Recent increases in HIV incidence among men who have sex with men (MSM) have been observed in most resource-rich countries [1–4]. Novel prevention strategies are needed to address these increases . In contrast to findings from the randomized trials of circumcision among African heterosexuals [6–8], published data on the association of circumcision and HIV infection in MSM are few and conflicting [9–13].
There is evidence from Australia and other developed countries that when HIV-negative MSM practise unprotected anal intercourse (UAI) with an HIV-positive or HIV status unknown partner they increasingly do it in the insertive position, as a conscious HIV risk reduction strategy [14–17]. Should circumcision afford the insertive partner in UAI a similar protection as it does to African heterosexual men, the relative impact of circumcision as an HIV prevention intervention could be substantial.
Study participants were initially HIV-negative homosexual men who were recruited from a range of community-based sources between June 2001 and December 2004, and followed until June 2007. Details of recruitment sources and methods of the Health in Men (HIM) study have been described elsewhere . Ethics approval was granted by the University of New South Wales.
At 6-monthly interviews, detailed behavioural data were collected on UAI behaviour as previously described . Each year participants reported their preference for anal intercourse position (preference for insertive role, receptive role or no preference). Circumcision status was reported at baseline and self-reported circumcision status was validated by clinical examination in a subgroup of 240 consecutively presenting participants . Participants were tested annually for HIV with western blot confirmation. Identifiers were matched against the Australian national HIV register each year to identify HIV infections which occurred in those who tested outside the study or had been lost to active follow-up. The date of HIV infection was estimated in 36 participants as the midpoint between their last HIV-negative and first HIV-positive test. Western blot and seroconversion symptom data were available to assist in further defining the date of HIV infection in the remaining 17 seroconverters, as previously described . Annually, participants were offered testing for other sexually transmissible infections (STIs) as described in detail elsewhere  and they self-reported STIs diagnosed elsewhere in the previous year.
Statistical analyses were performed using STATA 10.0 (STATA Corporation, College Station, Texas, USA). For HIV incidence, total person-years were calculated as the time from study entry to the estimated date of seroconversion, or to the end of the study in June 2007 for those who remained HIV-negative. The exact binomial method was used to calculate 95% confidence intervals (CIs). Univariate and multivariate Cox regression was used to identify the association of circumcision status with incident HIV infection. Analyses were adjusted a priori for age, due to the strong association of age with circumcision status among HIM participants . The multivariate model was adjusted for all potentially serodiscordant UAI, defined as insertive or receptive UAI (with or without ejaculation) with HIV-positive or unknown status partners. STIs independently associated with HIV seroconversion in the HIM cohort  were considered in the multivariate model. Hazard ratios and their corresponding 95% CIs were calculated. Subgroup analyses were performed for men who reported insertive but not receptive UAI and men who reported a preference for the insertive role in anal intercourse. The attributable risk percentage associated with being uncircumcised was calculated for all participants and the subset of participants who reported a preference for the insertive role in anal intercourse .
From June 2001 to December 2004 a total of 1426 participants were enrolled, of whom 66% were circumcised. The median age at enrolment was 35 years (range 18–75 years), 68% were Australian-born and 74% were of Anglo ethnicity. Ninety-five percent of participants self-identified as gay or homosexual.
By the end of the HIM study in June 2007, the total follow-up time was 5160.7 person-years, and median follow-up time for participants was 3.9 years.
Fifty-three HIV seroconversions were identified in HIM participants during the study period, an incidence of 0.78 per 100 person-years (95% CI 0.59–1.02). Of incident HIV infections, 31 were identified on HIM study visit testing and 22 were identified by matching with the national HIV register. Excluding one participant with a surgically reconstructed prepuce who seroconverted to HIV, univariate analysis of the remaining 52 participants revealed no association of being circumcised with HIV seroconversion (hazard ratio 0.71, 95% CI 0.41–1.24, P = 0.234). Six participants who had no behavioural data available in the 12 months prior to seroconversion were excluded from multivariate analysis. After adjustment for behavioural risk factors the circumcision hazard ratio was little changed (Table 1). Controlling for STIs (anal gonorrhoea and anal warts) significantly associated with HIV seroconversion in the HIM study  had little effect on the multivariate association of circumcision and HIV seroconversion (data not shown).
Participants who reported insertive unprotected anal intercourse only
Relatively few follow-up periods (10.1% of total person-years at risk) were in men who reported insertive but no receptive UAI. There was a nonsignificant reduction in HIV risk among circumcised men who did not report any receptive UAI (n = 4, hazard ratio 0.40, 95% CI 0.05–2.97, P = 0.330).
Preference for insertive role in anal intercourse
In men who preferred the insertive position a median 100% of UAI acts [interquartile range (IQR) 92.1–100%] with potentially serodiscordant partners were in the insertive position. In total, 33.1% of total person-years at risk were in men with a preference for the insertive role in anal intercourse, and there were seven new HIV infections in 1710.1 person-years of follow-up (incidence 0.41 per 100 person-years, 95% CI 0.20–0.86). This compared with 12 new HIV infections in 885.5 person-years of follow-up among men who preferred the receptive role in anal intercourse [incidence 1.36 per 100 person-years, 95% CI 0.77–2.39]. Among men who preferred the insertive role, who included the four participants that did not report any receptive UAI, being circumcised was associated with a significant reduction in HIV incidence (hazard ratio 0.19, P = 0.049; Table 1). The association was strengthened when adjusted for age and potentially serodiscordant UAI (hazard ratio 0.11, P = 0.041; Table 1).
In 1211 HIM participants who reported their anal intercourse position preference at least twice at consecutive annual interviews, only 10 of 404 (2.5%) participants who initially preferred the insertive role reported a change in this preference at one or more subsequent interviews [change to receptive (n = 2) or no (n = 8) preference]. No participants reported a change in preference to the insertive role from an initial report of no preference or receptive role preference.
Among all HIM participants, the estimated proportion of HIV infections that could be attributed to being uncircumcised was 8.7%. Among participants who preferred the insertive role in anal intercourse, the estimated proportion of HIV infections that could be attributed to being uncircumcised was 75.7%.
Overall, HIV incidence was not significantly associated with circumcision in the HIM cohort of homosexual men. However, being circumcised was associated with a significant reduction in HIV incidence among the one-third of participants who reported a preference for the insertive role in anal intercourse and whose sexual behaviour closely and consistently reflected this preference.
The HIM study is the first prospective investigation of circumcision and HIV in homosexual men that has reported data in those men likely to have been infected by insertive anal intercourse. Another published longitudinal study assessed the association of circumcision and HIV incidence in MSM  and described an independent association of being uncircumcised with HIV seroconversion in an entire cohort in which almost 90% of participants were circumcised [adjusted odds ratio (AOR) 2.0, 95% CI 1.1–3.7]. A similar result was found in a Seattle clinic MSM population (AOR 2.0, 95% CI 1.0–4.0) . In contrast, two recently published US cross-sectional analyses [11,12], one of which was community-based , found no association of HIV infection with circumcision status. These studies were among those included in a recent meta-analysis of mostly unpublished data which reported no effect of circumcision on the odds of HIV infection in an analysis comprising over 50 000 MSM participants [odds ratio (OR) 0.95, 95% CI 0.81–1.11] .
Only one published study, of cross-sectional design, has assessed circumcision and HIV infection among subgroups of MSM wholly practising insertive UAI  and it did not find an association of circumcision status with prevalent HIV infection. Similarly the meta-analytic association of being circumcised with HIV infection among MSM who primarily engaged in insertive anal intercourse did not reach statistical significance (OR 0.71, 95% CI 0.22–2.28) . However, the prospective data from the HIM study suggest that circumcision may have a protective effect among men who predominantly practise the insertive role in anal intercourse. Only 10% of HIM participants were 100% exclusive in their practise of insertive anal intercourse, but over 30% of participants consistently reported a preference for the insertive role in anal intercourse over 1 or more years of follow-up. Consistency of practise would be an important consideration should circumcision be considered as an HIV intervention among MSM. It remains to be seen whether the consistency in insertive anal intercourse preference observed in HIM is maintained over a longer period.
The key limitation of our analysis was lack of study power due to the relatively small number of HIV infections identified in the HIM cohort and the low incidence of HIV infection among predominantly insertive men. Nonetheless, the reduction in HIV incidence among this group was significant, and this group comprised around one-third of all gay men enrolled. Strengths of our analysis include consideration of potential demographic and STI confounders [19,22], validation that self-reported circumcision status was accurate  and 6-monthly prospective collection of detailed UAI behaviour which minimized recall and prevalence-incidence bias and enabled precise adjustment for behavioural risk variables in the multivariate model.
As the minority of HIV infections in HIM occurred in those reporting no receptive UAI, and most Australian men are circumcised , circumcision is unlikely to have a major impact on HIV incidence in homosexual men in Australia. Nonetheless, ‘strategic positioning’ when HIV-negative gay men adopt the insertive role in UAI to reduce their HIV risk is occurring commonly among Sydney gay men [16,17]. This coupled with a rapidly declining prevalence of circumcision in Australian  and US  homosexual men means circumcision could play a more important role in reducing gay men's susceptibility to HIV infection in the future. Randomized trials are warranted before recommendations can be made regarding circumcision as an HIV prevention intervention among MSM populations, but the design of such studies is challenging. Study populations would require high HIV incidence, low baseline circumcision prevalence and large numbers of participants exclusively or predominantly practising the insertive role. Such attributes are necessary for sufficient study power to detect an association of circumcision status with the relatively infrequent outcome measure of HIV acquisition via insertive anal intercourse.
We thank all the participants, the dedicated HIM Study team and the participating doctors and clinics. We also thank Dr Kathy Petoumenos, and Associate Professor Matthew Law, PhD, of the National Centre in HIV Epidemiology and Clinical Research for statistical advice. Funding: During the study D.J.T. was supported by National Health and Medical Research Council Public Health Scholarship no. 351044 and, subsequently, by the Royal Australasian College of Physicians GSK Scholarship for virological research. The National Centre in HIV Epidemiology and Clinical Research and the National Centre in HIV Social Research are funded by the Australian Government Department of Health and Ageing. The HIM Cohort study was funded by the National Institutes of Health, a component of the US Department of Health and Human Services (NIH/NIAID/DAIDS: HVDDT Award N01-AI-05395), the Australian Government Department of Health and Ageing (Canberra) and the New South Wales Health Department (Sydney), and the National Health and Medical Research Council (project grant # 400944). The funding organizations had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; or in preparation, review, or approval of the manuscript. All authors had full access to all the data in the study. D.J.T. is the principal author and takes responsibility for the integrity of the data and the accuracy of data analysis.
1. Grulich AE, Kaldor JM. Trends in HIV incidence in homosexual men in developed countries. Sex Health 2008; 5:113–118.
2. Fisher M, Pao D, Murphy G, Dean G, McElborough D, Homer G, et al
. Serological testing algorithm shows rising HIV incidence in a UK cohort of men who have sex with men: 10 years application. AIDS 2007; 21:2309–2314.
3. Katz MH, Schwarcz SK, Kellogg TA, Klausner JD, Dilley JW, Gibson S, et al
. Impact of highly active antiretroviral treatment on HIV seroincidence among men who have sex with men: San Francisco. Am J Public Health 2002; 92:388–394.
4. Marcus U, Voss L, Kollan C, Hamouda O. HIV incidence increasing in MSM in Germany: factors influencing infection dynamics. Euro Surveill 2006; 11:157–160.
5. Jaffe HW, Valdiserri RO, De Cock KM. The reemerging HIV/AIDS epidemic in men who have sex with men. JAMA 2007; 298:2412–2414.
6. Bailey RC, Moses S, Parker CB, Agot K, Maclean I, Krieger JN, et al
. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007; 369:643–656.
7. Gray RH, Kigozi G, Serwadda D, Makumbi F, Watya S, Nalugoda F, et al
. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007; 369:657–666.
8. Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R, Puren A. Randomised controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 trial. PLoS Med 2005; 2:e298.
9. Buchbinder SP, Vittinghoff E, Heagerty PJ, Celum CL, Seage GR 3rd, Judson FN, et al
. Sexual risk, nitrite inhalant use, and lack of circumcision associated with HIV seroconversion in men who have sex with men in the United States. J Acquir Immune Defic Syndr 2005; 39:82–89.
10. Kreiss JK, Hopkins SG. The association between circumcision status and human immunodeficiency virus infection among homosexual men. J Infect Dis 1993; 168:1404–1408.
11. Mor Z, Kent CK, Kohn RP, Klausner JD. Declining rates in male circumcision amidst increasing evidence of its public health benefit. PLoS One 2007; 2:e861.
12. Millett GA, Ding H, Lauby J, Flores S, Stueve A, Bingham T, et al
. Circumcision status and HIV infection among Black and Latino men who have sex with men in 3 US cities. J Acquir Immune Defic Syndr 2007; 46:643–650.
13. Grulich AE, Hendry O, Clark E, Kippax S, Kaldor JM. Circumcision and male-to-male sexual transmission of HIV. AIDS 2001; 15:1188–1189.
14. Jin F, Prestage GP, Ellard J, Kippax SC, Kaldor JM, Grulich AE. How homosexual men believe they became infected with HIV: the role of risk-reduction behaviors. J Acquir Immune Defic Syndr 2007; 46:245–247.
15. Grov C, DeBusk JA, Bimbi DS, Golub SA, Nanin JE, Parsons JT. Barebacking, the Internet, and harm reduction: an intercept survey with gay and bisexual men in Los Angeles and New York City. AIDS Behav 2007; 11:527–536.
16. Crawford JM, Kippax SC, Mao L, Van de Ven PG, Prestage GP, Grulich AE, et al
. Number of risk acts by relationship status and partner serostatus: findings from the HIM cohort of homosexually active men in Sydney, Australia. AIDS Behav 2006; 10:325–331.
17. Van de Ven P, Kippax S, Crawford J, Rawstorne P, Prestage G, Grulich A, Murphy D. In a minority of gay men, sexual risk practice indicates strategic positioning for perceived risk reduction rather than unbridled sex. AIDS Care 2002; 14:471–480.
18. Jin F, Prestage GP, Mao L, Kippax SC, Pell CM, Donovan B, et al
. Transmission of herpes simplex virus types 1 and 2 in a prospective cohort of HIV-negative gay men: the health in men study. J Infect Dis 2006; 194:561–570.
19. Jin F, Prestage GP, Imrie J, Kippax SC, Pell CM, Donovan B, et al. Anal sexually transmissible infections and risk of HIV in homosexual men. J Acquir Immune Defic Syndr
20. Templeton DJ, Mao L, Prestage GP, Jin F, Kaldor JM, Grulich AE. Self-report is a valid measure of circumcision status in homosexual men. Sex Transm Infect 2008; 84:187–188.
21. Jin F, Prestage GP, McDonald A, Ramacciotti T, Imrie JC, Kippax SC, et al
. Trend in HIV incidence in a cohort of homosexual men in Sydney: data from the Health in Men Study. Sex Health 2008; 5:109–112.
22. Templeton DJ, Mao L, Prestage G, Kaldor JM, Kippax S, Grulich AE. Demographic predictors of circumcision status in a community-based sample of homosexual men in Sydney, Australia. Sex Health 2006; 3:191–193.
23. Riffenburgh RH. Statistics in medicine. 2nd ed. Burlington, MA: Elsevier; 2006.
24. Millett GA, Flores SA, Marks G, Reed JB, Herbst JH. Circumcision status and risk of HIV and sexually transmitted infections among men who have sex with men: a meta-analysis. JAMA 2008; 300:1674–1684 [Erratum appears in: JAMA
25. Richters J, Smith AM, de Visser RO, Grulich AE, Rissel CE. Circumcision in Australia: prevalence and effects on sexual health. Int J STD AIDS 2006; 17:547–554.