Geospatial patterns and multilevel determinants of cesarean section rates in Iran: a Bayesian spatial analysis using...

Area of Science:

• Public Health
• Epidemiology
• Spatial Statistics

Background:

• Iran exhibits exceptionally high Cesarean section (CS) rates, surpassing WHO recommendations.
• While individual factors influencing CS are known, provincial spatial variations and contextual determinants remain under-researched.

Purpose of the Study:

• To quantify multilevel predictors of CS.
• To examine spatial patterns and identify high-risk clusters of CS in Iran using advanced statistical modeling.

Main Methods:

• Cross-sectional analysis of 24,982 deliveries from the 2010 Iran Demographic and Health Survey (IrMIDHS).
• Employed multilevel logistic regression and Bayesian spatial logistic models (INLA with BYM2 structure) to analyze predictors and spatial autocorrelation.
• Model fit was assessed using DIC/WAIC, and spatial autocorrelation was measured by Moran's I.

Main Results:

• The national CS rate was 52.1%.
• Advanced maternal age, university education, and higher income were associated with increased CS odds, while contraceptive use showed a protective effect.
• The Bayesian spatial model demonstrated superior fit and significant spatial autocorrelation, revealing excess adjusted risk in northern/central provinces and reduced risk in southeastern regions.

Conclusions:

• CS in Iran is linked to individual socioeconomic factors and significant spatial heterogeneity.
• Bayesian spatial modeling effectively identifies high-risk CS clusters.
• Findings support the need for geographically targeted interventions to reduce unnecessary CS and improve obstetric care equity.