Intimate Partner Violence Linked to HIV Infection Among Ugandan Women

First published online:

Intimate partner violence is associated with subsequent HIV infection among rural women in Uganda, according to a longitudinal study.1 At their final assessment, 28% of participants reported having been sexually, physically or verbally abused by an intimate partner in the past year, and 59% reported having ever experienced such abuse. Women who had ever been abused by a partner were more likely than those who had not to be infected with HIV by the study’s end (adjusted incidence rate ratio, 1.6). The researchers estimate that if the association between intimate partner violence and HIV is causal, then 22% of HIV infections in the study would not have occurred in the absence of such violence.

The data are from surveys conducted as part of the ongoing Rakai Community Cohort Study, which follows about 12,000 participants from 50 communities in southwestern Uganda. Participants are interviewed and tested for HIV every 10–18 months; those who drop out of the study are replaced. Each survey round collects data on experiences of intimate partner violence in the previous year; collection of data on other aspects of intimate partner violence (e.g., frequency of abuse) varies by survey year. The surveys also collect demographic information and data on participants’ relationships (length, type and age differential between partners) and their sexual history and behavior (early sexual abuse, condom use, pregnancy intention and number of partners in the last year).

The authors used data on 10,252 women who had participated in at least two interviews in 2000–2009, been in a sexual relationship during all or part of that period and tested negative for HIV in 2000 (or at baseline if they entered after 2000). HIV infection was assumed to have occurred in the round in which a woman tested positive, unless test results were missing for the previous round, in which case time of infection was assumed to be the midpoint between the last negative and first positive test. Participants who reported having ever been abused completed an average of 5.7 years of follow-up, and those who reported no abuse an average of 5.2 years.

By the end of the study’s follow-up period, 59% of participants had ever experienced intimate partner violence. Seventy-seven percent of these women were in a polygamous or monogamous marriage, and 13% had never been married; among women who reported no intimate partner violence, the proportions were 57% and 33%, respectively. Most women, regardless of experience of abuse, reported having engaged in certain risky sexual behaviors, including using alcohol before sex and not using condoms, but the vast majority reported having had only one partner in the past year.

In the final interview they completed, more than one-quarter (28%) of participants reported having experienced at least one type of intimate partner violence in the past year; 22% reported verbal abuse, 17% physical abuse and 16% sexual abuse. Of those who had been abused by a partner in the past year, 32% had experienced two forms of abuse, and 31% had experienced all three. Overall, 59% of women reported having ever been a victim of intimate partner violence; 53% reported verbal abuse, 45% physical abuse and 35% sexual abuse.

In a multivariate analysis that adjusted for demographic and relationship characteristics, the risk of HIV infection was heightened among women who had ever experienced verbal abuse (incidence rate ratio, 1.7), physical abuse (1.6), sexual abuse (1.6) or any of the three (1.6). Rate ratios were lower, though still elevated, among women who in the past year had experienced physical abuse (1.4), verbal abuse (1.4) or any abuse (1.3). The associations between HIV and intimate partner violence tended to become stronger as the frequency of abuse increased: While there was no association between low levels of abuse (1–2 episodes) and subsequent HIV infection, the incidence rate ratio was elevated among those who reported 3–5 episodes (1.5) and reached 3.0 for those reporting more than 20 episodes over the course of a year. However, HIV risk did not differ according to the severity of abuse. Characteristics of participants’ sexual history (pregnancy intention and early sexual abuse) did not modify the relationship between HIV and intimate partner violence, nor did condom use in the past year or number of partners in the past year.

Sixty-three percent of women who tested positive for HIV had been exposed to intimate partner violence at some point before the round in which HIV infection was detected. The authors estimated that if the relationship between intimate partner violence and HIV is causal, 22% of HIV infections in the study population would not have occurred in the absence of intimate partner violence.

The authors note several limitations of the study, including having limited rounds of data on some measures. In addition, they lacked data on men; such information, if available, could have shed light on whether the heightened HIV risk among abused women is attributable to elevated HIV rates among men who perpetrate abuse. Strengths of the study include its longitudinal design, large sample size and inclusion of all three forms of intimate partner violence.

While acknowledging that the pathway between intimate partner violence and HIV is not adequately understood, the authors note that women who experience abuse appear to be at increased risk of contracting the virus. They suggest that intimate partner violence be discussed in HIV-prevention programs, a step they believe may "lead to referring women who are experiencing [intimate partner violence] or who have experienced past [intimate partner violence] for counseling, legal assistance, and other community services, as well as consideration for female-controlled HIV prevention interventions, such as pre-exposure prophylaxis."—H. Ball


1. Kouyoumdjian FG et al., Intimate partner violence is associated with incident HIV infection in women in Uganda, AIDS, 2013, 27(8):1331–1338.