Background Currently, postpartum depression (PPD) screening is mainly based on self‐report symptom‐based assessment, with lack of an objective, integrative tool which identifies women at increased risk, before the emergent of PPD. We developed and validated a machine learning‐based PPD prediction model utilizing electronic health record (EHR) data, and identified novel PPD predictors. Methods A nationwide longitudinal cohort that included 214,359 births between January 2008 and December 2015, divided into model training and validation sets, was constructed utilizing Israel largest health maintenance organization's EHR‐database. PPD was defined as new diagnosis of a depressive episode or antidepressant prescription within the first year postpartum. A gradient‐boosted decision tree algorithm was applied to EHR‐derived sociodemographic, clinical, and obstetric features. Results Among the birth cohort, 1.9% (n = 4104) met the case definition of new‐onset PPD. In the validation set, the prediction model achieved an area under the curve (AUC) of 0.712 (95% confidence interval, 0.690–0.733), with a sensitivity of 0.349 and a specificity of 0.905 at the 90th percentile risk threshold, identifying PPDs at a rate more than three times higher than the overall set (positive and negative predictive values were 0.074 and 0.985, respectively). The model's strongest predictors included both well‐recognized (e.g., past depression) and less‐recognized (differing patterns of blood tests) PPD risk factors. Conclusions Machine learning‐based models incorporating EHR‐derived predictors, could augment symptom‐based screening practice by identifying the high‐risk population at greatest need for preventive intervention, before development of PPD.

Objective To examine the association of maternal and/or paternal smoking during pregnancy with offspring cardio-metabolic risk (CMR) factors at adolescence and early adulthood, taking into account socio-demographic, medical and lifestyle characteristics of parents and offspring, as well as offspring common genetic variation. Methods We used a population-based cohort of all 17 003 births in Jerusalem during 1974–76, with available archival data on parental and birth characteristics. Measurements at age 17 were assessed at military induction examinations for 11 530 offspring. 1440 offspring from the original 1974–1976 birth cohort were sampled using a stratified sampling approach, and were interviewed and examined at age 32. Parental smoking during pregnancy (i.e. maternal, paternal and any parent) was primarily defined dichotomously (any number of cigarettes smoked daily by mother or father during pregnancy vs. non-smokers). Additionally, smoking was assessed by quantity of cigarettes smoked daily. Linear regression models were used to evaluate the associations of parental smoking during pregnancy with various offspring CMR factors, after controlling for potential confounders and for genetic variation in candidate genes. Results Prevalence of exposure to parental smoking in-utero (i.e. smoking of any parent) was 53.2% and 48.4% among the 17 years old and 32 years old samples, respectively. At age 17, smoking of at least one parent during pregnancy was significantly associated with weight (B=1.39), height (B=0.59), BMI (B=0.32) and pulse rate (B=−0.78) (p-values<0.001). At age 32, parental smoking, adjusted for covariates, was associated with 2.22 kg higher mean offspring weight, 0.95 cm higher mean offspring height, 0.57 kg/m2 higher BMI, and 1.46 cm higher waist-circumference (p-values≤0.02). Similar results, reflecting a dose response, were observed when maternal and paternal smoking were assessed by number of cigarettes smoked daily. Conclusions This prospective study demonstrates a potential long-term adverse effect of parental smoking during pregnancy on offspring health and calls for increasing efforts to promote smoking cessation of both parents before pregnancy.

IntroductionStudies demonstrate associations between changes in obesity-related phenotypes and cardiovascular risk. While maternal pre-pregnancy BMI (mppBMI) and gestational weight gain (GWG) may be associated with adult offspring adiposity, no study has examined associations with obesity changes.ObjectivesWe examined associations of mppBMI and GWG with longitudinal change in offspring's BMI (ΔBMI), and assessed whether associations are explained by offspring genetics.Design and MethodsWe used a birth cohort of 1400 adults, with data at birth, age 17 and 32. After genotyping offspring, we created genetic scores, predictive of exposures and outcome, and fit linear regression models with and without scores to examine the associations of mppBMI and GWG with ΔBMI.ResultsA one SD change in mppBMI and GWG was associated with a 0.83 and a 0.75 kg/m2 increase in ΔBMI respectively. The association between mppBMI and offspring ΔBMI was slightly attenuated (12%) with the addition of genetic scores. In the GWG model, a significant substantial 28.2% decrease in the coefficient was observed.ConclusionsThis study points to an association between maternal excess weight in pregnancy and offspring BMI change from adolescence to adulthood. Genetic factors may account, in part, for the GWG/ΔBMI association. These findings broaden observations that maternal obesity-related phenotypes have long-term consequences for offspring health.

No abstract