Methodologies to adjust for measured confounding in ITC: an overview of population adjustment approaches: in issue panel Bias in indirect treatment comparisons and evolving methodology: implications for Health Technology Assessment and beyond

Network meta-analysis (NMA) and indirect comparisons combine aggregate data (AgD) from multiple studies on treatments of interest, assuming that any effect modifiers are balanced across study populations. Population adjustment methods such as matching-adjusted indirect comparison (MAIC), simulated treatment comparison (STC), and multilevel network meta-regression (ML-NMR) aim to relax this assumption, and are becoming increasingly prevalent in HTA. These methods use available individual patient data (IPD) to adjust for differences in effect modifiers between studies in a connected network, and are also used incorporate single-arm studies or disconnected networks under much stronger assumptions.
We give an overview of population adjustment approaches, and their properties and assumptions. ML-NMR extends the standard NMA framework to coherently incorporate IPD and AgD studies whilst avoiding aggregation bias, and has several advantages over MAIC and STC. It can analyse networks of any number of trials and treatments, and in larger networks allows key assumptions to be assessed. Crucially, ML-NMR can provide comparisons in any target population for decision making. We illustrate with an example and compare results between the different approaches. A user-friendly R package multinma is available for performing ML-NMR analyses.