Distinguishing between primary and secondary racial identification in analyses of health disparities of a multiracial population in Hawaii.

OBJECTIVE: To examine the importance of distinguishing between primary and secondary racial identification in analyzing health disparities in a multiracial population.

METHODS: A cross-sectional analysis of 2012 Hawaii Behavioral Risk Factor Surveillance System (H-BRFSS). As part of the survey, respondents were asked to identify all their races, and then which race they considered to be their primary race. We introduce two analytic approaches to investigate the association between multiracial status and general health: (1) including two separate dichotomous variables for each racial group (primary and secondary race; for example, 'primary Native Hawaiian' and, separately, 'secondary Native Hawaiian'), and (2) including one combined variable for anyone choosing a particular racial group, whether as primary or secondary race ('combined race'; e.g. Native Hawaiian). Linear regression then compares the multiracial health disparities identified by the two approaches, adjusted for age and gender.

RESULTS: The 2012 H-BRFSS had 7582 respondents. The four most common self-identified primary racial/ethnic groups were White, Japanese, Filipino, and Native Hawaiian. Native Hawaiians were the largest multiracial group with over 80% self-identifying as multiracial. Health disparities for Native Hawaiians, Portuguese and Puerto Ricans were attenuated by 10% after accounting for multiracial status. Populations that self-identified secondarily as Japanese, Puerto Rican, Mexican, and other PI had significantly poorer self-reported health.

CONCLUSION: The analysis illustrates the importance of accounting for multiracial populations in health disparities research and demonstrates the ability of two approaches to identify multiracial health disparities in data sets with limited sample sizes. The 'primary and secondary race' approach might work particularly well for a multicultural population like Hawaii.