Sociodemographic trends associated with caesarean delivery in rural subdistricts of Bangladesh: a cohort study, 2005–2019

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Introduction: Although an appropriate population range for caesarean section (CS) has been debated, the WHO recommends national CS rates be between 10% and 15%. Yet rates are much higher in many low-income and middle-income countries, including Bangladesh. These studies also reveal wide disparities in CS use by geographic location and sociodemographic conditions. Nevertheless, few studies have the sample size and temporal precision necessary to model convergence and divergence trends over time. We assessed trends in CS segregated by key sociodemographic factors known to affect CS. Methods: We conducted our study using prospective pregnancy observation data from two distinct Maternal, Neonatal, and Child Health service areas: International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) and the government service areas in Matlab, Bangladesh. We evaluated sociodemographic variations and trends using the multivariable logistic regression model. Findings: The proportion of CS rose from 5.9% in 2005 to 55.7% in 2019, about a ninefold increase, implying a 3.3% average annual growth rate. The CS was positively associated with residence in the icddr,b service area (vs government service area), older age, higher assets, lower parity births and, to a lesser extent, with higher levels of schooling. Among factors positively associated with CS, we observed negative time interactions for maternal age, education and place of delivery, suggesting a convergence in CS rates between high-risk and low-risk groups in these categories. However, positive time interactions for parity and asset score indicate that disparities in CS rates persist and are widening over time. Conclusions: Transition towards near-universal CS, irrespective of sociodemographic status or service area, is likely to be the inevitable future for rural Bangladesh. If such high rates of CS are considered sub-optimal for mother-and-child health, then evidence is needed to support alternatives.

This publication has 33 references indexed in Scilit: