School of Medicine
Showing 11-12 of 12 Results
Postdoctoral Research fellow, SCRDP/ Heart Disease Prevention
Current Research and Scholarly Interests My main research interests are on the effect of social mobility and social policies on health inequalities. My dissertation work focused on the effect of the Korean War GI Bill on mental health; I showed that eligibility for the Korean War GI Bill reduced socio-economic disparities in depression markers among veterans, but found no evidence of a spillover to veteran wives. We also found that GI Bill eligibility is associated with improved lung function for veterans from socially vulnerable backgrounds, but not for the socially advantaged. I am working on two papers studying the impact of lifecourse social mobility on inequalities in biologic and physical markers of health; we found that the upwardly mobile have similar health to those with consistently high socio-economic status, while the downwardly mobile retain few benefits from their early-life advantage. These studies indicate that health and health inequalities are mutable across the lifecourse. In future work, I will examine the effect of the Vietnam War GI Bill and compulsory schooling laws on socio-economic and racial disparities in smoking behaviors and smoking-related morbidity.
In order to explore health inequalities as rigorously as possible, I have lead several methodological analyses. In one paper, we compared different matching methods (propensity score matching, PSM, and coarsened exact matching, CEM) for making causal inference from observational data, and found that CEM drastically out-performs PSM in balancing the multivariate distribution of matching covariates. We are now in the process of conducting a simulation analysis to determine how PSM and CEM perform under different common support and confounding scenarios. In other methodological work, we used factor analysis to create scales that measure childhood human, financial, and social capital; we psychometrically validate these markers, and compared them to other comprehensive operationalizations for quality of predictions. We found the validated measures out-performed other operationalizations, increasing statistical power and reducing bias. Through these studies, I show tangible ways to improve the quality of health research.