To the Editor: The authors of a study of HIV/AIDS incidence among Medicaid enrollees with serious mental illness that was published in the October issue concluded that for their sample “a severe mental illness diagnosis in the absence of a substance abuse diagnosis was not highly associated with increased risk of a new HIV/AIDS diagnosis” (1).

This conclusion concerns us because the study relied on rates of HIV diagnosis captured in Medicaid claims data. The article provides no information about the number of enrollees with severe mental illness, with or without a substance use disorder, who were tested for HIV in the first place. The authors underemphasize the potential detection bias this missing denominator creates in determining the incidence of HIV infection among all patients with serious mental illness, and especially among those without a substance use disorder. A review of the HIV testing literature found that among people with serious mental illness, those with drug problems or a history of injection drug use were more likely to be tested for HIV (2), and there is no reason to assume otherwise in the study in question. Furthermore, because the authors were unable to obtain data on risk behaviors, their adjustment for other factors associated with HIV infection among patients with serious mental illness was incomplete, regardless of substance use disorder. Therefore, they have constructed a confounding dichotomy and provided analyses that may lead providers and policy makers to incorrectly conclude incorrectly that patients with serious mental illness are not at increased risk of contracting HIV/AIDS.

Trying to separate persons with serious mental illness from those with a substance use disorder is, in itself, problematic. The high incidence of co-occurring psychiatric and substance use disorders in the general population and in treatment-seeking populations has led the Substance Abuse and Mental Health Services Administration to state that co-occurrence is the expectation rather than the exception in both addiction and mental health treatment systems (3). Moreover, a primary non–substance-related mental disorder often precedes and is a robust risk factor for the later onset of a substance use disorder (4). Insofar as having a substance use disorder is an important HIV risk factor among persons with serious mental illness, focusing on those who have already developed a substance use disorder would miss the critical opportunity to intervene in this population to prevent the onset of substance use disorders. We need to integrate prevention and treatment of non–substance-related mental illnesses and substance use disorders rather than find new ways of separating the two.

The question posed by these authors, “Is serious mental illness a risk factor?” cannot be answered with the partial data and analyses presented, and at this point in the AIDS epidemic it may not be the right question. Routine HIV testing that includes people with serious mental illness regardless of whether they have a current substance use disorder is likely to yield better estimates of the incidence of infection and to foster earlier and more comprehensive prevention and care.