*Southern Africa Labour and Development Research Unit, University of Cape Town, Cape Town, South Africa
†Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA
Correspondence to: Atheendar S. Venkataramani, MD, PhD, Department of Medicine, Massachusetts General Hospital, Charlestown HealthCare Center, 73 High Street, Charlestown, Boston, MA 02129 (e-mail: email@example.com).
The authors have no funding or conflicts of interest to disclose.
To the Editors:
Randomized clinical trials (RCTs) have shown that medical male circumcision (MC) substantially reduces the risk of female-to-male HIV transmission.1 Driven by this evidence base, a growing number of countries have adopted large-scale MC campaigns in an effort to prevent heterosexually acquired HIV in men. In South Africa, for example, >130,000 medical male circumcisions were performed during 2010.2
Encouraging adoption of MC will require educating millions of men and women about its protective benefits. The success of these policies will depend critically on what people infer from prevention information and the actions they take as a result. On the one hand, failure to internalize the HIV risk-reduction benefits of MC could discourage adoption. On the other hand, those who hear that male circumcision reduces the risk of female-to-male HIV transmission may also assume reductions in the male-to-female transmission risk, which runs counter to recent RCT evidence.3 Incorrect beliefs about MC and female HIV risk may help explain recent findings from Kenya, where over a third of women reported being less worried about HIV and that condom use was less necessary now that MC was available.4 They may also explain findings from South Africa, where women who had heard that MC reduces HIV risk among men reported lower self-perceived HIV risk, less condom use, and greater condom-free sex with partners of unknown serostatus.5
Whether individuals infer that prevention efforts such as MC provide HIV risk-reduction benefits for persons other than those immediately targeted is something that is relatively unexplored in the literature. We addressed this gap by examining whether Zulu individuals in South Africa believed that male circumcision reduced the risk of male-to-female HIV transmission. The Zulu population is relevant given the low rates of male circumcision6 and one of highest rates of HIV in South Africa,7 making Zulu men a key target in the nationwide MC effort.
Our data came from an opinion and beliefs poll conducted by the ikapadata research company (www.ikapadata.com) at the end of 2011. Specifically, we used information on 411 Zulu men and women aged 16–49 years (identified on the basis of reporting Zulu as their first language) residing in townships around Durban (280 individuals) and Johannesburg (131 individuals). These individuals were identified using quota-based sampling on age, gender, and socioeconomic class. This strategy was used given the difficulty of establishing a representative township sampling frame (the last population census was conducted in 2001 and township populations are known to change rapidly).
A short face-to-face survey was administered that consisted of 6 demographic questions, 15 questions about household characteristics and assets, and 5 questions about male circumcision. Regarding the latter, participants were first asked whether they had heard that MC reduces the risk of a man getting HIV. Those who answered yes were asked to report the year that they first heard this information, the source, and the degree to which they perceived MC protects a man from HIV infection (on a scale from 0 to 10, with 10 denoting complete protection). All participants were asked: “If a woman has unprotected sex with an HIV positive man, do you think her risk of getting HIV is greater, the same or less if he is circumcised compared to if he is uncircumcised?”
We computed descriptive statistics by gender for each of these questions. We also analyzed differences in responses to the female HIV-risk question by completion of high school; respondent age (split at the median); socioeconomic status (split at the median of an asset index created by summing yes/no answers to household ownership of a variety of durable goods, such as stoves, refrigerators, radios, and vehicles); having heard that MC protects men; and whether respondents believed that MC protects men by 60% or more.
In terms of sample characteristics, 45% of the respondents were male with the average age being 28 years. Fifty-nine percent of respondents reported completing high school. Most men (89%) and women (83%) had heard that MC reduces a man's risk of getting HIV. For the most part, individuals first heard this information in 2005 or later, but a significant minority (33% of men; 30% of women) reported hearing this information before 2005, before the publicizing of RCT findings linking male circumcision to reduced HIV risk. The primary sources of information fell within 4 categories: the media (TV, radio, newspapers), word of mouth (from family, friends, neighbors, etc), school, and health care professionals. The majority first heard this information from either the media (men: 52%; women: 53%) or via word of mouth (men: 34%; women: 24%). Almost half of the participants (men: 47%; women: 46%) believed that MC protects circumcised men by 60% or more.
Our core findings are presented in Table 1. Approximately 20% of respondents believed that a woman who has unprotected sex with an HIV-positive man has a lower risk of getting HIV if the man is circumcised compared to if the man has not been circumcised. There were no substantive or statistically significant differences in beliefs across gender, completion of schooling, age, and socioeconomic class groups. Differences in knowledge and beliefs about male circumcision and male HIV risk, on the other hand, seemed to matter when it came to beliefs about male circumcision and male-to-female HIV risk. Significantly fewer of those who had not heard that MC affords protection reported that MC directly reduces female HIV risk, whereas significantly greater numbers who believed MC offers men 60% or greater protection reported that MC directly reduces female HIV risk (28% of men; 33% of women). There were no substantive or statistically significant differences in beliefs about female HIV risk in relation to the timing of when participants first heard about MC and male HIV risk or information source (available on request).
Collectively, although our results indicate that the vast majority of Zulus were informed about the protective benefits of male circumcision for men, a significant number of men and women also believed that male circumcision directly reduces the risk of HIV infection for women. Furthermore, these beliefs were more common among individuals who believed that male circumcision offers men a relatively high degree of protection against HIV infection. These results could have important implications for prevention programs. In particular, misattribution of the benefits of preventive efforts could reduce the perceived need for safe sex behaviors, thus counteracting positive program effects. This is a more nuanced form of risk compensation.8 The aforementioned evidence from Kenya and South Africa, where women who heard about the protective benefits of circumcision perceived reduced personal HIV risk and engaged in riskier sexual behavior, underscores the fact that this is more than a theoretical possibility.
A limitation of this study is our relatively small and nonrepresentative sample. Comparing the present sample to 16- to 49-year-old Zulu men residing in the KwaZulu-Natal or Guateng provinces (where Durban and Johannesburg are situated, respectively) in the 2010 National Income Dynamics Study (www.nids.uct.ac.za), respondents in our data were nearly twice as likely to have completed high school (59% vs 31%). This, along with the fact that >80% of respondents had heard of the protective benefits of MC for men suggest that levels of misinformation about male circumcision and female HIV risk may even be underestimated in our study as participants were, on average, significantly more educated than the general population. More importantly, we lacked the data to ascertain the degree to which male circumcision is perceived to reduce HIV risk for women and to assess how these beliefs affect sexual behavior. Finally, there was no information about the MC status of either the male participants or the partners of female participants in the study. Future studies would do well to collect this information, as MC status may influence attitudes toward MC and HIV risk perceptions.
Regardless, the results point toward the need for prevention messages and HIV education that make explicit who would benefit from a given intervention and who would not. Ultimately, we recommend further research into what individuals learn from prevention messages and the subsequent behavioral response. This would not only help inform burgeoning preventive male circumcision efforts and potentially improve their efficacy but could also provide insights into how prevention programs more generally can be better designed to address the potential of risky behaviors arising from mistaken inferences.
The authors would like to thank ikapadata for access to their opinion and belief poll data and Jan Schenk and Amrik Cooper for their patience and assistance in clarifying aspects of respondent selection and data capture. The authors also thank 3 anonymous referees for helpful comments and suggestions on previous versions of their article. The authors state that all errors are their own. Brendan Maughan-Brown is grateful for funding from the NRF Research Chair in Poverty and Inequality Research and the University Research Committee at the University of Cape Town for his Postdoctoral Research Fellowship.
1. Siegfried N, Muller M, Deeks JJ, et al.. Male circumcision for prevention of heterosexual acquisition of HIV in men. Cochrane Database Syst Rev. 2009;2.
2. WHO/UNAIDS. Progress in Scale-up of Male Circumcision for HIV Prevention in Eastern and Southern Africa: Focus on Service Delivery. Geneva, Switzerland: World Health Organization; 2011.
3. Wawer MJ, Makumbi F, Kigozi G, et al.. Circumcision in HIV-infected men and its effect on HIV transmission to female partners in Rakai, Uganda: a randomised controlled trial. Lancet. 2009;374:229–237.
4. Okeyo T, Westercamp N, Bailey RC, et al.. What women think about male circumcision: perceptions of the female partners of recently circumcised men in Nyanza Province, Kenya [TUAC0401]. Paper presented at: XIX International AIDS Conference; July 24, 2012; Washington DC.
5. Maughan-Brown B, Venkataramani A. Learning that circumcision is protective against HIV: risk compensation among men and women in Cape Town, South Africa. PLoS One. 2012;7:1–9.
6. USAID. The Potential Cost and Impact of Expanding Male Circumcision in South Africa. Washington, DC: USAID; 2009.
7. Shisana O, Rehle T, Simbayi L, et al.. South African National Prevalence, Incidence, Behaviour and Communication Survey, 2008: A Turning Tide Among Teenagers? Cape Town, South Africa: Human Sciences Research Council Press; 2009.
8. Eaton LA, Kalichman S. Risk compensation in HIV prevention: implications for vaccines, microbicides, and other biomedical HIV prevention technologies. Curr HIV/AIDS Rep. 2007;4:165–172.