In South Africa, Some HIV Risk Factors Also Predict Intimate Partner Violence
Among young rural South African women, having experienced intimate partner violence is strongly associated with certain risk factors for HIV infection, according to an observational analysis.1 Women who had experienced such violence were more likely than others to have had at least three sexual partners in the past year (odds ratio, 2.1), to have had sex recently (1.8) and to have a more educated partner (1.5)—all factors also associated with elevated odds of HIV infection. Intimate partner violence was associated with increased odds of HIV infection before these HIV risk factors were taken into account, but not afterward.
Researchers analyzed information collected from sexually experienced women aged 15–26 years who lived in rural villages in the Eastern Cape province of South Africa and were participating in a trial of a behavioral intervention to prevent HIV infection. Most of the women were recruited from schools and lived with their families. At the start of the trial, they completed a questionnaire and gave a blood sample for HIV testing. The questionnaire asked about intimate partner violence (physical and sexual), as well as social and demographic factors, knowledge and attitudes about sexual and reproductive health and HIV, adverse childhood events (including sexual abuse), substance use, characteristics of the current or most recent main sexual partner, and sexual activity and practices.
The 1,295 women studied were 19 years old, on average. About 12% were infected with HIV, and 27% had experienced intimate partner violence, defined as more than one episode of physical or sexual violence perpetrated by an intimate partner. In an unadjusted analysis, relative to their counterparts who had not experienced such violence, women who had were more likely to be HIV positive (odds ratio, 1.6).
In a first bivariate analysis comparing HIV-positive and HIV-negative women, the positive women were older (20 vs. 19 years), and a larger proportion had ever been pregnant (34% vs. 20%). On average, the age difference between women and their partners was greater in the HIV-positive group (four vs. three years); moreover, larger proportions of HIV-positive women had partners who were at least three years older than they were (62% vs. 45%), who had at least a high school education (54% vs. 32%) and who earned money (45% vs. 35%). Women in the positive group had also had more partners in the past year (two vs. one), and a larger proportion had had at least three partners during that time (22% vs. 9%) and had been sexually active in the past three months (90% vs. 72%).
In a second bivariate analysis comparing women with and without a history of intimate partner violence, larger proportions of those who had experienced such violence had had a casual sexual partner (35% vs. 17%) and transactional sex with a casual partner (16% vs. 6%) or a main partner (28% vs. 20%). In addition, larger proportions of the women who had been assaulted by intimate partners had had sex in the past three months (82% vs. 71%) and three or more partners in the past year (19% vs. 7%). They also had higher mean scores on a variable measuring the experience of adverse events in childhood.
In a multivariate analysis assessing risk factors for HIV infection, women's odds of being HIV positive increased with each year of age (odds ratio, 1.4). In addition, four sex- and partner-related factors increased their risk of being positive—having at least three sexual partners in the past year (2.5), having had sex in the past three months (3.4), having a partner who was at least three years older (1.7) and having a partner with at least a high school education (1.9). Women who had experienced intimate partner violence did not have elevated odds of HIV infection in this adjusted analysis.
In a set of multivariate models testing associations between selected variables and the four sex- and partner-related HIV risk factors, women who had experienced intimate partner violence were more likely to have had at least three sexual partners in the past year (odds ratio, 2.1), to have had sex in the past three months (1.8) and to have a partner with at least a high school education (1.5). Other factors associated with an increased risk of having at least one of the four HIV risk factors included younger age (1.3), adverse childhood experiences (1.4), a greater number of years since sexual initiation (1.2–1.3), higher socioeconomic status (1.2), more egalitarian attitudes toward gender relations (1.4), having had transactional sex with a casual partner (1.9) and having been pregnant (1.9). On the other hand, women were less likely to have certain of these HIV risk factors if they were older (0.8), had higher resistance to peer pressure to have sex (0.7) and had a less controlling partner (0.8).
Taken together, the researchers assert, the results suggest that certain risky sexual behaviors—having more frequent sex and having more partners—mediate the association between intimate partner violence and HIV infection among young rural South African women. These results, they write, add to the evidence linking gender-based violence and HIV infection, and support the contention that whether it is direct or indirect, "there is an underlying, enduring, and strong association between experience of intimate partner violence and HIV risk practices." In conclusion, they say, "This suggests that undertaking efforts to promote gender equity and reduce levels of intimate partner violence are of critical importance for HIV prevention."—S. London
1. Jewkes R et al., Factors associated with HIV sero-status in young rural South African women: connections between intimate partner violence and HIV, International Journal of Epidemiology, 2006, 35(6):1461–1468.