Postdoctoral scholar at the Center for Health and Community
Department of Epidemiology and Biostatistics
University of California, San Francisco
Extensive empirical health research exploits variation in the timing and location of policy changes using differences-in-differences, interrupted time series, instrumental variables, and related designs. While these methods offer a promising approach to identify the causal effects of social policies and conditions on health, an important challenge is that multiple social policies are often adopted simultaneously in the same locations, creating clustering which must be addressed analytically for valid inferences. Limited research has evaluated the pervasiveness of social policy clustering, its consequences for causal effect estimation, or potential analytic solutions. In this talk, I describe research addressing each of these gaps. I begin by illustrating the value of quasi-experimental approaches with an investigation of the effects of gun shows on firearm injuries. I then discuss ongoing methodological work leveraging a systematic sample of 55 studies of the health effects of social policies corresponding to 13 unique social policy databases from diverse domains including poverty, family leave, immigration, and cannabis.