Quasi-experimental methods for pharmacoepidemiology: difference-in-differences and synthetic control methods with...

Area of Science:

• Pharmacoepidemiology
• Health Policy Evaluation
• Quasi-Experimental Designs

Background:

• Difference-in-differences (DID) and synthetic control methods (SCM) are increasingly used for policy evaluation.
• These quasi-experimental designs are vital for real-world effectiveness and safety studies in pharmacoepidemiology.
• A thorough understanding of their advantages and limitations is essential for robust application.

Purpose of the Study:

• To elucidate the benefits and drawbacks of DID and SCM in pharmacoepidemiology.
• To assess the utility of these methods for evaluating health policies and generating real-world evidence.
• To highlight challenges and opportunities in applying these designs to vaccine evaluation.

Main Methods:

• Review of quasi-experimental designs, specifically difference-in-differences and synthetic control methods.
• Analysis of key assumptions including parallel trends, stable weighting, absence of concurrent events, and no contamination.
• Case study presentations of three vaccine evaluation studies in pharmacoepidemiology.

Main Results:

• Quasi-experimental designs allow estimation of average treatment effects without measuring all confounders and enable population-level effect assessment.
• Key assumptions (parallel trends, no concurrent events, no contamination) are critical for valid estimation.
• Case studies illustrate specific challenges and opportunities in applying these methods to vaccine evaluations.

Conclusions:

• Difference-in-differences and synthetic control methods are feasible and valuable for pharmacoepidemiologic research.
• Further research is needed to optimize the identification and weighting of advantages and disadvantages in specific settings.
• These methods can significantly contribute to the evidence base for health policies and drug safety.