Adversity, psychological distress and sexual risk taking amongst 15-26 year olds in the Eastern Cape, South Africa |
Background: the subject of the mental health of young people in Sub-Saharan Africa has received very little research attention. Despite the fact that many in this region face ubiquitous material deprivation, childhood adversity, violence, AIDS and orphanhood which could result in distress, there is very little scientific understanding of the precursors of psychological distress. Understanding and intervening in young people is important as psychological distress is believed to precede one of the most common health problems facing young people, that is, risky sexual behaviours. This thesis aims to document and describe the prevalence of depressive symptoms in a sample of young people, develop an explanatory model for factors associated with depressive symptoms and distress, and explore links with risky sexual behaviours. Method: This thesis is based on two studies that were both conducted in the Eastern Cape Province of South Africa. Study 1 was an analysis of cross-sectional and longitudinal data collected amongst young people aged 15 to 26 who were enrolled in a cluster randomised controlled trial evaluating an HIV prevention intervention. The study sample was drawn from volunteers who normally resided in villages and townships in and around the area within 1.5 hours drive radius of Mthatha. Volunteers were mostly recruited through schools. In each of the seventy study sites, approximately twenty males and twenty females were included resulting in the enrolment of 2,801 volunteer participants. Quantitative data were collected by a fieldworker-administered questionnaire in 2003/4 and 2004/5, and analysed separately for men and women, using STATA IC 11.0. A crosssectional analysis of 2003/4 data was undertaken to investigate factors associated with depressive symptoms. Presence of depressive symptoms was established through a self report measure - the Centre for Epidemiological Studies on Depression Scale. Participants who scored above a cut-off point of 16 were considered to have depressive symptoms. Results for factors associated with the presence of depressive symptoms are presented in Paper I. The association between baseline depressive symptoms and sexual risk taking at baseline (2003/4) and twelve months later (2004/5) was explored. Here, participants with depressive symptoms were treated as an exposed group and were compared to those who scored below 16 on the symptom checklist within the same cohort. These results are presented in Paper II. In Study 2, phenomenological experiences of distress were explored. In-depth face-toface, one-on-one interviews with forty young people were conducted in isiXhosa between 2007 and 2008. Participants for Study 2 were recruited in Butterworth from the townships. The sampling approach was a non-random, non-probability approach, with participants who self-selected and volunteered on the basis of their interest in the research topic. The sample came from a town that was included in Study 1 and generally shared similar socio economic conditions, cultural experiences, and history as the other community sites that participated in Study 1. Criteria for inclusion into the study included having no known clinical diagnosis of a mental health problem. Participants were aged 16 to 26 years. Each interview took about forty five minutes to an hour. Data were analysed using the constant comparison approach and following recommendations for qualitative phenomenological studies. Findings from this study are presented in two papers. Paper III of this thesis presents a conceptual model based on discussions of perceived sources of distress by participants. Paper IV discusses the role of undisclosed paternal identity in causing distress. Findings: In Study 1, the prevalence of self-reported depressive symptoms was 21% among women and 14% among men. Child abuse and neglect (aOR 1.34 95% CI 1.116, 1.55), substance use (aOR 1.98 95% CI 1.17, 3.35), perceptions of less cohesion in the community (aOR 1.2395% CI 1.07, 1.40), intimate partner violence victimisation (aOR 2.2195% CI 1.16, 3.00) and sexual violence before the age of eighteen (aOR1.45 95% CI 1.02, 2.02) were associated with depressive symptoms in women. For men, factors associated with depressive symptoms were child abuse and neglect (aOR 1.61 95% CI 1.38, 1.88), having lost a mother (aOR 2.24 95% CI 1.25, 4.00), alcohol abuse (aOR 1.63 CI 1.13, 2.35), having been forced by a woman to have sex (aOR 2.36 95% CI 1.47, 3.80) and conflict in the current sexual relationship (aOR 1.07 95% CI 1.01, 1.12). Findings on the associations between depressive symptoms and risky sexual behaviours show that women with depressive symptoms at baseline were more likely to have dated a man five years or older than them in their lifetime (aOR 1.37 95% CI 1.03-1.83), had transactional sex (aOR 2.60 95% CI 1.37-4.92) and experienced intimate partner violence (IPV) at baseline (aOR 2.56 95% CI 1.89-3.46). Women with depressive symptoms were more likely to have experienced IPV a year later (aOR 1.67 95% CI 1.18-2.36) after adjusting for baseline IPV experiences. At baseline, in men, an association between depressive symptoms and perpetration of intimate partner violence (aOR 1.50 95% CI 0.98-2.28) and rape was evident (aOR 1.81 95% CI 1.14-2.87). Men who had depressive symptoms were also less likely to report correct condom use at last sex, at both baseline and twelve months later (aOR 0.50 95% CI 0.32-0.78 and aOR 0.60 95% CI 0.40-0.89). Study 2 showed that family-based adversity, most notably perceptions of mother’s distress, conflict over financial resources, undisclosed paternal identity and parental substance abuse caused distress in young people. A culture of silence in families on issues considered pertinent by participants, such as paternal identity appeared to intensify distress. From this study, sexual relationship problems, including intimate partner violence, an unacknowledged pregnancy, and violent transactional sexual relationship themes dominated women’s narratives of distress. In their narratives, men described violence and sexual risk taking as expressions of anger directed towards women. They described using substances and sexual philandering as ways to express distress, and as coping mechanisms, although they in turn became sources of distress. xiv Discussion, conclusions and recommendations: This research reports a high prevalence of depressive symptoms among young people in South Africa and supports international patterns of a higher prevalence in women than men. Findings presented in this thesis have important implications as they show that structural factors that cause different forms of disempowerment are implicated in some of the psychological distress experienced by young people. For instance, gender power inequity, violence, cultural expectations of respect from youth and women, compounded by financial dependence on relatives were sources of vulnerability especially in the face of maternal orphanhood. When ones mother was perceived to be under distress and honest and effective communication within families was lacking this caused distress. Depressive symptoms were associated with risky sexual behaviours commonly found among rural young people such as intimate partner violence, boys’ sexual victimisation by women, relationship conflict and involvement in transactional sex. Though HIV prevalence among young men is lower than in women in South Africa, consistent failure to use a condom at last intercourse among men with depressive symptoms may ultimately increase risk for HIV infection. Hence, sexual health youth-friendly clinics should be aware of the links between depressive symptoms and sexual expression. The strength of this thesis is in the mixed method approach to exploring psychological distress through qualitative and quantitative data. The quantitative study used a large sample and had a prospective component, which enabled the impact of depressive symptoms on sexual risk-taking to be studied in temporal sequence. This is valuable and unusual in a dataset. However, it has a volunteer sample and its findings cannot necessarily be generalised to all young South Africans. Nevertheless, there is no reason to expect the associations described to significantly differ from those that would be found in a non-volunteer sample. Qualitative research is inherently non-generalising, but the methodology used here enables a depth of understanding an exposure of nuance that is not attainable through quantitative methods. Prevention of depressive symptoms among young people in South Africa should start with interventions to reduce exposure to childhood adversity. Some of this distress could be reduced if relationship violence was prevented for women, pregnancies acknowledged and disputes resolved on time by the putative fathers. This research suggests that a reduction in young people’s depressive symptoms could have positive benefits for reduction of HIV risk taking behaviours.
Thesis (Ph.D.)--University of the Witwatersrand, Faculty of Health Sciences, 2012.