The association between substance use and HIV in two Sub-Saharan African countries, 2014 - 2016

Abstract

Background People Living With HIV/AIDS (PLWHA) are recognized to have an increased risk of substance abuse - alcohol heavy episodic drinking (HED), tobacco smoking, and illicit drugs. Substance use amongst PLWHA is associated with poorer antiretroviral treatment (ART) adherence, increased risk of cancers, and worse morbidity and mortality. Substance use is also associated with an increased risk of HIV infection amongst those who are HIV negative. The majority of this research is focused outside of the African continent and in high-income countries; thus, this study aims to contribute towards research in low-middle income African countries such as South Africa and Kenya. Aim This study aims to investigate the association between HIV and substance use in South Africa and Kenya, and the possible associated covariates. Methods The study population for this secondary data analysis included n=7919 men and women participants drawn between the ages of 40 and 60 recruited between August 2013 and August 2016. The Mann-Whitney test was used for continuous variables and all other categorical variables were analysed using a chi squared test or Fisher’s exact test. Predictive factors for the logistic regression models were informed by the literature review and conceptually modelled using a Direct Acyclic Graph. An overall logistic regression model was run which included both countries, and then separate logistic regression models were run for Kenya and South Africa. Results The prevalence of substance use in the overall sample was 54.30%. In Kenya it was 51.33% and in South Africa (SA) 55.60%. The prevalence of HIV in the overall sample was 20.94%. In Kenya HIV prevalence was 13.66%, and in SA 23.35%. PLWHA had a 1.22 greater odds of substance use than those who were HIV negative in the overall sample (p=0.012, 95%CI 1.04 - 1.43). In Kenya, PLWHA had a 1.49 higher odds of substance use (p=0.013, 95%CI 1.09 - 2.03). In SA, there was no significant association found between HIV status and substance use. In the Kenyan logistic model HIV status, sex, marital status and employment status were found to be associated with substance use. Full-time employment and marriage presented as protective factors against substance use in Kenya. In the SA logistic model age, sex, marital status and employment status were associated with substance use. Marriage was also a protective factor for the SA model. Self-employment and informal employment increased the odds of substance use by more than two-fold in the SA model (aOR=2.27, p<0.001, 95%CI 1.63-3.16; aOR= 1.50, p=0.0008, 95%CI 1.19-1.91 respectively). The factor with the largest odds for substance use across all three models was being male, with increased odds of 15.5 in the overall model (p<0.0001, 95%CI 13.4-17.9), 11.6 in Kenya (p<0.0001, 95%CI 8.65-15.5) and 17.7 in South Africa (p<0.0001, 95%CI 14.9-21.1). Conclusion Treatment of substance use disorders should become a staple in integrated routine ART care as the prevalence of substance use in PLWHA in this study and across the literature is high, especially in men The implementation of the substance use treatment should be tailored to the unique gendered and socioeconomic factors that are present in each country.