Re: "Cigarette Smoking and Breast Cancer Risk in Hispanic and Non-Hispanic White Women: The Breast Cancer Health Disparities Study" by Conner et al. (J Womens Health (Larchmt) 2016;25:299-310).

Comprehensive reviews of current epidemiologic research have found that researchers fail to differentiate between concepts of race and ethnicity and they are not transparent with their methods to assess these concepts. In issue 3 volume 25 of the Journal of Women’s Health, Conner et al. examines the effect of modification of menopausal status and alcohol consumption in regard to the relationship between cigarette smoking, ethnicity, and breast cancer risk. By compiling three population-based case–control studies, the researchers were able to add evidence to support an association between smoking and breast cancer risk and explain health disparities between U.S. Hispanic, Mexican, and U.S. non-Hispanic white women. While this study furthered knowledge in the field of epidemiology on breast cancer risk in an ethnic population, it erroneously included Native American women from the 4-Corners Breast Cancer Study (4-CBCS) in the Hispanic group, which was compared with Mexican and non-Hispanic white women. It is problematic to combine Hispanics and Native Americans into one ethnic group because they are affected by distinct health disparities. Variations in health outcomes between different racial and ethnic groups are due to biology, behavior, and environmental factors. As a result, Native American women should not have been combined within the same category as Hispanic women because the two groups experience different health disparities. Furthermore, Kaufman and Cooper note that racial and ethnic populations are socially distinct, and it is difficult to compare environmental variables that modify behavior, therefore it is implausible to combine two ethnic groups into one. Also, it is not clearly stated if women in this study were both Hispanic and Native American or only one of the two groups. When comparing different ethnicities, it is important to realize that each comprises their own specific culture, language, and religion. Even if the women were multiethnic (Hispanic and Native American), they will have different experiences than Hispanic women. By including Hispanic and multiethnic women in the same population group, the researchers are inaccurately comparing Hispanic women with non-Hispanic white and Mexican women. The other cohort study included in this research was the San Francisco Bay Area Breast Cancer Study, which included non-Hispanic white, African American, and Hispanic women. Conner et al., however, did not include African Americans in their study. It is perplexing then why they would not remove Native American women if they only comprised 2.5% of the 4-CBCS population. The addition of Native American women in the Hispanic group is a key limitation in this study. The study would be greatly improved if the researchers only compared U.S. non-Hispanic white, Mexican, and U.S. Hispanic women.