Quality of care and the demand for health services in Bamako, Mali: the specific roles of structural, process, and outcome components.

The public finance and foreign exchange crisis of the 1980s aggravated the unfavourable economic trends in many developing countries and resulted in budget cuts in the health sector. Policymakers, following the suggestions of World Bank experts, introduced user fees. Economic analysis of the demand for health care in these countries focused on the impact of price and income on health service utilisation. But the lesson to date from experiences in cost recovery is that without visible and fairly immediate improvements in the quality of care, the implementation of user fees will cause service utilisation to drop. For this reason, the role of quality of health care has been recently a subject of investigation in a number of health care demand studies. In spite of using the data from both households and facilities, recent studies are quite limited because they measure quality only by structural attributes (availability of drugs, equipment, number and qualifications of staff, and so on). Structural attributes of quality are necessary but not sufficient conditions for demand. A unique feature of this study is that it also considers the processes followed by practitioners and the outcome of care, to determine simultaneously the respective influence of price and quality on decision making. A nested multinomial logit was used to examine the choice between six alternatives (self-treatment, modern treatment at home, public hospital, public dispensary, for-profit facility and non-profit facility). The estimations are based on data from a statistically representative sample of 1104 patients from 1191 households and the data from a stratified random sample of 42 out of 84 facilities identified. The results indicate that omitting the process quality variables from the demand model produces a bias not only in the estimated coefficient of the "price" variable but also in coefficients of some structural attributes of the quality. The simulations suggest that price has a minor effect on utilisation of health services, and that health authorities can simultaneously double user fees and increase utilisation by emphasising improvement of both the structural and process quality of care in public health facilities.