• Priscilla Novak, Kester F Williams-Parry, and Jie Chen.
• Agency for Healthcare Research and Quality, University of Maryland (UMCP, College Park), 4200 Valley Drive, #2242, College Park, MD, 20742-2611, USA. Priscillanovakmph@gmail.com.
• J Racial Ethn Health Disparities. 2017 Aug 1; 4 (4): 607-614.

BackgroundIn 2010, the Affordable Care Act (ACA) extended eligibility for dependent coverage under private health insurance. Emerging evidence shows that young adults, including those with behavioral health disorders (BHDs), have benefited from this expansion.ObjectiveThe objective of this study is to explore the population characteristics of the remaining uninsured individuals with and without BHDs and to examine whether the factors that contribute to racial and ethnic disparities in the likelihood of being uninsured were different after the implementation of the ACA provision that extended insurance eligibility for young adults in 2010.DesignWe use cross-sectional data analysis.ParticipantsWe use a nationally representative dataset of the non-institutionalized civilian population in the Medical Expenditure Panel Survey from 2007 to 2012.MethodsWe compare population characteristics of the remaining uninsured individuals ages 19-25, before and after the implementation of the ACA expansion in 2010. We use multivariate logistic regression to estimate the predictors (such as family income and English proficiency) that are associated with the likelihood of having no health insurance. We utilize the Fairlie decomposition method to examine the factors that contribute to racial (non-Latino White (White) vs. non-Latino African-American (African-American)) and ethnic (non-Latino White (White) vs. Latino) differences in the probability of being uninsured. Finally, we apply our analysis among populations with and without BHDs respectively, to examine the differences in the predictors of being uninsured between these two groups.ResultsAmong individuals with BHDs, after adjusting for covariates, the estimated probabilities of being uninsured for Whites were 0.21 and 0.16 pre- and post- the ACA expansion, respectively. The predicted probabilities of being uninsured for Latinos were 0.29 and 0.26 and for African-American were 0.19 and 0.17 pre- and post- the ACA expansion, respectively. The ethnic disparity between Whites and Latinos was 19 %. The racial disparity was moderate. Reductions in the uninsured rate of individuals without BHDs were observed as well. However, the ethnicity disparity remained at 19 % and the racial disparity increased by 5 %. Overall, our decomposition model explained 63-89 % of the racial and ethnic disparities in insurance coverage. The major factor associated with the ethnic disparity among those with BHDs was the immigrant status of Latinos, and the major factor associated with racial disparity was geographic location. The major factor associated with the ethnic disparity among those without BHDs included the immigrant status of Latinos, lack of English proficiency, and geographic location, whereas the major factor associated with racial disparity was family income.DiscussionOur study presents health insurance coverage trends among racial and ethnic minorities after the implementation of insurance expansion, a major provision of the ACA. It is important for policy makers to be aware of differences among the remaining uninsured as they evaluate ways to improve healthcare access and affordability.

28 keywords...

hide…