Place-based approach to estimate drinking water contamination

Drinking water contamination is no longer limited to highly localized incidents; instead, it has become a nationwide public health issue. The traditional study of drinking water and its impact on health has been hindered by the logistical challenge of monitoring water quality. I adopt a place-based approach to address and fill this data gap.

UCMR3

The first to show that on a national level contaminated drinking water by PFAS for millions of Americans can be linked to the presence of wastewater treatment plants, military bases, and airports. It spurred a Senate-led initiative to study the health risks from PFAS-contaminated drinking water, which in conjunction with other research efforts, catalyzed the proposed federal drinking water standards in March 2023.

Funding for this work is from Harvard University and the Smith Family Foundation.

ML Models for Drinking Water Quality

We reviewed 27 articles that focused on using machine learning (ML) models to predict drinking water contaminants. Good performance metrics were reported for binary models that classified chemical concentrations above a threshold value by finding significant predictors. However obtaining good predictive performance for continuous models is still challenging. Improving continuous models is important for potential future use in epidemiological studies to supplement data gaps in exposure assessments for drinking water contaminants. While significant progress has been made over the past decade, methodological advances are still needed for this area.

Funding for this work is from National Institute of Environmental Health Sciences Superfund Research Program.

Leverage health big data to improve public well being

Researchers in various policy areas have asked "what works?" and adopted rigorous econometric methods to evaluate policy outcomes. I am hoping to bring that methodology to the environmental health area, evaluating what works? for whom? under what conditions? to support evidence-based public policies.

Evaluation of the Healthy Davis Together Project

Using a synthetic control method, a statistical method used to evaluate the effect of an intervention in comparative case studies, we evaluated the program’s effect on four key areas: (1) disease occurrence, (2) testing uptake, (3) health behaviors and perceptions, and (4) economic indicators and found it to be highly cost-effective.