Journal of HIV/AIDS and Infectious Diseases
Research Article

Innovative Approaches for Preventing HIV among Adolescent Girls in Uganda: Evidence from an Evaluation Study on Anti Cross Generation Sex in Secondary Schools

Received Date: January 28, 2014 Accepted Date: February 25, 2014 Published Date: April 16, 2014


Citation: Simon Sensalire , et al. (2014) Innovative Approaches for Preventing HIV among Adolescent Girls in Uganda: Evidence from an Evaluation Study on Anti Cross Generation Sex in Secondary Schools. J HIV AIDS Infect Dis2: 1-04


In Uganda, Cross Generational Sex (CGS) accounts for the disparity in levels of infection between men and women of young age. MoH and Macro[1,2] show that 7% of women aged 15-19 who had high risk sex in the 12 months preceding the survey had higher risk sex with a partner who is 10 or more years older in the past 12 months. Thus, there is a disproportionate gap in HIV prevalence between women (4%) and men (1%) of the same age group. It is widely believed that CGS is to some extent the cause of this disparity.


We conducted a longitudinal survey among secondary school girls aged 15-19 before and after anti CGS interventions. The sample was drawn from 50 schools under which a program of anti-CGS was implemented by PACE in the past 12 months within the most urban districts of Uganda of Kampala, Wakiso, Mukono and Mbarara. The sample was drawn from schools which were High Fees Schools (HFS) and Low Fees Schools (LFS). The two categories reflect the social economic background which is the driving factor for CGS in Uganda.


CGS was generally higher among LFS (10.9%) as compared to HFS (4.8%) even after similar anti CGS interventions


HIV programs should put priority to girls in LFS because they are more vulnerable to CGS as a result of their low economic status background.


Studies in different parts of Sub Saharan Africa indicates that young women aged 15-24 are three times more likely to be infected with HIV than males of the same age. The disparity in levels of HIV infection especially in countries in Sub Saharan Africa is of great concern. One explanation for this disparity in infections is age mixing in sexual relationships between older men and young women. Older men have higher rates of HIV than young men and the relationships with older men limit young women's power to negotiate safer sex, particularly since there is often exchange of money or gifts for sex. In addition, the socio-economic and power imbalances inherent in cross generational and transactional sexual relationships put young women at high risk of unintended pregnancy and sexually transmitted infections including HIV [3].

Cross generational sex among young girls is driven by the need to fulfill wants like lipstick, handbags, nail polish. It is sometimes, motivated by the hope to get married to a good, already reliable and stable person who in most economic situations of Africa are characteristic of men of higher ages.

In Uganda, the association between cross-generational sex, unsafe behaviors and HIV risk makes the phenomenon a priority concern. The data clearly indicate that in much of Uganda, young women bear the brunt of the AIDS epidemic. HIV prevalence among young women aged 15-19 is 4.8% compared to 2.3% among men [4]. In the 20-24 age groups, women's prevalence was 6.3% compared to 2.4% among men. Prevalence among men above 30 years or more peaks at 9.3%. It is widely believed that sex among young women (15-24) and men who are ten years or older is to some extent the cause of this disparity as older men infect the younger women [1]. The Uganda Demographic Health Survey (UDHS) 2006, showed that one in ten had sex with a man 10 or more years older [1] among the unmarried women aged 15-19 who had nonmarital sex in the last 12 months. Number of factors account for this phenomenon. According to Ntozi JP, Najjumba IM, et al. [5], the economic conditions of most families have affected the potential of parents to meet the growing demands of their children in a competitive environment. Thus, the socioeconomic imbalances put young women in situations of sexual relationships with men perceived to be economically stable. The economic value of sexuality is particularly pronounced for young women from poor families with less access to pocket money and other needs necessary to match the standards of their counterparts from well off families.

Secondly, older men appear to prefer young women and associate themselves with young educated ladies as sexual partners, partly because they are believed to be free of HIV/ AIDS infection [1]. Individual beliefs about cross generational sex may also be driving this behavior. These may include perceived social norms, social support, self efficacy, knowledge, susceptibility, severity, attitudes and subjective norms.

Today much is known about the dangers of cross generational sex, yet effects of existing interventions against cross generational sex remain largely unknown. Given the disproportionate rates of HIV among young women in Uganda, continued efforts are needed to better understand that the effect of interventions against cross generational sex and develop evidence based approaches to address it.

The overall aim of this study was to determine how interventions for cross generational sex influenced a change of behavior among young girls in secondary schools. The study sought to test the hypotheses that young women from low economic status families are more likely to engage in cross generational sex than young women from high economic status families and young women who are exposed to media campaigns against cross generational sex are less likely to engage in cross generational sex.

Description of the Intervention

PACE launched a three year BCC campaign project among female university students with the goal of contributing to a reduction in the incidence of HIV among girls aged 15-19 years in secondary schools. Specifically, the project aimed at i) reducing the number of unmarried female students engaging in cross generational sex by increasing risk perception of the dangers associated with cross generational sex ii) reducing the target group's perception that having a sugar daddy is normal or acceptable iii) increasing the girls skills and confidence to reject advances from older men iv) increasing the perception that there is emotional support from friends for rejecting cross generational sex v) increasing the perception of the negative outcomes for engaging in cross generational sex.

The project theorized that we would attain the above behavioral and outcome objectives using two strategies namely Interpersonal Communication (IPC) and Mass Media. The project also targeted community leaders as advocates and older men as perpetrators. Key messages communicated through the behavior change communication campaign where encouraging girls to value long-term prospects over short term gain asking community leaders to speak out against the practice to members in the community and stigmatizing the practice among older men.

The IPC activities included the use of peer education, peer social support clubs called 'Go-Getters' and role model presentations. These three were designed to reinforce each other. Peer education was conducted by a group of young girls in schools who were volunteers. The 'Go-Getter' clubs were formed to promote discussion of cross generational sex and its ills. These sessions were supported by PACE with discussion guides and the peer educators were trained to facilitate meetings. A 'Go-getter' concept was developed by PACE as a tool to portray the model student (who does not engage in CGS) as focused and determined to succeed without the undue advantage offered by cross generational relations (incomes).

Female role models who had been successful in the professional and personal lives were asked to present and discuss their personal lives and experiences with a group of girls at the university. The role model presentations included discussions about goal/target setting and implementation while taking into account challenges and opportunities. A total of 80 role model presentations were made. The numbers attending the presentations were ranged from 100 to 250 per session. The role model presentations aimed to reinforce the girls self confidence to reject cross generational sex and reinforce the message that short term gains from the old man's relations are associated with increased risk of HIV. The project also used mass media such as radio, TV, posters, billboards. Radio spots targeted both the young women and the old men.

Radio spots were aired with the following themes: The gifts can never be worth your lives and future. Respect yourselves. Say NO to Sugar Daddies; Men before you think about manipulating someone's daughter for their own selfish needs, just remember that someone else could be doing the same to you; Take a stand, Cross Generational Sex stops with you.

Two TV spots were produced in July 2007 and were aired for a period of 3 months. A total of 320 television Adverts were aired. The messages were 1) encouraging the girls to focus on the long-term as opposed to the short term gains 2)cross generational sex stops with you. Posters, billboards and T-shirts were produced with the following slogans 1) Say no to sugar daddies, cross generational sex stops with you (Billboard) ; 2) you might want these material's benefits but do you need HIV (Billboard /poster) ; 3) cross generational sex stops with you (T-shirts); 4) Go-Getters . am going places (Tshirts); 5) Would you let this man be with your teenage daughter, so why are you with him? (Bill board/Poster); 6) everybody has a role to play; Cross Generational Sex stops with you (Bill board). The billboards were placed at entrances to the universities while posters were placed within the university. T-shirts were given out to both males and females and were to act as 'mobile' posters.

Study design

A longitudinal study design was used to collect information on cross generational sex behavior among young women aged 15-19 for one and a half years. The design was appropriate to observe changes in sexual behavior among girls in high fees schools and low fees schools during the intervention period. Baseline information was collected among 2,410 girls and deliberate effort was made to interview (2387) at follow up from 50 schools in the most urban districts of Kampala, Wakiso, Mukono and Mbarara. The 2387 young women at follow up were a purposive sample and a subset of the baseline. At baseline, the selection was random in each school until the desired number was achieved. The baseline sample was determined Kish and Lesley and factored in the proportion of the student population in non candidate classes and the prevalence of cross generational sex.

The schools were categorized as Low Fees Schools (LFS) and High Fees Schools (HFS) which is a reflection of the economic background of young girls and a number of students who enroll in university after completion of secondary school Demographic characteristics such as age, were used as control variables to ensure the populations are comparable.

Data collection

We conducted both baseline and follow up studies in which respondents completed a self administered questionnaires at a central place within the school premises under the supervision of the research assistants. The duration of the interview is 1 hour an average. Prior to the completion of the questionnaire, the research assistants first elicited consent from the respondents and also explained thoroughly how the questionnaire should be filled in and the former were always present to clarify and guide the respondents in completing the interviews and clarify and answer any questions that were raised. The research assistants then edited the questionnaire while still at school with the respondent. Editing was done only to clarify responses and skip patterns and not modify or alter respondent responses. All ethical procedures were followed before data collection. Upon ethical approval from the IRB, (Makerere University School of Public Health), consent of parents was obtained from the respective schools of all respondents under 18 years. Students were informed that choice was primary consideration and there was no influence of school administration.

Statistical Analysis

We conducted reliability and factor analysis tests to determine, if the multi-items were unidimensional measures of a construct and were reliable at Cronbachs alpha >0.70. Thereafter, composite scores (means) were created. Individual scale items were used to represent construct in cases where the multi- item scale was not reliable. These were selected after a factor analysis was done to determine which items 'loaded' more on the selected factors as per the extracted components. Items were used to assess whether or not social norms, social support, self efficacy, outcome expectations influenced change in sexual behavior of young girls. Bivariate correlations, to ascertain multicollinearity were done for all opportunities, ability and motivation items/scales during analysis. In multivariable models, we included all variables associated with self-reported sexual behavior at p0. 05 into the logistic regression model to determine the drivers of cross generational sex in two categories of young girls in secondary schools. All analyses were conducted using SPSS and study was approved by an ethical review board at the School of Public Health, Makerere University in Uganda.

Reported changes in cross generation sex

Respondents (students) were asked whether they have ever had sex with a partner 10 years older or whether on any occasion in the past 12 months, were engaged in cross generational sex. We detected an increase in the proportion of students who have ever had sex with a man 10 years older or more in the two years (p0. 05).

Cross generational sex by school category

We detected significant difference in levels of cross generation sex practice in the two categories of schools. Most young girls in LFS practiced cross generational sex (10.9%) compared to HFS (4.8%).

Determinants of cross generational sex among young girls between the intervention period

Self efficacy: This measures the ability and confidence of a girl to resist advances from older men. We noted a significant increase (p< 0.001) in the number of young girls, who perceive that they have the ability and confidence to resist advances and gifts from older men.

Social norms: Young girls who think it's not common for girls to have older sexual partners less reported cross generational sex than girls who think it's normal for young girls to have sex with older men.

Susceptibility: Young girls who held that engaging in a sexual relationship with an older man puts one at high risk of contracting HIV were more likely not to engage in cross generational sex behavior (p=0.000)

Negative outcome expectations: Girls who reported that cross generational sex is associated with negative outcomes were more likely not to engage in cross generational sex. This was also a reflection of the high perceived severity of HIV among older men.

'Life skills' expressed 'how confident one could refuse to have sex with an older man'. More girls felt confident in refusing to have sex with an older man in 2011 as compared to that in 2010. The variation in the proportions of the risk group in the two years is highly significant.

Author Contributions

Conceived, designed and implemented the study, analyzed the data and in charge of all quality control and quality assurance.

1MoH, ORC Macro (2006) Uganda HIV/AIDS Sero-Behavioral Survey 2004- 2005. Maryland, USA. Centers for Disease Control and Prevention (U) Entebbe. U.S. Agency for International Development.
2UBOS and Macro International (2011) Uganda Demographic and Health Survey 2006. Calverton, Maryland, USA
3WHO (2008) The State of the AIDS Epidemic. Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO), Geneva, Switzerland.
5Ntozi JP, Najjumba IM, Ahimbisibwe F, Ayiga N, Odwee J (2003) Has the HIV/AIDS epidemic changed sexual behaviour of high risk groups in Uganda? Afr Health Sci 3:107-116.