Estimating Causal Effect of Two-Dose COVID-19 Vaccination on Hospitalization: A Propensity Score Matching Approach

Mahboobeh  Taherizadeh; Mohammad Taghi  Shakeri; Saeed  Akhlaghi; Ehsan Mosa  Farkhani

doi:10.18502/jbe.v10i1.17158

Mahboobeh Taherizadeh Department of Biostatistics, Faculty of Health, Mashhad University of Medical Sciences, Mashhad, Iran.
Mohammad Taghi Shakeri Department of Biostatistics, Faculty of Health, Mashhad University of Medical Sciences, Mashhad, Iran.
Saeed Akhlaghi Department of Biostatistics, Faculty of Health, Mashhad University of Medical Sciences, Mashhad, Iran.
Ehsan Mosa Farkhani Department of Epidemiology, School of Health, Mashhad University of Medical Sciences, Mashhad, Iran.

DOI: https://doi.org/10.18502/jbe.v10i1.17158

Keywords: Propensity score match- ing; Causal effect; Observational study; Logistic regression.

Abstract

Introduction: To estimate the effectiveness of two-dose COVID-19 vaccination in reducing hospitalization, accounting for complex confounding factors in observational studies.

Methods: Researchers applied propensity score methods to adjust for confounding variables, comparing their performance to traditional covariate adjustment methods. Multiple Logistic Regression and Propensity Score Matching were employed to analyze the data, ensuring a balanced comparison between vaccinated and unvaccinated groups.

Results: Both analytical methods demonstrated a significant reduction in the likelihood of hospitalization among vaccinated individuals. The adjusted odds ratios (OR) were 0.29 (95% CI: 0.26, 0.31) via logistic regression and 0.32 (95% CI: 0.30, 0.34) using propensity score matching.

Conclusion: The study confirms the effectiveness of two-dose COVID-19 vaccination in decreasing hospitalization. It highlights the importance of using meticulous approaches like propensity score methods to assess real-world impacts in complex observational data settings