Women in HIV serodiscordant relationships less likely to take PrEP consistently if they experience intimate partner violence

Experiencing intimate partner violence is associated with an increased risk of poor adherence to HIV pre-exposure prophylaxis (PrEP) among women in serodiscorant relationships in sub-Saharan Africa, investigators report in the online edition of the Journal of Acquired Immune Deficiency Syndromes. Overall, 16% of women experienced intimate partner violence (IPV) and this increased the risk of suboptimal adherence to PrEP by 50%, when measured by either pill count or plasma tenofovir concentrations.

“This is the first study to examine the association between IPV and PrEP adherence,” write the authors. “Women who reported IPV in the past 3 months had an increased risk of low PrEP adherence.”

PrEP has been shown to reduce the risk of infection with HIV in a range of populations, including gay men, heterosexual men and women and injecting drug users. The World Health Organization (WHO) therefore recommends PrEP as part of comprehensive HIV prevention programmes targeted at high-risk populations.

The effectiveness of PrEP is related to adherence. Intimate partner violence has been associated with higher HIV incidence, reduced condom use and suboptimal adherence to antiretroviral therapy. It is therefore possible that intimate partner violence may also affect adherence to PrEP.

Investigators from the recent Partners PrEP study therefore analysed data obtained from 1785 HIV-negative women in serodiscordant relationships and enrolled in the study. At monthly face-to-face interviews, the women were asked to report their experience of verbal, physical or economic intimate partner violence.

The investigators assessed the relationship between partner abuse and suboptimal adherence to PrEP. Two measures were used to assess adherence: pill count (less than 80% of doses defined as low adherence) and measurement of plasma tenofovir levels (low adherence defined as levels below 40 ng/ml). In-depth interviews with a subset of women provided insights as to how intimate partner violence affected adherence and also the strategies individuals used to maintain adherence to PrEP in the context of violent relationships.

Participants had a mean age of 33 years and 70% had earned an income in the previous three months. The vast majority (99%) were married. The mean relationship duration was 13 years and women had been in mutually disclosed serodiscordant relationships for a mean of 1.4 years.

During 35 months of follow-up, 288 women (16%) reported intimate partner violence at 437 study visits (0.7% of total). Of these women, 69% reported intimate partner violence at one visit, 20% at two visits, 7% at three visits and 5% at four or more visits. The most common form of intimate partner violence reported was verbal, followed by physical and economic.

Women reporting intimate partner violence in the past month were less likely than women reporting no partner abuse to have had recent sex with their study partner (69% vs. 81%) but more likely to report recent unprotected sex (22% vs. 13%). They were also more likely to have had partners who reported sex with another partner (20% vs. 15%).

Those reporting intimate partner violence were similar in most respects to women who did not report violence from their partner.

Adherence as assessed by pill count was high (95%) among most women, regardless of reported intimate partner violence. Pill count suggested adherence below 80% at 7% of study visits and 32% of plasma tenofovir measurements were below optimum levels.

Overall, women were 50% more likely to have suboptimal adherence to PrEP if they had experienced intimate partner violence in the previous three months. This association was consistent regardless of whether adherence was measured by pill count (aRR, 1.51; 95% CI, 1.17-1.89, p = 0.001) or plasma concentrations of tenofovir (aRR, 1.51; 95% CI, 1.06-2.15, p = 0.02).

However, the impact of intimate partner violence on adherence ceased to be significant after three months.

When types of intimate partner violence were considered separately, the investigators found a significant relationship between suboptimal adherence and verbal abuse (aRR = 1.65; 95% CI, 1.17-2.33, p = 0.005) and  low adherence and economic partner violence (aRR = 1.48; 95% CI, 1.14-1.92, p = 0.003). The relationship between poor adherence and physical partner violence was not significant, but increased frequency of physical abuse from a partner was associated with lower adherence to treatment (p < 0.001).

There were a total of 48 new HIV infections among the women. However, experiencing intimate partner violence did not significantly increase the risk of seroconversion.

Seven women discussed intimate partner violence during in-depth interviews with staff. Reasons why abuse from partners affected adherence included stress and forgetfulness, running away and leaving medication behind and partners throwing pills away. Strategies to overcome these challenges and maintain high adherence included sending children to retrieve pills that had been thrown away, or explaining events to clinic staff who were able to offer replacement therapy.

“Efforts to target PrEP towards women with IPV should recognise the low risk of adherence, and interventions should be evaluated to promote PrEP adherence in the context of violence,” conclude the authors. “Some women in our study reported strategies to maintain adherence in the face of IPV, and lessons from these examples of resilience could help in developing successful interventions. Such interventions could increase the benefit of PrEP by promoting effective use in a population at high risk of HIV.”