Health Disparities
Empirical support for a mechanism-driven approach to categorizing Multiracial participants in depression disparities research Sarah Forthal* Sarah Forthal Forthal Columbia University Mailman School of Public Health
Studies suggest that Multiracial American adults may have higher depression rates than those identifying with any single racial group. Approaches to categorizing Multiracial participants in racial disparities research may affect the detection and interpretation of this finding, yet these decisions often lack theoretical grounding. We (1) developed an approach to categorization that varies by hypothesized depression disparity-driving mechanisms and (2) tested statistical relationships between categorization schemes and hypothesized mechanisms to provide empirical evidence that supplements theory in guiding researchers’ choices. We used the National Epidemiologic Survey on Alcohol and Related Conditions-III, a nationally representative cross-sectional study of US adults conducted 2012-2013 (N=36,309). We compared standardized effect sizes representing the relationship between race and 9 depression disparity-driving mechanisms identified in the literature to assess sensitivity to scheme and compared R², AIC, BIC, and 95% CI widths to assess statistical performance. Figure 1 demonstrates the sensitivity of estimates to Multiracial categorization. Estimates for nearly all mechanisms varied by scheme, particularly for the Multiracial, Hispanic, and American Indian or Alaska Native groups, demonstrating the significance of Multiracial categorization for estimating disparities. Grouping all Multiracial participants into one “non-Hispanic Multiracial” group performed best for the mechanisms of racial identity affirmation and belonging, income, education, and traumatic experiences; disaggregating the Multiracial group (“disaggregated Multiracial”) performed best for discrimination; and grouping Multiracial participants with their lowest-income single race group (“hypodescent”) performed best for adverse childhood experiences, physical health, and life stressors. Aligning categorization schemes with underlying mechanisms may improve the validity of disparities research.
