Challenges to Reducing Discrimination and Health Inequity Through Existing Civil Rights Laws

Abstract

Fifty years after the passage of Civil Rights Act, minority healthcare remains separate and unequal. We combine insights from Civil Rights Law and research on racial-disparities to understand whether stronger enforcement of existing Civil Rights laws would improve minority healthcare today, or whether complementary approaches are also necessary. Despite earlier success, modern challenges to improving minority healthcare are different than those confronted during de jure segregation. We review these challenges and the potential effectiveness of existing Civil Rights legislation in overcoming them. We conclude that enforcement could be strengthened by executive orders that strengthen existing laws, but Congressional action would be required to allow private individuals to bring suits against discriminatory providers. We contrast the relative benefits of this approach to wider non-litigation-based solutions. We conclude that a combination of the two approaches would better address the challenge of improving minority healthcare in the 21^st century.

Fifty years after the passage of the Civil Rights Act, minority healthcare remains separate and unequal.^1,2 The Act achieved singular success during the 1960s, but the modern challenges to improving minority healthcare are different than those confronted during the era of de jure segregation. De facto segregation still survives: in Medicare, roughly 20% of U.S. hospitals treat 80% of all African American heart attack patients and 40% of hospitals have no African-American heart attack patients. One study estimates that over half of the overall racial disparity in survival after heart attacks may be attributed to the lower performance of minority-serving hospitals.³ Peter Bach and colleagues have noted that physicians who treat African American patients are less likely to be board certified and lack access to important resources.⁴ In addition to minorities being treated by different providers with fewer resources, the National Academy of Medicine (formally, the Institute of Medicine) has drawn attention to implicit provider bias arising from ‘bias, discrimination and stereotyping’ that may cause a provider to treat patients differently by race.⁵ In recent years, the Obama Administration initiated certain measures to strengthen enforcement of Title VI of the Civil Rights Act.^6,7,8 The Affordable Care Act (ACA) also attempted to update Title VI of the Civil Rights Act to account for the increasing integration between the financing and delivery of health care. Specifically, § 1557 of the ACA expanded federal civil rights protections to encompass Medicaid and Medicare Advantage plans. But many gaps remained even with the ACA—a point that is reiterated by the tremendous uncertainty surrounding the future of the ACA after the November 2016 election. As Congress reconsiders health-policy and contemplates enacting policies that aren’t necessarily tied to the ACA chassis, we explore whether greater enforcement of historical Civil Rights Laws and expansion of the scope of these laws-- are up to the task of improving minority healthcare today or whether additional approaches to take on modern-challenges, are also necessary.⁹

To inform this discussion, we review these challenges, and examine the potential effectiveness of litigation-based solutions such as greater enforcement and expansion of Civil Rights legislation, paying special attention to how the behavior of practicing physicians is evaluated by Courts. We contrast the relative benefits of this approach to wider non-litigation-based solutions—such as policies to provide greater resources to minority-serving providers—and conclude that a combination of the two approaches with a tilt towards the non-litigation solutions, would better address the challenge of improving minority healthcare in the 21^st century.

The Civil Rights Act: Background and Scope

Title VI of the 1964 Civil Rights Act prohibits race discrimination by private institutions that receive ‘federal financial assistance’. It is enforced by withholding federal funds from institutions that discriminate against persons on the basis of race, color or national origin. Given the Federal government’s role in the Medicare and Medicaid programs, the leverage and scope of Title VI legislation is, in principle, enormous. When these programs first started, their budgets were a fraction of what they are today. Yet the threat of being barred from these programs encouraged more than 1,000 hospitals to integrate their facilities within four months of enactment. The salutary effect on minority healthcare was extraordinary, even for minority populations not covered by Medicare.¹⁰ Integration is estimated to have saved 6,000 African American infants in the first five years after the reform.¹¹

The integration of hospital facilities in the 1960s only addressed the most blatant forms of discrimination. Other insidious forms remained, including the denial of admitting privileges to African American physicians, prepayment requirements for African American patients, and the relocation of inner city hospitals to predominantly white suburbs. In legal suits against discriminatory admitting privileges and prepayment requirements, plaintiffs challenged hospital policies that overtly distinguished between races. Civil right lawyers call these “disparate treatment” cases and they were very successful at winning them. In suits involving hospital relocations, however, plaintiffs did not challenge overtly discriminatory behavior, but rather activities that may not facially distinguish across races but that had a “disparate impact” on minorities. Here, litigation was far less successful.

The reason for the success of some suits and failure of others stems from the manner in which plaintiffs must satisfy their burden of proof in civil rights cases. At the start, the plaintiff must present some evidence suggesting discrimination. If the defendant—a physician or hospital--can then offer a non-discriminatory explanation for the evidence, the plaintiff must convince the court that the explanation is a pretext for discriminatory intent or, in the case of disparate-impact claims, that a reasonable result from the activity in question could have been achieved in a less discriminatory fashion. This can be a difficult gauntlet for plaintiffs, but perhaps far more so in the disparate-impact setting.^12,13,14 With disparate-treatment claims, plaintiffs do face some evidentiary burdens insofar as they must show (through direct or often indirect, statistical evidence) that race formed the basis of the differential treatment as opposed to some correlated characteristic of the plaintiff. However, their ability to satisfy the final prong—that is, to show that the defendant’s explanation is a pretext for discrimination—is made simpler by the fact that the provider actually treated blacks and whites differently, unlike the situation in disparate impact cases. The challenges facing plaintiffs in disparate impact cases are often insurmountable. This is especially true in cases involving issues of geography—e.g., in challenges to hospital relocations. Defendant hospitals in relocation cases argued that they moved because they were losing money in the inner city. Plaintiffs lost because they could not show that the hospital actually moved because it wanted to avoid minority patients or that the costs savings and quality improvements sought by the hospitals could have been attained through other means. Unfortunately, today’s obstacles to adequate minority healthcare more closely resemble cases involving disparate impact than disparate treatment.¹⁵

Benefits from Greater Scope and Enforcement

Given this legal background, is greater enforcement of existing Civil Rights laws the solution to improving minority care? We present four reasons to be wary, but we also point to ways in which Title VI and § 1557 litigation might be strengthened.

1. Scope of Civil Rights Law

Title VI legislation applies to all entities receiving “federal financial assistance”—and so includes hospitals, nursing homes, long-term care facilities. Importantly, it does not extend to physician care. Soon after the passage of Medicare, the Federal Government exempted physicians from the scope of the civil rights law by failing to classify Medicare Part B or Medicaid hospital payments to physicians as ‘federal financial assistance. At the time Medicaid and Medicare were enacted, this omission was minor: in 1970, physician and outpatient care accounted for 5.4 percent of spending on personal health care services, yet by 2010, they accounted for 40 percent of this spending.¹⁶ Although the statutory language of § 1557 of the ACA—which incorporates Title VI—suggests that federal civil rights protection extends to Medicare Part C and Medicaid, the implementing rule issued by the Department of Health and Human Services indicated that Medicare Part B would continue to not constitute “federal financial assistance,” 45 C.F.R. 92 (2016). Moreover, § 1557, may be in jeopardy if the ACA is repealed wholesale under a new Congress and Administration. As a result, minority patients may still be unable to successfully use Title VI to hold many physicians legally responsible for bias, discrimination and stereotyping during their care.

This gap in federal civil rights coverage means that a minority patient who is a reasonable candidate for, say, an angioplasty but does not receive one can sue the hospital, but not the physician, if she was providing care under Medicare Part B. A natural solution would be for the Administration to expand Title VI or § 1557 to include physician services by defining physician payments as “federal financial assistance” that are subject to civil rights regulation, and this could be accomplished through an Executive Order or agency regulation, and would not require the support of Congress. Expanding the scope of Title VI to cover physicians would not itself surmount the evidentiary obstacles faced in Title VI actions and discussed above. At the very least, however, such an expansion would allow plaintiffs to challenge any disparate-treatment practices of physicians and achieve comparable successes to those achieved in the hospital arena.

2. The Geography of Healthcare

Title VI cannot address the root causes of health care inequality. The disparate medical treatment of minority patients is, to a large extent, the product of geography. Figure 1 (Panels A and B) illustrates that different hospitals and physicians treat black and white patients even for a common condition such as Acute Myocardial Infarction (AMI). We created this figure using data from the Medicare program which insurers all Americans over the age of 65, and focused on heart-attacks because virtually everyone gets treated for this condition, as opposed to chronic conditions which may go undiagnosed. More specifically, the steepness of the blue line in Figure 1 indicates that hospital admissions (and physician visits) for black patients are largely concentrated on a modest number of hospitals (physicians), while the plateau of the blue line near the 100 percent mark relatively early in the graph suggests that many hospitals (physicians) serve predominately white patients. That different providers treat minority patients is unsurprising given the concentration of African Americans in the southeast United States and in highly segregated cities. But hospitals with a disproportionate share of minority have lower performance on standard Hospital Compare metrics (Figure 2),¹⁷ and this difference in performance extends to care in an outpatient setting, such as recommended care for diabetic patients, which is paid for my payers.¹⁸ In short, a significant source of disparate treatment is that minorities are disproportionately treated by lower quality providers that treat all their patients less-well, regardless of race.

Figure 1.

A: Differences in hospitals that treat black and white patients with a heart attack

The sample is fee-for-service Medicare beneficiaries between 2005 and 2013. Figure reports the cumulative percentage of all index admissions for heart-attacks that are accounted for by hospitals treating fee-for-service Medicare beneficiaries. Hospitals are ranked by the number of black patients that they treated—for example, the hospital ranked 1 treated the largest number of black patients.

B: Differences in physicians who treat black and white patients with a heart attack

The sample is fee-for-service Medicare beneficiaries between 2005 and 2013. Figure reports the cumulative percentage of physicians who treated heart-attacks patients within 30 days of the index event for any reason. Physicians are ranked by the number of black patients that they treated; the physician ranked 1 treated the largest number of black patients.

Figure 2.

Differences in quality of hospitals that treat black and white patients for heart attack and pneumonia

Figure reports the average quality scores achieved by hospitals of varying degrees of minority status, where quality is measured by the average percentile of performance on quality measures for the treatment of heart attacks and pneumonia (using measures from the Centers of Medicare and Medicaid’s Hospital Quality Alliance program).²⁷ Whereas just 9 percent of non-African Americans are seen in hospitals whose caseload is more than 20 percent African American, over 51 percent of African American patients are discharged from such hospitals. The figure shows that moving from a hospital with the typical composition of African Americans (6 percent of a hospital’s patients) to one that predominantly serves minorities (20 percent of a hospital’s patients) results in more than a 30 percent decrease in the quality of care.

Title VI is ill-suited to address this problem. First, it only makes illegal discrimination within a hospital. It cannot be used to challenge the quality of care at a minority-serving serving hospital that fails to perform at the level of another, non-minority-serving hospitals. Second, Title VI has limited power to force hospitals to serve minority neighborhoods. While it does not permit a plaintiff to force, say, a suburban hospital to relocate to the inner city, in theory it permits a plaintiff to stop a hospital from fleeing the inner city. In practice, however, these suits have faltered because of the high burden of proof, as noted earlier. Thus, Title VI cannot improve the quality of hospitals that remain in inner cities, let alone improve the quality of care in the deep-South relative to higher performing regions.

3. Penalties and Performance Improvement

One penalty for violating Civil Rights statutes is to bar a hospital from receiving further Federal funds. This penalty is rarely enforced, and for good reason. It could leave inadequately funded minority-serving hospitals with even fewer funds and adversely affect the quality of care received by all patients who receive care at such facilities, whether white or black. A more common penalty is an injunction or settlement requiring a hospital to reform its practices. For instance, plaintiffs with limited English proficiency have successfully secured settlements against hospitals with inadequate Spanish and other foreign language services.¹⁹ While this could be viewed as a victory, many of these facilities were unable to secure the funding to implement reforms they promised in those settlements; neither Medicare nor private insurance pay for interpretation and translation, and only a handful of Medicaid programs do.²⁰ Thus greater enforcement of Title VI with injunctions may prove to be less transformative, and possibly, more punitive than commonly believed.

4. Private Suits and Compensation

The Supreme Court has limited minority access to Title VI. A 2001 decision, Alexander v. Sandoval, barred private persons from bringing Title VI suits to challenge practices that have a disparate impact on minorities. Following Alexander, only the government itself has authority to litigate such practices. The Department of Health and Human Services’ rule implementing § 1557(a) of the ACA appears to have reversed this Supreme Court ruling by providing a private right of action for claims of disparate impact on the basis of race, 45 C.F.R. 92 (2016). Should § 1557 be repealed as part of a broader ACA repeal, of course, Alexander would continue to deny an individual right of action under Title VI. Private individuals could nonetheless present complaints to the Office of Civil Rights in the Department of Health and Human Services, which monitors Title VI compliance and can initiate suit. However, the Office of Civil Rights is notoriously under-resourced and can only challenge a fraction of activities that harm minority patients.

Even when minority patients are permitted to bring Title VI suits, they are often unable to obtain compensation for the wrongs they have suffered. Under the Supreme Court’s 1983 decision in Guardians Association v. Civil Service Commission of the City of New York, plaintiffs can obtain monetary damages under Title VI only if they can prove that they suffered intentional discrimination. But intentional discrimination is uncommon. Many researchers believe provider prejudice is largely a consequence of “implicit discrimination” (of the type measured by Implicit Association Tests). Despite considerable academic research on implicit discrimination, it does not qualify a patient for compensation because courts believe it falls short of intentional discrimination.^21,22

In order to compensate for these limitations on the private enforcement of Title VI, the Administration could expand the resources available to the Office for Civil Rights in the Department of Health and Human Services so it could bring more Title VI suits where the evidence for intentional discrimination is more compelling. Such suits, ideally initiated by CMS pilot programs that permit evaluation, could be directed towards hospital systems that put more resources into hospitals located in white areas instead of black areas. In addition, Congress could also protect private enforcement efforts by retaining § 1557 even if it repeals other provisions of the ACA.

Congress could also broaden the scope of remedies available to wronged patients. Specifically, it could authorize monetary compensation even in cases of non-intentional discrimination, including disparate impact cases. It bears repeating, however, that these remedies may exacerbate the problem that minority-serving providers already lack resources. Having to pay damages to one minority patient may reduce the quality of care for other minority patients at the facility. Counseling in the other direction is the plausible but speculative view that even if cases brought against physicians are difficult to win, the threat of litigation may increase diligence and reduce disparities caused by implicit discrimination.

Renewed DOJ Enforcement Efforts

In July, 2014, the Department of Justice announced a new initiative geared towards revamping Title VI enforcement.⁵ This DOJ initiative sets forth some steps that may lead to enhanced enforcement activity: initiating new Title VI training programs, forming new channels of communication and coordination among relevant agencies and proclaiming heightened expectations by the DOJ for agencies to pursue litigation against hospitals should efforts to secure voluntary compliance fail. It remains to be seen, of course, whether these renewed efforts have stimulated further reductions in observed disparities. Moreover, these enhanced enforcement efforts lack the aggressive reforms envisioned above that go to the heart of the enforcement difficulties—e.g., extending Title VI to even Medicare Part B physicians. It also remains to be seen whether such efforts will continue into the new Administration under President-Elect Trump.

Benefits from Non-Litigation Solutions

In contrast to the legal reforms discussed above, there are a number of complementary non-litigation based solutions for improving minority care. Some of these have already been put into motion as a result of the general provisions of the ACA. The ACA’s expansion of health insurance coverage will disproportionately help minority patients, because they are disproportionately likely to be uninsured; prior to the passage of the ACA the uninsurance rate was 12% for non-Hispanic whites, 21 percent for blacks and 31 percent for Hispanics. Exchange plans grant newly-insured patients access to a wider network of providers than safety-net providers, who are mainly Medicare-participating hospitals with emergency departments that are required by the Emergency Medical Treatment and Labor Act to serve all comers. Moreover, the ACA’s efforts to improve the quality of care, through Accountable Care Organizations and alternative reimbursement schemes, will disproportionately help patients who are presently being cared for by providers whose performance is laggard. In other words, one unintended consequence of the ACA may be to curb the ‘separate and unequal’ character of minority healthcare.

Of course if the coverage provisions of the ACA are repealed, much of this progress may be undermined. Moreover, even should the ACA survive, it is not a panacea for racial disparities in healthcare. The majority of the insurance expansion under the statute comes from an expansion of the Medicaid program. Several states have chosen not to participate in the expansion; a decision that disproportionately hurts minorities because of their over-representation in lower income groups and non-expanding states. Moreover, Medicaid offers smaller provider networks than exchange plans. So while a Medicaid expansion would improve minority healthcare, exchange plans with wider networks would likely have provided even more choice to minority patients. Finally, while the ACA’s insurance market reforms have the potential to provide minorities with greater access to medical providers that previously served a predominantly white population, any such potential will be limited to the extent that one’s choice of medical provider is still predominantly determined by geography—that is, by the geographic distance between that provider’s location and one’s own residence.

One alternative solution is to improve the quality of minority-serving providers. This could start with creating report-cards on disparities, with the goal of relying on market learning by patients, providers and payers to reduce disparities.²³ For such report cards to be effective they would have to measure how the same provider treats patient of different races. Such report cards would require that attention be paid to the substantial statistical challenges of robustly measuring rates at smaller providers. There is a nascent effort to build such report cards, but their effect is unknown.¹⁸ Another approach would be to invest in programs that reduce implicit-bias, acknowledge that the science of tracing implicit bias to actual treatment decisions, and then reducing the implicit-bias is far from established. Minority serving ACOs may adopt these programs and have strong incentives to create and evaluate them.

These efforts could be strengthened by new efforts to target quality-improvement efforts at providers with disproportionately minority patient populations—a solution that perhaps best confronts the geographic realties behind care disparities. As an example, the government (possibly CMS) could provide resources to elevate the performance of the 20 largest minority-serving hospitals, which are responsible for over 10 percent of all African American hospital discharges. This assistance would help both African American and non-African American patients treated at these hospitals, but disproportionately help African American patients in the general population. Assistance does not have to be solely financial, but financial assistance is necessary for these hospitals to hire better physicians and purchase equipment. Such efforts are similar to those of the ‘environmental-justice’ movement where the focus is on improving communities as defined by geography or race, as opposed to improving outcomes for an individual. Unfortunately, the ACA took a step in the opposite direction from that suggested here: the ACA called for various reductions to the Disproportionate Share Hospital (DSH) program, which allocates additional funds to hospitals that treat a large number of Medicaid and indigent patients.

One may object that throwing money at the problem will not improve patient outcomes. However, experience with the DSH program to date gives us reasons to be optimistic. When states do not offset these funds by reductions in state funds for poor-serving hospitals, evidence shows that the DSH program significantly improves health outcomes for neonatal health and heart attack patients.²⁴ For an additional level of surety against waste, the Department of Health and Human Services could offer matching grants on implementation of reforms or outcomes. In contrast to the lack-luster performance of pay-for-performance in general, there is evidence that pay-for-performance initiatives significantly increase the performance of hospitals that serve indigent populations.²⁵

The choice between litigation and non-litigation based strategies is a difficult one. One is not obviously superior to the other. As hospitals employ more physicians and consolidate with downstream providers such as nursing homes and long-term care facilities, hospitals become responsible for a larger portion of total-care, and that increases the benefits from litigation directed at hospitals. Some litigation-based solutions (such as expanding the scope of the Civil Rights Act to cover all physician services) do not require Congressional approval, which is important given that body’s recent paralysis. But litigation remains inherently uncertain, slow, and costly. Allowing minority-serving hospitals to be sued may weaken the already-precarious financial situation of those providers, worsening care for minorities. Any effort to reform the statute would have to confront vocal provider lobbies who are unlikely to accept greater scrutiny of their practices, especially at a time when physicians are beleaguered by fee-cuts in Medicare, health-information technology implementation and malpractice litigation.²⁶ Despite these concerns, a robust Civil Rights statute may improve minority healthcare by bolstering minority patients’ trust in the healthcare system, physician diligence, and through these channels, increasing their minority patients’ willingness to seek care.

A different set of problems affects non-litigation based solutions. They will cost money, and the current focus on ‘bending the cost curve’ has had the chilling effect on all increases in healthcare spending. Yet, there is less uncertainty with non-litigation solutions relative to litigation based ones. They also offer the promise of being designed in a manner that can be evaluated, and scaled up or down depending on their success.

With health care reform once again in Congress, this may be the right time to discuss proposals aimed at seriously targeting the sources of health care disparities. A series of pilot programs could help uncover the optimal mix of litigation and non-litigation based solutions, but the legislative uncertainty around the future of CMMI means that there is a fair bit of uncertainty around who will initiate these pilots. Some states may rise to the challenge if Congress delegates reform responsibilities to states. An important thing to note is that litigation and non-litigation strategies are not mutually exclusive. While greater enforcement of existing Civil Rights legislation offers some promise for improving minority healthcare, complementary gains are possible with non-litigation based options. What is beyond doubt, however, is that persistent gaps in minority healthcare remain a Civil Rights issue.