Using Respondents' Uncertainty Scores to Mitigate Hypothetical Bias in Community-Based Health Insurance Studies

Abstract Community-based health insurance (CBHI) is being considered as an effective way to reach the poor people to access adequate health care services in developing countries. Assessing what the poor are willing to pay is of paramount importance for policy-making. The contingent valuation method, which relies on a hypothetical market, is commonly used for this purpose. But the presence of the hypothetical bias which is most often inherent in this method tends to bias the estimates upward, and compromises policy-making. This paper uses respondent uncertainty scores in an attempt to mitigate hypothetical bias in CBHI in one rural setting in Cameroon. Uncertainty scores are often employed in single dichotomous choice surveys. An originality of the paper is to use such approach in a double-bounded dichotomous choice survey. The results suggest that this instrument is effective at decreasing the mean willingness-to-pay (WTP) which ranges from 973.63 CFA francs/person/month ($2.07) to 804.82 CFA francs/person/month ($1.71). This mean is the conservative mean WTP. Furthermore, the results of the study indicate that the insurance is a normal good, and 50% of rural households are willing to pay on average 2.62% of their monthly income for the proposed health insurance. This shows that there is a viable demand for CBHI. It also comes out from the results that the age, usual practice when the rural populations get sick, involvement in association, income, and the belonging of households to the villages called Mbiem, Fonègom and Famla are determinants of WTP. As shown by the relatively high mean WTP values, such schemes may provide a reliable income flow for health care providers and warranty easy access to high quality care by the poor in the rural areas. The policy implications are discussed. Count word for the abstract: 284 words No conflict of interest