Determinants of Clients Satisfaction in HIV Testing and Counseling: A Multilevel Logistic Regression Analysis on Ethiopian Service Centers

Abstract

HIV testing has been a cornerstone of HIV prevention efforts since 1985 when an accurate and reliable test for the HIV antibody became available. This study investigated determinants of client satisfaction in HIV counseling and testing at Hawassa city. The data were collected from eleven service centers administered by Government, NGO and private sector using Stratified random sampling design. Due to Hierarchical nature of the data, multilevel logistic regression models were used to analyze the data. The descriptive analysis showed that the HIV counseling and testing service satisfaction distribution among the eleven centers ranges from 39.7% to 67.1% with about 52.4% average satisfaction. The Multilevel analysis showed that about 2.4% of the variation in clients’ satisfaction was due to unobserved service center characteristics (with a VPC of 0.024). The random intercept model showed that income, age, prior expectation, actual service rating of client level predictors and client-physician interaction, convenience of service center, waiting time of the service center level predictors were found statistically significant. The Random coefficient/slope model with income as a random factor across service centers resulted that except prior expectation, all the predicators significant in the random intercept model were still statistically significant. Generally, the study identified clients’ income, age, prior expectation, actual service rating, client-physician interaction, convenience of service center, waiting time in the service centre as a determinants of clients satisfaction at the HIV counseling and testing.

Keywords: Multilevel analysis, HIV testing, Client satisfaction, Ethiopia