MIXED MODEL APPROACH FOR ANALYSING SPATIAL DISTRIBUTIONS AND HIV/AIDS STIGMATIZATION IN NIGERIA

Abstract

Stigma is often attached to things people are afraid of. Ever since the first case of acquired immune deficiency syndrome (AIDS) in the early 1980’s, people living with Human immunodeficiency virus (HIV) have been stigmatized. People living with HIV/AIDS (PLWHA) isolate themselves from their homes, in public and their places of work. This work aims at identifying factors that are associated with HIV/AIDS stigmatization and also mapping the geographical variation across the states of Nigeria. Data were obtained from the 2013 Nigerian Demographic and Health Survey (DHS). Stigmatization was determined as a composite variation from responses to questions on whether a respondent would allow to be taught by an HIV patient, if the respondent would buy vegetable from the patient and if the respondent would care for a HIV positive family member. A generalized linear mixed model was used to analyze the spatial data while Integrated Nested Laplace approximation (INLA) was used for the Bayesian inference. Result of the analysis reveals that the more educated and older a respondent is, the less he/she will stigmatize against a person living with HIV/AIDS. It also explores the geographical variations of PLWHA being stigmatized in Nigeria. Findings revealed that the respondents from the Northern parts of Nigerian didn’t stigmatize as much as respondents from the Southern parts. Furthermore, mass media played an important role because respondents that listened to radio, watched television and read newspapers did not stigmatize against PLWHA.