Intimate partner violence associated with subsequent HIV infection in Uganda

Ugandan women who have been subject to violence from a sexual partner are more likely than other women to go on to acquire HIV, according to a large, longitudinal study from the Rakai cohort, published in the May 15 issue of AIDS. Women who had experienced more severe forms of violence, more frequently, or over a longer period of time had greater risks of HIV infection.

Violence against women is a serious and common human rights and public health problem, which causes significant morbidity and mortality. 'Intimate partner violence' (IPV) is one form of violence and has been defined as "behaviour within an intimate relationship that causes physical, sexual, or psychological harm, including acts of physical aggression, sexual coercion, psychological abuse and controlling behaviours".

Several (but not all) cross-sectional studies have identified an association between intimate partner violence and HIV infection. However, these 'snapshot' studies are unable to tell us what comes first – the violence or the HIV.

More reliable are prospective, longitudinal studies and two such studies have previously been published. Data from a South African cohort found that experience of physical or sexual IPV increased the risk of HIV infection (incidence rate ratio 1.51). In a seven-country African cohort, physical or verbal IPV appeared to increase the risk of infection but this was not statistically significant (incidence risk ratio 1.69).

In order to better understand the possible role of intimate partner violence in women’s vulnerability to HIV, Dr Fiona Kouyoumdjian and colleagues examined a decade’s worth of data from a cohort in Rakai, Uganda. There were 10,252 female participants who took part in interviews on multiple occasions and were HIV negative at the first interview. On average, women stayed in the cohort for five and a half years.

Rakai is a rural district in Uganda and agriculture was the main occupation for the majority of participants. Half the cohort were under the age of 24 and two thirds had less than seven years of schooling. Only one-in-five participants had never been married.

Most of the women (58.9%) had experienced intimate partner violence (IPV) at least once in their lifetime. In the past year, 16.0% had experienced sexual IPV, 16.8% physical IPV and 22.4% verbal IPV. Women often suffered more than one form of violence.

Just under one-in-ten women became HIV positive while in the cohort. The HIV infection rate was higher for women who had ever experienced intimate partner violence than for women who had not experienced IPV.

After statistical adjustment, women who had ever experienced IPV had an incidence rate ratio of 1.55 (95% confidence interval 1.25-1.94). This is comparable to the results seen in the two previous prospective studies.

Similar statistically significant results were seen for ever experiencing each form of intimate partner violence (physical, sexual, verbal) as well as for experiencing some forms of IPV in the past year.

When IPV was recorded as being 'severe' rather than 'minor', the HIV risk tended to be greater (e.g. ever experiencing severe physical violence, 1.96 [95% CI 1.46-2.63]).

There was a stepwise relationship between the number of times a woman experienced intimate partner violence and her increased risk of HIV infection. For example, women who reported five or more IPV events while in the cohort had an incidence risk ratio of 1.82 (95% CI 1.06-3.10); women with more than 20 events had a risk ratio of 3.03 (95% CI 1.83-5.01).

The association between violence and HIV infection could not be explained by condom use or by partner numbers. When these factors were controlled for, the association remained the same.

The researchers calculated that the population attributable fraction of infections that are associated with IPV is 22.2%. In other words, if intimate partner violence could be eliminated, there would be 22% fewer HIV infections in this group of women.

What is the link between violence and HIV infection?

The Rakai data, as well as the previous studies, appear to identify an association between intimate partner violence and subsequent HIV infection. But what are the mechanisms that link these two events? How is it that violence raises the risk of infection?

One possible mechanism is that forced sex could cause physical trauma and so increase the risk of transmission. However, the finding that it is all forms of violence, and not just sexual violence, that are associated with infection suggests that this cannot be the sole mechanism.

Alternatively, the experience of intimate partner violence may have an impact on women’s ability to negotiate safer sex or willingness to have more risky sex, either with the perpetrator of violence or with other partners. However, the Rakai data on condom use and partner numbers do not support this.

Another possible explanation of the link is not that violence increases the risk of HIV, but that HIV increases the risk of violence. For example, a woman suffers violence after disclosing her HIV status to her partner.

Finally, perhaps men who are violent may be more likely to have HIV, because some norms of masculinity encourage both men's control of women using violence and their sexual risk-taking.

Whatever the explanation, women attending health services for HIV testing and counselling should be asked about intimate partner violence and referred to appropriate services, say the authors. Moreover, more should be done to prevent violence, "as a means of both stopping the psychological and physical consequences of IPV and potentially of preventing HIV".