and claims for services in other years. [End of section] Appendix II: Comments from the Centers for Medicare & Medicaid Services: DEPARTMENT OF HEALTH & HUMAN SERVICES: Centers for Medicare & Medicaid Services: Administrator: Washington, DC 20201: DATE: JUL 27 2004: TO: Leslie G. Aronovitz: Director, Health Care-Program Administration and Integrity Issues: General Accounting Office: FROM: Mark B. McClellan, M.D., Ph.D.: Administrator: SUBJECT: General Accounting Office Draft Report:"Comprehensive Outpatient Rehabilitation Facilities: High Medicare Payments in Florida Raise Program Integrity Concerns" (GAO-04-709): Thank you for the opportunity to review the General Accounting Office (GAO) draft report entitled"Comprehensive Outpatient Rehabilitation Facilities: High Medicare Payments in Florida Raise Program Integrity Concerns" (GAO-04-709). A comprehensive outpatient rehabilitation facility (CORF) is defined in 42 CFR 485.51 as a"nonresidential facility that is established and operated for the purpose of providing diagnostic, therapeutic, and restorative services to outpatients for the rehabilitation of injured, disabled, or sick persons, at a single fixed location, by or under the supervision of a physician." Pursuant to 42 CFR 410.105 (c), the services must be furnished under a written plan of treatment that is established and signed by a physician before treatment is begun. The plan must be reviewed at least every sixty days by a facility physician who, when appropriate, consults with the professional personnel providing the services. Upon that review, the physician must certify or recertify that the plan is being followed, the patient is making progress in attaining the rehabilitation goals, and the treatment is having no harmful effect on the patient. Medicare permits a beneficiary to receive care from more than one CORF as long as there are different plans of treatment for each CORF. There could be two different episodes of care, which would require different plans of treatment and also different CORFs. The beneficiary has the right to decide from which provider to receive those services. However, 42 CFR 485.51 states that a CORF should be able to provide all necessary therapeutic, diagnostic and restorative services at a single fixed location. GAO Recommendation: The GAO recommends that the Centers for Medicare & Medicaid Services (CMS) direct the Florida claims administration contractor to medically review a larger number of Comprehensive Outpatient Rehabilitation Facilities (CORFs) claims. CMS Response to the GAO Recommendation: The Centers for Medicare & Medicaid Services (CMS) recognizes the importance of using medical review to ensure correct payment. The goal of the medical review program is to reduce payment error by identifying and addressing billing errors concerning coverage and coding made by providers. In order to meet this goal, our Program Integrity Manual provides that contractors have the authority to review any claim at any time. However, the claims volume of the Medicare program prohibits review of every claim. Resources dictate that in attempting to make only correct payments, contractors make deliberate decisions on the best uses of limited resources to maximize returns. We agree that disproportionately high payments made to CORFs indicate a need for medical review of these providers, and believe the contractor is already taking appropriate steps to address this problem. From our discussions with First Coast Service Options (FCSO), the Medicare contractor in the state of Florida, we understand that they currently place every new CORF on medical review, and continually monitor new CORFs, through prepayment review, until they maintain at least an 80 percent correct payment rate through prepayment review and display no aberrancies. Furthermore, FCSO provides education to all CORFs on prepayment review on an ongoing basis. Additionally, once a provider is removed from prepayment review, FCSO periodically checks to see if the provider's billing patterns are still acceptable. If FCSO observes an anomalous claim submission pattern, FCSO typically obtains a small postpayment sample to identify any new problems. The FCSO is prepared to take additional steps to address CORF issues if necessary. [End of section] Appendix III: GAO Contact and Staff Acknowledgments: GAO Contact: Rosamond Katz, (202) 512-7148: Acknowledgments: In addition to the contact named above, Jennifer Grover, Rich Lipinski, and Hannah Fein made key contributions to this report. FOOTNOTES [1] Physical therapy treatments--such as whirlpool baths, ultrasound, and therapeutic exercises--are designed to improve mobility, strength, and physical functioning, and limit the extent of disability resulting from injury or disease. Occupational therapy helps patients learn the skills necessary to perform daily tasks and function independently. Speech-language pathology services include the diagnosis and treatment of speech, language, and swallowing disorders. [2] Medicare does not cover maintenance therapy--therapy services performed to maintain, rather than improve, a beneficiary's level of functioning. Examples of maintenance therapy are when a patient's restoration potential is insignificant in relation to the therapy required to achieve such potential, when it has been determined that the treatment goals will not materialize, or when the therapy performed is considered to be a general exercise program. [3] The conditions under which Medicare will pay for outpatient therapy services provided by a CORF were established by the Omnibus Reconciliation Act of 1980, Pub. L. No. 96-499, § 933, 94 Stat. 2599, 2635. [4] Rehabilitation medicine is the treatment of individuals with disabling conditions and diseases, designed to yield improvement in function, level of independence, and quality of life. [5] While CORFs must be able to provide all of the therapy and related nontherapy services required by each patient's treatment plan, they are not permitted to provide nontherapy services alone; such services must be delivered as a component of each patient's rehabilitative therapy treatment. Furthermore, although each CORF chooses which of these services to offer, a facility cannot accept a patient unless it can provide all required services. [6] States included in the study were Florida, Louisiana, Michigan, New Jersey, Pennsylvania, and Texas. See HHS/OIG, Six-State Review of Outpatient Rehabilitation Facilities, pub. A-04-99-01193 (Washington D.C.: March 2000). [7] Part B covers physician services and payments to other licensed practitioners, clinical laboratory and diagnostic services, surgical supplies and durable medical equipment, and ambulance services. Part A covers hospital and certain other services. [8] Medicare pays 80 percent of the payment amount with a 20 percent coinsurance payment required from the beneficiary. [9] In part, turnover in the industry may be the result of a new payment system for Medicare outpatient therapy services that took effect in 1999. That year, facility providers were switched from a cost-based reimbursement system--under which payments were based on the cost to the provider of delivering services--to a fee schedule--where payments are based on pre-established amounts. Previously, the fee schedule had only applied to therapy provided by physician practices and independent practitioners. [10] These changes in Medicare payments since 1999 reflect, in part, the industry's varied responses to the new Medicare payment rules for outpatient therapy. A 2001 study by the Urban Institute reported that CORF payments per patient for therapy services declined 55 percent nationwide, from $1,642 in 1998 to $743 in 1999--the first year under the fee schedule payment system. The study also showed that this adjustment year was followed by a 61 percent "rebound" in average payments of $1,199 in 2000. See S. Maxwell, C. Baseggio, and M. Storeygard, Part B Therapy Services Under Medicare in 1998-2000: Impact of Extending Fee Schedule Payments and Coverage Limits, (Washington, D.C.: The Urban Institute, September 2001). [11] Payments for physical therapy services made up the largest share of Medicare payments to Florida CORF facilities. Physical therapy payments climbed from 49 percent of all Medicare payments to CORFs in 1999 to 64 percent in 2002. Payments for occupational therapy also increased, changing from 18 percent of payments in 1999 to 32 percent in 2002. The CORF service for which there was a substantial decline in payments was respiratory treatments, which changed from 29 percent of Florida CORF payments in 1999 to 2 percent of payments in 2002. [12] On average, Medicare fees per unit of therapy provided by Florida CORFs were about the same as the fees per unit of service furnished by other provider types. In 2002, fees averaged about $22 (CORFs), $22 (SNF OPD), $21 (hospital OPD), and $19 (rehabilitation agencies). [13] Five MSAs with elevated CORF payments were in southern Florida (Fort Lauderdale, Fort Myers-Cape Coral, Miami, Naples, and West Palm Beach-Boca Raton); two were in the middle of the state (Orlando and Tampa-St. Petersburg-Clearwater), and one was in the northwest (Panama City). CORFs operating in Panama City had the highest per-patient payments, but treated only 37 Medicare patients in 2002. In contrast, CORFs in Miami provided services to 6,069 Medicare patients that year. [14] While grouping beneficiaries with the same diagnosis allows for comparison of similar patients, some patients in each grouping are likely to have higher levels of health care needs than others. [15] Just as CORFs consistently provided high levels of therapy services, hospital OPDs were uniformly low across the four common diagnosis categories in the amount of therapy services delivered. A CMS official remarked that this may be driven by the traditional practice of hospitals to employ the same set of therapists for both their inpatient and outpatient care. Because Medicare's payment system for inpatient hospitalization provides a set payment amount, based on patient diagnosis, for each hospital stay, hospital-based therapists may be accustomed to discharging inpatients from therapy quickly. They may approach outpatient care in much the same way, despite the fact that Medicare pays on a fee schedule for therapy patients treated in hospital outpatient departments. [16] Recent hospitalization records are one indicator of patient health care needs. To compare differences in predicted patient health care use across outpatient therapy provider types, we used CMS's Principal Inpatient Diagnostic Cost Group (PIP-DCG) model, which uses demographic information and hospitalization data to predict health care expenditures for each beneficiary. Among all Florida beneficiaries who received outpatient therapy services during 2002, 26 percent were hospitalized during 2001. [17] We found two exceptions to this finding. First, among patients with neurologic disorders, those treated by CORFs appeared to be similar in health status to patients treated by other types of providers. Second, patients treated by CORFs were shown to require slightly less health care services than patients using SNF OPDs. The patients using SNF OPDs comprise only 5 percent of all Florida beneficiaries who received facility-based outpatient therapy services. [18] We used the PIP-DCG score developed for each beneficiary in combination with 2002 claims payment data to conduct an analysis of covariance. We found that differences in average payments were statistically significant at the .01 level across comparative provider types for every diagnosis category except for beneficiaries with neurologic disorders and amputations. However, the overall R-Square for the analysis was 0.18, which indicates that much of the difference we found in average payments across provider types remains unexplained by patient demographics and prior hospital diagnosis. [19] The investigation included claims submitted by CORFs and rehabilitation agencies. [20] Although Medicare does not prohibit beneficiaries from receiving the same type of services from multiple providers during the same day, contractor staff indicated that such a situation raises questions about the medical necessity of services provided. [21] An outpatient therapy company that owned several CORFs in Florida was later investigated by the Department of Justice. A settlement in December 2002 for $600,000 resolved allegations of billing for medically unnecessary services, falsifying patient and facility records, and providing services outside the state in which the facilities were licensed to operate. [22] These reviews were conducted on claims submitted from all provider types. [23] Other therapy providers subject to these reviews also reduced the amount of therapy services billed to Medicare. Rehabilitation agencies billed Medicare $1.5 million for this group of beneficiaries in 2000, $827,000 in 2001, and $709,000 during 2002. The Florida contractor denied 36 percent of the amount billed by rehabilitation agencies in 2001, and 48 percent in 2002. [24] An official from the CMS regional office with oversight responsibility for Florida reported that some facilities marketing themselves as "specialized" CORFs have had problems complying with Medicare requirements. Specifically, the regional office found that some CORFS were paying a fee to providers (such as therapists or psychologists) to have the provider's name included on the initial CORF application for Medicare certification. However, after certification was granted, these providers never worked for the CORF; in fact, the facilities were only providing specialized services and not the core CORF services required by Medicare. [25] Includes payment for therapy services and other types of services provided by CORFs, such as physician and nursing services, psychological services, and pulmonary treatments. In 2002, 2 percent of all Medicare payments to Florida CORFs were for nontherapy services. [26] Overall, we found considerable variation in Medicare per-patient payments across Florida CORFs. The top quartile of CORFs received payments of $2,070 or more, while the lowest quartile received payments of $982 or less. The median per-patient payment across all Florida CORFs was $1,498. [27] One Medicare rule offers CORFs unique operating circumstances that may contribute to providing services over longer periods of time. CORFs may provide therapy services for 60 days before the patient's physician must reevaluate the patient and certify that continuing therapy services would result in continuing improvement of patient function. 42 C.F.R. § 410.105 (c)(2)(2003). In contrast, other facility-based outpatient therapy providers must reevaluate patients every 30 days. 42 C.F.R. § 424.24 (c)(4). [28] This was the latest year for which complete CMS claims data were available. [29] To ensure that each beneficiary included in our study received services from only one type of outpatient rehabilitation therapy provider during 2002, we also examined therapy claims from physician practices and therapists in independent practice. Beneficiaries who received services from more than one type of facility-based provider, or from a facility-based provider and a nonfacility-based provider (such as a physician's office), were excluded from our analysis. In addition, we limited the analysis to beneficiaries whose annual therapy payments were $100 or more. [30] We based our diagnosis categories on an approach originally developed by rehabilitation researchers. See Joan L. Buchanan, J. David Rumpel, and Helen Hoenig, "Outpatient Institutional Rehabilitation Services 1987-1990: Who Provides Them and How Do They Compare?" (Santa Monica, CA: RAND/UCLA/Harvard Center for Health Care Financing Policy Research, 1993). [31] Florida beneficiaries with more than one condition listed as their primary diagnosis on therapy claims in 2002 were assigned to the multiple diagnosis category. [32] The PIP-DCG model is an algorithm that uses base-year inpatient diagnoses, along with demographic factors, to predict total health spending in the following year. CMS has used the PIP-DCG model to determine relative risk factors and predict health expenditures for beneficiaries enrolled in its Medicare+Choice program and, as a result, has risk adjusted payments to participating health plans. GAO's Mission: The Government Accountability Office, the investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through the Internet. GAO's Web site ( www.gao.gov ) contains abstracts and full-text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as "Today's Reports," on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select "Subscribe to e-mail alerts" under the "Order GAO Products" heading. Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office 441 G Street NW, Room LM Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000: TDD: (202) 512-2537: Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov Automated answering system: (800) 424-5454 or (202) 512-7470: Public Affairs: Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, D.C. 20548: