Data disaggregation as a means to improved health research and policy-making

Researchers and policy makers have initiated a recent push towards improved data availability to inform evidence-based policy making. Often the goal of research and policy is to examine differences in outcomes across pre-determined demographic, socio-economic, or geographically distinct populations, and to address the root causes of any documented disparities in these outcomes. In particular, basic socioeconomic differences across racial and ethnic groups must be recognized and studied in order to adequately address health and welfare disparities. Using only average group characteristics in this type of analysis may mask important differences for heterogeneous groups or communities and result in unsuccessful policies.

To adequately address such questions, data must be available at a sufficiently disaggregated level so that the populations of interest are identifiable and there is sufficient statistical power to produce credible empirical estimates. The question of what level of disaggregation is suitable in different settings is not trivial. On September 26 the Economic Studies program at Brookings organized a conference to highlight and discuss issues related to data disaggregation as a means to improve the quality of empirical work and policy effectiveness with an emphasis in health and well-being.

Download full agenda here. 

Download an event summary here.