No abstract

Background Because of the strong link between childhood obesity and adulthood obesity comorbidities, and the difficulty in decreasing body mass index (BMI) later in life, effective strategies are needed to address this condition in early childhood. The ability to predict obesity before age five could be a useful tool, allowing prevention strategies to focus on high risk children. The few existing prediction models for obesity in childhood have primarily employed data from longitudinal cohort studies, relying on difficult to collect data that are not readily available to all practitioners. Instead, we utilized real-world unaugmented electronic health record (EHR) data from the first two years of life to predict obesity status at age five, an approach not yet taken in pediatric obesity research. Methods and findings We trained a variety of machine learning algorithms to perform both binary classification and regression. Following previous studies demonstrating different obesity determinants for boys and girls, we similarly developed separate models for both groups. In each of the separate models for boys and girls we found that weight for length z-score, BMI between 19 and 24 months, and the last BMI measure recorded before age two were the most important features for prediction. The best performing models were able to predict obesity with an Area Under the Receiver Operator Characteristic Curve (AUC) of 81.7% for girls and 76.1% for boys. Conclusions We were able to predict obesity at age five using EHR data with an AUC comparable to cohort-based studies, reducing the need for investment in additional data collection. Our results suggest that machine learning approaches for predicting future childhood obesity using EHR data could improve the ability of clinicians and researchers to drive future policy, intervention design, and the decision-making process in a clinical setting.

Hispanic women have a higher prevalence of weight associated complications in pregnancy. This ethnic disparity is likely related to behavior patterns, social circumstances, environmental exposures, and access to healthcare, rather than biologic differences. The objective was to determine associations between sociodemographic characteristics, health behaviors, and psychosocial stressors and gestational weight gain (GWG) in low-income Hispanic women. During pregnancy, information on sociodemographic characteristics, health behaviors, and psychosocial stressors were collected. Linear regression estimated mean differences in GWG by selected predictors. Multinomial logistic regression estimated odds of inadequate and excessive GWG by selected predictors. Five-hundred and eight women were included, 38% had inadequate and 28% had excessive GWG; 57% with a normal pre-pregnancy BMI had inadequate GWG. Compared to women with normal BMI, women with overweight or obesity were more likely to have excessive GWG (aRRR = 1.88, 95% CI: 1.04, 3.40 and aRRR = 1.98, 95% CI: 1.08, 3.62, respectively). Mean total GWG was higher among women who were nulliparous (ß = 1.34 kg, 95% CI: 0.38, 2.29) and those who engaged in ≥3 h of screen time daily (ß = 0.98 kg, 95% CI: 0.02, 1.94), and lower among women who were physically active during pregnancy (ß = −1.00 kg, 95% CI: −1.99, −0.03). Eating breakfast daily was associated with lower risk of inadequate GWG (aRRR = 0.47, 95% CI: 0.26, 0.83). Depressive symptoms and poor adherence to dietary recommendations were prevalent, but none of the psychosocial or dietary variables were associated with GWG. In this cohort of primarily immigrant, low-income, Hispanic women, there were high rates of poor adherence to diet and physical activity recommendations, and a majority of women did not meet GWG guidelines. Modifiable health behaviors were associated with GWG, and their promotion should be included in prenatal care.