DEMOGRAPHIC AND BEHAVIORAL RISK FACTORS ASSOCIATED WITH HIV INFECTION IN ETHIOPIAN ADOLESCENTS

HIV is still a global issue, particularly among young adolescents. According to UNICEF, they carry the burden of the disease at approximately 5% of all people living with HIV and about 11% of new adult HIV infections. The DHS program found that Ethiopia in particular has one million people who have HIV. The purpose of this analysis was to determine the effect of socio-demographic and behavioral risk factors, relating to both a sexual and a non-sexual context on HIV outcomes in adolescents in Ethiopia among individuals aged 15-24. The research question of “what are contributing factors to HIV infection in 15-24-year-olds in Ethiopia?” was answered by the use of the Population-Based HIV Impact Assessment (PHIA) public dataset. The PHIA Ethiopia data use manual, the adolescent questionnaire consisting of the child interview data set, and Statistical Analysis System (SAS) coding were used. In order to identify risk factors that have the most robust point estimate effect on HIV outcomes among adolescents in Ethiopia, this study analyzed datasets compiled by the Population-Based HIV Impact Assessment (PHIA) Project that were gathered to guide the global HIV response. PHIA is a household-based cross-sectional survey on HIV prevalence that was implemented in two waves, one in 2015 and the other in 2017, to detect trends in population-based data. This project consisted of two comprehensive questionnaires that were distributed to each household. Bivariate analysis was conducted to assess if there is a statistically significant association with each predictor and the outcome of interest. Risk factors and variables included as potential correlates of HIV infection in this population will include sociodemographic characteristics, sexual activity and related factors, and clinical characteristics. The correlation between each individual risk factor and the outcome of interest, HIV infection, was determined through characterizing frequencies of demographic variables and risk behaviors. Using a 95% confidence interval, the statistical tests of chi-square and t-tests were performed to identify the p-values respective to each factor. All correlations were combined and coded into one multivariable model to visualize if the associations persist with the outcome, adjusting for various confounding variables such as age and gender. All analyses were conducted using SAS. The results showed that the prevalence of HIV among adolescents and young adults in Ethiopia was 0.82%, with women having close to 4 times the odds of having HIV compared to men, having multiple partners (more than three) substantially increasing the odds of being HIV positive, a diagnosis of tuberculosis substantially decreasing the odds of being HIV positive, and a diagnosis of an STD other than HIV had substantially decreased odds of being HIV positive