Young adult black men and women in South Africa differ with regard to the association between HIV-related stigma and testing behavior, a cohort study finds.1 Women who held HIV-related stigmatizing attitudes (reflecting negative moral judgment of infected individuals) were more likely than other women to have been tested for the virus (odds ratio, 3.0). In addition, women who perceived that others viewed and treated HIV-infected people negatively or who had witnessed others discriminating against infected individuals had reduced odds of having been tested (0.6 and 0.4, respectively). None of these stigma-related measures was associated with HIV testing among men, however.

Researchers analyzed data from the Cape Area Panel Study, which followed a representative sample of adolescents in Cape Town as they moved into young adulthood. The current analysis used data from the study’s 2009 wave, which was conducted when participants were aged 20–30, and was restricted to HIV-negative black respondents who had had sexual intercourse. During interviews, participants reported whether they had ever been tested for HIV and provided information on their demographic, social and other characteristics. In addition, they answered questions pertaining to three dimensions of stigma—holding stigmatizing attitudes, perceiving stigma in the community and directly observing enacted stigma (i.e., witnessing action indicative of stigma). For each dimension, participants answered three questions; responses were analyzed both in a binary fashion (whether participants had answered “yes” to any of the three questions) and in a continuous fashion (by tabulating the number of “yes” responses). The researchers computed descriptive statistics for the sample and performed bivariate and multivariate analyses to assess associations between HIV-testing history and each stigma dimension.

The analytic sample consisted of 553 men and 674 women. On average, respondents were 25 years old, had 10 years of education and lived in households with a monthly income of about US$100. Only 4% of men and 12% of women were married.

Overall, 55% of men and 38% of women held at least one stigmatizing attitude (e.g., believing that “HIV is a punishment for sleeping around”); 57% and 62%, respectively, perceived that others stigmatized HIV-positive individuals (e.g., by treating them badly or unfairly); and 35% and 45% had directly observed others treating HIV-positive persons in a manner consistent with stigma (e.g., rejecting them as family members). Larger proportions of men than of women answered “yes” to each of the questions pertaining to stigmatizing attitudes, whereas larger proportions of women than of men reported having observed enacted stigma.

Overall, 63% of men and 69% of women had been tested for HIV. In multivariate analyses among men, no associations were apparent between any of the dimensions of stigma and the odds of having been tested, regardless of whether the dimension was assessed in a binary fashion or continuous fashion. Men’s odds of testing increased with number of years of education (odds ratio, 1.1), and they were higher among men who correctly answered both rather than neither of two HIV knowledge questions (1.6) and those who knew someone with HIV (1.4–1.5, depending on whether the model treated stigma as a binary or continuous variable). Relative to peers who believed that they had no risk of infection, men who did not know their risk were less likely to have been tested (0.5).

Women, however, had sharply elevated odds of having been tested for HIV if they held any stigmatizing attitude (odds ratio, 3.0), and the odds rose with the number of attitudes endorsed (2.0 for each item). On the other hand, women were less likely to have been tested if they perceived stigma in others (0.6) or had observed others enacting stigma (0.4), and the odds fell as the number of questions answered affirmatively rose (0.7 for each). Women’s likelihood of having been tested was elevated if they knew someone with HIV (1.4–1.5, depending on the model); it was lower if their perceived risk of infection was low (0.6), moderate or high (0.2–0.3) or unknown (0.5), rather than none.

The findings were generally similar in additional analyses that examined each stigma question individually. They were also much the same after exclusion of participants who had been tested for HIV by 2006 (to assess associations with recency of testing) or of participants who reported having had an STI or having ever been pregnant (as HIV testing is routinely offered to those groups).

The study may have been limited by unmeasured confounders, its cross-sectional nature, self-report bias, attrition, questionable generalizability to older persons and the potential influence of repeated questioning about HIV over time, according to the researchers. The results nonetheless underscore the need to investigate various dimensions of stigma individually, and to examine associations with HIV infection separately for men and women, they maintain. The seemingly counterintuitive finding that women with stigmatizing attitudes had elevated odds of having been tested for HIV may reflect that such women had assumed that their test result would be negative, the investigators speculate; the observed sex differences may be related to the much higher prevalence of infection among women than among men, and to women’s lower ability to avoid stigma, in this setting. “Our findings…point towards the need for interventions designed specifically to limit the degree to which stigma in the broader social environment discourages the uptake of HIV testing among young women,” the researchers conclude.—S. London


1. Maughan-Brown B and Nyblade L, Different dimensions of HIV-related stigma may have opposite effects on HIV testing: evidence among young men and women in South Africa, AIDS and Behavior, 2014, 18(5):958–965.