I. Background

African Americans and Latinos are the two largest racial/ethnic minority groups in the United States, accounting for nearly 25% of the U.S. population and representing about 84% of the minority population in 1999.   Today, they represent nearly equal shares of the U.S. population. While the two population groups differ in a number of respects, particularly in their diversity of ethnic origins and language, they share a commonality of experiences in the United States. Both populations reside largely in racially segregated neighborhoods and have poverty rates three times those of whites. Both have cultural beliefs and practices that sometimes conflict with western medicine and, thus, may result in a lack of confidence in the medical system. Both have faced a history of discriminatory policies and practices that have limited their health care access and compromised their trust in the health system. In addition, changes in federal policy and large demographic shifts in our nation's cities have had direct effects on the medical care that is available to both population groups. 

Improving access to “mainstream” medical care was an implicit, if not explicit, goal of Medicaid and Medicare, programs enacted in 1965 to expand health insurance coverage to low-income and elderly Americans.   Since providers were required to comply with the 1964 Civil Rights Act, these new programs had the direct effect of reducing financial barriers to care, as well as indirectly reducing racial barriers to care. Title VI of the Act prohibits discrimination by any facility receiving federal funds. Numerous studies have documented the important role of Medicaid and Medicare in reducing differentials in care between low-income and upper-income Americans across racial and ethnic groups. 

Concurrent with federal efforts to reduce financial barriers to care were initiatives designed to expand the supply of health care resources in *17 low-income communities. The Community and Migrant Health Centers Program and the National Health Service Corps were among the major initiatives of the “War on Poverty” that helped to expand the supply of health providers in medically underserved areas.   Not surprisingly, most medically underserved areas were in low-income neighborhoods, and many were also in racial/ethnic minority communities. Additionally, as a result of litigation in the 1960s that explicitly defined the “free care” obligation of hospitals built with federally provided construction funds, access to private hospital-based sources of care improved for those unable to pay. 

In the 1970s another demographic shift occurred that also had an impact on the health care system. Many of the inner-city hospitals found that their mostly white, middle class patient-base had moved to suburban communities and a large, low-income, mostly minority patient-base remained in inner city communities.   In addition, many private physicians were reluctant to establish practices in low-income communities, thus increasing the importance of urban hospitals as sources of outpatient care for the poor. 

The rise of managed care in the late 1980s and throughout the 1990s produced another major shift in the health delivery system. Increasing health care costs were driving both public and private payers to search for alternatives to control the rate of growth in health spending and many believed that managed care was the solution. Reflecting national trends, nearly two-thirds of privately insured African Americans and Latinos were enrolled in a managed care plan by 1996.   Managed care enrollment among the publicly insured was slightly lower, but still approximately 45% of African Americans and 35% of Latinos covered by Medicaid were estimated to be enrolled in a managed care plan in 1996.   This shift to managed care likely increased the number of patients using private physicians rather than hospital-based providers and clinics for their care, particularly among the Medicaid population.

Health insurance, whether obtained through a managed care or fee-for-service plan, has become the primary means used to pay for medical care, and is an important determinant of an individual's ability to obtain care.   Compared to those with coverage, the uninsured face greater obstacles to receiving care and to developing an ongoing relationship with a health provider. In addition, studies have found that type of insurance is a strong determinant of whether individuals have a usual source of medical care.   People who lack insurance are significantly less likely to have a usual source of care and are more likely to rely on institutional providers such as hospitals or clinics for their care than persons who are insured.

A recent study by Weinick et al. provides evidence that racial/ethnic *18 differences persist in the share of the population lacking a usual source of care.   In 1996, roughly 16% of whites compared to 20% of African Americans and 30% of Latinos report not having a usual source of medical care. Especially troubling is that the gap between Latinos and whites without a usual source of medical care widened between 1977 and 1996 (figure 1). For African Americans and whites, the gap remained about the same. 

  Factors related to type of insurance also influence where care is obtained. For example, Medicaid beneficiaries, regardless of race, have faced challenges in accessing “mainstream” sources of care such as private sector office-based physicians. In 1993, for example, Medicaid payment rates for private sector office-based physicians were 73% of Medicare rates and about 47% of private rates.   Low payment rates, concerns about practicing in high-poverty areas, and bureaucratic hassles have been cited as major reasons for low participation in Medicaid among private physicians. 

*19 While we do know that insurance affects where individuals go for care, very little is known about whether race/ethnicity has an effect, independent from insurance status, on the site of care. Cornelius et al. documented that insurance status and race/ethnicity are separately associated with the usual sources of medical care.   African Americans and Latinos were more likely than whites to use hospital outpatient departments (OPDs), community-based clinics, and emergency rooms (ERs) as regular sources of care in the 1980s. Also, the publicly insured and the uninsured were more likely than the privately insured to obtain care from these sources (figure 2). What was unknown was whether the findings by race were largely a function of differences in the health insurance coverage of the population groups.

Given the role of insurance coverage today, a comparison of the major sources of medical care used by whites, African Americans, and Latinos, controlling for type of insurance, is a critical first step in understanding whether race is independently associated with where medical care is obtained. This is particularly important in light of the fact that minority Americans are more likely to be uninsured or covered by Medicaid than whites.   Thus, comparisons by race, unadjusted for differences in *20 insurance coverage or other population characteristics, such as age or income, can lead to a misinterpretation of the effects of race/ethnicity on the site of medical care.