Determinants of Health Insurance Uptake Among Women in Kenya: an Application of Discriminant Analysis

Abstract

The current high utilization of out of pocket payments by a majority of Kenyans to settle their medical bills has continued to ensure that the poor and the vulnerable in the society cannot access essential health care services. Studies have shown that having a health insurance cover can greatly reduce the over-reliance on out of pocket financing. Despite studies showing that Kenya’s population under health insurance coverage has grown, the population of women with health insurance schemes has continued to fall below the national average. The goal of this study was to examine determinants of health insurance uptake among women in Kenya using the discriminant analysis approach. The study used data from KDHS collected in the year 2014.After fitting a discriminant analysis model using the step-wise procedure, all the eight predictor variables, namely age,marital status, education level, employment,wealth quintile,place of residence,household size and access to media were found to be significant in discriminating as to whether a woman was insured or uninsured. The classification accuracy of the discriminant model was 86.9 per cent, and the model was found to be statistically significant and hence using the eight predictor variables, one can be able to classify a woman as to whether she is insured or uninsured.