Methods for claims-based pharmacoeconomic studies in psychosis

Area of Science:

• Health Economics
• Pharmacoepidemiology
• Psychiatric Research

Background:

• Claims data are vital for pharmacoeconomic studies comparing drug costs.
• Methodological limitations in analyzing claims data can bias health resource use comparisons, particularly in psychosis treatment.

Purpose of the Study:

• To review methodological approaches for analyzing claims data in pharmacoeconomic studies.
• To identify and address limitations in comparing healthcare costs of competing drugs for psychosis.

Main Methods:

• Review of different methodological approaches for claims data analysis.
• Discussion of intent-to-treat and mirror-image designs and their limitations.
• Emphasis on drug treatment episodes for health resource use measurement.

Main Results:

• Drug treatment episodes improve health resource use comparisons over traditional designs.
• Financial measures of health resource use are preferred over volume, but require careful selection.
• Selection and period bias can impact between- and within-group comparisons, respectively.

Conclusions:

• Careful control of patient factors (disease severity) and trends can mitigate bias.
• Log transformation of skewed healthcare data has limitations and can distort findings.
• Selecting appropriate methods for claims data analysis is essential for valid pharmacoeconomic investigations.