OBJECTIVE: The aim of the present study was to assess changes in cesarean section (CS) rates over 3 years at a public tertiary care hospital using the Robson Ten Group Classification System (RTGCS). The specific objectives were to compare population distribution across Robson groups, evaluate overall, group-specific, absolute and relative CS rates and assess data quality and completeness. METHODS: This analytical cross-sectional study was conducted at a public tertiary care hospital in Rawalpindi, Pakistan and spanned over two phases. Phase 1: June 2019 to November 2019 and phase 2: August 2022 to January 2023. For analysis, RTGCS was applied. Robson's report tables (RRTs) of two phases were generated to calculate and compare CS rates. Statistical analysis included the calculation of P values (considering <0.05 as significant) by chi-square and Fisher exact tests, odds ratios (ORs), relative risks (RRs) and 95% confidence intervals (CIs), using established online tools. RESULTS: The total number of women included in study during phase 1 were 5437, and in phase 2 were 3762. Overall CS rate increased from 31.06% (phase 1) to 40.09% (phase 2) (P < 0.001). In both phases, Robson group 3 was the largest (33% and 25.2% in phases 1 and 2, respectively) and group 9 was the smallest (<1%). The highest absolute contribution towards CS rate was made by group 5, followed by groups 10 and 2 in both phases. Combined contribution of groups 5, 10, and 2 was 70.5% in phase 1 and 64.9% in phase 2. Group 9 had the highest group-specific CS rate in both phases (93.7% and 100%). In phase 1, it was followed by groups 5 (80.4%), 6 (78.2%) and 7 (72%), while in phase 2 by groups 6 (89.6%), 5 (85.2%), and 8 (72.2%). In CS rate, a statistically significant increase was noted in Robson groups 1, 3, 5, 6, 8, and 10, while a decline was observed in group 4 only. CONCLUSION: The overall CS rate significantly increased over time. This rise was mainly due to a shift in the obstetric population in high-risk groups; 6, 8, 9, 10, and group 5. Conversely, the reduced CS rates in groups 2 and 4 suggest improved induction practices. The data quality was satisfactory.
Bilqis et al. (Sat,) studied this question.