About this Event
This course introduces methods for combining machine learning (ML) and causal inference to improve health economic evaluations and decision-making. Participants will learn how ML can be used for prediction, estimating treatment effects, and designing stratified treatment policies. The course will also demonstrate practical implementation of these methods in R, with accompanying code provided.
Faculty
- David Glynn, PhD, Postdoctoral Researcher, University of Galway, Ireland
- Julia Hatamyar, PhD, Research Fellow, Centre for Health Economics, University of York, UK
- Noemi Kreif, PhD, Assistant Professor, University of Washington, Department of Pharmacy, USA
Course Overview
The purpose of this course is to present Causal ML as a tool for research prioritization, treatment stratification, and integration into decision models. The course will cover predictive ML, estimation of average and heterogeneous treatment effects, and policy learning approaches.
Learning Objectives
- Differentiate between predictive and causal questions
- Understand the role of machine learning in causal inference
- Understand the role of machine learning in estimating heterogeneous treatment effects
- Understand how ML outputs can be integrated into decision models
Pre-Course Preparation
- Some understanding of regression models (linear/logistic).
- Familiarity with concepts like bias, variance, and confounding.
- Awareness of randomized controlled trials vs. observational data challenges.
- Familiarity with cost-effectiveness analysis, QALYs, and decision modeling basics.