Select publications related to current research priorities

Predictive Modeling to Facilitate Targeted Intervention Strategies

This line of research applies machine learning methods to develop models that predict suicide, PTSD, and other adverse psychiatric outcomes at critical intervention points. By identifying individuals at high-risk for these outcomes, such models may facilitate development and testing of targeted prevention or early intervention strategies.

Development of Interventions that Target Transdiagnostic Mechanisms

This line of research applies a translational approach to understand transdiagnostic mechanisms of anxiety, stress, and addiction, and leverages technology to intervene on these mechanisms. Virtual reality and mobile health technology is used to target mechanisms more precisely with the aim of enhancing treatment effects while increasing the efficiency and accessibility of interventions.

Integrating Genomic Data to Estimate Risk for Adverse Psychiatric Outcomes

This line of research seeks to enhance the clinical potential of polygenic risk scores from large GWAS studies by applying statistical methods that incorporate complex mixtures of exposures and outcomes.

Rigorous Methods for Causal Inference

When it comes to making causal inference, we use rigorous methods including randomized controlled design (including large cluster-randomized multi-site trials) and doubly-robust statistical approaches that leverage machine learning to estimate intervention effects with observational data. We strive to maximize open science practices including pre-registration and data sharing policies that protect participant privacy.

Complete list of publications