Predictors of early sexual debut amongst youth in the inner city of Johannesburg, South Africa

Abstract

Introduction: South African adolescents and youth face multiple sexual and reproductive health (SRH) challenges, including HIV, teenage pregnancy and sexually transmitted infections (STIs). Early sexual debut (ESD), defined as having had first sexual intercourse at age 14 years or younger, is associated with increased risk of HIV, STIs, pregnancy, and risky sexual behaviours such as low condom use and multiple partners later in life. Despite being a focus of SRH and HIV prevention programming, national surveys have reported an increase in ESD over the last nine years. The aim of this study was to determine the prevalence and factors associated with ESD among youth (aged 15 –24 years) in the inner city of Johannesburg. The first objective was to describe the population of youth in the inner city with ESD. The second objective was to determine the prevalence of ESD among this population and the third objective was to determine the factors associated with ESD among this population. Methods: This study was a cross sectional secondary analysis of data collected for the Linking Information for Testing (LIFT) study. The primary study was a cross-sectional, respondent-driven, community survey among youth aged 15 –24 years in the inner city of Johannesburg. A descriptive analysis of the study population was undertaken. The overall prevalence of ESD was determined by tabulating experience of ESD by the explanatory variables. The factors associated with ESD were explored through inferential analysis using a univariable and then multivariable binary logistic regression analysis. Results: There were 797 participants included in the analysis. The prevalence of ESD in the study population was 15,5% with significantly more males (27,2%) experiencing ESD compared to females (1,5%). In the adjusted analysis, being aged 15 –19 years, of male sex, of a racial group other than Black African, having a male caregiver before the age of 15 years, and having a secure household food source were found to be significantly associated with an increased odds of ESD. The level of education completed, employment status, type of primary female caregiver, presence of mental health symptoms, perception of community safety, presence of peer support and whether the first sex was wanted or not were not found to be significantly associated with ESD. Discussion: The results of this study confirm the prevalence of ESD among youth in Johannesburg at around 15%. In keeping with other local literature, males and those aged 15 –19 years had a greater odds of ESD. Although there was no significant association, the findings are suggestive that enrolment in school continues to have a protective benefit on ESD. The study results indicate that having no primary male caregiver was protective of ESD, highlighting that adolescents from stable, two-parent families may not necessarily be at lower risk and that there may be relevant factors beyond family structure which were not explored in this study. The finding that those with a secure household food source had an increased odds of experiencing ESD compared to those with food insecurity is contrary to generally established hypotheses and warrants further investigation. This finding should be interpreted with caution given that no temporal relationship could be established. A limitation of this study was the inability to conduct a sex stratified analysis. Conclusion and recommendations: The prevalence of ESD among youth in Johannesburg requires it to remain a priority for both research and health programming. A unique finding of this study was the protective effect of having no male caregiver on ESD, a finding that warrants further research to better understand the role of caregivers in ESD. Nationally representative, longitudinal adolescent and youth data is required to better evaluate the multiple, complex factors contributing to ESD and to provide guidance to inform specific SRH interventions relevant to the South African context