Methods for Population-Adjusted Indirect Comparisons in Health Technology Appraisal

Area of Science:

• Health Economics
• Biostatistics
• Pharmacoeconomics

Background:

• Standard indirect comparisons and network meta-analyses assume no differences in effect modifiers across trials.
• Reimbursement agencies increasingly use population-adjusted indirect comparisons (PAIC) with individual patient data.
• Methods like Matching-Adjusted Indirect Comparison (MAIC) and Simulated Treatment Comparison (STC) are gaining traction but lack extensive testing.

Purpose of the Study:

• To describe assumptions for PAIC and demonstrate their application in target populations.
• To distinguish between anchored and unanchored comparisons, noting the stronger assumptions of unanchored methods.
• To provide recommendations for the appropriate and statistically valid use of PAIC in health technology appraisal.

Main Methods:

• Review and description of assumptions underlying population-adjusted indirect comparisons.
• Categorization of PAIC into anchored and unanchored comparisons based on comparator arm presence.
• Discussion of supporting analyses required for robust and transparent results.

Main Results:

• PAIC methods allow for treatment comparisons within specific target populations using individual patient data.
• Unanchored comparisons, lacking a common comparator arm, rely on assumptions considered infeasible.
• Clear guidelines are proposed for the application of PAIC to ensure validity and consistency.

Conclusions:

• PAIC methods offer a way to adjust for population differences in indirect treatment comparisons.
• Careful consideration of assumptions, particularly for unanchored comparisons, is crucial.
• Further simulation studies are recommended to assess the robustness of PAIC methods.