Propensity Score Matching: A Step-by-Step Guide to Coding in R and Application in Observational Research Studies

Area of Science:

• Medical research methodology
• Epidemiology
• Health services research

Background:

• Observational studies are crucial for research when randomized controlled trials are not feasible or ethical.
• However, observational studies are prone to bias due to their non-randomized nature.
• Propensity score matching (PSM) is a statistical technique used to reduce bias in observational studies.

Purpose of the Study:

• To provide a comprehensive, step-by-step guide for implementing Propensity Score Matching (PSM) in observational research.
• To demonstrate the application of PSM using a relevant clinical example.
• To highlight methods for assessing and ensuring covariate balance after matching.

Main Methods:

• The guide covers PICO question development, database selection, data processing, and PSM coding techniques.
• It illustrates PSM using an example of cholecystectomy timing and outcomes in pregnant patients with cholecystitis.
• Methods discussed include various matching techniques, calculation of standard mean differences, and visualization using Love Plots.

Main Results:

• Propensity score matching (PSM) was applied to create balanced intervention and comparison groups.
• Covariate balance was assessed mathematically and visually to confirm the effectiveness of PSM.
• The Average Treatment Effect on the Treated (ATT) was calculated to evaluate the intervention's impact.

Conclusions:

• PSM is an effective method for mitigating bias in observational studies, enhancing the reliability of real-world evidence.
• This guide provides a practical framework for researchers to apply PSM in their studies.
• The successful application of PSM in the cholecystectomy example demonstrates its utility in clinical research.