Received Mar 9; Accepted Aug 8. This article has been cited by other articles in PMC. Abstract Background Kisumu has shown a rising HIV prevalence over the past sentinel surveillance surveys, and most new infections are occurring among youth. We conducted a qualitative study to explore risk situations that can explain the high HIV prevalence among youth in Kisumu town, Kenya Methods We conducted in-depth interviews with adolescents aged 15 to 20, held 4 focus group discussions, and made 48 observations at places where youth spend their free time.
Results Porn video shows and local brew dens were identified as popular events where unprotected multipartner, concurrent, coerced and transactional sex occurs between adolescents. Video halls - rooms with a TV and VCR - often show pornography at night for a very small fee, and minors are allowed. Forced sex, gang rape and multiple concurrent relationships characterised the sexual encounters of youth, frequently facilitated by the abuse of alcohol, which is available for minors at low cost in local brew dens.
The desire for love and sexual pleasure also contributed to their multiple concurrent partnerships. A substantial number of girls and young women engaged in transactional sex, often with much older working partners.
Condom use was irregular during all types of sexual encounters. Conclusions In Kisumu, local brew dens and porn video halls facilitate risky sexual encounters between youth. These places should be regulated and monitored by the government. Our study strongly points to female vulnerabilities and the role of men in perpetuating the local epidemic.
Young men should be targeted in prevention activities, to change their attitudes related to power and control in relationships. Girls should be empowered how to negotiate safe sex, and their poverty should be addressed through income-generating activities. Background Globally, the HIV epidemic is increasing faster amongst young women than young men and nowhere is this trend more apparent than in sub-Saharan Africa [ 1 ]. The main form of HIV transmission is heterosexual sex.
The contrasting HIV prevalence between boys and girls is a pattern observed in many parts of sub-Saharan Africa [ 4 , 5 ]. Kisumu town is found by the shores of Lake Victoria, is the capital of Nyanza province and the third largest town in Kenya. Although more recent figures for Kisumu are slightly lower, the numbers are still alarmingly high, and higher than in other parts of the country [ 9 , 6 ].
Cultural norms such as wife inheritance and widow cleansing, polygamy, "jaboya" in which female fishmongers develop sexual relationships with fishermen and middlemen in exchange for fish , and "chira" a curse that comes from breaking certain taboos and traditions , continue to have a powerful hold on people in this lakeside province [ 10 , 13 , 14 ].
We conducted a qualitative study on the sexual behaviour of young people in Kisumu, to further explore results from an earlier population survey on factors determining the differential spread of HIV in four African cities: This study showed that girls in Kisumu had older sexual partners than boys and higher rates of herpes simplex type 2, which are both risk factors for HIV transmission. But most girls reported very few sexual encounters, and HIV prevalence was very high even among girls reporting one lifetime partner and few episodes of sexual intercourse.
This may be due to underreporting, but also may indicate high transmission during loss of virginity [ 5 ]. Our aim was to deepen our understanding of the dynamics of sexual interactions of adolescents, in order to explain the high HIV prevalence among Kisumu youth in general and specifically among girls.
We triangulated data from in-depth interviews, focus group discussions FGD , and observations, to generate a holistic description of the contexts and dynamics of sexual interactions among youth.
Methods We conducted in-depth interviews with adolescents aged in Kisumu, held 4 FGDs, and performed 48 observations at places where youth spend their free time. For the in-depth interviews, a convenience sample of 75 boys and 75 girls aged years were interviewed at their households. Using the sampling framework of the multicentre study, quota sampling was used to ensure diversity in age, socio economic status SES of household, and education [ 9 ].
A qualitative interview guide was developed in English and translated into Swahili and Luo. Trained fieldworkers pre-tested the guide in communities neighboring the study sites. The interviews were held in Luo, Swahili or English by same-sex interviewers, and tape-recorded; they took about 45 minutes to one hour. Four focus group discussions FGDs were held by same-sex interviewers in preparation for the in-depth interviews i.
A topic guide was used, that was pre-tested for face and construct validity. The FGDs mainly focused on youth's attitudes, risk perception and socio-cultural norms regarding sexuality.
Each FGD had 8 to 12 participants aged , and the discussions took about one and half hour. We sought permission from the persons in charge and ensured confidentiality of all collected information. The field workers tried to get an inside view of reality without the participants' knowledge that they were being observed. Short notes were taken during the 2-to-3 hour observations when possible, and detailed notes were compiled afterwards describing the physical setting, the activities taking place, socio-demographics of participants estimated age, gender , and their verbal and non verbal behaviour.
The audio-taped interviews and FGDs were transcribed verbatim and translated into English where necessary [ 16 ]. Data analysis of the in-depth interviews, FGDs and observations followed grounded theory principles, which allows analytical themes to emerge during the process of re reading transcripts and exploring and coding responses [ 17 ].
The following themes relating to risky sexual behaviour emerged from the interviews, FGDs and observations and are discussed below: Where quotes are used in the Results section, they are from the in-depth interviews unless indicated otherwise. The average age of adolescents interviewed in the study was A year-old girl from a low-SES area stated: Sixteen of the 28 girls who mentioned the age of their first partner reported that he was 25 years or older.
All girls with a current partner had a partner who was over 20 years old ranging from 2 to 17 years older. During the observations, older men were often seen with girls as young as Of the sexually active adolescents, over half of the girls reported having had partners; over half of the boys indicated having had partners and about a third reported more than 10 girlfriends. A few boys found it difficult to count all sexual partners: Some of the bicycle taxi transporters reportedly had sex daily, with different partners.
Girls mostly reported serial monogamy and rarely stated outright that they had concurrent partners. In contrast, many sexually active boys casually reported having overlapping partnerships: These 'disco funerals' are parties held by the relatives of a person recently deceased, in order to raise funds for the funeral.
The disco funerals are characterised by loud music, singing, dancing, bidding games and risky sexual behaviour. Because we have described these events in which Kisumu youth engage in risky sexual behaviour in a separate paper, we do not elaborate on it here [ 18 ].
A third of the youth interviewed 49 report porn video shows either from personal attendance or as an influence on youth sexuality. Some also report watching porn at home or at the home of a friend: Video show halls are basically rooms with a television and VCR, they are popular leisure spots, where youth pay to watch movies. The fieldworkers did a dozen of random observations in video halls: Most attendees were young men, but some girls also attended.
The movies ranged from non-violent to violent pornography, and the scenes revolved around group sex, anal sex, and oral and vaginal intercourse. Youth said that the owner of a video hall most times disguises the announcement of a featured film and writes "on-por": During one observation, several adolescents engaged in sex in the darkness of the hall. Reports of forced sex were many: Some girls were lured into secluded places such as a boy's cube [separate living quarter for boys], and were forced to have sex: Some boys were said to waylay prostitutes and force them to have sex: But these groups of boys don't have any cash, they just get them and hijack them yah," FGD, out-of-school boys.
Few adolescents reported condom use at first sex 12 girls and 15 boys out of 96 sexually active adolescents. Condom use was neither common nor consistent: I use a condom but not every time. For most girls, non-use was related to their limited ability to request condom use, ideals of intimacy and pleasure, and mixed messages about safety: Some young people also held the belief that very young boys and girls could not have STIs.
Some adolescents only used condoms during the unsafe period in the menstrual cycle. Others reported they usually stop using condoms in a new relationship when they have become more used to each other or, as a few reported, after getting a negative HIV test.
Heavy alcohol intake and in particular illicit brews and drugs reinforced and mostly accompanied these sexual risk behaviours, as illustrated by this year-old girl: During interviews over a third 56 of youth interviewed mostly boys said they have gone to drink in these local brew dens.
The brew was often made by widows who depend on the trade for their livelihood. It was reported that in environments where a parent s made local brews, girls started having sex at an early age and some progress to trade sex for money: One girl said in an interview that she had her first sexual encounter at the age of fourteen, after her older sister described to her what the men were doing to her when she went to serve them namely having sex for money.
The younger girl got pregnant and gave birth at the age of fifteen. Another girl stated that she had sex several times with a client who promised to build her widowed mother a house.
Many interviewed youth reported to have had transactional sex. There were reports of young women going to local brew dens or disco funerals hoping the men would buy them drinks or give them money, in exchange of sexual favours. Often only a little money was required: Bicycle-taxi operators were reportedly popular, offering a few shillings, some food, or free transport to young women in exchange for sex: Only occasionally did a boy receive gifts from his girlfriend s. An year-old girl from a low-SES area described how she decided to have sex with her year old boyfriend: Many young men acknowledged that their ability to provide for their girlfriends affected both the longevity and exclusivity of their relationships: Eight of the 75 interviewed boys reported that that they had had contacts with a prostitute.
Not much money was involved in such encounters: Discussion In this qualitative study on sexual behaviour of youth in Kisumu, Kenya, we found that the majority had sex at a young age, sometimes with multiple and concurrent partners, mostly without using a condom. Drugs and local alcohol often facilitated these encounters [ 19 - 22 ].
Findings point strongly to the role of men in perpetuating the HIV epidemic forcing sex, gang rape, multiple concurrent relationships. Many young people were exposed to pornography in video halls, which seemed to increase their sexual risk behavior. Peer influence was a great motivator for these risk behaviors.
There are a number of limitations that should be considered when interpreting the results. Because we used purposive quota sampling, it cannot be concluded that our results are fully representative for the total Kisumu youth population.