Editorial and reviews 
Volume 354, Number 9192 20 November 1999
Previous Next
Male circumcision and HIV infection: 10 years and counting

Daniel T Halperin, Robert C Bailey

Lancet 1999; 354: 1813-15

Department of Community Health Systems, and Medical Arthropology Program, University of California, San Francisco, CA, USA (D T Halperin PhD); and Division of Epidemiology and Biostatistics, School of Public Health, and Department of Anthropology, University of Illinois at Chicago, Chicago, IL 60612, USA (R C Bailey PhD)

Correspondence to: Dr Robert C Bailey (e-mail: rcbailey@uic.edu)

Epidemiological findings
Biological mechanisms
Public-health response
Response on the ground
Time for action

A decade has passed since publication of Cameron and colleagues' prospective study1 that showed a greater than eight-fold increased risk of HIV-1 infection for uncircumcised men. Today, many observers of the AIDS pandemic are puzzled by the glaring discrepancies in HIV seroprevalence between different countries and regions, despite the presence of what seem to be similar risk factors. For example, the November, 1998 UNAIDS/ WHO Report on the AIDS Epidemic concludes, "It is not fully understood why HIV infection rates take off in some countries while remaining stable in neighbouring countries over many years."

We argue that since Cameron and colleagues' landmark study, the epidemiological and biological evidence that links lack of circumcision with HIV transmission has become compelling and that lack of male circumcision is one of the main causes of many regional discrepancies in rates of HIV infection. Furthermore, as increasing numbers of men in some traditionally non-circumcising communities seek safe affordable circumcisions to avoid AIDS and other sexually transmitted diseases (STDs), it is time for the international health community to add male-circumcision services to the current limited armamentarium of AIDS prevention measures in countries with a high prevalence of heterosexually transmitted HIV and STDs.  

Epidemiological findings

In their study of the risk factors for HIV infection among 422 men who visited commercial sex workers in Nairobi, Kenya, Cameron and colleagues found that men who were not circumcised had a 8·2-fold increased risk of seroconversion, compared with circumcised men.1 Subsequently, six more prospective studies from four countries investigated the relative risk of heterosexual HIV-1 infection in uncircumcised men. Four studies reported significant relative risks that ranged from 2·3 to 4·5 after multivariate analyses, and in the other two prospective studies multivariate risk ratios were 3·0 or greater, but were not significant.2 Of 38 cross-sectional studies, 27 from eight countries found a significant association between lack of male circumcision and HIV infection, five found a trend towards an association, five found no association, and one reported an increased risk of infection in men who had been circumcised.2,3 In 1994, Moses and colleagues3 established that, on the basis of the information available at the time, the association between lack of male circumcision and HIV infection met all but three of Hill's criteria for making causal inferences; an additional 17 studies from eight countries have since been published. That circumcision is partially protective has been documented even in settings in which circumcised men have higher risk profiles for HIV transmission (eg, more sexual partners, alcohol use, and some STDs).4,5

More recently, while epidemiologists have been investigating the protective effect of male circumcision against HIV infection, a wider public discussion has ensued with regard to why, 20 years into the pandemic, some countries continue to retain fairly low HIV seroprevalence, whereas in other places, sometimes even neighbouring regions, rates of infection are many times higher (table). For example, rates of HIV-1 infection continue to be much lower in the Philippines (0·06% of the adult population), in Bangladesh (0·03%), and in Indonesia (0·05%) than in Thailand (2·2%), India (0·8%), and Cambodia (2·4%). Such dramatic discrepancies, sometimes on the order of ten-fold to 50-fold, have been attributed to differences in surveillance systems or the implication is made that governments, non-governmental organisations, and international agencies have mounted more effective prevention programmes in countries that have low HIV seroprevalence. Yet the pandemic has been raging for too long and surveillance and prevention efforts have been too pervasive for such a widespread, consistent pattern to be so easily explained.

<20% circumcised Seroprevalence* >80% circumcised Seroprevalence*
Sub-Saharan Africa      
Zimbabwe 25·84 Kenya 11·64(dagger)
Botswana 25·10 Congo (Brazzaville) 7·64
Namibia 19·94 Cameroon 4·89
Zambia 19·07 Nigeria 4·12
Swaziland 18·50 Gabon 4·25
Malawi 14·92 Liberia 3·65
Mozambique 14·17 Sierra Leone 3·17
Rwanda 12·75 Ghana 2·38
    Gambia 2·24
    Guinea 2·09
    Benin 2·06
South and southeast Asia    
Cambodia 2·40 Pakistan 0·09
Thailand 2·23 Philippines 0·06
Myanmar 1·79 Indonesia 0·05
India 0·82 Bangladesh 0·03
Nepal 0·24    
Countries within each geographical region have similar risk factors for heterosexual HIV infection (eg, multiple sexual partners, widespread STDs, low condom use). Countries excluded from table because: (1) estimated proportion of men circumcised >20%, <80%; (2) similar risk factors for heterosexual HIV epidemic not present; (3) insufficient information.
*June 1998 UNAIDS/WHO % estimates. (dagger)Excluding the predominately non-circumcising region around Kisumu, western Kenya, where HIV-1 seroprevalence is 34·9%, national seroprevalence would be about 8%.
HIV-1 seroprevalence in sub-Saharan African and south/southeast Asian countries by estimated proportion of men circumcised


Biological mechanisms

For over a decade researchers have suggested that the foreskin provides a vulnerable portal of entry to HIV and other pathogens.6 The highly vascularised prepuce has been discovered to contain a higher density of Langerhans cells--primary target cells for sexual transmission of HIV--than cervical, vaginal, or rectal mucosa.7 Other scientists and clinicians have noted that the foreskin is more susceptible to traumatic epithelial disruptions during intercourse, which allows additional vulnerability to HIV.1 An intact foreskin also exposes a man to greater risk of ulcerative STDs, such as chancroid, syphilis, and herpes, that are known cofactors for HIV infection.2

Although the increased risk of HIV transmission in uncircumcised men is independent of genital-ulcer diseases,1,6 a cycle of amplification occurs in which these diseases enhance transmission of HIV--HIV infection increases the frequency of genital-ulcer disease, and lack of male circumcision augments the transmission of both genital-ulcer diseases and HIV. This amplification was highlighted by Cameron and colleagues,1 who found that all but one of the 24 seroconversions occurred in men who had a genital-ulcer disease, were uncircumcised, or had both risk factors.1 Since the probability of female-to-male transmission of HIV-1 is otherwise very low, as rare as one per 9000 acts of unprotected vaginal intercourse in the absence of facilitating risk factors,8 a widespread heterosexual AIDS epidemic is unlikely. However, when a large proportion of men are uncircumcised and STDs are common, conditions are ideal for an explosive epidemic of HIV infection.9

Male circumcision, were it to be adopted by a substantial proportion of men within customarily non-circumcising societies, could have a huge impact on the HIV pandemic in developing countries. If the relative risk of HIV-1 infection for uncircumcised men is 2·5 (near the low end of the risks found in the prospective studies) in a country where 20% of men are not circumcised, which is roughly the situation in countries such as Nigeria and Indonesia, the proportion of heterosexual HIV-1 infections in men attributable to lack of circumcision is 23%.10,11 On the other hand, if 80% of men are not circumcised, as is roughly the case in Zambia and Thailand, an estimated 55% of HIV-1 infections in men are attributable to lack of circumcision. In populous regions such as South Asia where a large population of men are uncircumcised, the number of infections attributable to lack of male circumcision could soon reach into the millions.

Public-health response

In the face of such compelling evidence, we would expect the international health community to at least consider some form of action. However, the association between lack of male circumcision and HIV transmission has met with fierce resistance,12 cautious scepticism,13 or, more typically, utter silence, which is evidenced by a dearth of public-health information on the issue. For example, the Johns Hopkins Media/Materials Clearinghouse has been unable to identify among its comprehensive collection of over 30000 health communication materials a single pamphlet, poster, or flyer that mentions lack of male circumcision in relation to HIV/AIDS.

Circumcision is a surgical procedure that outside of the USA, Canada, Australia, and South Korea is restricted mainly to specific cultural and religious groups. Perhaps because circumcision is usually imbedded in a complex web of deeply held cultural values and religious beliefs, many health professionals have been hesitant to integrate it with other HIV and STD prevention strategies. Yet these same health professionals have seldom hesitated to promote use of condoms or attempt other sweeping changes in sexual behaviour--practices that may be equally charged with deeply rooted cultural and religious meanings. Although promotion of changes in sexual behaviour and condom use remain two of the three cornerstones of HIV prevention in most developing countries (STD treatment now constitutes the third), male circumcision remains largely unexplored, at least by the international public-health community.

Response on the ground

In east and southern Africa, increasing numbers of people are becoming aware of the differences in prevalence of AIDS and STDs between circumcised and uncircumcised men, and they are taking action. Male circumcision is increasingly recommended by traditional healers.14 Private clinics that specialise in male circumcision, many of which are run by people with minimum or no medical training, are sprouting up in Tanzania, western Kenya, Rwanda, and Uganda,2,9 and many advertise their services as a way to alleviate chronic STD infection and AIDS. Young men and adolescents in east and southern Africa are increasingly electing circumcision--both the medically safe procedure and more precarious non-clinical methods--in regions where traditionally they have avoided the practice. In a recent Ugandan study, 23% of non-Muslim men not belonging to any traditionally circumcising ethnic group reported that they were circimcised.5 In western Kenya, 60% of uncircumcised men stated that they would prefer to be circumcised;15 similarly, researchers in Tanzania concluded that "ethnic group or religious denomination are no longer the sole determination of male circumcision".16 As the leader of a South African traditional healers' organisation has reflected, "When tradition and the health of our people are in conflict, it is tradition we must sacrifice."14

Time for action

By avoiding this issue althogether, medical professionals and public-health authorities may inadvertently be harming the very individuals whom they are trying to help. As increasing numbers of men and boys turn to circumcision as perceived protection from AIDS, many will be exposed to harm by untrained practitioners who use unsafe methods. Yet, contrary to some popular misconceptions, safe and inexpensive male circumcision is routinely performed in developing countries in clinical settings. The procedure is normally performed on an outpatient basis with local anaesthesia, and most men return to light work activities the next day.

The hour has passed for the international health community to recognise the compelling evidence that show a significant association between lack of male circumcision and HIV infection. It is time to take the following actions: to provide communities with accurate, balanced information so that individuals can make informed choices; to provide the training and resources needed to offer safe, voluntary male circumcision in which pain is kept to a minimum; and to begin investigations of the feasibility of acceptable male-circumcision interventions in communities with high HIV and STD seroprevalence where circumcision has traditionally not been practiced. To our knowledge, these actions have so far been adopted only in western Kenya on a trial basis.15

Finally, an important caution is in order. Offering male-circumcision services as a way to prevent HIV transmission will be counterproductive if men opt for the procedure believing it will fully protect them from AIDS. As the results of Kelly and colleagues' study17 suggest, although male circumcision before age 20 has a protective effect, when the procedure is preformed on adults it may not reduce the risk of HIV transmission. One explanation for this finding, in addition to the possibility of infection before circumcision, is adoption of circumcision by those individuals who are already at increased risk of infection. Male-circumcision interventions must not be perceived by individuals or communities as a substitute for other HIV and STD prevention strategies. Rather, information about and services for male circumcision should be integrated with existing AIDS prevention and reproductive health programmes in places with a high prevalence of HIV infection.

We thank Jack Goldberg, Ronald Hershow, Moira Killoran, Jay Levy, Steven Paul, Malcolm Potts, Paul Volberding, Stanley Yoder, and two reviewers for helpful comments.


1 Cameron DW, Simonsen JN, D'Costa LJ, et al. Female to male transmission of human immunideficiency virus type 1: risk factors for seroconversion in men. Lancet 1989; 2: 403-07.

2 Moses S, Bailey RC, Ronald AR. Male circumcision: assessment of health risks and benefits. Sex Transm Infect 1998; 74: 368-73.

3 Moses S, Plummer FA, Bradley JE, Ndinya-Achola JO, Nagelkerke NJD, Ronald AR. The association between lack of male circumcision and risk for HIV infection: a review of the epidemiological data. Sex Transm Dis 1994; 21: 201-10.

4 Seed J, Allen S, Mertens T, et al. Male circumcision, sexually transmitted disease, and risk of HIV. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 8: 83-90.

5 Bailey RC, Neema S, Othieno R. Sexual behaviours and other HIV risk factors in circumcised and uncircumcised men in Uganda. J Acquir Immune Defic Syndr Hum Retrovirol (in press).

6 Simonsen JN, Cameron DW, Gakinya MN, et al. Human immunodeficiency virus infection among men with sexually transmitted diseases. N Engl J Med 1988; 319: 274-78.

7 Hussain LA, Lehner T. Comparative investigation of Langerhans' cells and potential receptors for HIV in oral, genitourinary and rectal epithelia. Immunology 1995; 85: 475-84.

8 Padian NS, Shiboski SC, Glass SO, Vittinghoff E. Heterosexual transmission of HIV in nothern California: Results from a ten-year study. Am J Epidemiol 1997; 146: 350-57.

9 Caldwell J, Caldwell P. The African AIDS epidemic. Sci Am 1996; 274: 62-68.

10 Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. New York: Van Nostrand Reinhold, 1982.

11 Moses S, Plummer F, Bradley J, et al. Male circumcision and the AIDS epidemic in Africa. Health Transition Rev 1995; 5: 100-03.

12 Fleiss PM, Hodges FM, Van Howe RS. Immunological functions of the human prepuce. Sex Transm Infect 1998; 74: 364-67.

13 de Vincenzi I, Mertens T. Male circumcision: a role in HIV prevention? AIDS 1994; 8: 153-60.

14 Green EC, Zokwe B, Dupree JD. Indigenous African healers promote male circumcision for prevention of sexually transmitted diseases. Trop Doct 1993; 23: 182-83.

15 Bailey RC, Poulussen R. Phase I and II of a trial intervention integrating male hygiene and circumcision with reproductive health in Nyanza Province, Kenya. Chicago: University of Illinois at Chicago School of Public Health, 1999.

16 Urassa M, Todd J, Boerma JT, Hayes R, Isingo R. Male circumcision and susceptibility to HIV infection among men in Tanzania. AIDS 1997; 11: 73-80.

17 Kelly R, Kiwanuka N, Wawer MJ, et al. Age of male circumcision and risk of prevalent HIV infection in rural Uganda. AIDS 1999; 13: 399-405.