Health Care Shortage Areas: Designations Not a Useful Tool for Directing Resources to the Underserved (Chapter Report, 09/08/95, GAO/HEHS-95-200). GAO reviewed the Department of Health and Human Services' (HHS) systems for identifying geographical areas where access to medical care is limited, focusing on: (1) how well the systems identify areas with primary care shortages; (2) how well the systems target federal funding to the underserved; and (3) whether the HHS proposal to combine the systems would lead to improvements. GAO found that: (1) the two HHS systems do not reliably identify areas with primary care shortages or help target federal resources to the underserved; (2) the systems have widespread data and methodology problems which severely limit their ability to pinpoint needy areas; (3) both systems tend to overstate the need for additional primary care providers because they do not consider all of the categories of providers already in place; (4) the Health Professional Shortage Area System (HPSA) does not consider the extent to which available resources are being used; (5) the Medically Underserved Area System (MUA) is limited in its ability to identify underserved areas and populations; (6) neither system identifies the specific subpopulations that have difficulty obtaining medical care; (7) while the systems can sometimes accurately identify needy areas, they do not provide the necessary data to determine which programs are best suited to those areas; (8) the proposed consolidation and streamlining of the systems is not likely to solve system problems, since the underlying causes of the problems have not been addressed; (9) it may be more cost effective to modify individual programs and application processes to identify where needs exist and the appropriate program to meet those needs and to target resources better; and (10) HHS officials believe that they need to maintain a national shortage designation system to monitor primary care access, but HHS has another initiative underway that could serve those purposes. --------------------------- Indexing Terms ----------------------------- REPORTNUM: HEHS-95-200 TITLE: Health Care Shortage Areas: Designations Not a Useful Tool for Directing Resources to the Underserved DATE: 09/08/95 SUBJECT: Health care services Physicians Disadvantaged persons Health resources utilization Data bases Data integrity Evaluation methods Geographic information systems Health care personnel IDENTIFIER: HHS Health Professional Shortage Area System HHS Medically Underserved Areas System HHS Community Health Centers Program Medicare Incentive Payment Program National Health Service Corps Scholarship Program HHS Community Scholarship Program Rural Health Clinic Services Program PHS Title VII Health Professions Education and Training Grant Program PHS Title VIII Health Professions Education and Training Grant Program Indian Health Scholarship Program HHS Title III Mental Health Clinical Traineeship Program HHS Title X Family Planning Services Training Program ************************************************************************** * This file contains an ASCII representation of the text of a GAO * * report. Delineations within the text indicating chapter titles, * * headings, and bullets are preserved. Major divisions and subdivisions * * of the text, such as Chapters, Sections, and Appendixes, are * * identified by double and single lines. The numbers on the right end * * of these lines indicate the position of each of the subsections in the * * document outline. These numbers do NOT correspond with the page * * numbers of the printed product. * * * * No attempt has been made to display graphic images, although figure * * captions are reproduced. Tables are included, but may not resemble * * those in the printed version. * * * * A printed copy of this report may be obtained from the GAO Document * * Distribution Facility by calling (202) 512-6000, by faxing your * * request to (301) 258-4066, or by writing to P.O. Box 6015, * * Gaithersburg, MD 20884-6015. We are unable to accept electronic orders * * for printed documents at this time. * ************************************************************************** Cover ================================================================ COVER Report to Congressional Committees September 1995 HEALTH CARE SHORTAGE AREAS - DESIGNATIONS NOT A USEFUL TOOL FOR DIRECTING RESOURCES TO THE UNDERSERVED GAO/HEHS-95-200 Health Care Shortage Areas (108218) Abbreviations =============================================================== ABBREV DSD - Division of Shortage Designation HHS - Department of Health and Human Services HPSA - Health Professional Shortage Area IMU - Index of Medical Underservice MUA - Medically Underserved Area MUP - medically underserved population NHSC - National Health Service Corps OTA - Office of Technology Assessment USIA - United States Information Agency Letter =============================================================== LETTER B-259394 September 8, 1995 Congressional Committees This report was prepared at our own initiative in response to the growing number of federal programs using health care shortage areas to determine who can apply for federal assistance. Our report discusses how the Department of Health and Human Services' systems for identifying health care shortage areas are currently used to target resources to the underserved, and Department proposals to combine these systems. We include recommendations to the Congress that could result in a better match of federal program resources to needy communities, and eliminate funding where there is not a demonstrated need for federal assistance. Mark V. Nadel Associate Director, National and Public Health Issues List of Addressees The Honorable Bob Packwood Chairman The Honorable Daniel P. Moynihan Ranking Minority Member Committee on Finance United States Senate The Honorable Nancy L. Kassebaum Chairman The Honorable Edward M. Kennedy Ranking Minority Member Committee on Labor and Human Resources United States Senate The Honorable Michael Bilirakis Chairman The Honorable Henry A. Waxman Ranking Minority Member Subcommittee on Health and Environment Committee on Commerce House of Representatives The Honorable William M. Thomas Chairman The Honorable Pete Stark Ranking Minority Member Subcommittee on Health Committee on Ways and Means House of Representatives EXECUTIVE SUMMARY ============================================================ Chapter 0 PURPOSE ---------------------------------------------------------- Chapter 0:1 Many Americans live in places where there are barriers to obtaining primary health care. These locations range from isolated rural areas to inner-city neighborhoods. In fiscal year 1994, the federal government spent about $1 billion on programs for alleviating access problems in such locations. To work effectively, these programs need a sound method of identifying the type of access problems that exist and focusing services on the people who need them. The Department of Health and Human Services uses two main systems for identifying such locations. One designates Health Professional Shortage Areas (HPSAs), the other Medically Underserved Areas (MUAs). Over half of all U.S. counties are designated as HPSAs or MUAs, and over another fourth have HPSAs or MUAs somewhere within their borders. As part of the broad federal effort to improve access to care, GAO reviewed the two systems to determine (1) how well they identify areas with primary care shortages, (2) how well they help target federal funding to benefit those who are underserved, and (3) whether they are likely to be improved under Department proposals to combine them. GAO's review included evaluating the systems' criteria for identifying health professional shortages and medical underservice, measuring the accuracy and timeliness of the data in the databases and in a statistical sample of HPSA applications, and discussing the systems with managers who use them to allocate program resources. BACKGROUND ---------------------------------------------------------- Chapter 0:2 The primary care HPSA system focuses on whether an area has a critical shortage of physicians available to serve the people living there. A HPSA can be a distinct geographic area (such as a county), a specific population group within the area (such as the poor), or a specific public or nonprofit facility (such as a prison). The system was first used in 1978 to place National Health Service Corps employees and those providers receiving scholarships or repayment of student loans in exchange for service in shortage areas. Its use has expanded to nearly 30 other programs, each having a different strategy to improve access to care. In fiscal year 1994, combined funding for programs using the HPSA system was about $473 million. The MUA system identifies areas or populations with shortages of health care services using several factors in addition to the availability of health care providers. These factors include infant mortality rate, poverty rate, and percentage of population aged 65 or over. Developed at about the same time as the HPSA system, the MUA system has been used for a more limited range of programs--mainly to identify areas eligible for federally funded community health centers. The Community Health Center program, with fiscal year 1994 expenditures of about $663 million, is still the system's main user. Federal programs use the HPSA and MUA systems in varying degrees as a screen to determine eligibility for federal funding. RESULTS IN BRIEF ---------------------------------------------------------- Chapter 0:3 The HPSA and MUA systems do not effectively identify areas with primary care shortages or help target federal resources to benefit those who are underserved. For programs relying on the systems for these purposes, there is little assurance that federal funds are used where most needed. Data and methodology problems are widespread, severely limiting the systems' ability to pinpoint the extent of need in underserved areas. For example, the HPSA methodology may be overstating the need for additional physicians in HPSAs by 50 percent or more, and the designations in both systems are often based on inaccurate or outdated information. Even when the systems accurately identify needy areas, they often do not provide the information needed to decide which programs are best suited to the area's particular need. As a result, a program without additional screening processes may be applied that does not directly benefit the specific subpopulation with insufficient access to care. An example is the Medicare Incentive Payment program, which provides bonus payments to all physicians treating Medicare patients in geographic HPSAs, even though a different group than Medicare patients--such as migrant farmworkers--may be those actually underserved. The Department's proposals for combining and streamlining the systems are unlikely to solve the problems we identified. But fixing the systems is not the only option--and probably not the best one. Instead, all but one of the individual programs already have criteria and application processes in place that may be more easily modified to identify where a need exists and whether the program is an appropriate remedy. PRINCIPAL FINDINGS ---------------------------------------------------------- Chapter 0:4 SYSTEMS DO NOT EFFECTIVELY IDENTIFY SHORTAGE AREAS -------------------------------------------------------- Chapter 0:4.1 The HPSA methodology for identifying the extent of primary care shortages is flawed. It tends to overstate the need for additional primary care providers because it omits several important categories of providers already in place. For example, it does not count National Health Service Corps providers, U.S.-trained foreign physicians (unless they are permanent residents), and nonphysicians such as physician assistants, nurse practitioners, and nurse-midwives. These omissions have substantial effects. For example, adding just physician assistants and nurse-midwives known to be practicing in countywide HPSAs would decrease the number of providers said to be needed in such HPSAs by at least 22 percent. Similarly, the MUA methodology does not accurately identify the geographic areas and populations that have the greatest health service shortages. When the methodology was initially developed in the mid-1970s, independent testing showed that rural areas designated as MUAs did not differ greatly from non-MUAs in terms of access to care. The methodology has remained virtually unchanged; but since 1986, the law authorizes the Department of Health and Human Services to designate underserved populations that do not meet the requirements of the MUA methodology, if so recommended by a state governor on the basis of unusual local conditions. SYSTEM DATA ARE NEITHER ACCURATE NOR TIMELY -------------------------------------------------------- Chapter 0:4.2 GAO estimates that about 20 percent of geographic HPSAs were designated in error or without sufficient supporting documentation in the application file. For example, the number of available physicians listed in some applications was understated by up to 50 percent when compared with information in other sources (such as physician directories) for the area. HPSAs are also not being reviewed on a timely basis to determine if they still qualify or should be dropped from designation. Department policy calls for HPSAs to receive a comprehensive review every 3 years, but 31 percent of current HPSAs have not been reviewed within this time period. The list of MUAs has gone substantially unchanged since it was established in 1976. Although new MUAs have been added, the overall list has not been reviewed systematically to update scores or to propose areas for dedesignation since 1981. GAO's review of current countywide MUAs showed that if the designations were to be reviewed using 1990 data, almost half would lose their designation. SYSTEMS DO NOT EFFECTIVELY TARGET FUNDING TO THE UNDERSERVED -------------------------------------------------------- Chapter 0:4.3 In many shortage locations, access to care is a problem for only part of the population. For example, most residents in a city may have adequate access to care, but the poor may not. However, most HPSA and MUA designations do not identify the specific subpopulations having difficulty obtaining access to care. Instead, they identify an area only by its geographic boundaries. This approach presents a problem because, unlike the Community Health Center program, other assistance programs do not go beyond the MUA or HPSA designation to identify who is underserved, and why. As a result, a disconnect can occur between the reason for underservice and the remedy provided. This disconnect is particularly apparent for the Medicare Incentive Payment program. The program, which pays a 10-percent bonus on Medicare billings, was established in 1987 to address concerns that low reimbursement rates could discourage physicians from accepting Medicare beneficiaries as patients. Current evidence indicates that this is no longer a significant problem. However, in 1994, this program provided almost $100 million in bonuses to physicians in all HPSAs, despite the lack of evidence that Medicare beneficiaries in these areas have difficulty obtaining access to care. GAO believes the use of this program as a remedy for underservice merits close scrutiny. SYSTEMS DO NOT MERIT UPGRADING -------------------------------------------------------- Chapter 0:4.4 The Department has efforts under way to address some of the problems with the two systems, but for several reasons, GAO questions whether these efforts will provide significant benefits to the federal programs using them. First, while the proposed changes may streamline the systems' administrative processes, the more significant problems of identifying underserved populations and the type of federal assistance needed will remain. Addressing these problems could be difficult and costly, in part because the data needed to verify primary care capacity in locally defined service areas are often unavailable at the national level. Second, for all programs except the Medicare Incentive Payment program, the Department already has alternative criteria and application processes in place that would appear to be more easily modified for targeting federal resources. Third, neither of these systems is a suitable match for the Medicare Incentive Payment program because neither specifically identifies or addresses Medicare-related demographics. Department officials said that maintaining a national shortage designation system for some other purposes (such as general planning and monitoring with regard to health care shortages) would be useful. However, the Department has another effort under way that may address these issues. A cooperative agreement between the Department and the states has provisions for identifying underserved populations and improving their access to existing health care delivery systems by integrating federal assistance with state and local resources. RECOMMENDATIONS TO THE CONGRESS ---------------------------------------------------------- Chapter 0:5 GAO recommends that the Congress remove legislative requirements for HPSA or MUA designations as a condition of participation in federal programs. Instead, GAO recommends that the Secretary of Health and Human Services be directed to develop and use program-specific criteria that will best match the type of program strategy with the type of access barrier existing for specific underserved populations. GAO also recommends that the Congress direct the Secretary to suspend funding for the Medicare Incentive Payment program until the Department can ensure that funding is specifically targeted to situations in which Medicare beneficiaries have a demonstrated difficulty accessing a physician because of low Medicare reimbursement rates for primary care services. AGENCY COMMENTS ---------------------------------------------------------- Chapter 0:6 GAO requested written comments on a draft of the report from the Department of Health and Human Services, but did not receive the comments in time for publication. However, GAO discussed an earlier draft of the report with the Department's management officials responsible for the HPSA and MUA systems. These officials offered observations about GAO's analysis and findings, and their comments were incorporated as appropriate. INTRODUCTION ============================================================ Chapter 1 The Department of Health and Human Services (HHS) spends about $1 billion a year on programs for improving access to health care for areas with shortages of primary care physicians and health care services. Many of these programs depend heavily upon systems to identify and designate specific areas and populations that are underserved. HHS has two such systems: The Health Professional Shortage Area (HPSA) system identifies underservice caused by a shortage of health professionals, and the Medically Underserved Area (MUA) system more broadly identifies areas and populations not receiving adequate health services for any reason, including provider shortages. About 88 percent of all U.S. counties contain HPSAs, MUAs, or both (see fig. 1.1). Figure 1.1: U.S. Counties With HPSAs and MUAs (See figure in printed edition.) (See figure in printed edition.) Source: GAO analysis of data from HHS' HPSA and MUA databases. DESCRIPTION OF THE HPSA AND MUA SYSTEMS ---------------------------------------------------------- Chapter 1:1 A primary care HPSA is an area designated by HHS as having a critical shortage of primary health care providers.\1 These include areas with no providers or areas with an insufficient number of providers to serve the population living there. Designation as a HPSA is generally based on the following: the specified geographic area must be rational for the delivery of health services;\2 the area must have a population-to-provider ratio of at least 3,500 to 1 (or 3,000 to 1 under certain circumstances); and adjoining areas must have provider resources that are overused, more than 30 minutes travel time away, or otherwise inaccessible. In 1994, there were 2,538 primary care HPSAs reporting a need for 5,133.8 full-time-equivalent physicians. About 45 million people reside in these HPSAs. HHS designates primary care HPSAs in one of three ways: a general shortage of providers within a geographic area, such as an entire county or group of census tracts; a shortage of providers willing to treat a specific population group (such as poor people or migrant farmworkers) within a defined area; or a shortage of providers for a public or nonprofit facility such as a prison or hospital.\3 As shown in figure 1.2, most primary care HPSAs are geographically designated. Of the geographic HPSAs, 845 comprised an entire county and 1,107 comprised other types of self-defined geographic service areas.\4 Figure 1.2: Types of Primary Care HPSA Designations (See figure in printed edition.) Note: Percentages do not add to 100 because of rounding. Source: GAO analysis of data from HHS' HPSA database. An MUA is an area designated by HHS as having a shortage of health care services. One major difference between an MUA and a HPSA is that underservice in a HPSA is measured primarily as a shortage of health care providers, while underservice in MUAs is measured using other factors as well. Qualification as an MUA is based on four factors of health service need: primary care physician-to-population ratio, infant mortality rate, percentage of the population with incomes below the poverty level, and percentage of the population aged 65 and older. As of June 1995, 1,455 U.S. counties were designated in their entirety as MUAs, and an additional 1,037 counties had at least 1 MUA designated within them. According to HHS officials operating the system, there were about 3,100 MUAs in all. Like HPSAs, MUAs can be designated for all people within a geographic area or can be limited to a particular group of underserved people within the area. Most MUAs have been designated for geographic areas rather than population groups (exact figures are not available). Unlike the HPSA system, however, the MUA system does not allow individual facilities to be designated as underserved. -------------------- \1 Separate HPSA systems are used to identify and track provider shortages for dental and mental health care. \2 For example, a rational service area may be defined as a county, or group of contiguous counties, whose population centers are within 30 minutes travel time of each other. \3 Hospitals are only eligible for designation when they have insufficient resources to treat underserved populations from an existing area or population HPSA. \4 A geographic service area can be a portion of a county, portions of multiple counties, or an urban neighborhood. HOW THE HPSA AND MUA SYSTEMS ARE USED ---------------------------------------------------------- Chapter 1:2 The current HPSA system was developed in 1978 as a means to designate areas for placement of National Health Service Corps (NHSC) providers. NHSC awards students scholarships or loan repayment for medical education and training in exchange for service in areas with critical physician shortages.\5 The types of primary care health professionals that could participate in the program included physicians, nurse practitioners, physician assistants, and certified nurse-midwives. In 1994, NHSC reported having 1,147 physicians and 482 physician assistants, nurse practitioners, and nurse-midwives working in HPSAs. Besides NHSC, two other programs now require that a location be designated as a HPSA to be eligible for participation. Like NHSC, the Community Scholarship Program addresses provider shortages by awarding grants to HPSAs for local scholarships in the health professions. The Medicare Incentive Payment program attempts to ensure that physicians treat Medicare patients by paying a 10-percent bonus on all Medicare billings generated from a practice located in a geographic HPSA. The MUA system was developed about the same time as the HPSA system but independently of it. Authorized by the Health Maintenance Organization Act of 1973, the MUA designation has been applied primarily in identifying areas eligible to participate in the Community Health Center program. This program awards grants for the operation of community health centers and migrant health centers in qualifying areas. In fiscal year 1994, HHS provided support for about 627 grantees providing services at more than 1,600 sites. Centers that serve a designated MUA area or population also are eligible for cost-based reimbursement under the Medicare and Medicaid programs. Another 100 health centers (the so-called "look-alikes") meet all requirements of the Community Health Center program and receive Medicare and Medicaid cost-based reimbursement, but do not receive Community Health Center grant support. Nearly 30 other programs use the HPSA or MUA systems to some degree, though none rely on a HPSA or MUA designation alone to decide who can apply for federal assistance. For example, nonphysician providers may qualify for cost-based reimbursement under the Rural Health Clinic program if they are located in a state-defined underserved area, HPSA, or MUA. The remainder of programs in this category are health professions education and training programs. Programs under titles VII and VIII of the Public Health Service Act give funding preference to schools that place graduates in medically underserved communities. Other programs, using various designations of underservice, award scholarships or grants for obligated service or training. Together, the various programs that use the HPSA and MUA systems accounted for more than $1 billion in funding and expenditures in fiscal year 1994. Table 1.1 summarizes the various programs and their funding levels. Table 1.1 Federal Programs Allocating Funds Using the HPSA and MUA Systems Number of individu al Fiscal year Basic program Benefit provided programs 1994 funding ---------------- ---------------- -------- ------------ Programs requiring HPSA designation to apply for federal funds ---------------------------------------------------------- National Health Awards 3 $126,720,000 Service Corps scholarships and provides loan repayment for service in a HPSA. Medicare Provides 10- 1 98,332,938 Incentive percent bonus Payments payment on all program Medicare billings in geographic HPSA. Community Awards grants to 1 478,000 Scholarship HPSA Program communities for health professions scholarships. Programs requiring MUA designation to apply for federal funds ---------------------------------------------------------- Community Health Awards grants 1 663,000,000 Center program for operation of community health centers. Federally Provides cost- 1 \a qualified based health center reimbursement "look-alike" for Medicare and Medicaid services provided by a federally qualified health center. Programs requiring HPSA, MUA, or some other designation as a medically underserv ---------------------------------------------------------- Title VII/VIII Training 24 151,834,000 Health programs may Professions provide Education and preference or Training Grant priority to Programs schools placing graduates in underserved communities. Rural Health Provides direct 1 77,010,536 Clinic program cost-based reimbursement for Medicare and Medicaid services provided by nurse practitioners, certified nurse- midwives, and physician assistants. Indian Health Scholarships for 1 7,702,000 Professions service in HPSA Scholarship or other locale Grant Program with large Indian population. Title III Mental Trainees perform 1 5,943,000 Health Clinical obligated Traineeship service in HPSA or other site with specific need for psychiatric services. Title X Family Gives priority 2 4,500,000 Planning for grants to Services institutions Training that place Program providers in HPSAs. ========================================================== Total 36 $1,135,520,4 74 ---------------------------------------------------------- \a Sources at the Health Care Financing Administration told us that funding for federally qualified health center "look-alikes" could not be broken out from funding for all federally qualified health centers. -------------------- \5 Federal intervention was considered justified only if the number of health care providers was significantly less than adequate, indicating that the needs of these areas were not being met through free market mechanisms or reimbursement programs. DESIGNATING AND UPDATING HPSAS AND MUAS ---------------------------------------------------------- Chapter 1:3 Any person, agency, or community group may request designation of an area, population group, or facility as a HPSA. Copies of each new request are received and reviewed by the Bureau of Primary Health Care's Division of Shortage Designation (DSD). State representatives from the health department, medical society, and the governor's office are also asked to review and comment within 30 days. DSD staff then check the application data against national and state sources. They also resolve conflicts among applicants, commenters, and data sources to the extent possible. HHS must by law review annually each designated HPSA to decide if it is still experiencing a shortage of health care providers. DSD does this by giving a list of HPSAs to each state and asking the state to update the information. In addition, Bureau policy requires HPSAs to provide data to DSD every 3 years to support their continued need for the designation. HPSAs not providing these updates are to be proposed for dedesignation in the Federal Register. MUAs are designated on a much different basis. The Department of Health, Education, and Welfare\6 designated the original lists in 1975 and 1976 by applying the four criteria (population-to-physician ratio, infant mortality rate, poverty rate, and percentage of population that is elderly) to all U.S. counties, minor civil divisions, and census tracts. All areas that ranked below the county median combined score for the four criteria were designated as MUAs. MUA designations have been added since then on the basis of newer data and the same cutoff score. Since 1986, HHS has also been able to designate new MUAs under an exception process if requested to do so by a state's governor on the basis of unusual local conditions. MUAs also differ from HPSAs in that there is no requirement to update the designations regularly. HHS officials managing the MUA and HPSA systems told us that DSD no longer reviews the list of MUAs to decide whether any should be dedesignated. -------------------- \6 The Department of Health, Education, and Welfare is the predecessor agency to HHS. HHS PLANS TO COMBINE AND REVISE THE TWO SYSTEMS ---------------------------------------------------------- Chapter 1:4 HHS has an effort under way to combine and revise the HPSA and MUA systems. According to HHS officials, this action is being taken to reduce redundancies and differences in the application and administrative processes of the two systems. While HHS officials told us that no changes would be made before 1996, a draft working document says that HHS' goal is to replace the existing systems with one that is consistent for all primary care programs, has simpler data-gathering requirements, uses relevant indicators of need, and will not disrupt services in existing areas. OBJECTIVES, SCOPE, AND METHODOLOGY ---------------------------------------------------------- Chapter 1:5 We reviewed the HPSA system as part of the broad federal effort to improve access to care, and as a follow-up to a congressionally mandated review of the role of federal health education and training programs in achieving this purpose.\7 While separate HPSA systems identify and track provider shortages for primary care, dental care, and mental health care, our review of the HPSA system focused on primary care HPSAs. We chose this focus because the HPSA primary care system is by far the most heavily used for identifying areas eligible for federal funds. We included the MUA system in our review because of current HHS efforts to combine it with the HPSA system. To review the extent to which the HPSA and MUA systems identify areas with primary care shortages, we reviewed past evaluations of the criteria and methodology for designating primary care HPSAs and MUAs and discussed the results with responsible HHS officials, identified the number of primary care providers in HPSAs and compared it to the number reported by the HPSA system, selected a random sample of primary care HPSA applications and reviewed whether designations were appropriate and accurately reflected in the HPSA database, and compared how often the HPSA and MUA data were updated with requirements in the law and HHS policy. To determine the extent that the HPSA and MUA systems provide information needed to target federal funding appropriate to meet the needs of underserved populations, we analyzed the types of designations requested by communities for their underserved populations and determined the extent to which the designations identify who is underserved in each HPSA and the reasons for underservice. To determine whether proposed changes to the HPSA and MUA systems would improve them, we discussed the purpose of the proposed changes with HHS representatives operating the HPSA and MUA systems. We also asked federal, state, and program participants how much the proposed changes would help them identify underserved populations and provide assistance appropriate to meet their needs. Further details of our scope and methodology are presented in appendix I. We did our work from November 1994 through June 1995 in accordance with generally accepted government auditing standards. We requested written comments on a draft of this report from HHS, but we did not receive them in time for publication. We did discuss an earlier draft of the report with HHS management officials responsible for the HPSA and MUA systems. They made observations about our analysis and findings, and we incorporated their comments in the report where appropriate. -------------------- \7 Health Professions Education: Role of Title VII/VIII Programs in Improving Access to Care Is Unclear (GAO/HEHS-94-164, July 8, 1994). SYSTEMS DO NOT RELIABLY MEASURE THE EXTENT OF PRIMARY CARE SHORTAGES ============================================================ Chapter 2 Neither the HPSA nor the MUA system reliably measures the extent of shortages in primary health care, providing little assistance to federal programs in directing the $1 billion spent each year for alleviating underservice. We identified two main reasons for the lack of reliability. First, both systems have methodological problems, such as omitting important categories of primary care providers from the calculations. In the HPSA system, for example, these omissions may be overstating the need for additional physicians in shortage areas by 50 percent or more. Second, both systems rely on data that are often inaccurate or outdated. On the basis of these data problems alone, we estimate that about 20 percent of HPSA designations are in error or lack adequate supporting documentation. Although we did not develop such an estimate for MUAs, many MUA designations are very old and may be invalid. For example, about half of the U.S. counties designated as MUAs would no longer qualify for designation if updated using 1990 data. SYSTEMS DO NOT CONSIDER ALL PRIMARY CARE RESOURCES IN MAKING SHORTAGE DETERMINATIONS ---------------------------------------------------------- Chapter 2:1 Both systems rely on a population-to-physician ratio in establishing the need for additional primary care providers. The HPSA system bases its shortage determinations on a population-to-primary care physician ratio of 3,500 to 1,\8 which identified a need for 5,134 physicians in shortage areas in 1994. The MUA system, which uses a population-to-primary care physician ratio as one of four factors in its underservice score, is less dependent on the ratio. However, in making their calculations, both systems exclude two categories of primary care physicians already providing services in the shortage areas: NHSC and federal physicians. The systems exclude federally salaried NHSC providers and privately salaried providers who are fulfilling an NHSC service obligation in exchange for health professions scholarships or loan repayment. There were 1,147 such physicians in 1994--the equivalent of about 22 percent of the shortage identified in HPSAs. Other providers employed by federal entities such as the Indian Health Service and the Bureau of Prisons are also excluded. There is no centralized accounting for the total number of federally salaried physicians. U.S.-trained foreign physicians with J-1 visa waivers. Such waivers allow noncitizens who complete their residency training in the United States to remain and practice if they are needed in underserved areas.\9 While the total number of such physicians practicing in underserved areas is unknown, their numbers are substantial. For example, the Appalachian Regional Commission and the Department of Agriculture approved at least 538 J-1 visa waivers for foreign physicians willing to practice in shortage areas in 1993 and 1994 alone.\10 This is equivalent to about 10 percent of the reported shortage of providers in HPSAs. Both systems also exclude several other categories of providers that deliver primary care services: Nonphysician providers. These include nurse practitioners, physician assistants, and nurse-midwives. Comprehensive data on the number of such providers in HPSAs and MUAs are not available. However, NHSC reported having 485 physician assistants, nurse practitioners, and nurse-midwives of its own practicing in HPSAs in 1994. Data provided by health professional associations in 1993 showed at least 369 nonfederal physician assistants and nurse-midwives practicing in HPSAs.\11 In total, these two groups may be the equivalent of between 8 and 17 percent of the shortage reported by the HPSA system.\12 Specialist physicians who provide primary care. Other research shows that specialists such as general surgeons may provide a substantial amount of primary care in areas where the population base is insufficient to support a full-time specialty practice. Further, a 1991 study illustrated that the availability of a full range of specialists to rural communities almost doubles the number of people needed to support a family physician practice from 2,000 to 3,990.\13 In urban areas, the oversupply of physicians in various specialties is reportedly causing them to provide an increasing amount of primary care services.\14 Current data on the extent to which specialist physicians are providing primary care in HPSAs and MUAs are not available. However, our review of a sample of 23 single-county HPSAs showed that most had specialist physicians in addition to primary care physicians providing patient care, averaging 1 physician for every 1,968 people.\15 HHS has various reasons for excluding these categories of health care providers. These reasons were published in the Federal Register in 1980 and 1983 to explain or clarify the HPSA criteria and were confirmed by more recent discussions with HHS officials. HHS' rationale for excluding NHSC and federal providers is that they probably would not serve in the HPSA without the service obligation or federal employment, and counting them could cause a community to lose its HPSA status. For similar reasons, HHS regulations exclude foreign physicians unless they are permanent residents. Although HHS originally planned to count nonphysician providers as 0.5 of a physician full-time-equivalent in the HPSA population-to-provider ratio, it excluded them from the final methodology because their scope of practice varies by state, and communities using them for care may be penalized in trying to establish a rural health clinic.\16 HHS does not count specialist physicians because HHS believes the law allows it to count only primary care physicians. While we understand HHS' rationale for excluding these providers from the designation calculations, we do not agree with it for several reasons. Omitting these providers has such a substantial cumulative effect that the true extent of primary care available in underserved areas cannot be determined if they are excluded. If the 1,147 NHSC physicians, 538 J-1 visa waiver physicians, and 854 physician assistants and nurse-midwives mentioned earlier were included in the HPSA calculations, the reported need for additional providers would be reduced by up to 50 percent.\17 If more complete data were available for all provider categories, this percentage could be substantially higher. Excluding primary care providers from the system also makes it difficult for federal and state agencies to coordinate their efforts in addressing underservice. For example, in 1994, NHSC had 19 providers in West Virginia in response to the HPSA system's reported shortage of 54 primary care physicians there. However, this need for 54 physicians did not reflect the presence of the NHSC providers or that other federal agencies had also assisted in placing 97 foreign physicians in the state's HPSAs in 1993 and 1994 alone. Finally, understating the number of primary care providers severely limits the usefulness of the system as a screen to identify which communities should be eligible for additional program benefits. For example, NHSC records show that 15 percent of the 576 providers placed in HPSAs in 1994 were in excess of the number needed for dedesignation, while other HPSA vacancies went unfilled. The excess numbers of NHSC providers placed in these HPSAs ranged from one to six.\18 -------------------- \8 This ratio can be dropped to 3,000 to 1 in areas where high need is indicated, such as areas with high poverty or high infant mortality rates. \9 Under a J-1 visa, foreign medical graduates can enter the United States for residency training if they return to their own country to practice medicine for at least 2 years after their training is completed. In effect, the waiver cancels the requirement to return to their own country, allowing them to practice in the United States for up to 6 years under a renewable visa until permanent resident status is achieved. \10 The Appalachian Regional Commission and the Department of Agriculture requested the largest number of J-1 visa waivers, according to officials of the United States Information Agency (USIA) administering the waiver process. While the agencies' records showed that 538 waivers had been approved, data from USIA showed an additional 181 physicians that had not yet been entered by the Department of Agriculture in its database. See appendix I for further discussion. \11 Data from the American Academy of Physician Assistants and the American College of Nurse Midwives showed that at least 4,203 physician assistants and certified nurse-midwives were practicing in counties with HPSAs in 1993. However, our analysis includes only those practicing in single-county HPSAs because the number providing care to underserved populations in other types of HPSAs is unknown. \12 Opinions differ on the physician full-time-equivalency that should be attributed to nonphysician providers. Counting these providers as a 0.5 full-time-equivalent in comparison to a full-time physician would result in a range of 8 to 17 percent. \13 L.L. Hicks and J.K. Glenn, "Rural Populations and Rural Physicians: Estimates of Critical Mass Ratios, by Specialty," Journal of Rural Health, Vol. 7, No. 4, Supplemental (1991). \14 See conference proceedings for the Health Resources and Services Administration National Conference at the Washington-Dulles Airport Renaissance Hotel, Virginia, on March 27-28, 1995, Estimating Medical Speciality Supply and Requirements in a Changing Health Care Environment: The Technical Challenge (Rockville, MD: Health Resources and Services Administration). See also B. Starfield, Primary Care: Concept, Evaluation, and Policy (Oxford University Press, 1992). \15 Fifteen of the 23 HPSAs in our sample had specialist physicians in addition to primary care physicians with total population-to-physician ratios ranging from 300 to 1 to 3,298 to 1. \16 Rural health clinics may only be established in HPSAs, MUAs, or state-designated underserved areas. If counting nonphysician providers precludes designation, nonphysician providers would not be eligible to establish a rural health clinic and receive direct cost-based reimbursement from Medicare and Medicaid. Dedesignation does not affect rural health clinics once they are established. \17 As discussed in appendix I, this estimate was derived by subtracting the total number of providers practicing in HPSAs from the total number reported as needed by the HPSA system. The range is estimated at about 40 percent to 50 percent if physician assistants, nurse practitioners, and nurse-midwives are counted as 0.5 to 1.0 full-time-physician-equivalent. \18 Many NHSC placements were in areas needing less than 1 full-time physician to dedesignate the HPSA. In these instances, we rounded the needed full-time-equivalent to 1 before comparing the number of physicians needed in a community with the number NHSC placed there. SYSTEM METHODOLOGIES ARE ALSO FLAWED IN OTHER WAYS ---------------------------------------------------------- Chapter 2:2 Both systems have other problems with their methodologies that make it difficult to identify and measure underservice. For the HPSA system, an ongoing concern is that it does not assess the extent that existing primary care resources in the community are being used. For the MUA system, although a number of methodological weaknesses were reported in the past, the methodology has not been revised. HPSA METHODOLOGY DOES NOT CONSIDER THE EXTENT THAT AVAILABLE RESOURCES ARE BEING USED -------------------------------------------------------- Chapter 2:2.1 The HPSA methodology has no mechanism for measuring the extent that existing primary care resources are insufficient to meet the demand for care. HHS officials said that data for this purpose are unavailable using current sources and would exclude the health needs of people who cannot afford to seek care from a provider. However, past studies of the HPSA criteria and methodology have pointed out that such a mechanism is needed because many factors influence the extent that communities use primary care resources at a rate above or below the 3,500-to-1 ratio.\19 For example, one county in our sample was designated as needing 1 additional physician even though the HPSA application showed that 42 of the 80 physicians surveyed within that HPSA were willing to take new patients.\20 Of those willing to accept new patients, over half reported no patient waiting time for appointments--another indication of additional capacity. Assessing the extent that existing primary care services are insufficient to meet the demand for care would also provide a better indication of whether a provider shortage exists in these areas or whether there are other barriers in accessing existing primary care resources. In the previous example, only 15 of the 42 physicians with additional capacity would accept all new patients; the remainder would only accept patients with certain types of health insurance. In HPSAs such as this that appear to have barriers to accessing underutilized capacity, it may be more appropriate to give incentives for expansion of services rather than adding more providers. For example, states are increasingly placing Medicaid patients in managed care in an effort to make these underserved populations more attractive to the existing physician workforce. -------------------- \19 Evaluation of Health Manpower Shortage Area Criteria, Mathematica Policy Research, Inc., HRA contract #232-78-0156 (1980), and M.L. Berk, A.B. Bernstein, and A.K. Taylor, "The Use and Availability of Medical Care in Health Manpower Shortage Areas," Inquiry, Vol. 20, Winter 1983, pp. 369-80. \20 The methodology counts the full-time-equivalent each physician currently spends in patient care. This may understate physician resources in areas where physicians are willing to work full time, yet do not have enough patients to do so. MUA METHODOLOGY IS LIMITED IN ABILITY TO IDENTIFY UNDERSERVED AREAS -------------------------------------------------------- Chapter 2:2.2 The MUA methodology has a number of flaws that limit its ability to accurately identify geographic areas and populations that have the greatest shortages of health care services. The methodology--an index of medical underservice--was developed within a short time frame using a process that involved limited empirical testing. Because the developers could not agree on a definition of "medical underservice," a mathematical model was developed to predict experts' assessments of service shortages. Subsequent evaluations of the model, however, found little significant difference in the availability of health services between areas that were designated as MUAs and areas that were not. These evaluations, which pointed out other methodological limitations as well, are summarized in appendix II. The MUA designation methodology has remained virtually unchanged since its development, despite improvements in U.S. health status and resources. The methodology uses the same four criteria to determine the MUA index score, and the cutoff score for MUA status (set at or below the median score for all U.S. counties in 1975) remains the same. The only changes to the methodology were made in 1981, when the weights for the infant mortality rate and the population-to-physician ratio were adjusted slightly. In 1986, the law was amended to allow the Secretary of HHS to designate MUAs that score above the cutoff, if the state's governor recommends designation based on "unusual local conditions which are a barrier to access or availability of personal health services." Since 1986, about 100 new medically underserved areas and populations have been designated on the basis of this exception process. SYSTEM DATA ARE NEITHER ACCURATE NOR TIMELY ---------------------------------------------------------- Chapter 2:3 Numerous problems exist with the accuracy and timeliness of the data used to obtain and maintain HPSA and MUA designations. Many HPSA applications do not contain the data necessary to support the designation, and data in the HPSA application often differ from those in the HPSA database. The reliability of the HPSA system is also compromised by data that have not been updated as required by law and HHS policy. For the MUA system, because it has no requirements for periodic review and updating, little has been done to keep the system's information current. DATA FOR HPSA DESIGNATION OFTEN INCOMPLETE OR INACCURATE -------------------------------------------------------- Chapter 2:3.1 We estimate that about 380 of the 1,952 geographic HPSA designations were made in error or without adequate supporting documentation.\21 HPSAs qualify for designation on the basis of three main factors: a population-to-primary care physician ratio that equals or exceeds 3,500 to 1, an insufficient number of providers in adjacent areas to provide care, and evidence of being a rational service area. Our review of a random sample of 46 geographic HPSA applications found 17 instances in which data in the file did not support one or more of these three factors. Examples follow: Substantial differences existed in the number of physicians reported by some communities and the number obtained by HHS's Division of Shortage Designation (DSD) from other sources. DSD verifies the number of physicians reported by applicants against data available from the professional associations. Any discrepancies and their subsequent resolution are required to be documented in the HPSA file. However, unresolved physician counts in some HPSA applications varied by as much as 50 percent. These differences were enough to preclude HPSA designation in each case. For example, one HPSA needing fewer than 8.7 physicians to qualify for designation reported having 7 physicians, while the American Medical Association directory showed 14 physicians practicing in the area. Some HPSA applications did not include data supporting that the number of providers in nearby communities was insufficient to provide care. DSD uses a population-to-physician ratio of 2,000 to 1 in contiguous areas within 30 minutes of travel time from the HPSA population center for this purpose. However, some HPSA files did not show that resources in all contiguous areas were considered. For example, one applicant reported the number of physicians in a town over 30 minutes away, but did not report the number of physicians practicing in a town within 30 minutes travel time. In some HPSA applications, there was no documentation to support the presence of rational service areas for primary care delivery. For example, one single-county HPSA was so large that distances between its population centers exceeded the 30-minute criteria for travel time to care. In another example, two separate service areas asked to be combined and enlarged to maintain the designation for one service area that no longer met the HPSA criteria. DSD was able to provide additional information to support 8 of the 17 designations GAO questioned. DSD officials said it was unclear why five of the remaining nine HPSAs had been designated, and said that they would follow up to resolve the discrepancies and propose dedesignations as necessary. In the other four cases, they provided additional information that we considered but still found to be insufficient to support the designation. We attempted to project the financial impact of federal funding provided to these areas, but because of program data limitations were unable to do so with an acceptable degree of statistical confidence. Another problem is that once the HPSA application data are verified, there is still no assurance that they will be entered or accurately reflected in the HPSA database. Of the 46 HPSA applications in our sample, 14 had discrepancies between the verified data and the data existing in the database for population, physicians, poverty rates, or differences in travel distances or times to the nearest source of care. Although these differences did not seem great enough to cause any of the 14 to lose their designation as a HPSA, some may be great enough to affect eligibility for placement of NHSC scholars or loan repayors.\22 We were unable to determine the effect these data entry errors or omissions had on the NHSC program because, as explained in the next section, HHS sometimes uses data other than those in the HPSA database to prioritize HPSAs for placement of NHSC providers. -------------------- \21 This is the median error using a 95-percent confidence interval. The range of error lies between 131 and 633 of the 1,952 geographic HPSAs. \22 Each HPSA is scored using four factors: population-to-primary care physician ratio, poverty rate, rates for infant mortality or low birth weight, and distance to care outside the HPSA. HHS establishes cutoff scores each year to determine which HPSAs are eligible for placement of NHSC scholars and loan repayors. HPSA SYSTEM DATA ARE NOT UPDATED ON A TIMELY BASIS -------------------------------------------------------- Chapter 2:3.2 HPSAs are not being reviewed on a timely basis to determine if they still qualify for federal assistance or should instead be dropped from designation. Federal law requires HPSA designations to be reviewed annually, a task that DSD has delegated to the states. However, DSD annually obtains data from the Bureau of the Census and the National Center for Health Statistics for many of the HPSA fields. DSD uses these current data instead of the older system data to identify which HPSAs have the greatest need for NHSC providers, but it does not use the current data to update its database. DSD officials said they did not use the data to update the database because doing so may cause some HPSAs to become dedesignated. DSD considers it inappropriate to dedesignate a HPSA until it can conduct a complete review of all data submitted by each HPSA during the formal update cycle. DSD's policy calls for each HPSA to submit an update application to them every 3 years for DSD review and verification that the HPSA designation is still valid. However, DSD is conducting these reviews, at best, every 5 years. Currently, about one-third of the HPSAs have not been updated in more than 3 years and should be updated or deleted, according to DSD policy (see fig. 2.1). Figure 2.1: Years Since Last Update of Individual HPSAs, as of August 1994 (See figure in printed edition.) Note: Percentages do not add to 100 because of rounding. Source: GAO analysis of data from HHS' HPSA database. DSD officials said they have extended the update period from 3 to 5 years because they have not been able to keep up with the backlog of HPSA applications. However, even the 205 HPSAs that did not reapply for HPSA designation within the past 5 years have not been dropped from the system.\23 Delaying the update process means that HPSA designation is continued for communities no longer requesting it. Communities generally do not request dedesignation when federal assistance is no longer necessary; instead, they simply do not reapply for designation during the update cycle. However, when the designation for these outdated HPSAs is still on the books, federal programs may continue to provide them with resources, perhaps to the detriment of those HPSAs with current designations. The following examples illustrate this problem: NHSC policy is to place providers in HPSAs updated in the last 5 years. However, in 1994, 9 percent of the NHSC providers were placed in HPSAs that had not been updated for 5 years or more. Twenty-three percent were placed in HPSAs that had not been updated in the past 3 years. Under the Medicare Incentive Payment program, Medicare pays bonuses to physicians in all designated HPSAs regardless of when the HPSA was last updated. Although we were not able to determine how much bonus money was paid in HPSAs that had not been updated in the past 3 years, more than $98 million was paid to physicians in HPSAs in 1994, and one-third of all HPSAs were more than 3 years old. -------------------- \23 As we were concluding our review, DSD officials told us they were in the process of proposing withdrawal of these designations. MUA DESIGNATIONS HAVE NOT BEEN REVIEWED SINCE 1981 -------------------------------------------------------- Chapter 2:3.3 There is no required schedule for periodically reviewing and updating MUA data and designations, and even less has been done to keep this system current than for the HPSA system. According to DSD officials, no systematic attempt to update the MUA designations has been made since 1981, when existing designations were reviewed against newer data. They told us that at that time, areas that no longer qualified as MUAs with the newer data were not always dedesignated, however, to avoid disrupting existing community health center services. Community health centers are required to serve MUAs or medically underserved populations to receive federal grant support. Essentially, once an area or population has been designated, it remains designated until the state's governor requests dedesignation. Since 1990, this has happened only once, when three counties in North Dakota were proposed for dedesignation in 1994. To show what might happen if designations were updated, we compared an application of the MUA methodology to 1990 data for all U.S. counties with DSD's 1995 list of MUA-designated counties. We found that about 740 counties would qualify as MUAs on the basis of 1990 data, compared to about 1,380 counties that DSD now has designated.\24 Although according to an HHS official there may be other reasons--such as continued eligibility for community health center funding--not to delete some old designations, maintaining the system with such obviously outdated information provides further evidence of the system's unreliability in identifying medically underserved areas. -------------------- \24 The total 1,380 DSD-designated MUA counties cited here differs from the 1,455 MUA counties mentioned earlier because of variations in how counties were defined in the two data sets we compared. To make the comparison, we excluded variations in Alaska and Virginia. See appendix I for details. CONCLUSIONS ---------------------------------------------------------- Chapter 2:4 The HPSA system does not accurately measure the existing capacity of communities to provide primary care services to its populations or the additional number of providers needed for this purpose. Shortages in many communities are overstated because the HPSA criteria do not recognize differences in the types of health care providers used to obtain care, or consider the extent that federal resources are already provided. The system's reliability is also questionable because HHS has difficulty verifying and updating the HPSA data in a timely manner. Continued reliance on the inaccurate and outdated MUA system likewise has resulted in designations that are not valid indicators of primary health service shortages, or where federal program funding is most needed. The next chapter discusses other aspects of the HPSA and MUA systems that hinder effective targeting of federal resources to the underserved. DESIGNATION SYSTEMS DO NOT PROVIDE INFORMATION NECESSARY TO TARGET FUNDING TO THE UNDERSERVED ============================================================ Chapter 3 Even when the HPSA and MUA designations identify needy areas, they generally do not provide the type of information needed by federal programs to target assistance best suited to meet a location's particular needs. Because most HPSAs are defined as general geographic areas, the designation does not identify the specific part of the population that has difficulty accessing a primary care provider or the underlying reason for this access problem. Similarly, although the MUA system can be used to designate specific underserved populations, most designations encompass everyone within a broad geographic area. As a result, federal programs relying on the designations to identify the type and scope of assistance needed may not provide assistance to those actually underserved in these areas. A case in point is the Medicare Incentive Payment program, which spent over $98 million in 1994 without any assurance that funds were used to improve access for Medicare beneficiaries in geographic HPSAs. DESIGNATIONS OFTEN DO NOT IDENTIFY DEMOGRAPHICS OF THE UNDERSERVED ---------------------------------------------------------- Chapter 3:1 Most HPSA designations do not provide information about the HPSA community beyond defining that a shortage of providers exists somewhere within the geographic area. Over three-fourths of all HPSAs in 1994 were geographically designated. Such a designation assumes that everyone within the general geographic area is underserved because the population-to-primary physician ratio exceeds a standard of 3,500 to 1. Only one-fourth of HPSAs were designated for specific types of underserved populations or the facilities that treat them. Unlike geographic designations, these designations provide some indication of the types of access problems that exist in the community. For example, there are seven categories of population-based HPSAs, primarily designated for specific poverty populations such as the homeless or Medicaid-eligible, but which also include designations related to cultural or language barriers experienced by migrant farmworkers or immigrants. Facility HPSAs are primarily used to designate shortages for prison populations but may also include public or nonprofit medical facilities. While designation as a geographic HPSA implies that federal assistance is needed to address access problems for all residents of the HPSA, HHS and state officials agree that specific subpopulations within the area may be those actually at risk. While HHS and state officials believe that underservice may affect entire populations living in areas with no physicians or in remote rural areas, only 12 percent of the underserved populations live in such areas.\25 The remaining underserved populations live in urban areas or rural areas nearby. Access in these areas may more likely be a problem for specific subpopulations, such as the poor. The MUA system has similar problems. By combining the weights for four factors into a single MUA score, the system produces scores that are difficult to interpret and tend to obscure an area's specific needs. While communities may request designations for specific populations with shortages of health care services, DSD officials told us this medically underserved population (MUP) designation was not used much until the 1980s. Following program amendments in 1986 that permitted state governors to request designations, about half of those new designations have been for underserved populations. -------------------- \25 This percentage includes HPSAs the system identifies as having no physician full-time-equivalents and rural HPSAs that are not identified as being located near an urbanized area as defined by the Department of Agriculture's urban-rural continuum codes. EXISTING DISINCENTIVES FOR IDENTIFYING UNDERSERVED POPULATIONS -------------------------------------------------------- Chapter 3:1.1 While the HPSA system allows designation for various types of underserved populations, there are several disincentives to request them instead of the geographic designation. First, communities with geographic designations can participate in all federal programs, while the programs available to population HPSAs are more limited. For example, the 10-percent bonus on Medicare billings is available to all physicians in a geographic HPSA, but not to those providing care in HPSAs designated on the basis of a poverty population. Second, the application process for population designations takes longer and is more difficult. Population designations require the applicant to conduct a physician survey to determine the proportion of services available to the underserved population and to explain why access to care is a problem. These requirements do not exist for geographic designations, which must only provide a population-to-physician ratio.\26 Finally, individual program requirements for geographic HPSAs are more flexible. For example, in HPSAs designated for poverty populations, 80 percent of the patients treated by an NHSC provider must live below the poverty level, but in a geographic HPSA, NHSC providers can treat anyone living within the defined geographic area. To ensure access to the broadest range of federal assistance, HHS officials encourage communities to use the geographic designations if possible, even when a specific underserved population can be identified. As a result, population HPSAs appear to be designated only as a last resort for communities not meeting the criteria for geographic designation. Our review of HPSA withdrawals and designations made in 1993 also showed that population designations are often used to maintain HPSA designation for areas no longer qualifying on the basis of geography. For example, of the 66 HPSAs that lost their geographic designation in 1993, about a third were redesignated on the same day as population HPSAs. -------------------- \26 Communities applying for geographic designation are required to conduct a physician survey if the number of physicians exceeds the minimum standard. In such cases, the community must demonstrate that it falls below the standard because some physicians are working only part time. DESIGNATIONS HAVE NOT CHANGED TO REFLECT THE NEEDS OF PROGRAMS USING THEM -------------------------------------------------------- Chapter 3:1.2 The general nature of most designations does not reflect the need of many federal programs to target assistance to specific populations or circumstances. Over the years, a variety of federal assistance programs have been created to address underservice identified by the HPSA and MUA systems. Initially, these programs served a broad purpose, requiring only that the HPSA and MUA systems designate the geographic areas that required additional providers or services. The NHSC program, for example, placed providers in all types of urban and rural shortage areas, regardless of who was underserved or whether underservice was caused by an undesirable geographic location, an inability to support a physician practice because of sparse or poor populations, or cultural or language differences of migrant farmworkers or immigrants. As new programs were added, they became more specific about the types of populations they served and the scope of assistance they provided. An example is the Medicare Incentive Payment program, which was expected to assist Medicare patients having difficulty obtaining access to a physician because of the low reimbursement rates for primary care services. Another example is the Rural Health Clinic program. Recognizing that many isolated rural communities are unable to support a physician practice, this program provides cost-based Medicare and Medicaid reimbursement to nonphysician providers such as nurse practitioners and physician assistants providing care in these areas without direct physician supervision. However, the HPSA designation system has not been changed to serve the narrowed scope of these programs. This has raised concerns that programs using the geographic designations to determine the type and scope of assistance needed in communities, instead of identifying the specific needs of underserved population within them, may result in misdirecting hundreds of millions of dollars in program resources. This change over time is of less concern with regard to MUA designations, because fewer new programs use them. Moreover, the Community Health Center program, which is the chief user of the MUA system, relies on MUA designations only as a screen for eligibility to apply for funding, according to the program's Director. Grant funds to support existing and new community health centers are allocated on the basis of reported performance and detailed need criteria within the community. Outdated MUA designations still may be used, however, to certify rural health clinics in areas that no longer have serious health service shortages. PROGRAM INTERVENTIONS MAY BE MISDIRECTED ---------------------------------------------------------- Chapter 3:2 When the access problems of specific underserved populations are not identified, it is difficult to determine what kind of federal intervention would be effective--and conversely, to avoid funding "solutions" that do not address the real need. In regard to the MUA system, for example, a study published shortly after its implementation expressed concerns that the methodology did not adequately capture variations in ability to obtain physician services between rural and urban areas, and among populations of different racial and cultural compositions. Consequently, the study concluded that programs using the MUA system could misallocate resources away from those most in need of federal assistance.\27 According to HHS officials operating the HPSA system, they are responsible only for determining whether primary care physician shortages exist. The specific programs using the HPSA system should determine who is underserved in geographic HPSAs and whether their programs are appropriate to address the access problems that exist there. However, to date the programs have relied on the HPSA designations for this purpose and have not developed mechanisms to determine whether their strategies are appropriate for the underserved population in each HPSA. They have not targeted or tailored their programs for individual HPSA needs. Some examples follow. The NHSC program requires that providers placed in HPSAs serve 80 percent of the HPSA population. While designations for the Medicaid or migrant populations require that these specific populations be treated, there is no mechanism to ensure that these same populations would be identified and treated by an NHSC provider in a geographic HPSA.\28 The Medicare Incentive Payment program pays all physicians in geographic HPSAs a 10-percent bonus on Medicare billings even if Medicare patients are not those actually underserved in the HPSA, and even if low Medicare reimbursement rates are not the cause of underservice. The Rural Health Clinic program provides cost-based reimbursement for Medicare and Medicaid services provided in any rural HPSA or MUA, even if the rural health clinic will not accept the entire HPSA population as patients. According to program managers at the Health Care Financing Administration, there is no requirement to distribute rural health clinic services throughout the underserved area or for rural health clinics to accept patients regardless of ability to pay for services. We did not directly audit these programs to determine the extent that program controls were adequate to prevent misdirection of resources for underserved populations in HPSAs and MUAs. However, we did find evidence that such problems exist--especially in the case of the Medicare Incentive Payment program. -------------------- \27 John E. Kushman, "The Index of Medical Underservice as a Predictor of Ability to Obtain Physicians' Services," American Journal of Agricultural Economics (Feb. 1977), pp. 192-7). \28 While NHSC providers in a geographic HPSA must agree to see anyone requesting care, they are not required to seek them out as patients. Therefore, their location or hours of practice may be a barrier to access for some underserved populations, such as migrant farmworkers. MISAPPLICATION OF THE MEDICARE INCENTIVE PAYMENT PROGRAM -------------------------------------------------------- Chapter 3:2.1 At present, there is no evidence that the Medicare Incentive Payment program is targeted to improve access to care for Medicare beneficiaries, even though over $98 million was paid to physicians in 1994 for this purpose. Neither the HPSA system nor the program identifies the extent that Medicare beneficiaries are underserved in geographic HPSAs or that low reimbursement rates cause access problems for them. The Medicare Incentive Payment program was established in 1987 subsequent to concerns expressed by the Physician Payment Review Commission\29 that low Medicare reimbursement rates for primary care services may cause access problems for Medicare beneficiaries in rural HPSAs. Under the program, all physicians providing services to Medicare beneficiaries in a rural or urban geographic HPSA are eligible for a 10-percent bonus on Medicare billings. The premise on which this program was created may no longer be valid in that the basis for Medicare reimbursement has changed since 1987.\30 In its 1995 report to the Congress, the Physician Payment Review Commission found no evidence that provider shortages or low Medicare reimbursement rates cause health care access problems for beneficiaries in rural areas.\31 Close to half of the $98 million spent under the program in 1994 was paid to about 82,000 rural physicians. While the Commission found some evidence of a link between living in urban HPSAs and access-to-care problems, beneficiaries cited the cost of services not covered by Medicare and a lack of transportation as the primary causes of access difficulties. These problems are unlikely to be solved by providing a bonus on Medicare billings. The remaining half of the $98 million spent under the program in 1994 was provided to about 96,000 physicians in urban areas. Further, the HHS Inspector General has questioned the appropriateness of applying the program in HPSAs because it provides bonuses to specialist physicians as well as primary care physicians, while the HPSA system only identifies areas with primary care physician shortages. The Inspector General reported that 45 percent ($31 million) of the Medicare incentive payments made in fiscal year 1992 went to specialist physicians who provided little or no primary care.\32 Among primary care physicians, the Inspector General concluded that Medicare incentive payments rarely have a significant effect on their decisions to practice in underserved areas. Bureau of Primary Health Care officials agreed that the HPSA system is not structured to effectively identify areas where the Medicare Incentive Payment program should be implemented. However, they do not believe they should modify the HPSA system for this purpose. Rather than add a designation for underserved Medicare populations, they suggested that the Health Care Financing Administration devise another system. While recognizing that the HPSA system is inappropriate, officials at the Health Care Financing Administration said that use of the HPSA system is mandated by law and that they do not have an alternative system that would effectively allocate funding under this program. -------------------- \29 The Physician Payment Review Commission was established in 1986 to advise the Congress on reforms in physician payment under the Medicare program. \30 The Health Care Financing Administration has been implementing changes to the physician fee schedule since physician payment reform measures were passed in the Omnibus Reconciliation Act of 1989. These changes generally increased reimbursement rates for primary care services and for services in rural areas. \31 The Physician Payment Review Commission found that Medicare beneficiaries, as a group, do not appear to have particular problems securing access to care. The Commission analyzed the results of the Health Care Financing Administration's 1993 Medicare Current Beneficiary Survey and reported that 97 percent of Medicare beneficiaries surveyed said they had no trouble with access to care during the previous year. The Commission did not address the continued need for the Medicare Incentive Payment program. However, it did support retaining and expanding the program in its 1994 report to the Congress. \32 Medicare Incentive Payments in Health Professional Shortage Areas: Do They Promote Access to Care? HHS Office of Inspector General, Report No. OEI-01-93-00050 (June 1994). CONCLUSIONS ---------------------------------------------------------- Chapter 3:3 While designating HPSAs on a strictly geographic basis may be appropriate for areas with no providers or rural areas remote from other sources of care, such a designation provides limited benefit in targeting assistance in areas where specific subpopulations are at risk. In addition, although