University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > The Protective Effect of Circumcision
The protective effect of circumcision on HIV incidence in rural low-risk men circumcised predominantly by traditional circumcisers in Kenya: two-year follow-up of the Kericho HIV cohort study
Global Health Sciences Literature Digest
Published July 13, 2007
Journal Article

Shaffer DN, Bautista CT, Sateren WB, Sawe FK, Kiplangat SC, Miruka AO, et al. The protective effect of circumcision on HIV incidence in rural low-risk men circumcised predominantly by traditional circumcisers in Kenya: two-year follow-up of the Kericho HIV cohort study. J Acquir Immune Defic Syndr 2007 Aug 1;45(4):371-379.


To examine the association between circumcision and HIV infection in a cohort of adult agricultural workers and dependents after two years of follow-up, specifically considering socio-demographic and culturally relevant characteristics that may influence the relationship between circumcision and incident HIV infection in males.

Study Design

This was a 2-year follow-up sub-analysis within the "HIV and Malaria Cohort Study Among Plantation Workers and Adult Dependents in Kericho, Kenya," which is a 3.5-year prospective observational cohort study aimed at estimating HIV prevalence, incidence, co-morbidities, molecular epidemiology, and vaccine feasibility and acceptability.


A large tea plantation on the outskirts of Kericho, Kenya in the southern Rift Valley Province, which has a relatively low prevalence of HIV (5.3%) compared to urban areas in Kenya (10.0%).


Adult plantation workers and dependent volunteers aged 18 to 55 years (n=2,801) were recruited for study participation. This sub-study excluded all women (n=1,081) and HIV-infected men (n=195). Another 147 were lost to follow-up or had incomplete data, resulting in 1,378 men for analysis.


There was no intervention in this observational cohort study.

Primary Outcomes

The primary outcomes for this study were HIV incidence at two years of follow-up, and associated socio-demographic and behavioral factors.


Baseline characteristics: Of the 1,378 men, 80.4% reported being circumcised, and 74.1% were from the Luo tribe. Circumcised men compared with uncircumcised men were more likely to report having a high school education or greater (50.8% v 40.3%, p=0.002), to have never or only once been married (23.2% v 60.1% v 21.5% and 55.2%, p=0.037), and to have a smaller age difference between them and their spouse (p=0.046). Uncircumcised males also had more years of sexual activity compared to circumcised males (p=0.037).

Two-year HIV incidence: Thirty incident HIV cases occurred among 2,689 person-years of follow-up, for an overall incidence rate of 1.12 (95% CI: 0.75-1.59) per 100 person-years. The incidence rates differed between circumcised and uncircumcised males, 0.79 (0.46-1.25) and 2.48 (1.33-4.21), respectively. These rates corresponded to a statistically significant hazard ratio of 0.31 (0.15-0.64). However, after adjusting for socio-demographic characteristics, the HR became non-significant at 0.55 (0.20-1.49). When not controlling for tribe, the HR remained significant at 0.34 (0.16-0.73). Also, after controlling for both socio-demographic (without tribe) and behavioral/HIV risk characteristics, circumcision remained associated with a protective effect (HR=0.32, 0.15-0.68).

Luo vs. non-Luo men: The Luo males were more likely than non-Luo males to be older than 35 years (36.4% v 29.6%, p=0.043), practice traditional African religions (10.0% v 5.8%, p=0.021), have a >10 years age difference with their spouse (26.8% v 20.6%, p=0.035), and report having sex with a commercial sex worker (17.5% v 11.7%, p=0.022). HIV incidence was significantly higher among Luo men (HR=3.14, 1.73-5.21) compared to non-Luo men (HR=0.71, 0.41-1.16). Regardless of circumcision status, Luo men were 4.55 (2.21-9.35) times more likely to become HIV-infected during the two years of follow-up. Circumcised Luo men were 5.22 (1.19-22.99, p=0.029) times more likely to become HIV-infected compared to circumcised non-Luo men.

Circumcised men: Most (73.9%) of circumcised males had the procedure performed by a traditional circumciser, and 62.1% had the procedure done between the ages of 12 and 19 (mean age 12.7 years). Time since circumcision was not significantly different among those HIV-infected compared to those who were not infected.


The authors conclude that circumcision offers protection from HIV infection in low-risk adult men living in rural Kenya, where circumcision is common and predominantly performed by traditional circumcisers.

Quality Rating

According to the relevant aspects of the Newcastle-Ottawa criteria for evaluating the quality of longitudinal observational studies, this study was of high quality. However, the authors note that the study was limited by relying on self-report of circumcision and the lack of power to detect small differences within circumcised versus uncircumcised strata when considering potentially confounding variables.

In Context

The results of this study are consistent with the results of the three randomized trials in sub-Saharan Africa showing a decreased incidence of HIV among circumcised men(1,2,3) as well as several prospective studies in East Africa.(4,5,6) In contrast to these studies, however, men in this study were largely circumcised by traditional healers.

Programmatic Implications

This study is the only prospective cohort study that separated men traditionally circumcised from those circumcised in a medical clinic, and provided two years of follow-up in a rural, lower-risk population in Kenya. The observed benefit of circumcision performed by traditional practitioners is encouraging, although the results should be confirmed by a randomized controlled trial. Furthermore, proper training and safety evaluations for all practitioners, as well as cultural acceptability of circumcision methods, will be important for public-health-based circumcision initiatives. Additionally, the loss of significance for the protective effect of circumcision when controlling for tribe is a concern, and warrants further investigation. The authors speculate that this effect reflects the small number of incident cases and the limited power of the study to detect true differences in incident HIV within tribes after stratification by circumcision status. Alternatively, other non-measured factors associated with tribe, such as wife inheritance, may also explain this effect. This study highlights the importance of tribal and cultural practices and their potential modification of HIV risk. These factors should be taken into account for the development of successful comprehensive HIV prevention programs in Kenya.


  1. Bailey R, Moses S, Agot K. A randomized controlled trial of male circumcision to reduce HIV incidence in Kisumu, Kenya: progress to date. August 2006; XVI International AIDS Conference; Toronto.
  2. Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R, Puren A. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med 2005 Nov;2(11):e298.
  3. 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 Feb 24;369(9562):657-66.
  4. Lavreys L, Rakwar JP, Thompson ML, Jackson DJ, Mandaliya K, Chohan BH, et al. Effect of circumcision on incidence of human immunodeficiency virus type 1 and other sexually transmitted diseases: a prospective cohort study of trucking company employees in Kenya. J Infect Dis 1999 Aug;180(2):330-6.
  5. Baeten JM, Richardson BA, Lavreys L, Rakwar JP, Mandaliya K, Bwayo JJ, et al. Female-to-male infectivity of HIV-1 among circumcised and uncircumcised Kenyan men. J Infect Dis 2005 Feb 15;191(4):546-53.
  6. Gray RH, Kiwanuka N, Quinn TC, Sewankambo NK, Serwadda D, Mangen FW, et al. Male circumcision and HIV acquisition and transmission: cohort studies in Rakai, Uganda. Rakai Project Team. AIDS 2000 Oct 20;14(15):2371-81.