The Relationships Between Stress, Psychosocial Resources, and Mental Health and Adherence Outcomes among Perinatally HIV-Infected Adolescents in South Africa

Adolescents living with perinatally acquired HIV (APHs) in sub-Saharan Africa (SSA) constitute a significant population group that is experiencing poor HIV treatment outcomes (CIPHER Global Cohort Collaboration, 2018). Compared to younger children and older adults within the SSA sub-region, APHs experience poorer health outcomes, including retention in care, virologic treatment failure, and mortality rates (Anderson et al., 2019). Among countries in the SSA sub-region, South Africa has 360,000 adolescents living with HIV (ages 10-19), a disproportionate burden, accounting for 20% of the global adolescents living with HIV (UNAIDS, 2019a). The public health costs of suboptimal antiretroviral treatment (ART) adherence in adolescence are huge in terms of quality of life, life itself, and monetary value. Suboptimal adherence often results in high hospitalization costs for families and public health systems (Cluver et al., 2016; Kalichman et al., 2009; Paterson et al., 2001). Additionally, it denies APHs a good quality of life by placing them at high risk of disease-related complications such as cardiac, respiratory, cognitive, and chronic renal impairment, which often results in deaths (Lowenthal et al., 2014; Sainz et al., 2015; Rylance et al., 2016). Therefore, utilizing the Transactional Model of Stress and Coping and the Social Support theory (Glanz & Schwartz, 2008; Lazarus & Folkman, 1984; Berkman et al., 2000), three research questions were posed in this study. The first research question asked: To what extent are HIV-related stressors (i.e., HIV-related stigma on Caregiver, HIV-related stigma on APH, and caregiver report on the burden of child\u27s illness on the family) significantly associated with psychological functioning (i.e., APH report on depressive symptoms, anxiety, and of self-esteem) and ART adherence (i.e., APH and caregiver report) among APHs? The second research question asked: To what extent are psychosocial resources (i.e., HIV knowledge and self-efficacy, informal social support, and formal social support) associated with psychological functioning and HIV medication adherence (i.e., APH and caregiver report) among APHs? Finally, the third research question was exploratory and addressed: To what extent do psychosocial resources (i.e., positive parenting and the ability to cope with adversity) moderate the relationship between HIV-related stigma (i.e., both APH and caregiver) on APH and psychological functioning and medication adherence (i.e., APH and caregiver report) among APHs? This cross-sectional study used baseline data from the VUKA Family Program, an intervention study conducted with a convenience sample of 315 perinatally HIV-infected adolescents, ages 9 to 15, and their caregivers (the parent study) in KwaZulu Natal, South Africa, between 2014 and 2019. A preliminary analysis of the raw data (N=315) was conducted to evaluate the extent of missing data on all independent and dependent variables. Two dependent variables (i.e., anxiety and depressive symptoms) and five independent variables had over 5% of missing observations. Further analyses were conducted to establish the reasons for missing data and if the assumption of missing at random (MAR) is reasonable. For questions 1 and 2, multiple imputations by chained equations (MICE), an approach to creating datasets based on a set of imputed models, with one model for each of the variables (including covariates) with missing data, were performed in STATA (Schafer & Graham, 2002; White et al., 2011). For question 3, a hybrid approach to MICE was performed for each of the 28 regression models with an interaction term (Royston, 2005; White et al.,2011). Data analysis for the first and second research questions involved descriptive and bivariate analyses. Using the multiple imputed data, twenty-one regression models were performed for the first research question to identify significant risk factors (HIV-related stressors) associated with psychological functioning and HIV medication adherence. Fifty-six regression models were performed for the second research question to identify protective factors associated with psychological functioning and HIV medication adherence using the multiple imputed data. Ordinary Least Squares or Negative Binomial regression models were performed depending on the nature and distribution of the outcome variable. For the exploratory third research question, 28 regression models with an interaction term were performed to test the potential moderating influences of psychosocial resources (i.e., ability to cope with adversity and positive parenting) on the relationship between stressors and mediation adherence and psychological functioning. Results from the univariate analysis found that the adolescents\u27 mean medication adherence prevalence rates using the APH and caregiver report on the medication adherence scale were 79% and 83 %, respectively. Additionally, based on APH and caregiver reports on the number of missed doses, the adolescents\u27 medication adherence rates were 76% and 80 %, respectively. Regarding levels of psychological functioning, 49% of the APHs had moderate to high levels of anxiety; 96% of the APHs were in the elevated depressive symptoms category, and 98% of the sample were in the high to very high self-esteem category. Results identified caregiver and adolescent HIV-related stigma as the significant risk factors for inadequate psychological functioning among APHs. Adolescent HIV-related stigma was found to be significantly associated with high levels of anxiety and depressive symptoms and low levels of self-esteem. In addition, caregiver and adolescent HIV-related stigma were found to be significantly associated with high levels of anxiety. The burden of child HIV illness on the family was not significantly associated with anxiety symptoms. However, none of the three HIV-related stressors were significantly associated with medication adherence measures. Five psychosocial resources were found to be significantly associated with psychological functioning measures among APHs. More specifically, HIV knowledge, ability to cope with adversity, and positive parenting were significantly associated with high levels of self-esteem. The ability to cope with adversity was significantly associated with low anxiety symptoms. Medication management autonomy was significantly associated with low levels of depressive symptoms. Support group attendance was significantly associated with low levels of self-esteem, and HIV knowledge was significantly associated with high levels of depressive symptoms. However, none of the psychosocial resource variables were significant protective factors for medication adherence. The ability to cope with adversity was found to have statistically significant moderating effects on two regression models with an interaction term. More specifically, the ability to cope with adversity has a statistically significant moderating effect on the relationship between adolescent HIV-related stigma and psychological functioning among APHs. Additionally, the ability to cope with adversity has a statistically significant moderating effect on the relationship between caregiver HIV-related stigma and psychological functioning among APHs. Results found no statistically moderating effects of positive parenting on psychological functioning and medication adherence. There were no statistically moderating effects of the ability to cope with adversity and positive parenting on medication adherence either. This study contributed to knowledge about adolescents living with a perinatal HIV infection, their levels of psychological functioning, medication adherence, the adverse effects of HIV-related stressors, and their psychosocial resources and needs. This is the first study to test whether coping with adversity and positive parenting was significantly associated with psychosocial functioning and ART adherence. Furthermore, this study identified the ability to cope with adversity as a significant buffer against the adverse effects of adolescent and caregiver HIV stigma on psychosocial functioning. Study findings have implications for research that seeks to inform policies and practices that enhance psychological functioning, reduce HIV stigma, and provide for the psychosocial needs of adolescents living with a perinatal HIV infection while receiving treatment and care in publicly funded urban clinics of South Africa and sub-Saharan Africa