The relationship between household wealth and HIV prevalence in Ethiopia

Abstract

Epidemiologic research shows that socioeconomic status influences different health outcomes including HIV/AIDS. Although poverty (low socioeconomic status (SES)) and HIV prevalence are correlated at the global level, the association between SES and HIV prevalence is rather mixed in sub-Saharan Africa. Differences in findings could be attributed to context and thus context specific evidence is needed to develop interventions that could have greater impact in those settings. However there are few studies that investigate the association between SES and HIV prevalence in Ethiopia. In 2011, the Central Statistics Agency (CSA) in collaboration with Federal Ministry of Health and ORC-MACRO collected a broad range of demographic, socioeconomic and health data on a representative sample of the population of Ethiopia. This included information on HIV status, demographic and socioeconomic variables (age, gender, religion, marital status, place of residence, household wealth, education, occupation and others) and behavioural risk factors. The present study took advantage of this dataset to describe the relationships between SES and demographic factors and HIV prevalence in the Ethiopian adult population, and to explore the relationship between household wealth and HIV prevalence. Part A of this dissertation (Protocol) describes the characteristics of the Ethiopian Demographic and Health Survey (EDHS2011) dataset, gives details on the sampling and data collection in the original study, and delineates the methodology of the secondary analysis. Part B (Literature review) illustrates the main findings of the conflicting epidemiological literature on the socioeconomic determinants of HIV/AIDS in sub-Saharan Africa and presents a summary of the major studies on wealth, education and place of residence as risk factors for HIV prevalence. Part C (Article) presents the methodological details, results, and possible interpretations of the analyses carried out on the EDHS2011 dataset. The estimated prevalence of HIV in the Ethiopian population aged between 15 to 49 years was 1.47% (95% CI: 1.25% to 1.68%). The analysis showed that household wealth and education were the main socioeconomic status determinants and were independently associated with higher HIV prevalence in Ethiopia, though having education beyond high school was protective against HIV. Data also suggested that living in urban areas, religion and age were the main demographic determinants of HIV prevalence in Ethiopia. Behavioural factors, especially having multiple sexual partners and condom use in the last 12 months were more prevalent both among HIV positive individuals and among more educated and relatively wealthier individuals. It is possible that these factors could be involved in the causal pathway between household wealth and HIV prevalence. The results confirm the pattern of association between education and HIV prevalence in sub-Saharan African countries undergoing epidemiological transition. Those with higher educational attainment had lower HIV prevalence compared to those with no education as the epidemic matured. The evidence generated in this study can be used to develop and update prevention strategies in order to target areas which have higher HIV prevalence.