Assessment of the validity of the measurement of newborn and maternal health-care coverage in hospitals (EN-BIRTH): an observational study

Summary Background Progress in reducing maternal and neonatal deaths stillbirths is impeded by data gaps, especially regarding coverage quality of care hospitals. We aimed to assess the validity indicators newborn health-care around time birth survey routine facility register data. Methods Every Newborn-BIRTH Indicators Research Tracking Hospitals was an observational study five hospitals Bangladesh, Nepal, Tanzania. included women their babies who consented on admission hospital. Exclusion critiera at were no fetal heartbeat heard or imminent birth. For uterotonics prevent post-partum haemorrhage, early initiation breastfeeding (within 1 h), bag-mask ventilation, kangaroo mother (KMC), antibiotics for clinically defined infection (sepsis, pneumonia, meningitis), we collected time-stamped, direct observation case note verification as gold standard. compared reported via hospital exit surveys registers standard, pooled using random effects meta-analysis. calculated population-level ratios (measured observed coverage) plus individual-level metrics. Findings 23?471 births 840 mother–baby KMC pairs, verified notes 1015 admitted antibiotic treatment. Exit-survey-reported 99·9% (95% CI 98·3–100) with 100% (99·9–100), but underestimated (84·7% [79·1–89·5]) vs 99·4% [98·7–99·8] observed), ventilation (0·8% [0·4–1·4]) 4·4% [1·9–8·1]), (74·7% [55·3–90·1] 96·4% [94·0–98·6] observed). Early overestimated (53·2% [39·4–66·8) 10·9% [3·8–21·0] “Don’t know” responses concerning clinical interventions more common after caesarean Register (77·9% [37·8–99·5] 99·2% [98·6–99·7] (4·3% [2·1–7·3] 5·1% [2·0–9·6] (92·9% [84·2–98·5] [99·9–100] (85·9% (58·1–99·6) 12·5% [4·6–23·6] Inter-hospital heterogeneity higher register-recorded than report. Even same design, accuracy varied between Interpretation Coverage health had low specific interventions, except self-report KMC, which high sensitivity a ward corner could be considered further assessment. Hospital design completion are less standardised surveys, resulting variable quality, good best performing sites. Because approximately 80% worldwide take place facilities, standardising information systems has potential sustainably improve Funding Children's Investment Fund Foundation Swedish Council.