Eliciting the Economic Value and Preferences for Health Insurance Reform in Nigeria: Evidence from a Discrete Choice Experiment

Abstract

National health insurance systems across the world are confronted with mounting demands for reforms to make them fit for purpose and deliver as planned. There is, however, a dire paucity of empirical evidence from which policymakers can draw inferences to make informed reform decisions. Such evidence needs to detail the economic cost as well as consumers’ preferences for reforms. Using orthogonal coding to generate three attributes and a full factorial design with all three attributes having three levels each to generate 27 possible scenarios with only 9 scenarios presented to each respondent to reduce the cognitive burden, conditional logistic regression was used to estimate McFadden’s discrete choice model in a discrete choice experiment conducted in Nigeria. McFadden’s R2 value of 0.5346 indicates that the model is well fitted. The Chi-squared statistic of 1057.13 shows that the estimated model has the required explanatory power and statistical significance for the attributes with confidence level set at 95%. Results show that respondents are willing to pay for health insurance reform, opting for health insurance types with attributes that they deem would meet their needs, rather than settle for status quo health insurance. Respondents were willing to pay as high as 24% of annual income to access insurance types that have comprehensive benefits package, which indicates that risk pooling could potentially fund universal health insurance in Nigeria.