|
Medicare As A Catalyst For Reducing Health Disparities
As the nation’s largest purchaser and regulator of health care, Medicare is positioned to be a leader in reducing racial and ethnic health disparities. Its leverage was demonstrated in 1966—the year of Medicare’s inception—when hospitals desegregated as a condition for receiving Medicare reimbursement. Since then, Medicare has contributed to dramatic improvement in the health of the elderly and disabled minority population, although disparities between minority and white beneficiaries remain. A National Academy of Social Insurance study panel is exploring how Medicare could use its leverage to reduce disparities, for both its beneficiaries and the rest of the nation.
Medicare has been instrumental in reducing disparities in health coverage between racial and ethnic minority groups and whites. From birth to age sixty-five, racial and ethnic minorities are much less likely than whites to have health insurance. Upon reaching age sixty-five, virtually all Americans—white and minority—are eligible for Medicare, and minorities constitute a disproportionate share of those who qualify for Medicare on the basis of long-term disabilities, accounting for 32 percent of disabled beneficiaries under age sixty-five (compared with 18 percent of aged beneficiaries).1 Medicare also provides the same benefit package to all beneficiaries—rich and poor; minority and white. Furthermore, the vast majority of health care providers and institutions participate in Medicare, which, in theory, should help ensure that minority beneficiaries have equal access to services. Nonetheless, even among Medicare beneficiaries, marked disparities persist in treatment and health status, although they are smaller than the disparities that minority beneficiaries experience before becoming entitled to Medicare. The 2002 Institute of Medicine (IOM) report Unequal Treatment found sizable racial and ethnic health care disparities among Medicare beneficiaries, even after adjustment for socioeconomic differences and other health care access–related factors.2 Minority beneficiaries also fall short of whites on many measures of health status. Blacks, for example, have shorter life expectancy at age sixty-five than whites (by 1.8 years), and black and Latino beneficiaries are more likely than whites to have chronic conditions, such as hypertension and diabetes.3 Additionally, although many outcomes (such as life expectancy at age sixty-five) have improved for minority beneficiaries, the same outcomes have also improved for whites, so that the relative disparity has actually increased. Of course, much of the continuing disparity cannot be attributed directly to Medicare. Some disparities are the result of factors other than health care. Socioeconomic status (SES), for example, is highly correlated with health status, and minorities, on average, have lower SES. Substandard housing, poor nutrition, and smoking all have long-term impacts on minority groups’ health. So does inadequate access to health care prior to Medicare eligibility. In addition, Medicare and its providers are part of the U.S. health care system, not a distinct health system unto themselves. To the extent that biases pervade the system, they affect minority Medicare beneficiaries as well. How Medicare contributes to disparities. Some aspects of Medicare may, nonetheless, contribute to health disparities. Many minority beneficiaries face out-of-pocket spending for deductibles, coinsurance, and uncovered services, which they are less likely than whites to be able to afford. In addition, minority beneficiaries are less likely to have supplemental insurance, because fewer are covered by employer-sponsored retiree insurance, and Medigap is unaffordable for many.4 Moreover, various issues in the administration of Medicare and Medicaid potentially affect health disparities, "either by tolerating or tacitly countenancing access, treatment and quality differentials or by failing to act affirmatively to minimize the possibility of differentials."5 Those that may affect minority beneficiaries include the following. (1) Federal Medicare regulations that permit Medicare-participating physicians to select patients at will: Physicians participating in Medicare are not required to accept all patients, allowing for discrimination based on race, ethnicity, and supplemental insurance. Physicians and health plans are also not required to participate in Medicaid (although hospitals and nursing homes are), which could have adverse effects on minorities because of their higher representation in Medicaid. (2) The local medical review process under fee-for-service (FFS) Medicare (the review of claims by Medicare contractors to determine whether a service is covered) and the utilization review process of Medicare Advantage (MA, formerly Medicare+Choice) plans: These review processes may disadvantage minorities to the extent that people engaged in the process are unaware of, or unsympathetic to, the special needs of minority beneficiaries, or because minority beneficiaries are less likely to have providers prepared to advocate strongly on their behalf. (3) Eligibility requirements for the Medicare Savings Program and Medicaid: Although the Medicare Savings Program (Qualified Low-Income Medicare Beneficiaries and Specified Low-Income Medicare Beneficiaries) and Medicaid help low-income beneficiaries pay some or all of their Medicare premiums and deductibles, more than half of beneficiaries who are eligible are not enrolled.6 Many do not apply for these programs because they do not know about them, do not know where to apply, or are intimidated by the documentation required. Such barriers disproportionately affect minority beneficiaries, as they are more likely than white beneficiaries to qualify and may be more hesitant to apply at a government welfare office. The NASI study panel. Because of its conviction that reducing health care disparities requires continuing and focused attention, its commitment to social insurance programs, and its belief that Medicare can and should help improve the U.S. health care system, the National Academy of Social Insurance (NASI) has convened a study panel to address the issue of how Medicare might be changed to reduce disparities, for its beneficiaries and throughout the health care system.7 As of this writing, the study panel is still at a relatively early stage in its work but has already identified some preliminary conclusions and areas for further exploration.8 The NASI study panel believes that Medicare should take the lead in reducing disparities. As a social insurance program, Medicare should ensure that all beneficiaries who contributed to the Medicare Part A Trust fund, regardless of race or ethnicity, receive the best possible care. As a federal program, Medicare has a duty to uphold civil rights laws and ensure equal access to care for all. And these issues will become more important in the future, as the proportion of minorities among the elderly population, and thus among Medicare beneficiaries, continues to increase. By 2030 minorities are expected to account for 26 percent of the Medicare population age sixty-five and older.9
Medicare’s leverage as the largest U.S. purchaser and regulator provides a unique opportunity to reduce disparities. As a purchaser, Medicare provides coverage to forty-one million aged and disabled people, of whom one in five are members of racial and ethnic minority groups.10 Its reimbursement policies and coverage policies are widely imitated. Private insurers, for example, tend to follow Medicare approval for coverage of new medical technologies. Medicare has also been instrumental in moving the health care system toward prospective payment systems and electronic claims payments. In its regulatory role, Medicare influences the safety and quality of care, directly through its conditions of participation for most institutional providers and more indirectly through the accreditation standards of the Joint Commission on Accreditation of Healthcare Organizations (JCAHO). The Centers for Medicare and Medicaid Services (CMS) has also imposed a set of quality improvement standards in its requirements for the MA program, which is at least as rigorous as the most well-developed private accreditation standards (such as JCAHO and URAC).11 In the Part B FFS sector, Medicare relies less heavily on regulatory requirements, focusing instead on voluntary educational and feedback efforts and claims review, but the results there have been notable as well.12 Medicare’s administrative data set is the richest U.S. source of health care information. Although not as comprehensive clinically as one might like, this data structure is efficient and complete because it is developed from claims submitted for payment.13 The CMS also conducts the Medicare Current Beneficiary Survey (MCBS), which collects socioeconomic and health information from a sample of beneficiaries; the Health of Seniors (HOS) survey; and the Consumer Assessment of Health Plans (CAHPS) for MA and FFS Medicare, and it contracts for Medicare Health Plan Employer Data and Information Set (HEDIS), all of which have the ability to link with each other and the CMS’s administrative database to capture the broad picture of beneficiaries’ health. Although improvements in these data would greatly support further efforts, they are available to researchers and have been instrumental in documenting disparities as well as quality of care. The CMS also uses a variety of tools to test new ideas. For instance, its Cooperative Cardiovascular Project (begun in 1993) developed quality indicators and demonstrated improvements in adherence to clinical guidelines. Building on this project, the Quality Improvement Organizations (QIOs) worked in communities to target twenty-two quality indicators for improvement. Recent studies have documented the effectiveness of some of these efforts.14 Current projects include making bonus payments to hospitals with better health outcomes, to physician group practices to coordinate care to chronically ill beneficiaries, and to providers of capitated disease management services for use of evidence-based protocols.
Medicare’s history began with a monumental impact on disparities, as it forced hospitals to desegregate even before it began paying benefits. Until 1965 most hospitals across the county were segregated—de jure in much of the South, de facto in the rest of the country. Black physicians were not granted privileges in "white" hospitals, and black patients were not admitted in many "white" hospitals or were segregated by floor or room. Southern segregation was condoned and fostered by the Hill-Burton Act, which provided federal funds for the construction of racially separate hospitals. In 1965, as a condition of participation in Medicare—to be eligible for Medicare reimbursement—hospitals were required to desegregate. Although they met with some resistance, federal officials were steadfast in requiring hospitals to comply. More than 1,000 hospitals integrated their medical staffs, waiting rooms, and hospital floors in a period of less than four months.15 Nothing Medicare has done since has had this much impact on racial inequality, but a number of other efforts have been worthy of note. (1) The CMS has made great strides in improving the availability and accuracy of race/ethnicity data. Until 1993, "black," "white," and "other" were the only race/ethnicity designations in the CMS’s administrative database. In 1994 "Native American," "Asian American or Pacific Islander," and "Hispanic" were added, with a revision in 1997 that separated "Asian American or Pacific Islander" into two categories. Since most demographic data about Medicare beneficiaries are obtained and maintained through the Social Security Administration’s master beneficiary record file, the updating of data is a gradual process, but the CMS has made notable efforts to complete and update those data.16 (2) The CMS has charged the QIOs with designing interventions to reduce disparities among FFS Medicare beneficiaries. Each QIO is required to implement a project that identifies factors contributing to racial, ethnic, or rural disparities and design a plan to reduce them, for six targeted clinical conditions.17 Also, for its 2003 and 2004 national quality assessment and performance improvement project, the CMS required health plans to implement a quality improvement plan, with a focus on clinical health care disparities among racial and ethnic minority groups or culturally and linguistically appropriate services.18 (3) The U.S. Department of Health and Human Services (HHS) has also issued guidance to providers regarding patients of limited English proficiency. In 2000, under an Executive Order that every federal agency develop guidelines, HHS issued its own guidance document.19 According to the document, providers must provide "reasonable steps" to ensure meaningful access to their programs and activities by people with limited English proficiency.20 The guidance describes factors that providers should consider in determining and fulfilling their responsibilities to such people under Title VI of the Civil Rights Act.21 Although these and other CMS efforts are a step in the right direction, a strengthened, multiprong strategy will be necessary to greatly reduce racial and ethnic disparities in health.
How should Medicare focus its efforts? The NASI study panel is investigating opportunities for Medicare to address racial and ethnic health disparities through quality improvement, payment strategies, better collection and use of data, civil rights enforcement, and cultural competency and language initiatives. At the outset, though, it should be noted that the changes to Medicare that would have the most impact on reducing disparities would be general changes in its benefit and cost-sharing structures—which would have other positive impacts as well. In general, minority beneficiaries have poorer health status and lower incomes than white beneficiaries; hence, they would benefit disproportionately from general program improvements. These could include reducing or eliminating the Part B premium, deductibles, and copayments and enhancing Medicare’s benefit package to pay for services that Medicare does not cover (such as eyeglasses and long-term care). Also, because health care access and health status prior to Medicare eligibility are major determinants of beneficiaries’ health, improved access to care for other age groups should reduce health disparities. It could be argued that universal insurance for the entire population would have the largest impact on disparities of any strategy. In the absence of sufficient political will for universal coverage, incremental approaches such as making Medicare available to people ages 55–64, expanding the Medicare Savings Program, and ensuring the availability and affordability of Medigap coverage for disabled beneficiaries could all have a strong impact on disparities. These more modest measures would obviously have political and economic implications, but they would also benefit minorities and nonminorities alike. Quality of care. Even in the absence of systemic improvements in Medicare, however, many steps could be taken to reduce disparities. With the continuing erosion of the willingness of courts and administrative agencies to aggressively enforce civil rights statutes and the growing preoccupation with quality in the health care sector, existing efforts to reduce disparities have concentrated on quality-of-care strategies, and these approaches have considerable promise. Physician bias has been a major focus, because physicians, like the rest of society, hold stereotypes based on race and ethnicity, and a number of studies suggest that health care providers’ diagnostic and treatment decisions are influenced by these stereotypes.22 The IOM report Unequal Treatment recommends the use of evidence-based guidelines to promote consistency and equity of care.23 While improving care for all patients, evidence-based medicine may have the greatest impact on minorities, for whom the gap between the status quo and optimal care may be greatest. Nonetheless, efforts to achieve more widespread compliance with evidence-based guidelines have been halting and variably successful, and the belief that wider compliance will reduce disparities remains, to this point, a promising but minimally tested hypothesis. In this regard, making data available to the QIOs and other researchers should help evidence-based guidelines evolve. In addition, Medicare could improve the system within which evidence-based medicine is used. Such strategies might address information systems and other tools that support patient education, self-management, and disease management at the provider, practice, and health plan levels. They should ensure that minority beneficiaries have a designated primary care provider and that the primary care physician and specialists coordinate their care. The CMS could also support physician practices’ management of care, with specific strategies targeted to small to medium-size practices (33 percent of private-practice physicians work in solo practices and 26 percent work in practices of two to four physicians).24 Renewed emphasis on increasing beneficiary enrollment in MA provides further opportunity to evaluate the quality of care provided to minority beneficiaries in a managed care environment, as compared with FFS. Although the literature is mixed on whether managed care plans provide better quality of care than FFS Medicare, it appears that disparities in the use of preventive care services and primary care are reduced in managed care plans.25 MA plans may have greater capacity to serve minority beneficiaries because they can create systems to better coordinate care, notify members of the need for appointments, and provide health education. Nevertheless, quality of care must be monitored to ensure that minorities are not disadvantaged in obtaining services and treatments. Payment approaches. Payment approaches are another potential strategy for reducing health disparities. One option is to target diseases that are prevalent among minority beneficiaries (or procedures that are underused by minorities) and increase payments for such treatments. For example, reimbursement for office visits for diabetes care is relatively low, prompting complaints that reimbursement does not cover the time it takes to do high-quality patient and family history taking, education, and counseling. Under this rationale, raising reimbursement for an out-patient diabetes visit should result in better quality of care. A counterpoint, however, is that high reimbursement has not leveled the playing field for cardiovascular surgeries: Despite relatively high reimbursement rates for such procedures, minority beneficiaries are still undertreated compared with whites.26 Another possible approach is to pay bonuses to providers who meet indicators established specifically to reduce disparities. For example, to raise the percentage of minority beneficiaries who receive flu shots, the CMS could pay a bonus to providers for achieving specific goals. However, a number of issues would need to be addressed to make this work, including that fact that some physicians have a large number of minority patients, while others have very few; and some physicians have a substantial number of minority patients with lower socioeconomic status, while others have more middle-class minority patients. To ensure that the bonus approach does not penalize providers with large numbers of these two types of patients, risk adjustment is likely to be necessary. Data collection and usage. Data collection and analysis are central to reducing disparities. Health plans are not required to collect race, ethnicity, and SES information on their enrollees, and although HHS has the authority to require it, it has not done so.27 Examples of how data have been used to reduce racial and ethnic disparities are available from other sectors, such as mortgage lending and law enforcement. In the case of mortgage lending, for example, the 1975 Home Mortgage Disclosure Act (HMDA) required the collection of data on race on home mortgage applications. Until 1989 it had little effect on lending practices because the data were not widely publicized or collected in a publicly accessible format. In 1989, after HMDA was amended to make the data public, social scientists could assess the extent of disparities and advocacy groups and then put pressure on lenders. A recent study by Peter Bach and colleagues found that a small number of providers (22 percent) account for most (80 percent) physician visits by black patients and that these patients receive lower quality of care.28 This reinforces David Smith’s measures of the degree of segregation in health care (for hospital, outpatient, and nursing home care) and suggests patterns of segregation that have been the target of interventions in other areas, including housing and education.29 Medicare providers could also be required to examine enrollment and utilization patterns and take steps to improve access to their services. Additional conditions of Medicare participation could, for instance, require hospitals, health plans, and other providers to maintain data on patients’ race, ethnicity, and educational attainment (as a proxy for SES); analyze the data and select services or conditions prevalent among minority beneficiaries; and then establish a remedial strategy. HEDIS measures could be reported by race and ethnicity by health plan (they are not now). The CMS also could analyze its own administrative data and use the findings for programmatic and research purposes. Civil rights enforcement. Title VI of the 1964 Civil Rights Act prohibits discrimination by any program or activity receiving federal financial assistance. It applies to both intentional and unintentional discrimination on the part of physicians, providers, and other points in the health care system. It can be applied to a range of discriminatory concerns, including access to health care, redlining, physician staff privileges, participation in managed care, and interpreter services. Federal agencies are authorized to set nondiscrimination standards, investigate claims of discrimination, and terminate federal assistance to any entity found to have violated the law. Since the desegregation of hospitals in 1965, the federal government has not used civil rights laws aggressively to reduce disparities. It has never required the desegregation of Part A providers other than hospitals (such as nursing homes), and Part B providers are still not required to comply with Title VI, as the Medicare beneficiaries, not the physicians who care for them, are—at least in legal theory—the recipients of federal financial assistance. The HHS Office for Civil Rights (OCR) has kept a low profile these past three decades and has been underfunded and largely ineffective at driving change.30 In 2001 the Supreme Court Alexander v. Sandoval case eliminated the right of individuals to sue for unintentional discrimination, making federal agencies the sole enforcers of Title VI.31 Despite the increased role that the OCR would need to assume to maintain enforcement levels, it was given no additional resources. Although Title VI is the best known, other civil rights laws are also potential tools for reducing disparities. These include Section 504 of the Rehabilitation Act of 1973, which prohibits discrimination against people with disabilities by programs receiving federal financial assistance, and Title III of the Americans with Disabilities Act (ADA), which prohibits discrimination against people with disabilities in places of public accommodation. Unlike the Rehabilitation Act and Title IV, the ADA included physicians’ offices and other medical care providers in the "public accommodation" definition (without regard to whether the providers participate in federal programs). Both the Rehabilitation Act and the ADA permit enforcement through private rights of action as well as by government. The NASI study panel will consider whether Title VI could be revised using the ADA’s definition of public accommodation and whether Title VI should be amended to allow enforcement through private suits in disparate impact cases. It will also examine the Emergency Medical Treatment and Active Labor Act (EMTALA) as another legal avenue for reducing disparities. More aggressive use of state nondiscrimination laws to address disparities will also be addressed. Cultural competency and language. HHS’s initiatives on language competence in hospitals could also be part of a much broader effort to ensure some modicum of linguistic and cultural competency among providers participating in Medicare. Early efforts at training medical students and house staff in cultural competency have produced some promising results; what is now needed is to generalize from those limited experiences to a set of policies that could apply more broadly to the provider community.32 All-encompassing issues. An important question underlying these issues is whether Medicare should attempt to ensure parity for health outcomes as well as for health care. The study panel continues to debate this. It acknowledged that improved health outcomes is one of the goals, if not the ultimate goal, of health care and that attention to outcomes allows for programmatic targeting of serious and prevalent health conditions, the monitoring of the effectiveness of treatment, and the ability to modify programs accordingly. However, evidence connecting health care and outcomes is often tenuous, and because Medicare’s charge is primarily as a payer of clinical care, many of the factors that influence health outcomes are outside of Medicare’s scope. Many panel members believe that if Medicare adopted an overall strategy of encouraging population-based approaches to prevention, health promotion, and health maintenance, disparities in health care and health would be reduced. Another complex issue is whether and how Medicare should incorporate local diversity and heterogeneity into its national strategy. Under the assumption that a community-based approach is needed to effectively target minority populations, the CMS’s QIOs were charged with developing and implementing locally based projects intended to reduce health system and sociocultural barriers to health care for underserved populations.33 The QIOs engaged community, regional, and statewide partners to assist in the development of culturally appropriate educational materials and dissemination of intervention tools, with faith-based institutions used frequently for education. A report of the QIOs’ 1999–2002 efforts stated that although the work was time-consuming, improvement was seen at the local level in quality-indicator performance.34 Panel members agree that further thought and systematic research are needed to assess how Medicare could best target its efforts, given the character of local areas and the variation between and among different racial and ethnic groups. Medicare has not realized its potential as a catalyst in reducing health disparities. The CMS needs a comprehensive plan, including recommendations for improving data collection and analyses on disparities, systemic and sustained efforts to improve the quality of care provided to racial and ethnic minority groups, and the elimination of discriminatory practices—both intentional and unintentional. As history shows, race played—and continues to play—a decisive role in shaping the U.S. health care system and Medicare.35 Medicare has the responsibility and the potential to right these wrongs, for the sake of its beneficiaries, the health care system, and all Americans.
June Eichner (jeichner{at}nasi.org) is a senior research associate at the National Academy of Social Insurance in Washington, D.C. Bruce Vladeck is a principal at Ernst and Young in New York City. The study panel is funded with the generous financial support of the Robert Wood Johnson Foundation, the California Endowment, and the Kellogg Foundation through the Joint Center for Political and Economic Studies.
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Medicare’s Potential In...
Medicare’s Efforts To...
