Effect of Maternity Characteristics on Cesarean Section Rates in Belgium: A Robson Classification Approach.

Objective: To assess the effect of maternity unit characteristics on the cesarean section (CS) rate, using Robson's Ten-Group Classification System (TGCS) and considering the sociodemographic and medical characteristics of the mother.

Methods: The study, conducted in Belgium from 2011 to 2019, employed an analytical design utilizing a nationwide register of routine data and focused on hospital births. The CS rate was analyzed by neonatal intensive care unit (NICU) availability and by maternal unit size for TGCS groups 1, 2, and 5, which were the highest contributors to the total CS rate. Multivariable logistic regression models and generalized linear mixed-effects models were utilized to analyze the association between the maternity itself and CS.

Results: The overall CS rate was 20.8%, displaying a twofold variation across maternity units. This variation persisted irrespective of the presence or absence of a NICU and the maternity unit's size. Our findings highlighted a significant association between maternity unit characteristics (size and NICU availability) and the likelihood of performing CS in TGCS groups 1, 2, and 5. This association did not change after adjustment for sociodemographic and medical characteristics. However, the majority of odds ratios for maternity-related variables lost their significance in the multilevel analysis compared to simple logistic regressions.

Conclusions: The CS rate seems to be more influenced by the "philosophy" of the maternity unit than its specific size and NICU characteristics. Future research is needed to explore the underlying mechanisms of this association and to identify potential interventions that could reduce CSs performed without clinical indication in different settings.