Matching-Adjusted Indirect Comparison (MAIC)

Matching-Adjusted Indirect Comparison (MAIC) is a statistical method used in comparative effectiveness research, particularly in health technology assessments and clinical trials. This method is employed when direct head-to-head comparisons between treatments are unavailable, necessitating indirect comparisons using data from separate studies. The primary goal of MAIC is to adjust for differences in baseline characteristics between populations in different studies, which is crucial since variations in patient demographics, disease severity, and other factors can significantly influence treatment outcomes.

MAIC achieves this by reweighting individual patient data (IPD) from one treatment study to match the aggregated baseline characteristics reported in another study. By reweighting the IPD, MAIC ensures that the adjusted population closely resembles the comparator study’s population, facilitating more accurate indirect comparisons of treatment outcomes. This approach is particularly valuable when direct comparisons are infeasible due to the absence of head-to-head trials.

The process of MAIC involves several key steps. First, relevant studies for the treatments being compared are selected, ensuring that individual patient data is available for at least one treatment. Next, the baseline characteristics of the study with IPD are matched to the aggregated baseline characteristics of the comparator study. Weights are then applied to the individual patient data to create a synthetic population that closely matches the comparator study. Finally, treatment outcomes are compared using the reweighted data, enabling indirect comparisons between treatments.

MAIC is widely used in health technology assessments to compare the effectiveness and cost-effectiveness of new treatments against existing ones when head-to-head clinical trials are lacking. It is also utilized in regulatory submissions by pharmaceutical companies to provide comparative effectiveness data and assists clinicians and policymakers in making informed decisions about treatment options based on comparative effectiveness evidence.

However, MAIC has some limitations. It requires access to individual patient data for at least one treatment, which may not always be available. The method’s validity depends on the assumption that all relevant baseline characteristics are accounted for and appropriately adjusted. Additionally, the methodology is complex, requiring expertise in statistical analysis and detailed patient-level data.

Click HEOR TERMINOLOGY