I am a biostatistician in the Division of Biostatistics within the Department of Population Health with research interests in health services research, pragmatic and cluster randomized trials, and Bayesian methods for complex study designs. My work focuses on evaluating care delivery, continuity, and outcomes for older adults and other vulnerable populations across real-world healthcare settings, including home-, community-, and emergency-based care.
I collaborate extensively on large, multi-site studies that rely on longitudinal and linked clinical and administrative data. My recent work includes major NIH-funded trials such as PRIM-ER, a 33-site cluster-randomized stepped wedge trial evaluating emergency department–based palliative care using Medicare claims; EMPallA, a multi-site RCT comparing nurse-led telephonic versus specialty out-patient palliative care; and HAS-QOL, a national stepped-wedge trial aimed at improving quality of life for people living with dementia through enhanced home-based care. Across these projects, I have developed expertise in analyzing clustered and correlated data, defining utilization-based outcomes, and addressing methodological challenges related to missingness, time-varying processes, and implementation in pragmatic settings.
In addition to my trial work, I maintain an active portfolio of observational research using large administrative datasets, including studies of care continuity and health outcomes among people with opioid use disorder and justice-involved populations, with findings published in Addiction, Drug and Alcohol Dependence, and PLoS One.
I am the principal developer of the open-source statistical software simstudy, an R package created to generate simulated data across a wide range of study designs. The package and related methodological writing are featured on my blog
ouR data generation.
