Survival Analysis of Childbirth Using a Mixture Cure Frailty Model Publisher



Naderi A ; Foroushani AR ; Jafari AM ; Gubari MIM ; Yaseri M ; Hosseini M
Source: Journal of Biostatistics and Epidemiology Published:2025

Abstract

Introduction: Childbirth plays a crucial role in population growth and maternal health. In recent decades, many nations, including Iran, have experienced declining birth rates. Since childbirth is a recurrent event in a parent's life, it is useful to analyze it through the lens of recurrent event analysis. This methodological framework, commonly employed in biomedicine, allows for a nuanced examination of the relationship between multiple childbirth experiences and the potential for cured subjects. This study explores childbirth rates in Hamadan province. Methods: A total of 633 mothers who gave birth to their first child in 2012 at Fatemiyeh Hospital in Hamadan participated in this retrospective cohort study. Both mixture cure frailty models and simple frailty models were fitted. The analyses were conducted using the RSTAN package in RStudio version 26.2.4. Results: In this study, we analyzed the childbearing patterns of couples and found that the majority (60.6%) had two children. Additionally, we discovered that 49% of mothers and 55.9% of fathers had education levels below a diploma. The Kaplan-Meier (KM) curves indicated a cure pattern for families with three or more children, revealing that only 10.6% of individuals had three children, and a mere 0.8% had four. Furthermore, results from a mixture cure frailty model demonstrated that maternal education plays a crucial role in influencing childbirth probabilities. Conclusion: Based on the findings of this study, we recommend utilizing mixture cure frailty models rather than simple frailty models when the dataset contains individuals who are cured. Copyright © 2025 Tehran University of Medical Sciences.