LAUSR

Birth weight is a consistent predictor of morbidity and survivability in infancy and later life. This study aims to assess the accuracy of the mother’s perception of size at birth to predict low birth weight(LBW). This study used data from Nepal Demographic and Health Survey (NDHS 2016). Information about 5060 mother pairs was obtained from the NDHS dataset. However, birth weight data were available for 3095 children, and therefore they were only included in the further analysis. The predictive accuracy of the mother’s perception of size at birth to predict LBW was measured by sensitivity, specificity, positive predictive value, and negative predictive value. Factors associated with the discordance among the mother’s perception of birth size and birth weight were calculated using multinomial logistic regression analysis. The mother’s perception of birth size had low sensitivity (62%) and positive predictive value (46.7%) but high specificity (90.1%) and negative predictive value (94.4%) to predict the LBW. The overall agreement between birth weight(<2500gram vs > = 2500 grams) and the mother’s perceived size at birth (small vs average or above average) was 86% (Kappa = 0.45, 95%CI: 0.40–0.51), which is composed of a higher share of the agreement to identify non-LBW babies(79%) and a low share to identify LBW babies (7%). Among the five categories of mothers’ perception of size at birth and birth weight, the agreement was 67.2% (Kappa = 0.29, 95% CI: 0.26–0.33). Education status, ethnicity, multiple births, and sex of the newborn child were significantly associated with the discordance between the mother’s perceived size at birth and birth weight. A moderate agreement was found among the mother’s perception of birth size and birth weight. Mothers were more likely to correctly identify non-LBW babies compared to LBW babies based on their perception of size at birth. Efforts should be intensified to promote the practice of weighing the baby at birth.