Development of an infant mortality risk score in Uruguay

• Published in: European journal of public health 2020-09, Vol.30 (Supplement_5)
• Main Authors: Alegretti, M, Leon, I, Aleman, A, Cavallieri, F, Callero, W
• Format: Article
• Language: English
• Subjects: Apgar score; Archives & records; Birth certificates; Birth weight; Births; Childbirth & labor; Children; Data; Educational attainment; Infant mortality; Infants; Maternal and infant welfare; Mortality risk; Neonates; Public health; Regression analysis; Regression models; Risk; Risk levels; Software; Validity; Variables
• ISSN: 1101-1262; 1464-360X
• DOI: 10.1093/eurpub/ckaa166.925

Abstract Background The Ministry of Public Health of Uruguay incorporated a comprehensive home visit before 7 days of discharge to monitor children at risk of infant mortality. In this context, a precise risk stratification of newborns is needed to optimize the implementation of the home visit. Objective Implement a validated infant mortality risk score for Uruguay using the national electronic live birth certificate. Methods Electronic records of newborns from 2014 to 2017 were used to develop the score. The variables of the electronic live birth certificate were considered for the model and data of Infant mortality was obtained from the national mortality registry. A multivariate binary logistic regression model was estimated with a random sample of 80% of the cohort, the remaining 20% was the validation set. ROC curve analysis was performed. R software was used. Results The 2014-2017 birth cohort contains 187,388 records. 1307 children under one year died (IMR 6.97 per 1,000 births). The variables included in the final model were birth weight, APGAR score at 5 minutes, number of prenatal visits, maternal educational level and father living at home. The area under the curve (AUC) was 89%, CI 95% [87% - 91%]. Two cut-off points were defined: 0.4% and 2%. Less than 0.4% was considered low risk (IMR 1.4 per 1,000 births), between 0.4 and 2% was considered intermediate risk (IMR 7.1 per 1,000 births) and more than 2% was considered high risk (IMR 99.2 per 1,000 births). Conclusions The score identifies high-risk newborns at the time of entering the data in the electronic live birth certificate. This information could be used to plan and implement the home visit and other actions, according to the level of risk identified. Key messages In Uruguay, high-risk newborns can be identified using data collected routinely. The procedure could be applied in other countries with electronic birth certificate.