To investigate the associations between intimate partner violence, rape and HIV among South African men.
We tested hypotheses that perpetration of physical intimate partner violence and rape were associated with prevalent HIV infections in a cross-sectional household study of 1229 South African men aged 18–49. Violence perpetration was elicited in response to a questionnaire administered using an Audio-enhanced Personal Digital Assistant and blood samples were tested for HIV. A multivariable logistic regression model was built to identify factors associated with HIV.
18.3% of men had HIV. 29.6% (358/1211) of men disclosed rape perpetration, 5.2% (63/1208) rape in the past year and 30.7% (362/1180) of had been physically violent towards an intimate partner more than once. Overall rape perpetration was not associated with HIV. The model of factors associated with having HIV showed men under 25 years who had been physically violent towards partners were more likely to have HIV than men under 25 who had not (aOR 2.08 95% CI 1.07–4.06, p = 0.03). We failed to detect any association in older men.
Perpetration of physical IPV is associated with HIV sero-prevalence in young men, after adjusting for other risk factors. This contributes to our understanding of why women who experience violence have a higher HIV prevalence. Rape perpetration was not associated, but the HIV prevalence among men who had raped was very high. HIV prevention in young men must seek to change ideals of masculinity in which male partner violence is rooted.
Citation: Jewkes R, Sikweyiya Y, Morrell R, Dunkle K (2011) The Relationship between Intimate Partner Violence, Rape and HIV amongst South African Men: A Cross-Sectional Study. PLoS ONE 6(9): e24256. doi:10.1371/journal.pone.0024256
Editor: Sten H. Vermund, Vanderbilt University, United States of America
Received: March 4, 2011; Accepted: August 7, 2011; Published: September 14, 2011
Copyright: © 2011 Jewkes et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by UK Department For International Development (DFID). The grant was managed by their local partner Human Life Sciences Partnership (HLSP). Dr. Morrell participated in the project with funding from the National Research Foundation. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Gender-based violence and gender inequality are increasingly recognized as important structural drivers of the HIV epidemic in women , with research evidence from highly diverse cultural settings –. In attempts to explain this, authors have suggested that gender inequality and violence reduce women's ability to choose when and with whom to have sex and to protect themselves in sexual encounters. Other work suggests that men who are violent may be more likely to be HIV infected , . Gender theory argues for the existence in many settings of masculinities that emphasise dominance and control of women, and legitimate use of violence to achieve and demonstrate this –. These ideas of masculinity are predicated on patriarchy and heterosexual success. Associated with them is a configuration of practice valorises having multiple sexual partners, and toughness or nonchalance related to risk, the latter often translates into non-condom use. In South Africa the hegemonic African masculinity is framed around such ideas of heterosexual success and control of women and is believed to provide the predominant underlying connection between men's use of violence against women and HIV , .
There is considerable evidence that men who are more violent and controlling engage more often in risky sexual practices and are more likely to have sexually transmitted infections –. But there is yet no published research showing associations between perpetration of rape and HIV sero-status, and no population-based study from a country with a generalized HIV epidemic on associations between HIV and IPV. Furthermore the limited evidence conflicts. In one study of young South African men the prevalence of perpetration of rape of a non-partner and of more than one episode of physical or sexual intimate partner violence was very similar among men with and without HIV (in each case 12% v. 16%) . In contrast in India, married men were more likely to be HIV infected if they had perpetrated violence (aOR 1.91 95% CI 1.11, 3.27) . Both these analyses, however, were based on very small numbers as the HIV prevalence was low, being 2% and 0.4% respectively.
Understanding the association between gender inequalities and gender-based violence and HIV, including the role of men's sero-status in these, is important for HIV prevention. Furthermore understanding whether men who rape may have a higher prevalence of HIV is critical for provision of post-rape care and ensuring that post-exposure prophylaxis for HIV is available . To explore these issues in South Africa, a population based survey was conducted in two provinces with the aim of describing the prevalence and associations between rape, IPV perpetration and HIV sero-status among adult men and testing hypotheses that rape and IPV perpetration are associated with prevalent HIV infection.
Ethics approval was given by the Medical Research Council's Ethics Committee. The study was undertaken in 2008 in three districts in the Eastern Cape and KwaZulu-Natal provinces of South Africa. These form a contiguous area, and include rural areas with communally-owned land under traditional leadership, as well as commercial farms, small towns, villages, and a city, inhabited by people of all South African racial groups, several ethnic groups (predominantly Xhosa and Zulu) and socio-economic backgrounds.
The sample used a two stage proportionate stratified design to identify a representative sample of men aged 18–49 years living in the three districts. Using the 2001 census as the primary sampling frame, 222 census enumeration areas (EAs) were selected as the primary sampling unit, stratified by district and with numbers proportionate to district population size. The sample was drawn by Statistics South Africa. Households in each EA were mapped and twenty were systematically selected. In each household one eligible man was randomly selected to take part in the interview. Men were eligible for the study if they were aged 18–49 years and had slept there the night before. Eligible men were asked to complete a questionnaire and provide a blood sample for HIV testing. This analysis includes only those who agreed to provide blood. There was no replacement.
Of the 222 selected EAs, two (0.9%) had no homes, and in five (2.3%) we could not interview because permission from the local political gatekeepers was declined (1) or we could not access any eligible home after multiple visits at different times of day (4). In all the latter EAs, we established that many households were ineligible due to age or absence of a man. We completed interviews in 215 of 220 eligible EAs (97.7%), and in these in 1,737 of 2,298 (75.6%) of the enumerated and eligible households. However only 70.8% (N = 1,229) of men interviewed provided blood i.e. 53.5% of enumerated and eligible men. Men who did not give blood did not differ in age or race from those who did, but they were more highly educated (47.9% has completed school v. 37.6% of those who gave blood) and less likely to have perpetrated physical IPV (25.0% v. 30.7% of those who gave blood).
Questionnaires were administered in isiXhosa or isiZulu and English using APDAs (Audio-enhanced Personal Digital Assistants) and took 45–60 minutes to complete. They included categorical variables measuring age, education, race, employment and income. Health questions asked about circumcision and history of a genital ulcer. Alcohol consumption in the past 12 months was assessed through a question on frequency of having 5 or more drinks per drinking day.
A 12-item scale assessed men's power and control in their main relationship with a female partner, after Pulerwitz et al , as adapted for South Africa by Dunkle et al . These items were summed to derive a score (Cronbach's alpha 0.78). A typical question was “when I want sex I expect her to agree”. Men were asked about lifetime perpetration of physical intimate partner violence using the modified WHO violence against women instrument . Specific acts of violence were asked about in five items ranging from slapping to threats with or use of a weapon. We have followed previous practice of categorising physical intimate partner violence into none or one episode versus more than one , , , .
We asked about frequency of condom use in the past year and about the number of primary and non-primary partners in the past year and lifetime. Concurrency was assessed by asking about having a khwapheni (an indigenous category of definitionally concurrent non-primary partner). We asked about transactional sex with women, defined as sex that was primarily motivated by a desire for material gain on the part of the woman. This was defined as providing food, cosmetics, clothes, transportation, items for children or family, school fees, somewhere to sleep, handyman work, or cash . Interviewees were asked if they had ever had “sex with a prostitute”.
Rape of women and girls was assessed using seven questions adapted for the study from previously used items, and further validated through cognitive interviewing . A typical item was “How many times have you slept with a woman or girl when she didn't consent to sex or after you forced her?”. Questions asked about rape of a current or ex-girlfriend or wife, rape of a non-partner, gang rape and rape of a woman who was drunk. Men were then asked two questions about perpetration of rape of a man or boy. We also asked about the age of the youngest victim raped.
The HIV tests were conducted on dried blood spots. These were tested with a screen ELISA (Genscreen, Bio-Rad, Steenvorde, France) and positive results confirmed with a second ELISA (Vironostika, bioMérieux, Marcy d'Etoile, France). They were analysed by the National Institute for Communicable Diseases, which participates in a programme supported by the Center for Communicable Diseases Control that has shown these methods can optimally identify HIV-1 from dried blood spots.
The men signed informed consent for the interview and separately for blood, if they agreed to a sample being taken. To recognise their research contribution, they were given R50 (US $6.5). The questionnaire asked men to disclose having engaged in a range of criminal acts. The nature of South African law meant that men could only be protected from possible legal repercussions through complete anonymity. To do this we had to keep no identifying information on interviewees, and this meant that we could not make return visits to offer HIV test results. HIV testing is free and widely available in South Africa and all men received a leaflet and were encouraged to go to their nearest clinic for HIV testing.
Data were downloaded from the APDA memory cards and merged with the HIV test results. The resulting dataset provided a self-weighted sample. Analyses on participants who agreed to give blood (n = 1229) were carried out using Stata 10.0. All procedures took into account the two stage structure of the dataset, with stratification by district and the EAs as clusters. The social, demographic characteristics, sexual and violent practices and established HIV risk factors were summarised as percentages (or means) with 95% confidence limits, using standard methods for estimating confidence intervals from complex multistage sample surveys (Taylor linearization). Pearson's chi was used to test associations between categorical variables. No efforts were made to replace missing data.
To account for clustering of men within EAs, a random effects logistic regression model was fitted to identify factors associated with having HIV. Candidate variables for the model included social and demographic characteristics of the men, their sexual practices, relationship control, intimate partner violence (>1 episode v. none or one), any rape perpetration, substance abuse, having had anal sex, having had a genital ulcer and circumcision. The variables were entered into the model in groups and backwards elimination was used (p<0.2) to derive a final model. A term for stratum was included in the model. Variables for the final model were retained at p≤0.05. We tested for interactions between retained variables and found a significant interaction between perpetration of physical IPV and age; therefore present the association between physical IPV and HIV stratified by age of respondent.
Overall 18.3% (225/1229) of men tested positive for HIV. Perpetration of rape of women was disclosed by 29.6% (358/1211) of men, 5.2% (63/1208) had raped in the past year and 30.7% (362/1180) of ever partnered men had been physically violent towards an intimate partner on more than one occasion. Figure 1 shows the age specific prevalence of HIV and a steep rise is seen to a peak in the 30–34 year age group, 40% of whom are infected. The figure shows that the age specific prevalence among men who have ever raped is very similar to that of all men. Men who have been physically violent towards a partner on multiple occasions have a higher age specific prevalence up to the 30–34 year age group and thereafter it is a little lower than that for all men. 95% confidence intervals are not presented in this figure, but for every data point they overlap.
The social and demographic characteristics of the whole sample and men with and without HIV are shown in Table 1. Most of the men in the study were Black African (85%), and more than half were under 25 years of age. Just over a third (38%) had completed high school or had some tertiary education, and only just over half (54%) had worked in the past year, and this was generally in low paid work as about 80% of those working earned under R2000 (US $270) per month. Men with HIV were older than other men and over half were aged 25–34 years. More of them were Black African (95%). Men with HIV were less well educated than those without HIV, and more were cohabiting, rather than married or single. There was little difference in income and the proportion working.
High risk sexual practices were reported by a high proportion of all men (Table 2). However more men who were HIV positive had ever had transactional sex and sex with more than 20 partners, but a very similar proportion had consistently used a condom in the past year, drank alcohol heavily, had had sex with a prostitute/sex worker or had a khwapheni (concurrent partner). There was little difference in the proportion reporting ever having had anal sex with a man. More men with HIV had had a genital ulcer and fewer had been circumcised. There were some differences in relationship equity, with fewer men with HIV reporting having a relationship that was in the most equitable tertile.
Table 3 presents age-adjusted odds ratios for the association between HIV sero-positivity and perpetration of rape and physical intimate partner violence. Most of the measures of rape perpetration lacked evidence of association with HIV sero-status, however men who had been physically violent towards a partner were more likely to be HIV infected (aOR 1.50 95%CI 1.04, 2.17 p = 0.03).
Factors associated with having HIV are shown in Table 4. Being Black African and having ever had a genital ulcer were strongly associated with having HIV, and being circumcised was protective. Young men (under 25 years) who had been physically violent towards a female partner had twice the odds of having HIV as men of that age who had not been violent (aOR 2.05 95% CI 1.05–4.01, p = 0.034). There was no association between IPV perpetration and HIV among men 25 and older.
Among adult South African men, HIV prevalence was significantly higher in men who had been repeatedly physically violent towards their partner compared to those who had not, however HIV prevalence was not elevated among men who had raped. After adjusting for other factors associated with HIV infection, being physically violent remained associated for younger men. Whilst this could have been a chance finding, the fact that it mirrors that of Maman et al among women in Tanzania, among whom the association between HIV and violence was seen in younger women , makes this less likely. Thus our hypotheses were partially supported by our data.
Whilst we cannot draw conclusions about temporality of associations, the findings of the association between physical partner violence against women and HIV sero-status are in line with the research which has found evidence of a clustering of men's violent and sexually risky practices –. Although similar clustering has also been found around the practice of rape perpetration . Authors have argued that these practices have a common root in a unifying ideal of masculinity , and whilst this is highly plausible the interpretation of our findings on rape perpetration remain a challenge. Nonetheless changing this gender-inequitable ideal must be an important focus of HIV prevention programmes, because there is strong evidence that women in more gender inequitable relationships are at higher risk of HIV incident infections and our findings suggest that so doing will benefit to men and women alike . The findings of the Stepping Stones evaluation in South Africa provide evidence that this can be done, male violence perpetration and risk taking can be reduced through jointly targeted intervention, with impact on sexually transmitted infections .
Not with standing the absence of association between perpetration of rape and HIV, our findings provide little comfort for rape survivors. The HIV prevalence among all men was very high, even if it was not elevated among those who had in the past or recently perpetrated rape. In South Africa post-exposure prophylaxis for HIV after rape is a very important part of post-rape care for survivors who are HIV negative.
The high HIV prevalence among men in the study was as expected in this population. In the country as a whole, for example, in 2008 the prevalence of HIV among African males aged 25–49 was 24% . The high rates of disclosed perpetration of all forms of rape and partner violence was also comparable to findings of research with South African women  and men , . The age structure of the men in the study differed a little from that found in all South African households in the 2003 South Africa Demographic & Health Survey, with fewer men aged 35–49 than the population (17% versus 33% in SADHS) . Thus some caution is needed in generalising prevalences from the study to all men age 18–49 years in the population. Generalizability may be limited by the fact that the men who gave blood for HIV testing differed in some ways from those who refused, however adjusting the model for the socio-demographic variables that differed between those who did and did not test did not alter the findings.
A strength of the study was the use of APDAs for data collection, as these provides a confidential environment in which disclosure of anti-social and illegal behaviour could be enabled. Through removing the face to face component to interviews, APDAs greatly reduced the performative aspect of interview responses and so gave us more confidence that there would not have been a problem of over-reporting. Under-reporting, either deliberately, or through re-construing acts of force as those of persuasion, is a recognised problem in research on rape. It is impossible to estimate the magnitude of this in the study. Recall bias is a potential problem with any study of this nature. Self-completion of the questionnaire and the option of skipping questions resulted in some missing data on some items. We have not replaced missing values. We did not retain information on the number of eligible men per household and so were not able to weight the analysis for this, but we have not reason to believe this would have made much difference to the estimates of association .
This study provides further evidence of the need to engage with men to promote gender equity and reduce partner violence as part of HIV prevention. Whilst other studies have shown that gender inequities and violence are drives of the HIV epidemic in women , this study shows how harmful violent masculinities are for the health of men. These findings confirm the importance of engaging men to promote less violent and more respectful and gender equitable relations with women as part of the agenda of HIV prevention.
We thank the men who agreed to complete the interviews; Statistics SA for drawing the sample; Scott Johnson and his team from the University of Kentucky for development of the APDA systems; the National Institute for Communicable Diseases (NICD) for testing the blood; and all the project staff.
Conceived and designed the experiments: RJ YS RM KD. Performed the experiments: RJ YS RM KD. Analyzed the data: RJ. Contributed reagents/materials/analysis tools: RJ KD RM. Wrote the paper: RJ YS RM KD. Project manager: YS. Principle investigator: RJ.
- 1. Joint United Nations Programme on HIV/AIDS (UNAIDS) (2009) Joint Action for Results. UNAIDS Outcome Framework 2009–2011. Geneva: UNAIDS.
- 2. Jewkes RK, Dunkle K, Nduna M, Shai N (2010) Intimate partner violence, relationship gender power inequity, and incidence of HIV infection in young women in South Africa: a cohort study. The Lancet 367: 41–48.
- 3. Gilbert L, El-Bassel N, Schilling RF, Wada T, Bennet B (2000) Partner Violence and Sexual HIV Risk Behaviors Among Women in Methadone Treatment. AIDS and Behav 4: 261–269.
- 4. Decker MR, Seage GR 3rd, Hemenway D, Raj A, Saggurti N, et al. (2009) Intimate partner violence functions as both a risk marker and risk factor for women's HIV infection: findings from Indian husband-wife dyads. J Acquir Immune Defic Syndr 51(5): 593–600.
- 5. van der Straten A, King R, Grinstead O, Vittinghoff E, Serufilira A, et al. (1998) Sexual coercion, physical violence, and HIV infection among women in steady relationships in Kigali, Rwanda. AIDS and Behavior 2(1): 61–73.
- 6. Dunkle KL, Jewkes RK, Brown HC, Gray GE, McIntyre JA, et al. (2004) Gender-based violence, relationship power and risk of prevalent HIV infection among women attending antenatal clinics in Soweto, South Africa. The Lancet 363: 1415–1421.
- 7. Wood K, Jewkes R (2001) ‘Dangerous’ love: reflections on violence among Xhosa township youth. In: Morrell R, editor. Changing men in Southern Africa. Pietermaritzburg: University of Natal Press. London: Zed Press. pp. 317–336.
- 8. Jewkes R, Morrell R (2010) Gender and sexuality: emerging perspectives from the heterosexual epidemic in South Africa and implications for HIV risk and prevention. Journal of the International AIDS Society 13: 6.
- 9. Bourgois P (1996) In search of masculinity - Violence, respect and sexuality among Puerto Rican crack dealers in East Harlem. British Journal of Criminology 36(3): 412–427.
- 10. Decker MR, Seage GR 3rd, Hemenway D, Gupta J, Raj A, et al. (2009) Intimate partner violence perpetration, standard and gendered STI/HIV risk behaviour, and STI/HIV diagnosis among a clinic-based sample of men. Sex Transm Infect 85(7): 555–60.
- 11. Dunkle K, Jewkes R, Nduna M, Levin J, Jama N, et al. (2006) Perpetration of partner violence and HIV risk behaviour among young men in the rural Eastern Cape. AIDS 20: 2107–14.
- 12. Jewkes R, Dunkle K, Koss MP, Levin J, Nduna M, et al. (2006) Rape perpetration by young, rural South African men: prevalence, patterns and risk factors. Social Science and Medicine 63: 2949–61.
- 13. Abrahams N, Jewkes R, Hoffman M, Laubscher R (2004) Sexual violence against intimate partners in Cape Town: prevalence and risk factors reported by men. Bulletin of the World Health Organisation 82: 330–337.
- 14. El-Bassel N, Fontdevila J, Gilbert L, Voisin D, Richman BL, et al. (2001) HIV risks of men in methadone maintenance programs who abuse their intimate partners: a forgotten issue. J Subst Abuse 13: 29–43.
- 15. Silverman JG, Decker MR, Kapur NA, Gupta J, Raj A (2007) Violence against wives, sexual risk and sexually transmitted infection among Bangladeshi men. Sex Transm Infect 83: 211–15.
- 16. Jewkes R, Dunkle K, Nduna M, Levin J, Jama N, et al. (2006) Factors associated with HIV sero-positivity in young, rural South African men. International Journal of Epidemiology 35: 1455–1460.
- 17. Kim JC, Martin LJ, Denny L (2003) Rape and HIV post-exposure prophylaxis: addressing the dual epidemics in South Africa. Reprod Health Matters 11(22): 101–12.
- 18. Pulerwitz J, Gortmaker SL, DeJong W (2000) Measuring Sexual Relationship Power in HIV/STD Research. Sex Roles 42(7/8): 637–660.
- 19. World Health Organization (2000) WHO Multi-Country Study on Women's Health and Domestic Violence: Core Questionnaire and WHO Instrument - Version 9. Geneva: World Health Organization.
- 20. Sikweyiya Y, Jewkes R, Morrell R (2007) Talking about Rape: Men's responses to questions about rape in a research environment in South Africa. Agenda 74: 48–57.
- 21. Jewkes R, Nduna M, Levin J, Jama N, Dunkle K, et al. (2008) Impact of Stepping Stones on HIV, HSV-2 and sexual behaviour in rural South Africa: cluster randomised controlled trial. British Medical Journal 337: a506.
- 22. Maman S, Mbwambo JK, Hogan NM, Kilonzo GP, Campbell JC, et al. (2002) HIV-positive women report more lifetime partner violence: findings from a voluntary counselling and testing clinic in Dar es Salaam, Tanzania. American Journal of Public Health 92(8): 1331–1337.
- 23. Shisana O, Rehle T, Simbayi LC, Zuma K, Jooste S, et al. (2009) National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey, 2008: A turning tide among teenagers? Cape Town: Human Sciences Research Council.
- 24. Department of Health (2004) South Africa Demographic & Health Survey 2003. Preliminary report. Department of Health, Pretoria/ Measure DHS, Maryland, USA.
- 25. Korn EL, Graubard BI (1995) Examples of differing weighted and unweighted estimates from a sample survey. The American Statistician 49: 291–295.