maic-methodology

Use MAIC for indirect treatment comparisons when individual patient data is available for one trial and only aggregate data exists for the comparator, allowing you to align populations and estimate the comparison.

Select covariates for a matching-adjusted indirect comparison by identifying effect modifiers and prognostic factors, assessing overlap across trial data scenarios, and monitoring the effective sample size (ESS) to ensure reliability.

Anchored MAIC relies on a common comparator trial to connect treatments, while unanchored MAIC is used when no common comparator exists, requiring distinct methodological justifications based on the available data.

Interpret ESS (effective sample size) by evaluating weight diagnostics to determine feasibility; a severely reduced ESS indicates poor covariate overlap and low reliability for the indirect treatment comparison.

Reporting requirements for MAIC weight diagnostics include transparent checklists and templates that document the covariate selection strategy, weighting outcomes, anchored versus unanchored decisions, and sensitivity analyses.

MAIC requires individual patient data (IPD) for at least one trial to recalibrate weights against aggregate data (AgD); if both trials only have aggregate data, alternative indirect treatment comparison methods are needed.