Using Propensity Score Weighting to Enhance the Operating Characteristics of Power Prior in Leveraging External Data...

Area of Science:

• Biostatistics
• Clinical Trial Design
• Health Data Science

Background:

• The power prior method leverages external data to enhance traditional clinical studies.
• Integrating propensity scoring with the power prior method shows promise for improved operating characteristics.
• Existing methods may face challenges with covariate imbalance in external data integration.

Purpose of the Study:

• To introduce a novel propensity score-integrated power prior strategy using propensity score weighting.
• To address limitations of existing power prior methods, particularly in the presence of covariate imbalance.
• To offer an alternative approach for augmenting clinical studies with external data.

Main Methods:

• Development of a new power prior strategy incorporating propensity score weighting.
• Modification of the power prior formula by replacing the sufficient statistic with a propensity score weighted version.
• Evaluation through a simulation study comparing the proposed method with the original power prior and stratification strategies.

Main Results:

• The proposed propensity score weighting strategy demonstrates favorable operating characteristics compared to the original power prior method.
• The simulation study confirmed the effectiveness of the new strategy, particularly under conditions of covariate imbalance.
• The method offers a distinct approach from other propensity score-integrated power prior proposals.

Conclusions:

• The novel propensity score-integrated power prior strategy effectively enhances clinical studies by leveraging external data.
• This method provides a valuable tool for biostatisticians and researchers dealing with covariate imbalance.
• The proposed weighting strategy represents a significant advancement in the application of power priors in clinical research.