Methods for population-adjusted indirect comparisons based on individual and aggregate level data
David Phillippo, Senior Research Associate in Evidence Synthesis, University of Bristol
Department Research Seminars
Wednesday, 24 March 2021, 2pm to 3pm
Hosted by HERC
Date and Time: Wednesday 24th March 2021, 2pm (UK GMT)
To Join: This is a free event, which will be taking place online via Zoom. To register your interest in attending this talk please click HERE.
Healthcare decision making requires reliable estimates of relative treatment effects. In an ideal scenario, these are provided by high-quality randomised controlled trials (RCTs) comparing the treatments of interest, in a relevant target population. However, it is often the case that head-to-head RCTs are not available between all relevant treatments. Instead, standard network meta-analysis and indirect comparison methods can be used to estimate relative treatment effects between treatments of interest by combining aggregate data from multiple studies, assuming that any variables that interact with treatment effects (effect modifiers) are balanced across populations. Population adjustment methods aim to relax this assumption by adjusting for differences in effect modifiers using available individual patient data from one or more trials.
In this talk, I will give an overview of different population adjustment methods including Matching Adjusted Indirect Comparison, Simulated Treatment Comparison, and a new approach, Multilevel Network Meta-Regression. These methods will be illustrated using an applied example, and I will discuss the results of an extensive simulation study designed to assess the performance of the methods in a range of realistic scenarios under various failures of assumptions.
David Phillippo is Senior Research Associate in Evidence Synthesis at the University of Bristol. His research focuses on methods for evidence synthesis, Bayesian network meta-analysis (NMA), population adjustment methods for indirect comparisons, and assessing the impact of bias in clinical guidelines and decision making. He is the lead author of Technical Support Document 18 published by the NICE Decision Support Unit on population-adjusted indirect comparisons, providing guidance on the use of this new class of methods in NICE Technology Appraisals. He supports the development of NICE Clinical Guidelines through his involvement with the NICE Technical Support Unit based in Bristol, and was the 2019 recipient of the Guidelines International Network Najoua Mlika-Cabanne Innovation Award in recognition of his contributions to new methodology in guideline development. He is the author and maintainer of several freely-available R packages, including nmathresh for assessing sensitivity to biased evidence in NMA using threshold analysis, and multinma for performing NMA and multilevel network meta-regression with individual and aggregate data.