TEHReview

Review: Data-driven methodology for detecting treatment effect heterogeneity

Collaboration: UCL MRC-CTU and the Alan Turing Institute

Interest:

• Test for the presence of differential effects
• Identify subgroups that represent varied responses to treatment
• Estimate the effects for the determined patient subpopulations
• Characterise such subpopulations

Task: Comprehensive Simulation Study

Themes:

• Tree based methods
• Non tree based methods
• Metalearners