RECENTLY, THERE HAS BEEN increasing interest in the practice of male circumcision in Africa because of its association with HIV infection. Several authors considered male circumcision an important explanation for the uneven spread of HIV and AIDS in sub-Saharan Africa. Initial evidence was derived from macroanalyses of the geographic distribution of HIV infection and male circumcision in Africa. 1–3 Subsequently, studies of patients with sexually transmitted diseases (STDs) and, to a lesser extent, population-based studies have shown a relation between male circumcision status and HIV prevalence. 4–6 A recent review of 33 studies suggested that male circumcision particularly protects against HIV among high-risk groups in which genital ulcers and other STDs are driving the epidemic. 7 Recent results from a discordant couple study in Uganda showed a large protective effect of male circumcision. 8 In this context, an increasing number of authors discuss the promotion of male circumcision as a HIV-preventive measure. 7,9,10
Male circumcision is widely practiced around the world, mostly for religious or health reasons. In many parts of Africa, male circumcision practices are linked to culture and religion and signify a transition from one social status to another. Ethnographic data from Africa show marked regional differences in male circumcision practices within the continent. A large belt of noncircumcising tribes runs from southern Sudan through Uganda, western Kenya, Rwanda, Burundi, part of eastern Zaire, and western Tanzania to Malawi, Zambia, Zimbabwe, Botswana, and southern Namibia. 3,11 Most data regarding circumcision practices have been collected several decades ago as part of anthropologic studies of the colonial time (e.g., Murdock’s Ethnographic Atlas). 12 There are a few ethnic groups known to have completely changed their male circumcision practices, such as the Zulu in South Africa (in the 19th century, by order of their king) and the Akan (one of the few noncircumcising groups of West Africa). 3
While studying the risk of HIV infection among circumcised and noncircumcised men in northwest Tanzania, it was observed that traditional patterns of circumcision were changing. 6 A substantial number of men belonging to traditionally noncircumcising tribes have been circumcised. This paper describes these changes and presents an attempt to understand why they are taking place.
The study was carried out among the Sukuma ethnic group in Mwanza Region, northwest Tanzania. The Sukuma are the largest ethnic group in Tanzania and constitute the majority of the population in Mwanza and Shinyanga Regions. Traditionally, Sukuma do not practice male circumcision.
Quantitative data were derived from two sources. The first source is intake data from a cohort study among factory workers in Mwanza town, 13 which began in 1991. These data refer to the period from 1991 to 1994 (988 Sukuma men). The second source are data from a rural population study in one administrative ward (Kisesa), 20 km east of the regional capital Mwanza and 5 km from the factory cohort study site. Since 1994, this population has been followed up through demographic surveillance and epidemiologic surveys to assess the effects of community interventions on sexual behavior and HIV and AIDS, and the impact of the epidemic. Population-based surveys were conducted in 1994 to 1995 and in 1996 to 1997 among all men (and women) 15 to 44 years in Kisesa ward. 14 This analysis combines data from the male factory workers cohort study and the rural study. The studies also provide an opportunity to assess the quality of self-reported circumcision data.
In the rural area, 96% of the men were Sukuma, whereas in the urban factory population, 63% were Sukuma. Only members of the Sukuma tribe were included. In the urban area and the rural trading center, circumcision rates were several times higher than in the rural population.
Qualitative research used two methods. Focus group discussions were held with 13 groups, each consisting of five to eight participants: schoolboys (three groups), middle-aged men (two groups), older men (five groups) and women (three groups). All but one of the group discussions were held in rural settings, including a fishing village, a roadside settlement, and a typical rural community with little influence from outsiders. In addition, 16 in-depth interviews were conducted with schoolboys, adult men, health workers, a traditional healer, and a religious leader. In-depth interviews and focus group discussions took place in the local setting and were carried out by experienced project field staff in Swahili or, if necessary, the local language. All discussions were tape recorded, transcribed, and translated into English. Both the Swahili text and the verbatim-translated English text were consulted during the analysis.
Table 1 presents the determinants of circumcision among men in the first rural survey and in the urban factory cohort study. Many men were circumcised in their late teens or early twenties. Among men 15 to 19 years, only 10% were circumcised, compared with more than 20% of men 20 years and older. The relatively late age at circumcision is also shown in data from the male factory worker study and the second survey in the rural population, in which men were asked to recall at what age they had been circumcised. The mean age at circumcision was 17.4 years and 17.1 years among 329 Sukuma factory workers and 544 rural residents, respectively
Not surprisingly, the proportion of men circumcised was much higher among Muslim men than among non-Muslim men; however, 26% of Muslim Sukuma men had not been circumcised. Christian men had higher circumcision rates than men of other religions, including traditional religions. Circumcision is more common among the more educated; less than 10% of men with none or 1 to 4 years of formal education were circumcised, whereas 23% of men with 5 to 7 years primary school and 57% of Sukuma men with more than 8 years of education (implying at least some secondary education) had been circumcised.
The last two columns of Table 1 present results of a logistic regression model that includes all four variables. Controlling for other variables did not alter the relation between circumcision and the background variables, but the magnitude of some effects decreased. The most important change was the reduction of the effect of urban and roadside residence compared with rural residence after controlling for all other variables. In particular, controlling for religion and education resulted in reduction of the effects of urban and roadside residence compared with the unadjusted model. The higher rates of circumcision in urban areas and, to a lesser extent roadside settlements, can partly be explained by the large number of Muslims and men with secondary education that reside in these places, compared with the rural areas.
The two rounds of the rural Kisesa survey in 1994 to 1995 and 1996 to 97 showed an increase in circumcision rates among Sukuma men from 16.8% to 19.7% (number of participants: 2,606 and 2,776, respectively). The increase could not be attributed to changes in the age structure, residential pattern or survey attendance, or religious denomination. However, the analysis of 1,782 men who were present in both rounds showed that inconsistency in self-reported circumcision status between the first and second round was common (Table 2). Among 243 men who said they had been circumcised in the first survey, 65 (26.8%) said they were not circumcised during the second survey, which occurred 2 years later. Likewise, a sizable proportion of circumcised men reported themselves as not circumcised during the first round but circumcised at the second round. Of the 162 men who said they were circumcised at the second round, only 40 had been circumcised during the last 2 years (between the two survey rounds). If we exclude all men with inconsistent circumcision status, the incidence of circumcision during the 2-year period is 2.8% (40/1417).
Reasons for Changing Practices
Traditionally, the Sukuma men are not circumcised and there are no customary norms or rituals attached to the procedure. In the past it was shameful for a Sukuma man to be circumcised, as evidenced by the presence of a derogatory word for a circumcised man in Sukuma language (njilwa). The four possible factors contributing to the present popularization of circumcision practices—health, sexual pleasure, religion, and ethnic mixing—emerged from the in-depth interviews and group discussions.
In all focus group discussions, the association between circumcision and STDs was spontaneously mentioned. Circumcised men were considered less susceptible to STDs, and several informants claimed that men might circumcise to prevent contracting STDs. A young man in group discussion stated, “There are many reasons why it is important for the youths to circumcise. One is that it protects them from getting diseases. When you are not circumcised you can easily acquire STDs.” A woman attending an antenatal clinic focus group discussion stated that, “Uncircumcised men contract STDs more easily than the circumcised ones.” Finally, a participant in a discussion with school girls claimed, “It is very easy and common for uncircumcised man to contract STD from an infected woman.”
Almost all groups mentioned that STDs are more severe among uncircumcised men and that ulcers heal faster in circumcised men. Some claimed that Sukuma men tend to go for circumcision after contracting genital ulcers to facilitate the healing process. Also mentioned, though less frequently, was that noncircumcised men would be more infectious. A schoolgirl stated during a group discussion that, “An uncircumcised boy who has had sex with STD infected woman can easily infect you; if he proposes having sex with me I will definitely reject. He is dangerous. He may infect me as well.”
There was a strong belief that the foreskin creates favorable conditions for the growth and transmission of STDs, notably by maintaining high temperature. People also claimed that the foreskin secretes a dirty fluid that acts as a favorable medium for the growth of diseases. Male circumcision was also considered to enhance penile hygiene. The foreskin also may potentially be considered a source of bad smell. These points are illustrated by statements made by participants in different focus group discussions: “... when uncircumcised we keep a lot of dirt” (village youth); “... it helps to avoid the dirt under the foreskin”(adult man, rural village); “... circumcision facilitates the avoidance of the bad smell from the dirt that forms under the foreskin” (older man, fishing village); “... when you have the foreskin removed, it saves you the trouble of washing every time. This is the fact, and if you stop washing for, say, two days there will be a bad smell. But if you are circumcised, you need not bother” (adult man).
In focus group discussions and in-depth interviews with young people it was mentioned that circumcision enhances the sexual pleasure of both partners. They believed that it reduces friction during sexual intercourse and increases the woman’s sexual pleasure. Youth also expressed that the foreskin reduced sexual pleasure for men. Some likened the presence of a foreskin to wearing a condom: “Some boys say that to them it is less sensational to have sex with the condom on ... the same way as when one has the foreskin intact ... unlike a circumcised man they don’t experience maximum sensation” (from group discussion with secondary school girls).
Currently, approximately 4% of rural men in Mwanza Region are Muslim; most Muslims live in roadside settlements. In urban areas, Islam has more followers, and Islamic law necessitates that a male child be circumcised by the 40th day after birth. Arab traders introduced Islam to this part of Africa during the 19th century, mainly in trading centers. Circumcision could be obtained from a ngariba—an Arab expert who provided circumcision services for the Arab community. Later, circumcision services were also provided by hospitals, which probably led to the disappearance of the ngariba services in this area. Several Muslims in the group discussions and in-depth interviews revealed that they were not circumcised. They provided examples of how the practices of newly introduced religion are harmonized with preexisting belief systems, and in some cases circumcision was not considered compatible with their traditional religion (e.g., it provoked the ancestral spirits). There is no Christian influence favoring circumcision other than circumcision services, which may be provided by the church-owned health facilities.
Our study shows that male circumcision has become more popular among a traditionally noncircumcising tribe in northwest Tanzania where approximately one in five men are now circumcised. In most African societies where males are circumcised between the ages 6 and 15 years, the main reason is ritual practice. 3 Additional reasons (e.g., finding the circumcised penis esthetic, that the practice enhances fertility or sexual pleasure) have also been mentioned in a circumcising ethnic group in Nigeria. 15 Most men in our study were circumcised during late adolescence or in adulthood, mostly for health reasons. Circumcision is thought to enhance penile hygiene, reduce STD incidence among men, and shorten the duration or lessen the infectiousness of STDs. In some instances, enhancement of sexual pleasure was considered a secondary reason for male circumcision, in contrast with findings of studies from western countries, which emphasize reduction in sexual pleasure associated with removal of the sensitive foreskin. Few respondents mentioned male circumcision as a method of STD treatment. There was some anecdotal evidence of such a practice, 6 but this does not appear to be common.
How genuine is the increase in self-reported circumcision rates? Among men who were asked twice about their circumcision status with 2 years between questioning, there was considerable inconsistency. In the factory worker cohort study, the self-reported circumcision status was validated against subsequent physical examination in a 10% sample of the study population. 6 Upon physical examination, 31% of men who reported circumcision during the interview were not circumcised, and 6% of men who had reported not being circumcised were circumcised. Therefore, in the factory population, the actual prevalence of circumcision was lower than the reported rate (28% versus 34%). If the same misclassification rates would apply to a rural area with 10% of men circumcised, the effect would be the opposite—the “true” circumcision rates would be 12%. Therefore, even though the misclassification biases are not small, they do not refute our conclusion that male circumcision has increased considerably among Sukuma men. In analyses of the relation between male circumcision status and HIV, misclassification may present a problem and weaken the association.
Is it likely that changes such as those observed among the Sukuma population is or will occur elsewhere in Africa? In a homogenous noncircumcising society, it seems less likely that male circumcision will be accepted by a large part of the population. Sukuma society has been exposed to male circumcision through the influence of Islam and the mixing of noncircumcising and circumcising tribes. The nearest circumcising ethnic groups are the smaller tribes of the Jita, Kwaya, Zanaki, Ikizu, and Kurya, all living in Mara Region to the north of Mwanza Region. 16 Within Mwanza Region, no resident ethnic group is circumcising, and circumcision is not traditionally practiced to the south (Nyamwezi) and west (Haya and Ha). The growth of urban centers and the establishment of district capitals with government representatives from all over the country has led to increased mixing of circumcising and noncircumcising ethnic groups. For example, in a factory population in Mwanza town, 33% of men were found to be of circumcising ethnic groups. In fishing villages along the shores of Lake Victoria, this figure was 11%, and 4% of participants in a large rural survey were found to be of circumcising ethnic groups. 6 The mix of ethnic groups is most obvious in secondary schools, and has led to increasing acceptance of male circumcision. The discussions and interviews with schoolboys indicated positive attitude toward circumcision, linking it with modernity and hygienic practices. Both young and adult men stated that it is now difficult for a noncircumcised man to be accepted for sex by a woman from an ethnic group that practices male circumcision.
How desirable is the change in male circumcision practices? Poor penile hygiene, worsened by shortages of water and bathing facilities, is likely to be common and may lead to infections. The effect of circumcision on HIV transmission needs further study, but results increasingly suggest that male circumcision status is an important cofactor. 2,7,8,17 Male circumcision is likely to lead to an improvement in penile hygiene and contribute to a reduction in STDs, notably genital ulcers. No trials have been conducted to show the effect of male circumcision on HIV incidence, though studies in Kenya and Uganda suggested that a considerable proportion of men would want to be circumcised if such services were free of charge in hospitals. 10,11 This may also be the case among the Sukuma of Tanzania. The protective effect of circumcision on the risk of HIV transmission, however, may be smaller if circumcision is carried out at later age, as was suggested by the results of a study in Uganda. 18
Finally, health programs in Africa need to take into account the delicate and ever-changing balance between traditional systems and modernizing influences. The popularization of male circumcision among the Sukuma of Tanzania shows that traditional health beliefs and practices may change under influences caused by increased mobility of people (e.g., mixing of ethnic groups in schools, urban areas) and not by health programs per se. The increased mobility leads to more exposure to different health beliefs and practices, and may change practices at a societal level. This change should be taken into account in the design of health programs, and, in this case, of HIV and STD prevention programs.
1. Bongaarts J, Reining R, Way P, Conant F. The relationship between male circumcision and HIV infection in African populations. AIDS 1989; 3: 373–377.
2. Moses S, Plummer A, Bradley JE, et al. The association between lack of male circumcision and risk for HIV infection: a review of the epidemiological data. Sex Transm Dis 1994; 21: 201–210.
3. Caldwell JC, Caldwell P. The neglect of an epidemiological explanation for the distribution of HIV/AIDS in sub-Saharan Africa: exploring the male circumcision hypothesis. Health Transition Rev 1994; 4 (suppl): 23–46.
4. Cameron DW, Simonsen JN, D’Costa LJ, et al. Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men. Lancet 1989; 74: 368–73.
5. Moses S, Bradley JE, Nagelkerke NJD, et al. Geographical patterns of male circumcision practice in Africa: association with HIV seroprevalence. Int J Epidemiol 1990; 19: 693–697.
6. 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.
7. O’Farrell N, Egger M. Circumcision in men and the prevention of HIV infections: a ‘meta-analysis’ revisited. Int J STD AIDS 2000; 11: 137–142.
8. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type I. N Engl J Med 2000; 342: 921–929.
9. Bailey RC, Neema S, Otieno R. Sexual behaviors and other HIV risk factors in circumcised and uncircumcised men in Uganda. J AIDS 1999; 22: 294–301.
10. Halperin DT, Bailey RC. Male circumcision and HIV infection: 10 years and counting. Lancet 1999; 354: 1813–1815.
11. Conant FP. Regional HIV prevalence and ritual circumcision in Africa. Health Transition Rev 1995; 5: 108–112.
12. Murdock GP. Ethnographic atlas: a summary. Ethnology 1967; 6: 109–236.
13. Senkoro KP, Boerma JT, Klokke AH, et al. HIV incidence and HIV associated mortality among a cohort of factory workers in Tanzania, 1991–1996. J AIDS 2000; 23: 194–202.
14. Boerma JT, Urassa M, Senkoro K, Klokke A, Zaba B, Ng’weshemi JZL. Spread of HIV infection in a rural area in Tanzania. AIDS 1999; 13: 1233–1240.
15. Myers RA, Omorodion FI, Isenalumhe AE, Akenzua GI. Circumcision: its nature and practice among some ethnic groups in Southern Nigeria. Soc Sci Med 1985; 21: 581–588.
16. Dodge OG, Kaviti JN. Male circumcision among the peoples of East Africa and the incidence of genital cancer. East Afr Med J 1965; 42: 98–105.
17. Moses S, Bailey RC, Ronald AR. Male circumcision: assessment of health risks and benefits. Sex Transm Infect 1998; 74: 368–373.
18. Kelly R, Kiwanuka N, Wawer MJ, et al. Age of male circumcision and risk of prevalent HV infection in rural Uganda. AIDS 1999; 13: 399–405.