The Medicaid Expansion Gap and Racial and Ethnic Minorities With Substance Use Disorders

Abstract

We compared the race and ethnicity of individuals residing in states that did and did not expand Medicaid in 2014. Findings indicated that African Americans and Native Americans with substance use disorders who met new federal eligibility criteria for Medicaid were less likely than those of other racial and ethnic groups to live in states that expanded Medicaid. These findings suggest that the uneven expansion of Medicaid may exacerbate racial and ethnic disparities in insurance coverage for substance use disorder treatment.

The Patient Protection and Affordable Care Act’s¹ Medicaid expansion has the potential to connect many Americans in need of substance use disorder (SUD) treatment with insurance coverage. However, some racial and ethnic groups may be underrepresented in the Medicaid expansion population because they are more likely to live in states that have not expanded Medicaid.^2,3 Although racial and ethnic minority groups constitute 37% of the US population, they constitute 47% of individuals who meet new federal eligibility criteria for Medicaid.² The estimated percentage of individuals who meet the new Medicaid criteria and live in an expansion state is 34% among African Americans, 45% among Whites, 53% among Latinos, and 69% among Asians.

Yet, how Medicaid expansion may affect long-standing racial and ethnic disparities in insurance coverage for SUD treatment remains unknown.² Recent estimates have shown a decrease in the number of nonelderly adults who were uninsured, from 18.4% in 2013 to 15.7% during the first 3 months of 2014.⁴ However, little change occurred in the rate of the uninsured in nonexpansion states.⁴ Among expansion states, the most significant reduction in the rate of uninsured people was among working-age Latinos.⁵ Existing estimates do not include all major racial and ethnic groups and are not specific to individuals with SUDs. To address this gap, we estimated the proportion of people with SUDs residing in Medicaid expansion and nonexpansion states by race and ethnicity.

METHODS

We used the 2010 and 2011 waves of the National Survey on Drug Use and Health, a longitudinal survey that includes data on mental health and the use of tobacco, alcohol, and illicit drugs in the United States. Details about sampling and data collection procedures used in the National Survey on Drug Use and Health can be found elsewhere.⁶ For this study, we restricted the sample to childless, uninsured adults (aged ≥ 18 years) with reported personal income less than 138% of the federal poverty line. The total weighted sample included 16 617 respondents.

First, we examined differences in the total weighted number of individuals who met new federal eligibility guidelines for Medicaid under the Affordable Care Act in states that had or had not expanded Medicaid by March 30, 2014. We also examined differences among these individuals across race and ethnicity and need for alcohol use disorder and drug use disorder treatment. We categorized racial and ethnic groups as African American, Asian, Latino, Native American, White, and of more than 1 race. To test for statistically significant group differences (P < .01), we used the χ² test with Rao-Scott adjustment to account for the complex sample design.⁷ We completed all analyses using the Substance Abuse and Mental Health Service Administration’s (Rockville, MD) online Restricted-Use Data Analysis System.

RESULTS

In the weighted sample, 52.4% of respondents were male, 53.9% were younger than 35 years, 10.8% had a college degree, and 28.4% were employed full time. We found major racial and ethnic differences in the distribution of individuals who met new Medicaid eligibility criteria in expansion and nonexpansion states (Table 1). An estimated 47.3% of Whites, 40.0% of African Americans, and 30.0% of Native Americans met new eligibility criteria for Medicaid, compared with 81.1% of Asian Americans, 57.0% of Latinos, and 55.1% of individuals of more than 1 race. Differences in the distribution of racial and ethnic groups across expansion and nonexpansion states were statistically significant for all groups (P < .01).

TABLE 1—

Poor, Uninsured, Childless Adults Who Met New Federal Eligibility Criteria for Medicaid Under the Affordable Care Act by Race, Ethnicity, and Medicaid Expansion Status in State of Residence: 2010–2011 National Survey on Drug Use and Health, United States

Race and Ethnicity	Expansion, No. (%)	Nonexpansion, No. (%)	Total No.
African American	1 084 (40.0)	1 629 (60.0)	2 713
Asian	752 (81.1)	176 (18.9)	928
Latino	2 404 (57.0)	1 811 (43.0)	4 215
Native American	52 (30.0)	122 (70.0)	174
White	3 928 (47.3)	4 384 (52.7)	8 312
Mixed	151 (55.1)	123 (44.9)	274
Total	8 372 (50.4)	8 244 (49.6)	16 616

Note. Rao-Scott χ²: F_{5, 4500} = 18.41; P < .01. Numbers and percentages reflect the weighted estimate of the population.

We also found large racial and ethnic differences in the proportion of newly Medicaid-eligible individuals with SUDs residing in expansion and nonexpansion states. Fewer than half of Whites, African Americans, and Native Americans with alcohol use disorders who met new federal eligibility criteria for Medicaid resided in expansion states (Figure 1). In total, 46.3% of Whites, 39.2% of African Americans, and 26.2% of Native Americans with alcohol use disorders met new Medicaid eligibility criteria, as opposed to 80.9% of Asian Americans, 56.0% of mixed-race individuals, and 49.8% of Latinos. Our findings suggest similar disparities for individuals with drug use disorders. We found that 47.0% of Whites, 39.3% of African Americans, and 30.3% of Native Americans with drug use disorders met new Medicaid eligibility criteria, as opposed to 81.2% of Asian Americans, 56.7% of Latinos, and 54.2% of mixed-race individuals. Racial and ethnic differences across alcohol use disorder and SUD categories were statistically significant (P < .01).

FIGURE 1—

Percentage of poor, childless adults who met new federal eligibility criteria for Medicaid and resided in an expansion state by race and ethnicity: 2010–2011 National Survey on Drug Use and Health, United States.

Note. AUD = alcohol use disorder; DUD = drug use disorder. All estimates are significant at P < .01.

DISCUSSION

The key finding of this study is that African Americans and Native Americans will be underrepresented in the Medicaid expansion population. The findings are consistent with a recent estimate from the Kaiser Family Foundation.⁸ Before Medicaid expansion in 2014, African Americans and Native Americans already had lower rates of insurance coverage than did Whites. The uneven expansion of Medicaid across the United States may widen that gap, which is of special concern for Native Americans, who exhibit rates of SUDs nearly twice that of the general population.⁹ However, because the sample size of Native Americans was small (n = 174), this finding should be interpreted with some caution. By contrast, our findings indicate that Asian, Latino, and mixed-race Americans will be overrepresented in the expansion population. These findings largely reflect the fact that individuals from these groups are more likely to live in expansion states.

Our findings add to current reports that demonstrate how Medicaid expansion is likely to have a differential impact on coverage among individuals from different racial and ethnic groups. State and federal policymakers need to be cognizant of these consequences when devising enrollment and financing strategies for SUD treatment.