I am interested in developing and evaluating methods to assess the effectiveness and safety of health care interventions. I've lately been focusing on two particular types of problems.
My primary area of expertise is research synthesis, or
meta-analysis. Although research synthesis can be used to dramatically reduce
the cost of important discoveries, it is also the only practical way to answer some system-wide questions or empirically evaluate interventions for diverse and fragile populations. In this context, I have worked to develop methods to evaluate randomized trial generalizability, synthesize data collected
with randomized and observational designs, and assess treatment
on rare event outcomes.
More recently, I have been working on the problem of identifying and verifying subgroups for which a treatment is most effective and safe. In particular, I have been addressing multiplicity problems when there are a moderate number of possible subgroup identifiers (such as genetic SNPs).