Risk factors for unplanned sex among university students in Kampala, Uganda: a qualitative study

Abstract

There is inadequate in-depth information on the risk factors for human immunodeficiency virus (HIV) and sexually transmitted infections (STIs) among university students in Uganda, and more so of those that are unrelated to alcohol and drug use. ‘Unplanned sex’ is defined here as sexual intercourse that a person or individual decides out of free will as undesired. We conducted cross-sectional in-depth qualitative study with 23 students (52% female, age 19–25 years) from five universities in Kampala, Uganda as part of a larger survey group sampled from October 2008 through April 2009. Participants were selected based on their responses to select questions and themes of interest. Twenty-two students reported unplanned sex and one student seemed not to have had unplanned sex. Some university students reported environmental and personal factors as mitigating factors for unplanned sex: ‘We meet so many friends especially when we go to beaches and attend sports, having sex here comes unplanned and appears to be fun for everyone’ (Farida, 22 years old). Addressing factors associated with unplanned sex may contribute to prevention and control of HIV/STI and unplanned pregnancies among emerging adults in sub-Saharan Africa.

Keywords:

• emerging adults
• unplanned sex
• qualitative study
• Kampala
• Uganda

Introduction

There is inadequate in-depth information on the risk factors for human immunodeficiency virus (HIV) and sexually transmitted infections (STIs) among university students in Uganda, and more so of those that are unrelated to alcohol and drug use. Sex involving university students may be unplanned and may occur without condoms (Hingson, Heeren, Winter, & Wechsler, 2003; Klein et al., 2007). ‘Unplanned sex’ is defined here as sexual intercourse that a person or individual decides out of free will as undesired or spontaneous sexual activity that was not thought out beforehand or not prepared for in advance. Unplanned sex is a particular concern for sexually inexperienced individuals and can be associated with alcohol and drug use (Hingson et al., 2003), lack of condom use (Ma et al., 2007), and unwanted pregnancies (Ma et al., 2008). Although there has been a decline in HIV prevalence in Uganda, 50% of those living with HIV are youths below 25 years (Uganda Ministry of Health [MOH] & ORC Macro, 2006). These findings suggest that university students are at risk of HIV/STI infections.

In normal circumstances, age at first sex is a predictor of first risk exposure to HIV; in Uganda the youth first penetrative sex experience is at 15 years and there is a greater frequency of sex between ages 19 and 24 than at other ages (MOH & ORC Macro, 2006). In a nationally representative survey in Uganda, 31% of women and 39% of men aged 19–24 years demonstrated comprehensive knowledge about HIV, with higher knowledge scores among those with better education and higher wealth (MOH & ORC Macro, 2006).

University students may be seen as emerging adults who, no longer being ‘adolescents’ nor yet fully ‘adults’, at times may be in a state of confusion, especially where adulthood is viewed as being self-sufficient and independent from parents (Arnett, 2003). In some studies, emerging adults are seen as having a bicultural identity – copying both their cultural beliefs and other cultures (Schwartz, Zamboanga, Rodriguez, & Wang, 2007).

In some studies unprotected sex among university students is also associated with lack of social capital (Agardh et al., 2010). Social capital is simply trust in others, participation and sharing values (Agardh et al., 2010). Theoretically the aspects of social capital are interrelated to social participation, which involves sharing values (Agardh et al., 2010) like enjoying same genre of music in disco clubs. Health education and sensitisation for young people can be an important aspect in addressing these behavioural challenges (Kabiru & Orpinas, 2009).

Unplanned sex is complex but can be understood by an in-depth examination of individual experiences, especially for emerging adults (Arnett, 2003) like university students. Social cognitive theory is important in exploring individual behaviour in relation to their environment and personal factors. According to this theory, people acquire and maintain certain behavioural patterns, which if analysed may provide appropriate intervention strategies (Bandura, 1996). The environment, people and behaviour are constantly influencing each other (Bandura, 1996).

The exchange of gifts or material goods for sex is sometimes expected as a sign of appreciation in sexual relationships among African youth, and as such can be fittingly viewed in the ensuing context of a man's responsibility to provide for a woman. However, in other situations unplanned sex occurs because of desire for material gains (Moore, Biddlecom, & Zulu, 2007). Coerced sex may be frequent among individuals in the age group 19–25 as compared with other ages in Uganda (MOH & ORC Macro, 2006); however, there is no biological linkage between coerced sex and HIV risk among emerging adults in Uganda (MOH & ORC Macro, 2006).

In Uganda, university enrolment also marks the beginning of independent living by moving out of parents' homes into hostels on or off campus. There, students participate in numerous social activities and functions such as dance clubs, sporting events and binge drinking, which may lead some students into experimenting sexually (Mutungi, Facchini, & Barasa, 2008). These activities may facilitate unplanned, transactional, or coerced sex and unwanted sexual advances that can contribute to HIV transmission (Hingson et al., 2003; Wamoyi et al., 2010).

Pre-marital sex is frowned upon in Ugandan culture and penetrative sex is advocated as a preserve of marriage, traditionally happening after payment of a bride prize (Parikh, 2007). In some studies, ethnic diversity appeared to weaken the ability to enforce such traditional cultural norms (David, Priya, George, & Ken, 2006). Cohabiting without formalising a sexual relationship is now a common practice in Uganda, and efforts to reduce pre-marital sex have often been criticised for their messages like media advertisements and billboards against cross-generational sex, this potentially exposes youths to accompanying HIV infection risk (David et al., 2006).

In Uganda, 29% of young women and 33% of young men used condoms the first time they had sex. In addition, urban youth are more likely to use condoms at first sex (MOH & ORC Macro, 2006). In Uganda, the urban population had higher use of condoms at first sex in comparison with other similar populations elsewhere (MOH & ORC Macro, 2006). Evidence indicates that sexual intercourse within marriage or stable partnerships has a high risk of HIV infection, especially among women who practice unprotected sex with their partners (Parikh, 2007).

The study was conducted to obtain more information on unplanned sex among university students in Kampala, Uganda. A qualitative method was used for data collection; this was important because it enabled us acknowledge the concerns of the study population. The study was conducted from October 2008 to April 2009.

Background information

Uganda has a total of 28 government-aided and private universities. According to Uganda Bureau of Statistics, in 2010 there were 117,082 university students (50,819 women and 66,263 men) registered in public and private institutions of higher learning in Uganda. Most of the students in the universities are between 20 and 24 years old, indicating that most of the students are emerging adults (East African Community/AMREF Lake Victoria Basin Partnership [EAC/EALP], 2010). The national HIV prevalence of youths between 19 and 25 years is about 3.2%; however, according to the sero-behavioural survey among university students the prevalence in this group was below 1%. A majority of university students are non-residents and self-sponsored (EAC/EALP, 2010).

In a survey in Uganda, Kampala International University is reported to have most of the international students, followed by Makerere University and the Islamic University of Uganda, and most of the students in these universities reside within the communities surrounding the universities (EAC/EALP, 2010). The report also indicates three-quarters of university students are sexually active, with 18 years as the median age of first sexual encounter and 57% having sex 12 months before the study. Similarly, only 59% of the sexually active students reported having sex with a condom (EAC/EALP, 2010).

Research purpose

This study aimed to explore and to have an in-depth understanding of issues surrounding HIV and HIV risk, including practices, perceptions, stigma, and violence among university students in Kampala, Uganda; and to understand university students' perception of unplanned sex and factors that may have influence on such behaviour.

Methods

Setting

Kampala is the capital city of Uganda. It hosts the largest number of university students in the country and attracts a large number of foreign students from the East African region. The five recognised universities selected within Kampala district were Makerere University Kampala, Makerere University Business School, Kampala International University, Kyambogo University and Kampala University.

The Crane Survey was funded by the US Centers for Disease Control and Prevention and was conducted in partnership with the Makerere University School of Public Health and the Ugandan Ministry of Health in Kampala Uganda (Crane Survey report, 2010). This cross-sectional qualitative in-depth substudy was part of the Crane Survey, a survey designed to improve understanding of the epidemiology of HIV among risk groups in Kampala, Uganda, conducted from October 2008 to April 2009.

Theoretical model

We organised our risk assessment based on the constructs of social cognitive theory on human development (Bandura, 1986). The theory interprets human capabilities as facets of psychobiologic origin and can be sustained or enhanced by experience, and therefore diversity in social practices may leave some developmental process underdeveloped. This theory is based on a causation model, where behaviour is shaped and controlled by environmental or personal influences interacting at different levels (Bandura, 1989). In this manuscript we therefore took advantage of a survey among Kampala university students by analysing emerging themes to derive students' interpretation of unplanned sex, using social cognitive theory to explore factors and mode of intervention that may be important in addressing that risk.

Participant selection and data collection

This substudy purposively selected 23 students for qualitative substudy from the main survey of 694 university students recruited using respondent-driven sampling methods (Heckathorn, 2007a, 2007b). Out of the total recruited, 644 sexually active students undertook the audio-computer-assisted self-completed interviews (ACASI), as well as testing for HIV and selected STIs, and 54 had not had their sexual debut and neither undertook the ACASI nor were tested for HIV and STIs. However, these students were issued coupons for peer recruitment.

In this qualitative substudy we flagged the 23 selected students based on their responses to particular ACASI questions, including drug use, a prior HIV-positive test result, same sex or anal sex, sex work, or sex with lecturers. In-depth interviews took place during a separate appointment; not all students who were invited actually kept their appointments. Survey methods are described in more detail elsewhere (The Crane survey report, 2010). Eligibility criteria for participation in survey activities included being 19–25 years old, having had sex in the past six months, and having been enrolled at one of the five selected universities for at least one year. Qualitative data were collected over a period of six months (October 2008–April 2009) in English.

The interviewers had experience and training in qualitative research methods. The interviewers explained the purpose of the study and how confidentiality would be maintained before obtaining consent. The qualitative semi-structured interview guide explored the students' background, social networks, health-seeking behaviours, alcohol and drug use, perceptions of risk, HIV testing, condom use, abortion, sexual history, same-sex experiences, stigma and violence, recruitment patterns, and ACASI experiences. Interview length typically varied between 45 and 90 minutes.

Data management and analysis

Qualitative interviews were recorded on audiotapes in English and transcribed verbatim. Transcripts were then imported into Nvivo software (version 8.0; QSR International Pty Ltd, Victoria, Australia), a program designed to assist in qualitative data management and data analysis. Theme development, validation and code use were utilised as the primary analytic strategies with focus on descriptive thematic coding (Cooney, 2010). Themes were later summarised across all respondents. Interviewers and investigators read the transcripts and developed provisional coding of emerging themes and subthemes to eventually form a code book. Coding utilised a grounded theory approach from which concepts were derived (Cooney, 2010). The team used theoretical literature and emphasised emerging conceptual insights from data in relation to already existing codes. Script analysis was conducted across all interview data. A list of themes and subthemes was finalised and their results are presented in this manuscript. The analysis was sensitive to narrative approaches that explored the way participants constructed their experiences on themes of interest. The analysis team assessed the agreement, and any areas of difference were reviewed by the coders together with the other evaluation team members to validate data interpretations. Interpretation of students' narratives was done using social cognitive theory (Bandura, Caprara, Barbaranelli, Pastorelli, & Regalia, 2001).

Ethical considerations

Participants gave verbal informed consent for survey participation. Interviews were conducted considering participants' privacy. The protocol was approved by the review board of the Uganda Virus Research Institute, Uganda National Council for Science and Technology, and the US Centers for Disease Control and Prevention.

Results

We purposively selected 23 students (12 female and 11 male) from within the main survey for additional qualitative interviews. Fifty-two per cent of the respondents were female (n = 12) and 48% were male (n = 11). Respondents had a mean age of 22 years (female 22.1 years and male 22.5 years) with an age range of 19–25 years.

Table 1 illustrates self-reported behavioural characteristics by sex (‘Ever had unplanned sex’, ‘condom use at last sexual encounter’, ‘coerced sex’, ‘circumcised respondents’, ‘ever tested for HIV’, ‘ever taken drugs for pleasure’, ‘drank alcohol in last 30 days’, and ‘exchange of money for sex’). Only students enrolled for at least one year at one of the five participating universities in Kampala were selected. All students were sexually active, with 22 reporting unplanned sex and one student did not have unplanned sex. Students were asked for their experience on the variables in Table 1 in qualitative interviews

Table 1 Behavioural status through self-reported characteristics of university students by sex (n = 23).

Behavioural characteristic	Female	Male
Unplanned sexual encounter
No	0 (0.0)	1 (100.0)
Yes	12 (54.5)	10 (45.5)
Condom use at last sexual encounter
No	4 (57.1)	3 (42.9)
Yes	8 (50.0)	8 (50.0)
Coerced sex
No	7 (58.3)	5 (41.7)
Yes	5 (45.5)	6 (54.5)
Circumcised
No	0 (0.0)	8 (100.0)
Yes	0 (0.0)	3 (100.0)
Ever tested for HIV before
No	5 (38.5)	8 (61.5)
Yes	7 (70.0)	3 (30.0)
Drugs for pleasure
No	12 (70.6)	5 (29.4)
Yes	0 (0.0)	6 (100.0)
Drank alcohol before sex
None	9 (64.3)	5 (35.7)
About every day	0 (0.0)	2 (100.0)
At least once a week	3 (42.9)	4 (57.1)
Exchange of goods for sex
No	9 (56.3)	7 (43.7)
Yes	3 (42.9)	4 (57.1)
Age (years)	22.0 ± 1.6 (19–25)	22.5 ± 1.1 (19–25)
Total	12	11
Note: Data presented as n (%) or mean ±  standard deviation (range).

Prior to the qualitative survey participation, seven students reported never having used condoms at their last sexual encounter (four female and three male), 13 reported having had coerced sex (seven female and five male), and only eight HIV-negative males reported to have been circumcised. Out of those who responded on the question of the HIV test, more than one-half (n = 13) had never been tested for HIV before the study (five female and eight male). Seven students also reported having exchanged goods for sex (three female and four male). On drugs and alcohol, only six males reported having taken drugs for pleasure and less than one-half of the students interviewed reported having drunk alcohol before sex (three female and four male). Themes from students' unplanned sex experiences were interpreted based on the social cognitive theory that interlinks unplanned sex to personal factors and environmental factors (Bandura et al., 2001; Cooney, 2010) (Figure 1).

Figure 1 Conceptual framework of factors other than alcohol use associated with unplanned sex among university students in Kampala. Note: Adapted from Bandura et al. (2001) and Cooney (2010).

The emerging themes in this qualitative study were: minimal parental or guardian supervision; sex for pocket money, academic assistance, or course grades; peer pressure and excitement; sex for prestige or recognition; desire to maintain relationships and commitment; and suspicion of infidelity and revenge sex.

Minimal parental or guardian supervision

Minimal parental/guardian supervision refers to situations in which students' movements in and out of the university and social networks are not controlled by parents or guardians. Increased mobility as well as new and large social networks at universities provides opportunities for HIV/STI-related risk situations and behaviours. Some female students reported increased vulnerability in certain social and physical environments affecting their personal decision-making process, which could have been averted with parental or guardian supervision. Environments mentioned are guest houses, boys' rooms, sport galas and isolated recreation lakeside resorts (beaches). Students' miscalculation of personal factors exposed them to risks of unintended sex in exchange for favours like food and drinks or sometimes as reciprocity for such offers. Such situations may facilitate sexual encounters fitting the term ‘date rape’, although such expressions were not used by the interviewees:

This guy convinced me to pay him a visit. He gave me chips and drinks. Then he touched me all over without my consent, and I had sex unwillingly. (Fiona, 21 years old)

Other physical environments offering entertainment such as dance clubs, drinking halls and other social functions within university campuses were reported to be a factor for adventure sex that would occur spontaneously:

I went to a party and got attracted to someone, I surprisingly have sex with this guy I did not know. (Grace, 19 years old)

University students come from different parts of Uganda with varying socio-economic backgrounds. Moving into a student community usually means being in a new environment with its own different demands and standards of living:

You know coming from a restricted environment in secondary school to a free one at university, means enjoying it to the maximum. There is a girl I know in university from a rural area and they only give her 10,000 shillings [about 4 US dollars] per semester, so for her if she sees someone who is driving [a car], she is easily taken. (John, 22 years old)

Hostel location and the number of occupants in a room may encourage convenience sex and proof of masculinity by male students. This would be a push factor for unanticipated sex, especially for female students as narrated:

When you live alone, some boys come knocking at your door, and if you allow them in, they use all means to have sex with you even when you are not interested. (Shakira, 23 years old)

Sex for pocket money, academic assistance, or course grades

Sex for pocket money, academic assistance, and course grades refers to situations where students have unplanned sex to receive good grades or solve financial problems. Academic problems or the desire for academic success may impede their ability to avoid situations leading to sex; in such situations it can be difficult for students to decline their colleagues' sexual advances:

Normally after formal [academic course] discussions, we arrange to meet in their rooms for further consultation. The needs to perform well in class make[s] us have sex. (Anne, 20 years old)

Related to above experiences, some female students pointed out the desire for good grades as a motivation to accept sexual advances from their lecturers, usually much older than themselves:

When you want to pass exams and you are weak in a course, you have no choice but to have sex. (Jennifer, 22 years old)

Unequal power relations between students and their lecturers may affect their judgment and decision-making. University lecturers exercise great authority over students, and their pivotal role in determining final grades sometimes leads students to offer or provide sex in expectation of better examination results:

Lecturers would deny you grades until you give them sex in exchange of passing exams, and because of fear, you give in … (Catherine, 21 years old)

Other students described a need to improve their living standards or meet such expectations among their peers as leading to unplanned sexual relations with strangers, oftentimes significantly older than themselves:

If he gives you 300,000/ = ($150), or buys you a television set, or refrigerator, why not give him sex in return as a token of appreciation. (Jacky, 19 years old)

Peer pressure and excitement

Peer pressure here refers to situations where students engage in risk behaviours due to influence of friends or a group of individuals with whom they share common attributes. These friends and colleagues would behave and act in similar ways, especially in social activities and functions, affecting personal perceptions of other group members. Social events such as sports galas and parties were enabling environment for peer influence:

We meet so many friends especially when we go to beaches and attend sports, having sex here comes unplanned and appears to be fun for everyone. (Farida, 22 years old)

Friends and colleagues were reported to have had influence on students' decisions towards unintended sex. University students reported to belong to groups each with specific expectations of its members; for example, a trend or a fashion in dressing or the type of boyfriend one should get. As a student narrated:

Some friends will encourage you to go into an intended relationship because the person is attractive or famous. (Agnes, 20 years old)

When students, especially females, visit their friends at their families' homes, sex might arise due to their friends' planned set-up:

My female colleague set me up without my knowledge [with] her brother, and he forced me to have sex with him. (Joyce, 23 years old)

Students also described the role of peers in influencing unplanned sex behaviour. For example, a female student would tempt a male colleague in compromising situations, as described by one student:

A male classmate who performs well in class can make you look for all opportunities to have sex at all cost. (Candy, 20 years old)

Sex for prestige or recognition

Admission to university in Uganda is prestigious and may improve their social standing and facilitate relations with individuals outside the university. However, it may also provide a seductive environment for premarital sex, as mentioned by one student:

In my village we are very few at the university so girls would struggle to have you as a boy friend, even when you are not prepared you will have sex any time. (John, 21 years old)

At times students aspire for power and recognition on campus, which may lead to otherwise unintended risk-taking in order to gain financial advantages or votes:

I know of a female candidate at the university, who was forced to offer sex for votes; if they cannot use their bodies they will not win elections. (Mike, 22 years old)

Desire to maintain emotional relationships and commitment

The desire to maintain emotional relationships refers to sexual interactions between students engaged in steady or casual relationships, and as such may include unprotected sex at the demand of one partner. Unplanned sex set the stage for continued unprotected condom use, as narrated by one student:

At first it happened when we were not aware and we did not stop, and we had sex without condoms. We were playing and we did not know that it would happen. Since then we have never thought of using condoms again. (Anthony, 22 years old)

In other relationships, non-condom use may be a sign of commitment:

The guy would tell you to give him sex as a sign of love mostly without condoms. (Pamela, 19 years old)

Students who watched adult films were sexually aroused, especially if they watched with female colleagues. Student's experiences were influenced by interaction of personal, environmental and behavioural factors:

I think watching pornographic movies with a female may result into having unplanned sex. (Peter, 19 years old)

Suspicion of infidelity and revenge sex

Strained partner relations influenced by gender power relations were associated with students seeking ‘outside’ sex; as a way of ‘stress management’, male students would search for ‘a comfort zone’ in other relationships:

Sometimes it is just stress from a nagging partner, when you have misunderstanding with your partner, you ask yourself, should I really have one partner? You end up having sex with people you did not plan to relate with sexually. (Moses, 24 years old)

Students expressed breakdown in communication and misunderstandings in sexual relationship, which led to suspicion of infidelity and therefore revenge sex:

I had unprotected sex with a prostitute twice because my partner did not respond to my telephone call. (Chris, 21 years old)

Discussion

In our analysis of qualitative interviews with students of five major universities in Kampala, we identified six major themes related to unplanned sex: minimal parental or guardian supervision; sex for pocket money, academic assistance, or course grades; peer pressure and excitement; sex for prestige or recognition; desire to maintain relationships and commitment; and suspicion of infidelity and revenge sex. The qualitative results on risk factors related to unplanned sex concurred with the quantitative findings of the same group in the major study.

Our study adds to an understanding of unplanned sex in highly educated emerging adults and may help inform programmes to mitigate HIV/STI and pregnancy risk in this population.

University students seem to have different perceptions of circumstances for unplanned sex, and alcohol was not the only factor associated with this behaviour as mentioned elsewhere (Kabiru & Orpinas, 2009; Ma et al., 2007; Mongkuo, Mushi, & Thomas, 2010). Therefore, to understand students' risk behaviours and perceptions, we analysed emerging interactions from students' experiences using social cognitive theory where it was apparent that students' personal factors were inter-linked to environmental factors (Bandura et al., 2001) within and outside the university campuses. A theoretical analysis presents a better explanation of students' behaviour (Mongkuo et al., 2010).

In this qualitative study we found that unplanned sex may be linked to personal and environmental factors that together influence HIV/STI risk behaviour in this group. In addition, students' perception of unplanned sex in other contexts may be portrayed as coercion, date rape, adventure sex, or transactional sex (sex in exchange for academic help or other goods and services), which collaborated with other studies elsewhere (Wamoyi et al., 2010). It appears in some circumstances that unplanned sex was not necessarily for personal gain but a gesture of affection expected from men to women (Moore et al., 2007). We also found that their experiences met our operational definition of unplanned sex.

We examined personal factors including financial constraints, desire for academic success and maintenance of sexual relationships. Furthermore, psychosocial needs and partner relations may modify students' behaviours and increased the risk of unplanned sex. In comparison with other studies in Africa that emphasised alcohol use, we found social environmental factors, such as a lack of parental/guardian supervision, emotional distress and peer pressure, may be associated with risk behaviour (Brown, Schultz, & Parsons, 2000). In addition, environmental factors may make university students as emerging adults have a bicultural behaviour (Schwartz et al., 2007). Change of the living environment from home to university and a sense of independence make emerging adults more independent and vulnerable (Arnett, 2003).

A sense of belonging to the university community may have influence on students' sexual behaviours. Our observations are in agreement with a similar study in Uganda (Nobelius et al., 2010) where the presentation or acceptance of goods or gifts was associated with transactional sex (Hingson et al., 2003; Wamoyi et al., 2010). Unplanned sex may also serve to escape certain situations, illustrating the interaction between personal and environmental factors on students' behaviours.

A number of studies describe alcohol as an important risk factor for unplanned sex in young people (Hingson et al., 2003; Wamoyi et al., 2010). Our study's qualitative findings suggest additional factors may be at play as well. Our interviews also suggest that female students may be more vulnerable to unplanned sex than their male counterparts and that peer influence in the context of social events may be an important moderator as well, which has also been seen in other studies (Kabiru & Orpinas, 2009; Mongkuo et al., 2010; Zamboanga, Raffaelli, & Horton, 2006) – this may be as a result of gender power relationships. Furthermore, our qualitative findings related to condom use among students corroborate empirical observations from elsewhere in Africa (Atwoli, Mungla, & Ndung'u, 2011; Wamoyi et al., 2010).

Although unplanned sex is associated with unplanned pregnancy and HIV/STIs, intervention programmes have focused on modifying behavioural factors such as alcohol, drug and condom use (Atwoli et al., 2011; Hingson et al., 2003; Zamboanga et al., 2006). The availability and use of condoms are instrumental for having safe sex (Yotebieng, Halpern, Mitchell, & Adimora, 2009).

This study has some limitations. The design did not allow us to determine whether the identified themes are inter-related or merely demonstrate variability in behaviour. Further, our results cannot be generalised to Ugandan youth as our sample represents a highly educated demographic group. Although we only sampled students from five Kampala universities, students at these campuses have diverse geographic backgrounds. Our group of 23 interviewees may also not be representative of the larger sample of 644 students that participated in the survey as we purposively selected them based on specific answers to a range of quantitative interview questions. Although we tape-recorded interviews face to face in complete privacy, we cannot rule out that some interviewees may have been influenced due to prevailing social norms. Despite these limitations, our qualitative analysis of unplanned sex provides insights into Kampala university students' sexual behaviours and describes factors facilitating unplanned sex.

Unplanned sex may play an important role in HIV/STI transmission as an indicator of students' vulnerability (Urassa et al., 2008). It is therefore important to focus prevention efforts on factors promoting unplanned sex (Ma et al., 2007) and to design gender specific messages in this group. In addition, parent-supported or guardian-supported youth-friendly counselling services to emphasise gender equality and individual responsibility towards unplanned sex, and risk-reduction strategies, may provide additional benefits towards the prevention of unplanned sex. HIV programmes may want to consider these findings when addressing risk environments on and around student campuses.

Acknowledgements

The authors thank the Crane ∼ Survey staff and university students in Kampala for their time and efforts. They also thank David Katuntu, Sylvia Nakayiwa and Theo Wamala for their contribution. The Crane Survey Group included Wolfgang Hladik, Joseph Barker, David Serwadda, George Lubwama, Danstan Bagenda, Rachel King, Alex Opio, Tom Tenywa, Edith Nakku-Joloba, Michael Muyonga, John Ssenkusu, Avi Hakim, Kimberley Dills, Sylvia Nakayiwa, David Katuntu, Frank Kaharuza, and Jordan Tappero.

All authors contributed to the substudy design, study conduct, data analysis and interpretation, or manuscript writing, and approved the final version for publication. Funding was received from the US Government President's Emergency Plan for AIDS Relief. The findings and conclusions in this manuscript are those of the authors and do not necessarily represent the views of the US Centers for Disease Control and Prevention or the US Department of Health and Human Services.