Does health care utilization match needs in Africa? Challenging conventional needs measurement

textabstractAbstract. An equitable distribution of health care use, distributed according to people’s needs instead of ability to pay, is an important goal featuring on many health policy agendas worldwide. However, relatively little is known about the extent to which this principle is violated across socio-economic groups in Sub Saharan Africa (SSA). We examine cross-country comparative micro-data from eighteen SSA countries and find that (a) considerable inequalities in health care use exist and vary across countries, but that (b) identifying the extent to which these inequalities are unfair, i.e. do not correspond to inequalities in need, is not straightforward to ascertain with the conventional tools. These tools include rank-based measures such as the concentration index and the index of inequity. The two main concerns when using conventional tools to measure equity are (i) the reporting heterogeneity in self-reported health variables across socio-economic groups and (ii) the weak relationship between need and use. We show that the use of subjective self-reports of health leads to much lower measured degrees of socio-economic inequalities than those obtained using more objective indicators. This leads to an underestimation of the degree of inequity when using self-reported health measures. The observed weak relationship between indicators of ill-health and use of health care does not appear to provide an estimate of the adequate response to needs, which further puts a downward bias on equity measures. In all countries, apart from the more developed Mauritius, health care use is distributed according to wealth rather than to need. A better match of needs and use is realized in those countries with better governance and more physicians but, perhaps surprisingly, not those with greater urbanization. Given the importance of equity in many health policies worldwide, it is vital to develop more robust equity measures relevant to low income settings