Building a predictive model of low birth weight in low- and middle-income countries: a prospective cohort study

Abstract

Background: Low birth weight (LBW, < 2500 g) infants are at significant risk for death and disability. Improving outcomes for LBW infants requires access to advanced neonatal care, which is a limited resource in low- and middle-income countries (LMICs). Predictive modeling might be useful in LMICs to identify mothers at high-risk of delivering a LBW infant to facilitate referral to centers capable of treating these infants.

Methods: We developed predictive models for LBW using the NICHD Global Network for Women's and Children's Health Research Maternal and Newborn Health Registry. This registry enrolled pregnant women from research sites in the Democratic Republic of the Congo, Zambia, Kenya, Guatemala, India (2 sites: Belagavi, Nagpur), Pakistan, and Bangladesh between January 2017 - December 2020. We tested five predictive models: decision tree, random forest, logistic regression, K-nearest neighbor and support vector machine.

Results: We report a rate of LBW of 13.8% among the eight Global Network sites from 2017-2020, with a range of 3.8% (Kenya) and approximately 20% (in each Asian site). Of the five models tested, the logistic regression model performed best with an area under the curve of 0.72, an accuracy of 61% and a recall of 72%. All of the top performing models identified clinical site, maternal weight, hypertensive disorders, severe antepartum hemorrhage and antenatal care as key variables in predicting LBW.

Conclusions: Predictive modeling can identify women at high risk for delivering a LBW infant with good sensitivity using clinical variables available prior to delivery in LMICs. Such modeling is the first step in the development of a clinical decision support tool to assist providers in decision-making regarding referral of these women prior to delivery. Consistent referral of women at high-risk for delivering a LBW infant could have extensive public health consequences in LMICs by directing limited resources for advanced neonatal care to the infants at highest risk.

Keywords: Low birth weight; Low-income country; Preterm; Small for gestational age.

Fig. 1

CONSORT diagram depicting reasons for exclusion and outcome for analysis population

Fig. 2

Receiver operator characteristic (ROC) curves for the predictive models

Fig. 3

Calibration curves for the predictive models. The Y-axis is the true fraction of newborns who are low birth weight (LBW) and the X-axis is the model-predicted probability of being LBW. The worst performing model was k-nearest neighbors; the near-horizontal line for this curve indicates the model will predict a consistent LBW percentage of around 15% regardless of the true incidence of LBW. The best performing model was the linear support vector machine, which predicts nearly perfectly for the lowest incidence rates and begins to diverge around 40% incidence

Fig. 4

Permutation-based feature importance for the logistic regression model. The permutation-based importance was implemented in Scikit-Learn as permutation_importance method. This method randomly shuffles each feature and computes the change in the model’s performance. The features which impact the performance the most are the most important ones. The score is how the variable compares to other variables in the model. Thus, a high score for any level of a categorical variable indicates the entire variable is important. For clinical sites, the reference group is Belagavi, India. For maternal age, the reference group is 20–35 years. For maternal education, the reference group is University + . For parity, the reference group is parity of 1. For socio-economic status, the reference group is 66 + . For previous livebirth, yes is the reference group. For antenatal care visits, the reference group is 4 + visits