Comparative analysis of cesarean section rates using Robson Ten-Group Classification System and Lorenz curve in the main institutions in Japan.

Aim: The aim of this study was to clarify the indication for cesarean section (CS) using the Robson Ten-Group Classification System (RTGCS) and to clarify the center variation using the Lorenz curve in the main institutions in Japan.

Methods: The records of 68 702 deliveries, which were performed in 125 institutions, were extracted from the Japanese perinatal database in 2013 and the cases were classified using the RTGCS, which classifies deliveries into one of 10 groups on the basis of five parameters. The equality of the CS rate of each hospital was evaluated by the Lorenz curve and the Gini coefficient. The standard error (SE) and 95% confidence intervals (95%CI) for the Gini coefficient were determined by the bootstrap method. The institutions were divided into three categories depending on their scale: comprehensive center (CC, Category I), regional center (RC, Category II) and others (Category III).

Results: The overall CS rate was 37.3%. The difference between Categories I (42.6%) and II (34.3%) was significant (P = 0.02). The CS rates that were classified as RTGCS group 3 (multiparous, single cephalic, ≥37 weeks, with spontaneous labor) were higher in Category I (4.0%) than in Category II (2.7%, P = 0.01). The Gini coefficient of Category I (0.119 ± 0.015; 95%CI, 0.092-0.152) was significantly lower than that of Category II (0.189 ± 0.013; 95%CI, 0.16-0.217).

Conclusion: We clarified the indication of CS and center variation. These two types of methods are useful for the evaluation of medical intervention in the perinatal field.