Ethnicity, Depression, and Suicide

To the editor: This letter is in response to an article written by Maria A. Oquendo, M.D., et al. (1). The authors expanded on previous work by looking at the relationship between affective disorders and suicide (2). In using statistical procedures to standardize data from the Epidemiologic Catchment Area (ECA) study and the Hispanic Health and Nutrition Epidemiological Survey and suicide rates from the Centers for Disease Control’s National Health Statistics on Mortality, the authors exerted an admirable effort to tease out 1-year prevalence rates for major depression and suicide rates relative to depression for specific Hispanic ethnic groups. They found data that complicated the substantiated claim that major depression is positively associated with suicide. The results clearly revealed that Puerto Rican Americans have the highest rate of major depression while they have the lowest suicide rate relative to depression. These findings support research into why certain ethnic groups have protective factors against suicide, which is an important question in mental health policy and practice.

However, I would like to offer some critical remarks on the methods used in this study. The “ecological fallacy,” or the error of assuming an association between two characteristics of a community will generalize at the individual level (3), undermines these findings. The authors assumed that because depression and suicide are associated in the larger community, and in the literature, that they are also associated at the individual level. The question arises, were there no suicide indicators in the ECA study and/or the Hispanic Health and Nutrition Epidemiological Survey?

The conceptual problems are magnified in the authors’ use of a proxy measure—i.e., depression for suicide. From my understanding of the research question, the dependent variable of interest is suicide. It is a significant leap to use depression as a proxy measure for suicide. The finding that Puerto Rican men and women had significantly higher rates of major depression and the lowest rate of suicide relative to major depression challenges the choice of the proxy measure.

These limitations notwithstanding, the article offers new insight into an increasingly vital area of inquiry in mental health research. It is imperative that we begin to look at cultural experiences and the potential differences they impose on prevalence of illness, service use, and suicidality. However, it is equally important that we use careful methods.