The HIV epidemic remains a major health challenge all over the world. In 2013, the UNAIDS estimated 35million people were living with HIV globally; and in the same year, 2.1 million new HIV infections and 1.5 million AIDS related deaths occurred [1]. Recent studies have identified approaches for HIV prevention especially those that target the generalized epidemic in sub-Saharan Africa and includes Comprehensive condom and lubricant programming, harm reduction for people who inject drugs, behavioral interventions, prevention of transmission in health-care settings, ARV-related prevention, and voluntary medical male circumcision for HIV prevention [2–7]. Results from randomized clinical trials conducted in Kenya, Uganda and South Africa established that male medical circumcision greatly reduced the risk of HIV infection among men [3, 8, 9]. Further studies have shown how those already infected can be treated [2, 10] or how the effects of the virus on a person’s immunity can be minimized by avoiding reinfection [8, 11]. As a result of the findings of the randomized clinical trials which showed that medical male circumcision can be used as a prevention measure for HIV infection [3, 8, 9], the World Health Organization (WHO) and the Joint United Nations Programme on HIV/AIDS (UNAIDS) recently recommended that circumcision be added to current interventions to reduce the spread of HIV [12, 13]. Male circumcision is defined as the complete removal of the foreskin [14]. Traditionally, many African societies have embraced male circumcision as a rite of passage from childhood to adulthood, rather than a medical operation [14] and also for religious and to a lesser extent, for medical purposes [15]. While not all communities practice circumcision, studies investigating HIV transmission done among communities practicing circumcision have consistently found that circumcision significantly reduces the rate of HIV transmission at the community level [16, 17].

In many of these studies, the outcome has been that circumcised men are significantly less likely to acquire HIV when exposed sexually to HIV-positive women as compared to men with intact foreskins [3, 8, 9, 18]. Special phagocytic cells in the foreskin, Langerhans cells, are said to act as magnets to the virus which attaches to the cell and thus enters the body [19]. A recent review by Morris and Wamai termed the inner foreskin as a mucosal epithelium which is deficient in protective keratin but rich in HIV target cells [20]. Once exposed to infected mucosal secretions, infected T-cells combine with keratinocytes and then transfer HIV to Langerhans cells via dendrites just under the surface of the inner foreskin which then migrates to the basal epidermis and then pass HIV to T-cells leading to infection [20]. This is aggravated by occurrence of any inflammations and ulceration and also tearing of the foreskin during intercourse [20]. In addition, randomized trials of male circumcision showed that male circumcision significantly reduced the bacterial load by reducing both the prevalence and abundance of many coronal sulcus bacteria [21]. Nevertheless, male circumcision does not provide a complete prevention against HIV infection, and circumcised men can become infected, and can transmit the virus to their partners [2, 3, 8, 9].

However, there still remain gaps and unanswered questions relating to HIV transmission and circumcision. Although not supported by literature, one major concern is the behavioral disinhibition where circumcised men could feel more protected against HIV infection and hence may engage in a riskier behavior. Traditionally, and in some communities, young men are taken into seclusion and given advice touching on their responsibilities which includes their sexual behavior [14]. Though not well documented, its often assumed that these ‘guidance and counseling’ sessions inform or instill behavior change and risk reduction on HIV transmission among circumcised men as compared to the uncircumcised, who principally do not go through such sessions. However, cluster randomized trials done in South Africa to assess the interventions that could prevent men from sexual risky behaviors after circumcision showed that social cultural values dictated how men should behave among traditionally circumcised men [22] and hence resistance to change; while among those medically circumcised, a focused counseling session had short term effects on reducing risky sexual behavior [23]. During male circumcision rituals, the period of seclusion constitutes the most significant part where new attitudes, practices and behaviours are learnt which includes sexual issues [14]. Whereas sexual reserve and inappropriateness of promiscuity after reintegration into society are emphasized, in other contexts, boys are encouraged to have sex to prove their manhood even during the seclusion period, often referred to as ritual sex [14]. Ritual sex is practiced in many African communities and is associated with traditional circumcision where men are “cleansed through” and/or “trained on” sexual activities. Although ritual sex is conducted in some communities in Malawi, there is limited information regarding this practice. However, in Zambia, ritual sex is associated with traditional circumcision and use of condom is not emphasized [24, 25]. There is therefore a need to look at the association between type of circumcision and HIV infection in Malawi, as well as analyze the effect of ritual sex on HIV infection.

While some studies show that pre-pubertal circumcision is associated with reduced risk of HIV infection [26, 27], one study shows no significant association between age at circumcision and HIV infection [28]. Male circumcision in many African societies is done at puberty [29] ; however, there are communities that do not have a defined circumcision period and therefore some men are circumcised as infants, while others are circumcised at adulthood. When circumcised at old age, these men are already exposed to sexual activities for a long time, and have higher chances of being exposed to HIV as compared to younger men.

Though statistically insignificant, our paper shows that circumcised men in Malawi have a higher HIV prevalence rate (12 %) than the uncircumcised men (10 %). However, the circumcision in this context is mainly traditional circumcision as voluntary medical male circumcision was introduced in 2011, a year after these data were collected. This is contrary to the literature, which suggests that even traditional male circumcision is highly effective in HIV prevention [35]. However, our general model shows that other factors exhibited significant influence on the HIV status. These are age, marital status, number of lifetime sexual partners, type of residence and region. The mode of HIV transmission in Malawi is predominantly heterosexual and as such, it is also expected that individuals with higher levels of exposure to sexual intercourse would be more prone to HIV infection than those with lower levels of exposure to sexual intercourse. While our findings on age and number of lifetime sexual partners appear to confirm this assertion, our result on the effect of marital status deviates from this assertion. The never married are 1.3 times more likely to be HIV+ than the ever married men. Even though our study did not measure the actual number of times an individual had sexual intercourse, it is often assumed that the married have a higher coital frequency than the unmarried persons. It is also not very clear why men in the Northern and Central Regions have higher HIV prevalence than their counterparts in the Southern Region. However, urban areas have in many cases been found to have a higher HIV prevalence rate than rural areas [36, 37] and this is the case for Malawi as well.

Among the circumcised men (Model IV), age and number of lifetime sexual partners remain dominant as correlates of HIV status. Interestingly, however, this model appears to provide clues as to why there is no significant difference between circumcised and uncircumcised men in terms of HIV status. Circumcised men who ever had ritual sex are 2.4 times more likely to be diagnosed with HIV than circumcised men who have never had ritual sex. The context of ritual sex in Malawi is not immediately clear. However, evidence of ritual sex after traditional circumcision has been documented in Zambia, one of Malawi’s neighboring countries [24, 25]. This ritual sex is associated with traditional circumcision and exposes men to the risk of HIV infection as use of condom is not emphasized thereby negating the preventive role that male circumcision is supposed to play [24]. Since most of the circumcised men were circumcised by traditional methods in Malawi, there is a high likelihood that the observations made in Zambia [25] could also be at play.

Model IV also confirms observations from elsewhere that Catholics/Christians have higher HIV prevalence than Muslims [38]. Interestingly, circumcised men in the medium wealth index (or the middle class) have significantly higher odds of being diagnosed with HIV than the rich circumcised men. This seems paradoxical as rich men are often viewed as agents of HIV [39] as they are often associated with concurrent multiple sexual relationships [40].

There are some limitations associated with DHS data and that in a way may impact the results of this study. First, DHS data are cross-sectional and hence temporality of the association between male circumcision and HIV infection cannot be absolutely established. Secondly, male circumcision is self-reported, and thirdly, it’s impossible to establish when men were circumcised and when they became HIV+ and lastly is the dearth of information around ritual sex and traditional circumcision in Malawi.

Our study shows that there is a weak association between male circumcision and HIV infection in Malawi. Twelve percent of circumcised men were HIV positive compared to almost 10 % of uncircumcised men. However, most of the circumcised men in this study were circumcised traditionally as the policy to offer medical male circumcision was only introduced in 2011 (a year after the 2010 MDHS was conducted); and hence the results of this study refer mainly to traditional circumcision and should be interpreted with a lot of caution. In addition, the finding could be a pointer for the increased need for voluntary medical male circumcision in Malawi, and the need to shun away from the (unsafe) traditional male circumcision practices. Given that most circumcised males were circumcised using traditional methods and that HIV-positive status among circumcised men is significantly associated with ritual sex, it can be deciphered that there is a strong association between traditional circumcision (which may include ritual sex) and HIV infection in Malawi which may be connected to the actual reasons why HIV prevalence is higher among circumcised men than uncircumcised men in Malawi. Therefore, our study shows that traditional male circumcision does not act as a preventive measure to HIV infection in Malawi. However, we acknowledge that the main limitation of this notion is the dearth of information around ritual sex and traditional circumcision in Malawi.