Indirect Treatment Comparisons and Multi-Level Network Meta-Regression

This technical report assesses the current use of population adjustment methods in submissions to NICE, specifically focusing on the potential of multi-level network meta-regression (ML-NMR) to replace existing methods for time-to-event (TTE) outcomes. The analysis includes 31 technology appraisals (TAs) that met the inclusion criteria, with findings indicating that while matching adjusted indirect comparisons (MAICs) were frequently utilized, simulated treatment comparisons (STCs) were rarely reported. The report details a targeted literature review of TAs submitted to NICE from April 2021 to March 2024, emphasizing the methodologies used, including MAICs, STCs, and ML-NMRs. It outlines the limitations of current practices, such as the frequent reliance on unanchored analyses and the challenges posed by missing treatment effect modifiers. The findings suggest that while ML-NMRs can enhance the estimation of treatment effects, their application is limited by the prevalence of unanchored analyses in the reviewed TAs.