Factors associated with preterm birth: from logistic regression to structural equation modeling.

Area of Science:

• Public Health
• Epidemiology
• Biostatistics

Background:

• Preterm birth remains a significant global health challenge with complex contributing factors.
• Previous analyses using hierarchical logistic regression identified several associated variables.
• A need exists for advanced statistical methods to model intricate relationships influencing preterm birth.

Purpose of the Study:

• To apply structural equation modeling (SEM) to investigate variables associated with preterm birth.
• To analyze the mediating role of prenatal care adequacy in the relationship between various risk factors and preterm birth.
• To develop and test a theoretical model examining socioeconomic, psychosocial, and pregnancy-related factors.

Main Methods:

• Utilized data from a population-based case-control study in Londrina, Brazil (June 2006-March 2007).
• Employed structural equation modeling (SEM) with weighted least square mean and variance adjusted (WLSMV) and robust maximum likelihood (MLR) estimators.
• Constructed three latent variables: socioeconomic vulnerability, family vulnerability, and unwanted pregnancy.

Main Results:

• Socioeconomic vulnerability, family vulnerability, and unwanted pregnancy were indirectly associated with preterm birth.
• The primary pathway for these effects was through inadequacy of prenatal care.
• SEM effectively modeled direct and indirect effects, confirming the mediating role of prenatal care.

Conclusions:

• Inadequacy of prenatal care is a crucial mediator linking socioeconomic and family vulnerabilities, and unwanted pregnancy to preterm birth.
• Structural equation modeling provides a robust framework for understanding complex multifactorial health outcomes like preterm birth.
• Identifying these indirect pathways can inform targeted public health interventions to improve prenatal care and reduce preterm birth rates.