OBJECTIVES: In the absence of direct, head-to-head randomized evidence, indirect treatment comparisons (ITCs) are often conducted with aggregate data to estimate relative effectiveness and safety of competing treatments. Population-adjusted-ITCs (PA-ITCs), such as matching adjusted indirect comparisons, simulated treatment comparisons and multi-level network meta-regression can be useful when clinically meaningful patient characteristics are imbalanced between studies or in the absence of a common comparator. Although literature exists on methods for conducting PA-ITCs, a practical stepwise process for assessing the feasibility is absent. The aim is to provide a stepwise practical approach for assessing the feasibility of conducting well-informed PA-ITCs.

METHODS: Prior to conducting PA-ITCs, an assessment of the evidence to assess what characteristics are clinically meaningful and how the PA-ITCs should be conducted is required. Based on the recommendations provided in technical support documents on PA-ITCs and their stated assumptions, an approach to assess the feasibility of a PA-ITCs in a stepwise manner was developed.

RESULTS: A stepwise process is developed for assessing the feasibility of PA-ITCs. Assessing the feasibility commences at the stage of identifying relevant studies for potential inclusion into the analysis. Further steps include the assessment of outcome data availability, the availability of common comparators across studies and similarity in the PICOS (Population, Intervention, Comparison, Outcomes, and Study design) criteria. Most importantly, a careful assessment of potential (imbalanced) treatment effect modifiers and prognostic variables is also required and usually includes expert consultation.

CONCLUSIONS: By following these steps the researcher will be able to determine the most appropriate methodology for performing a valid PA-ITC. This process outlines the requirements, assumptions and provides practical advice in a simple step-by-step manner.