Health Policy Seminar
“Informing treatment stratification using causal machine learning and decision modelling”
Noemi Kreif, PhD is an Assistant Professor in Health Economics at the CHOICE Institute at the University of Washington. She is interested in advancing the use of causal inference and machine learning methods to improve health policy and economic evaluations, with a particular focus on making these approaches more relevant for real-world decision making. Noemi’s research often involves collaboration with colleagues from diverse fields, including global health, decision science, psychiatry, and computer science.
Her academic background includes a PhD in Health Economics from the London School of Hygiene and Tropical Medicine, where she developed an early interest in using machine learning methods for health economics and outcomes research. Before joining the University of Washington, she worked at the University of York, focusing on health economics and policy issues in low- and middle-income countries.
Currently, Noemi is exploring ways to combine causal inference and machine learning to analyze healthcare data more flexibly, aiming to inform personalized and equitable health policy decisions. She enjoys mentoring students and early-career researchers in modern methods for causal inference and computational statistics, and values the opportunity to help train the next generation of researchers.
The talk will focus on the application of machine learning in evaluating treatment effectiveness, particularly in its ability to personalize treatment recommendations by characterizing treatment effect heterogeneity. Drawing from a recently published study, the presentation will demonstrate how tools from causal inference, machine learning, and decision modeling can be integrated to assess the value of treatment stratification. Additionally, it will raise discussion points on the internal and external validity of such analyses, as well as the feasibility and fairness of utilizing algorithms to support treatment decision-making.
Date: Tuesday, May 6th
Time: 4:00pm – 5:00pm ET
Zoom Link: https://wellforce.zoom.us/j/99...
Passcode: 459637
