Methods in Evidence Synthesis seminar (MESS)

Summary: A key question for meta-analysis is to reliably assess whether treatment effects vary across different participant subgroups (referred to as an interaction). In addition, estimates of subgroup effects (effect of a treatment within specific covariate subgroups) is key information for clinical decision making to target treatments appropriately – which estimation of an interaction effect itself does not provide. In this talk I describe a framework for estimating interactions and subgroup effects in aggregate data meta-analysis. I describe the steps involved and apply the methods to two examples taken from previously published meta-analyses, in which detailed aggregate data were available.

Bio: Peter Godolphin is a statistician and senior research fellow in the meta-analysis programme at MRC Clinical Trials Unit at UCL. He works on meta-analysis projects predominantly in advanced prostate cancer, but also contributes to work in COVID-19 and women’s health, with these projects using a mix of aggregate data and individual participant data. In addition, he has an interest in developing new statistical methods in meta-analysis, currently focusing on approaches that determine whether the effect of a treatment varies based on participant-level factors (so called interaction or subgroup analysis).

Join Zoom Meeting: https://bristol-ac-uk.zoom.us/j/94559239881?pwd=L3UzenkvYjk4Y2hIZnR4a2M4WjFjUT09

Meeting ID: 945 5923 9881
Passcode: 931107