On Demand Webinar Series:
HEOR & RWE
Effect Modifiers In Indirect Treatment Comparisons; Steps To Ensure the Unbiased Identification to Inform Decision Making
Population-adjusted indirect treatment comparisons (PAICs) are commonly used in health technology appraisals to estimate relative treatment effects in presence of bias due to effect modification. Bias due to effect modification occurs when there are distributional differences in the effect modifiers (EMs) across the trials in a network. Standard network meta-analyses (NMAs) are based on aggregated data (AgD) and assume no EMs or that the distributions of EMs are balanced between trials. However, when this is not the case, different methods of PAICs such as matching-adjusted indirect comparison [MAIC], simulated treatment comparison [STC] and the newest method of Multi-Level Network Meta-Regression (ML-NMR) have been proposed to "correct" for the role of EMs by using available individual patient data (IPD) to adjust for differences in EMs through reweighting or regression adjustment.
Key Learning Points:
- How should EMs be identified from the literature following evidence-based principles? How to ensure the unbiased selection of EMs for inclusion in PAICs?
- How best to navigate across the different methodologies for adjusting for EMs in PAICs and to handle the nuances around each method such as multiple endpoints, sample size considerations and evidence sources?
- What is the potential of a novel indirect treatment comparison approach, called Network Meta-Interpolation which challenges the strong assumption of shared effect modification in NMAs and offer an alternative solution?
Different methods of population-adjusted indirect treatment comparisons (PAICs) have been proposed to “correct” for the role of effect modifiers (EMs). Recently, case-studies demonstrated the circumstances when these assumptions can be reliably justified which mainly depend on data availability (sample size), type of outcomes and other statistical considerations.
The second webinar in this series will address the following questions:
- How to reliably select EMs for inclusion in the PAICs?
- What to do when published evidence on EMs contradicts with evidence from trial individual patient-data?
- Should adjusting for EMs be the “norm” and when is best not to adjust for EMs?
- Grammati Sarri, Senior Research Principal, Head of RWAA External Research Partnerships, Cytel
- Michael Groff, Senior Research Consultant, Health Economics