Geospatial Analysis of HIV Prevalence in KwaZulu Natal, South Africa: Bayesian Spatial Hierarchical Models

Abstract

Background: Globally, South Africa has the highest number of people living with the human immunodeficiency virus (HIV) and the province of KwaZulu Natal (KZN) has the highest number of individuals who are HIV positive. However, a better understanding of the spatial heterogeneity of provincial estimates of HIV prevalence, specifically at localised level, could help advance HIV treatment and prevention strategies. The aim of the study was to assess the spatial heterogeneity of HIV prevalence and examine the individual-level characteristics of the prevalence of HIV at localised level using the Bayesian hierarchical spatial modelling technique. Methods: This was an analysis of data collected from 9812 men and women aged 15- 49 years participating in the HIV Incidence Provincial Surveillance System (HIPSS) from June 2014 to July 2015. To fit the Bayesian hierarchical spatial model to the HIV prevalence data, the integrated nested Laplace approximation (INLA) numerical method was employed. Results: Results revealed that there was a positive spatial autocorrelation in the wards. The Kulldorf’s spatial scan statistic identified one hot-spot cluster around Nadi, KwaMbanjwa and Zayeka areas and one cold-spot cluster around the Greater Edendale area. Gender, age group, education level, source of income and marital status, along with behaviours like alcohol use and having multiple sexual partners, were significantly associated with HIV prevalence. Also, being diagnosed with sexually transmitted infections (STIs) and TB increased the chances of getting infected with HIV. Conclusion: The detection of HIV hotspot cluster, the predictors of HIV transmission and the spatial distribution of HIV infection in uMgungundlovu Municipality is crucial for focused mitigations, outreach efforts, and resource allocation to populations in need, eventually advancing the efficiencies and integrity of public health schemes.