Widespread non-adherence to evidence-based maternity care guidelines: a population-based cluster randomised household survey

Objective To assess the quality of maternity care in an Indian metropolitan city. Study design Three-stage cluster randomised cross-sectional survey. Setting Sixty selected colonies of Delhi. Population One thousand eight hundred and one subjects (of 2286 eligible) were enrolled from 118 446 houses. Women who had delivered a live viable birth in the past 6 months were selected for the study. Methods In stage 1, 20 wards (of 150) were selected using a probability-proportionate-to-size systematic method. In stage 2, one colony from each income stratum (high, middle and low) was selected from each ward by simple random sampling. In stage 3, a house-to-house survey was conducted to recruit 30 women for administering a peer-reviewed and pilot-trialled questionnaire. Main outcome measures Caesarean section rate, induction rate and episiotomy rate. Results National health targets such as iron supplementation advice (>96%), tetanus vaccination (>81%), and ≥3 antenatal visits (>90%) were largely achieved across health care facilities but not in home deliveries. Interventions were lower in public than private hospitals: caesarean section [23.7% (20.2–27.7) versus 53.8% (49.3–58.3)], induction [20.6% (17.5–24.25) versus 30.8% (26.8–33.2)] and episiotomy [57.8% (52.3–63.1) versus 79.4% (71.0–85.9)]. Private hospitals achieved better labour support rates [1.1% (0.5–2.2) versus 14.6% (8.5–24.1)] and pain relief [0.9% (0.4–2.0) versus 9.9 (6.5–14.8)]. Pubic hair shaving [16.2% (11.5–22.5) versus 36.4% (29.9–43.4)], enema [20.2% (15.5–26.0) versus 57.3% (49.5–64.8)], and IV fluids during labour [44.0% (36.2–52.2) versus 38.7% (29.3–49.1)] were widely prevalent in public and private hospitals. Conclusion Present practices fall short of evidence-based guidelines, with relative overuse of interventions in private hospitals and deficiency of patient-centred practices such as labour support in public hospitals.