The finding: Using the total adult population of Sweden, this cross-sectional study estimated the point prevalence of blood-borne viruses (BBV)—HIV, hepatitis B (HBV), and hepatitis C (HCV)—in people with severe mental illness, determined the odds of each BBV relative to the general population, and identified independent risk factors for each BBV. After accounting for sociodemographic characteristics, findings among people with severe mental illness were as follows: 1) HIV prevalence was 0.24%, or 2.57 times higher; 2) HBV prevalence was 0.53%, or 2.29 times higher: and 3) HCV prevalence was 4.58%, or 6.18 times higher than in the general population. Among independent risk factors assessed—age, sex, immigration status, socioeconomic status, education and substance misuse—substance misuse conferred the greatest risk for all BBVs. With respect to the contribution of substance misuse to the risk of BBV, odds ratios were 1.61 for HIV, 1.28 for HBV and 1.72 for HCV, or “approximately four times the risk of HIV and HBV and 25 times the risk of HCV.” For all BBVs, prevalence was higher compared to the general population for each individual severe mental illness diagnosis: schizophrenia, schizoaffective disorder, bipolar disorder, and other psychosis. HIV/HCV coinfection prevalence was higher among people with severe mental illness than in the general population.

Strength and weaknesses: Whereas previous studies have relied on small convenience samples in clinical settings, this is the first full-population study of BBV prevalence and risk factors among people with severe mental illness and the only prevalence study in northern European countries. Previous research has examined disproportionate rates of cardiovascular and respiratory illness and cancer, but neglected the contribution of infectious diseases to the excess mortality seen among people with severe mental illness. This study also permits examination of the effects of individual risk factors by BBV and for each severe mental illness subgroup. This study may underestimate BBV prevalence since testing was not performed as part of routine care but rather occurred only when a clinical opportunity to do so arose. Additionally, the study’s cross-sectional design does not permit conclusions about the cause and effect relationships between BBVs and severe mental illness. Similarly, heightened BBV risk cannot be linked to particular symptoms or phases of illness, such as mania or active psychosis.

Relevance: The highly relevant findings of this study are thoughtfully discussed in the accompanying commentary by Wainberg and Dixon. They emphasize that this study underscores persistent inequalities in access to quality healthcare and identifies opportunities for improvement among people with severe mental illness. Despite convincing evidence to the contrary, the need for attention to sexual and substance-related risk behavior, at both clinical and policy levels, is too often overlooked in this population. Guidelines for routine testing, prevention and treatment strategies for BBVs are now widely available for general medical settings. This study supports the growing recognition that ending HIV may also depend on adequate resources allocated to psychiatric and substance-related settings for both enhanced training and education of personnel as well as improved treatments and interventions to reduce transmission of BBVs.

The finding: Using a prospective longitudinal design, this study followed 177 HIV-infected individuals in mid-range of illness from 1997 to 2004. Beck Depression Inventory (BDI) scores, CD4 cell count and viral load were obtained every 6 months over a 12-year period. Defining depression as a BDI score of 10 or higher, and after adjusting for age, baseline CD4 and antiretroviral use, mortality was significantly higher among depressed individuals: HR=2.044, CI=1.176-3.550, p=0.011. However, when adjusted for race and education, depression predicted worse survival for only non-African Americans and those individuals with a college education or higher. Neither antiretroviral use nor medication adherence accounted for the differences in effect on survival by race or education.

Strength and weaknesses: The study’s 17-year follow-up period occurred entirely during the period of combination antiretroviral therapies. The study’s inclusion of a diverse sample of both sexes also adds to previous work in this area that focused on women. It is the first study to demonstrate the moderating effects of race and education on survival in depressed vs non-depressed individuals. However, the study did not distinguish between all-cause and HIV-specific mortality. It also was not powered to detect effects in other subgroups such as Hispanics. The impact and role of other significant factors were not explored: IV drug use, drug dependence, and presence of HAND and other comorbid psychiatric diagnoses. Similarly, the authors emphasize the adverse effects of food insecurity, discrimination, stigma, exposure to violence and poverty in minority populations which were also not explored. Since the participants were volunteers, generalizability may be limited.

Relevance: Depression is one of the most commonly occurring psychiatric disorders in HIV and its impact on disease progression, medication adherence, quality of life and survival have been concerns since the early epidemic. However, given the growing disproportionate effect of HIV on African Americans and prevalence of social conditions such as homelessness and poverty, this study underscores the need for research in HIV as a syndemic. Depression and other psychiatric conditions such as PTSD, intimate partner and community violence, childhood trauma and substance use interact to influence the spread and effective management of HIV. This study suggests that HIV comprises parallel epidemics. For certain subgroups such as African Americans and individuals with lower educational attainment, identification and effective treatment of depression should occur within the wider context of substance use and social conditions such as exposure to violence, homelessness and poverty.