Increase in birthweight coverage of neonatal deaths is needed to monitor low birthweight prevalence in India: lessons from the National Family Health Survey

Abstract

Abstract Background Low birthweight (LBW), defined as birthweight < 2500gms, is the largest contributor to the malnutrition disability-adjusted-live-years in India. We report on the inadequacy of birthweight data, which is a significant barrier in the understanding of LBW epidemiology, to address malnutrition in India. Methods Data from the recent round of the National Family Survey (NFHS-5) were utilised. Birthweight of livebirths in the last 5 years was documented in grams either from the health card or based on mother’s recall. We computed the coverage of birthweight measurement availability and the extent of heaping (values of 2500, 3000 and 3500gms) by the place of delivery and by the survival of newborn during the neonatal period. Heaping of > 55% was considered as poor-quality birthweight data. LBW prevalence per 100 livebirths was estimated and extrapolated for under-reporting of birthweight. Findings are reported for India and its 30 states. Results Birthweight measurement coverage irrespective of the place of delivery was (89·8%; 95% CI 89·7–90) for India, and varied by 2 times among the states with the highest coverage in Tamil Nadu (99·3%) and the lowest in Nagaland (49·7%). Home deliveries had the least coverage of birthweight measurement (49.6%; 95% CI 49.0–50.1) as compared with public health facility (96.3%; 95% CI 96.2–96.3) and private health facility (96%; 95% CI 95.8–96.1) deliveries. This coverage was 66·5% (95% CI 65·2–67·7) among neonatal deaths as compared with 90.4 (95% CI 90.3–90.6) for livebirths who survived the neonatal period for India. The proportion of health card as the data source increased for livebirths born in year 2015 to year 2020 but then dropped for livebirths born in year 2021 (p < 0.001). The proportion of heaping was 52·0% (95% CI 51·7–52·2) in the recorded birthweight for India, and heaping > 55% was seen in 10 states irrespective of the type data source; and 3 states in addition had heaping > 55% in mother’s recall. LBW prevalence was estimated at 17·4% (95% CI 17·3–17·6) for India, and ranged from 4.5% in Nagaland and Mizoram to 22.5% in Punjab for livebirths for whom birthweight was available. We estimated LBW at 77.8% for whom birthweight was not available, and the adjusted LBW prevalence for all livebirths was estimated at 23.5% (95% CI 23.3–23.8) for India. Conclusions Without measuring birthweight for every newborn irrespective of the survival and place of delivery, India may not able to address reduction in low birthweight and neonatal mortality effectively to meet global or national targets.