Kristin L. Dunkle, Department of Behavioral Sciences and Health Education and Center for AIDS Research, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Room 540, Atlanta, GA 30322, USA.
A growing body of international research documents strong associations between gender-based violence and HIV, both in the general population and among high-risk subpopulations such as female sex workers. The causal pathways responsible are multiple and complex, thus conceptual clarity is needed to best inform population-based, clinical, and individually oriented interventions. Our brief overview is intended to provide an introduction to the research on the various mechanisms that link GBV to HIV risk. We review the evidence, describe the causal pathways, provide a conceptual framework, and outline prevention and intervention priorities at both the individual and population levels.
Gender-based violence (GBV1) affects 30–60% of women worldwide with significant consequences for mental, physical, and sexual health.[1, 2] Predominant among these health outcomes are sexually transmitted infections, including HIV. Over a decade of research around the world supports a multifaceted causal link between GBV and HIV infection. Beginning in the late 1990s, cross-sectional research from Africa and India consistently found that women who had experienced physical or sexual violence from male intimate partners (IPV) were more likely to have prevalent sexually transmitted infections, including HIV.[3-9] For example, in Tanzania, women seeking voluntary counseling and testing who had experienced partner violence were more likely to be HIV positive [adjusted odds ratio (aOR) = 2.39: 95% confidence interval (CI): 1.21, 4.73]; this association was exceptionally important for women under 30 who were about 10 times more likely to be HIV positive if they reported IPV. In South Africa, women seeking routine antenatal care who reported a history of physical or sexual IPV were 53% more likely to test HIV positive (aOR = 1.53: 95% CI 1.10, 2.04) and those experiencing high levels of gender power inequality in male–female relationships were 56% more likely to test HIV positive (aOR = 1.56: 95% CI 1.15, 2.11). In a study of over 28,000 married women in India, those who had experienced both physical and sexual IPV were over 3 times more likely to be HIV positive than those who reported no IPV (aOR = 3.92: 95% CI 1.41, 10.84). More recently, longitudinal research from South Africa affirmed that young women who had experienced either IPV and/or high levels of gender inequality in their sexual relationships with men were at elevated risk of acquiring HIV over a 2 year follow-up period, with increasingly severe violence associated with increasing risk of new HIV infection; this strongly suggests a causal link. Evidence from India echoes the evidence for the causal nature of these associations, with spousal sexual violence significantly associated with risk for incident sexually transmitted infections. In light of the evidence of causal association offered by these data, we review the pathways that link GBV to HIV both in the general population and within a group at uniquely high risk for both violence and HIV: women involved in sex work or female sex workers (FSWs) (Fig. 1).
What are the pathways from gender inequality and gender-based violence to HIV?
The first, and most obvious, pathway from GBV to HIV is direct infection through an act of sexual assault, with transmission facilitated by the genital and/or anal trauma that can accompany unwanted or forced sex.[12, 13] Yet, while HIV transmission via acts of rape unquestionably occurs, a growing body of evidence suggests that the increased HIV risk caused by violence at the population level is not limited to, or even primarily driven by, sexual assault. For example, women who report physical violence without accompanying sexual violence from male partners consistently show increased risk of STI and HIV.[4, 10, 14, 15] This pathway is most plausible biologically if men who are physically violent are also more likely to be HIV positive, and indeed this appears to be the case. Evidence from around the world suggests that men who perpetrate violence engage in higher levels of risky sexual behavior, including multiple and concurrent sexual partnerships, transactional sex and substance use.[3, 16-22] In turn, such men are also more likely to report STI symptoms,[19, 22] STI infection and to be confirmed as HIV positive in studies where serostatus data are available.[3, 24] Thus, a second important piece of the causal pathway from GBV to HIV is that male perpetrators of violence are more likely to be HIV positive. This increases the risk for survivors, particularly survivors of IPV who may be having repeated sexual contact with the perpetrator, even where the sex in the relationship is not overtly violent.
Violence can also increase HIV risk via multiple, interlinked, indirect pathways. Violence reduces victims' abilities to influence the timing and circumstances of sex, resulting in more unwanted sex and less condom use, including situations where women are coerced or pressured not to use condoms.[18, 25-27] Highlighting the confluence of these risks, research from India, which analyzed data from over 20,000 husband–wife dyads, confirmed that abused wives face increased HIV risk, not just because of greater likelihood of HIV infection among abusive husbands, but also because of elevated HIV transmission within abusive relationships. Experiences of violence can also influence subsequent patterns of sexual risk taking among survivors. In both developed and developing countries, past exposure to GBV and controlling behavior from a sexual partner is consistently associated with subsequent high-risk sexual behavior, including multiple and concurrent sexual partnerships, increased numbers of overall partners, lower levels of condom use, increased substance use and sex while intoxicated, and increased participation in transactional sex as well as commercial sex work.[4, 6, 15, 28-34] This risk increase in part reflects the psychological impact of violence, which can last many years after the violent acts. Women who have survived abuse may be more willing to engage in risky sex, and less able to refuse unwanted advances, especially when drunk, drugged, dissociating, seeking affection, or otherwise unable to resist controlling male partners.[33, 34, 36] Emerging evidence suggests a very similar link between violence and increased risk of HIV among MSM in Western countries.[37-46] All survivors of GBV are known to be at higher risk of revictimization or subsequent experience of additional violence.[47, 48] Because GBV increases risk of later GBV as well as HIV, and because the link between GBV and HIV risk seems to show a dose–response effect,[4, 10] revictimzation is an important phenomenon through which the link between GBV and HIV may be amplified.
Finally, we must consider the powerful social backdrop from which these risk pathways arise and within which they play out; this upstream social context plays a critically important role in promoting and perpetuating linked risks of GBV and HIV. Male perpetration of violence is underpinned by dominant social norms about masculinity, femininity, and sexuality. Qualitative research shows that the intersections of HIV, gender inequality and GBV lie in the patriarchal nature of most societies, especially in ideals of masculinity that are predicated on control of women and valorize male strength and toughness. These ideas are sometimes referred to as ‘hegemonic’ gender roles–the most dominant and highly valued ideas about how men and women are supposed act in a given cultural context. Violence against women, including IPV and rape, is one consequence of gender inequality; however, such violence also serves to reinforce and perpetuate gender inequality and at both societal and relationship levels.[49, 50] One key facet of hegemonic gender roles in most societies is what is sometimes called ‘compulsory heterosexuality’– the misguided idea that ideal masculinity and femininity are inherently heterosexual and that other sexual orientations are defiant, deviant, and/or pathological. Much like gender inequality fuels violence against women, heterosexism, homophobia, and transphobia fuel perpetration of violence against people perceived to be lesbian, gay, bisexual, transgendered, or queer (LGBTQ) and are in turn reinforced by such violence. Finally, some forms of violence between men, including much male-on-male sexual violence, are driven by and reinforce power hierarchies between men, particularly the dominance of stronger and more powerful men over men perceived as weaker or less masculine.[52-55] The multiple pathways from GBV to HIV thus emerge from self-reinforcing cycles of hegemonic gender roles and heterosexuality and violent acts. In addition to meeting the immediate needs of violence survivors and those at risk for abuse, addressing this upstream origin is a crucial goal for effective primary prevention of both GBV and HIV.
Unique Considerations for a High-Risk Population: Female Sex Workers
All of these dynamics are magnified in intensity among certain high-risk groups. Women involved in the trade of sex for money or other resources, also known as FSWs, are a uniquely high-risk population with regard to HIV, and demonstrate an almost 14-fold greater burden of infection. Moreover, a growing evidence base demonstrates that FSWs suffer alarming levels of physical and sexual violence perpetrated by partners, clients, police, and other actors.[28, 57-60] Consistent with evidence from the general population, violence against FSWs is associated with HIV infection as well as STI symptoms and STI infection.[28, 57]
The risk pathways described for the general population; that is, limited control over sex with high-risk partners, perpetuated by a culture of social norms that promotes abuse and sexual risk behavior, appear relevant for this population. FSWs face several additional and unique risks that should be noted with regard to violence and HIV. For example, severe sexual violence at initiation of sex work has been observed among sex workers, particularly among those forced or coerced into sex work[58, 61-63]; such experiences are associated with increased sexual risk, STI symptoms, and HIV infection. Sexual violence at initiation to sex work is often described as unprotected,[61, 62] particularly when clients perceive, or have been lead to believe, that younger women are sexually inexperienced. Sexual violence upon initiation into sex work conveys a clear message of disempowerment to women who face such experiences, thus conferring a long-term impact on sexual risk.
As with the general population, FSWs may abandon attempts to use condoms to prioritize their immediate needs for physical safety when threatened with violence or coercion. Overt or implicit threat of violence can operate as a coercive tactic for unprotected sex. Client aggression and threats of violence are described as intended to limit the ability of those involved to assert themselves or negotiate condom use. FSWs from a wide range of settings report client condom refusal and client insistence on unprotected sex. In addition, forced sex is often unprotected. FSWs exposed to violence are more likely to face client pressure for unprotected sex as well as overt client condom refusal. Violence has been linked with condom non-use, condom breakage[57, 65] and condom failure,[58, 69] suggesting that coercive and forceful dynamics can often underpin unprotected sex among FSWs.
FSWs also face unique pressures and forms of coercion–overt or implied pressure in the absence of physical force that may prompt them to prioritize their physical safety or immediate needs over HIV prevention. For example, sex workers often report economic coercion for unprotected sex, that is, pressure for unprotected sex in exchange for greater money or other resources.[23, 59, 66, 70, 71] While rarely quantitatively assessed, violence and coercion also appear to be associated with higher risk sex for FSWs. For example, qualitative investigation documents forced anal sex[67, 72] as well as coercive anal sex (for example, anal sex which cannot be refused for fear of not being paid). The limited epidemiological inquiry in this area illustrates that violence is associated with anal sex among FSWs, which in turn poses STI/HIV risk. Finally, illustrating the power of coercion in the absence of physical force, sex workers may also be forced into situations with multiple sex partners, whereby an arrangement with one client results in subsequent force or coercion into sex with additional men that the client has arranged.[28, 72, 73] Together these data illustrate the need to address the intersections of violence and HIV among FSWs, and consider additional sources of violence-related HIV risk for this highly affected population.
These data confirm that the causal pathways linking GBV and HIV are complex and multifaceted. Effective prevention is possible but requires similarly multifaceted approaches to be maximally effective. At the individual level, survivors of GBV are clearly a high-risk group who must be supported in seeking the full range of HIV prevention and treatment services. Appropriate clinic-based screening and referrals can connect violence survivors of all genders with violence-related support services, which can in turn improve mental health outcomes.[75-77] Investing in supporting survivors of abuse as they seek appropriate HIV testing, care, and treatment, and integrating violence-related support within HIV prevention and clinical care, is essential, particularly for high-risk groups such as FSWs.
At a population level, the unequal gender norms that give rise to violence against women and LGBTQ individuals also increase men's risk of contracting HIV and of passing it on to their partners. GBV, gender inequality, and homophobia/heterosexism are thus critical dynamics that help drive HIV transmission, including in high-risk contexts such as sex work and substance use. Best practice HIV prevention programmes, such as Stepping Stones[78-81] and IMAGE (Intervention with Microfinance for AIDS and Gender Equity),[82-86] therefore address both the broader context of gender inequality as well as the risk of infection through acts of violence, and in particular include efforts to engage men in accountable partnership in prevention.
We would like to acknowledge the contribution of fellow participants at the SSRC Scientific Research Planning Meeting: Sexual Violence and HIV Transmission in offering support for this writing. Dr. Dunkle's work was supported in part by P30 AI050409.
We use the term GBV to include most anti-LGBTQ violence and some types of violence between men.