New research from Berkeley Public Health shows that the way race is classified in epidemiological studies can have a substantial impact on effect estimates. The study, published in the American Journal of Epidemiology in January, builds on a body of research that finds that multiracial people are frequently excluded or deliberately misclassified to fit into discrete racial categories, co-author Tracy Lam-Hine, a UC Berkeley School of Public Health doctoral candidate, said.
“We find that these classification schemes are not valid measures,” Lam-Hine said. “Choices of one classification scheme over another may really change the findings of the study, even if it is not a study focused on race or ethnicity as a major focus. Thus researchers need to carefully think about why they’re using their chosen scheme, otherwise study results may be biased.”
Lam-Hine and others were working on a separate study, looking at various behavioral or demographic factors associated with experiencing HIV or PrEP-related stigma among clients of the San Francisco AIDS Foundation. Initial results from that research surprised them—they found that people of color seemed less likely to experience stigma than white clients. The team decided to explore and document that issue as well.
Lam-Hine said he hopes that the findings “push health researchers to be more thoughtful about how they choose to classify multiracial people, and to incorporate theory into those choices.”