Objectives: To assess the use of the categories of race, ethnicity, and national origin in recent public health research.
Methods: We reviewed all research articles on human populations published in the American Journal of Public Health from January 1980 through December 1989. Articles were classified by (1) mention of the categories, (2) use of the categories, (3) presence of explicit definitions, and (4) definitional criteria.
Results: Specific categories (e.g. 'black', 'Chinese', 'Hispanic') or generic categories (e.g. 'race', 'ethnicity', 'national origin') were mentioned in 461 (50.4%) of 914 articles on human populations. In most studies (65.1%), single categories (e.g. race or ethnicity) were considered; in 1.3% of studies, two terms (e.g. both race and ethnicity) were examined independently; in 1.3%, categories were used interchangeably; in 5.6% of the studies, combined categories (e.g. race-ethnicity) were used; and in 27.5% of the studies, specific population groups were named without reference to a generic category. Explicit definitions of categories were present in only 8.4% of the articles in which the categories were considered. Absence of explicit definitions and use of combined and interchangeable categories suggest a lack of clarity and conceptual consistency in research on race, ethnicity, and national origin-related topics.
Conclusion: To improve our assessment of differences in health status among racial, ethnic, and national origin groups, research involving these categories should assess their validity and should define concepts clearly, explicitly, and consistently. Such research would minimize misclassification, improve the interpretation of findings, facilitate comparison among studies, and enhance the understanding of causes underlying differences in health status among populations of different racial, ethnic, and national origins.