Prediction of gestational diabetes based on nationwide electronic health records

• Published in: Nature medicine Vol. 26; no. 1; pp. 71 - 76
• Main Authors: Artzi, Nitzan Shalom, Shilo, Smadar, Hadar, Eran, Rossman, Hagai, Barbash-Hazan, Shiri, Ben-Haroush, Avi, Balicer, Ran D., Feldman, Becca, Wiznitzer, Arnon, Segal, Eran
• Format: Journal Article
• Language: English
• Published: New York Nature Publishing Group US 01.01.2020; Nature Publishing Group
• Subjects: 631/114/2397; 692/163/2743; 692/499; 692/699/2743/137/1926; 692/700/459/1748; Accuracy; Area Under Curve; Biomedical and Life Sciences; Biomedicine; Cancer Research; Cohort Studies; Diabetes; Diabetes in pregnancy; Diabetes mellitus; Diabetes, Gestational - diagnosis; Diagnosis; Electronic Health Records; Electronic medical records; Electronic records; Female; Gestation; Glucose; Glucose tolerance; Health aspects; Humans; Infectious Diseases; Israel; Learning algorithms; Letter; Machine learning; Mass Screening; Medical records; Metabolic Diseases; Model accuracy; Molecular Medicine; Neurosciences; Population studies; Predictions; Pregnancy; Pregnancy complications; Pregnant women; Prognosis; Reproducibility of Results; Risk analysis; Risk factors; Risk reduction; ROC Curve; Surveys and Questionnaires; Israel
• ISSN: 1078-8956; 1546-170X; 1546-170X
• DOI: 10.1038/s41591-019-0724-8

Gestational diabetes mellitus (GDM) poses increased risk of short- and long-term complications for mother and offspring 1 – 4 . GDM is typically diagnosed at 24–28 weeks of gestation, but earlier detection is desirable as this may prevent or considerably reduce the risk of adverse pregnancy outcomes 5 , 6 . Here we used a machine-learning approach to predict GDM on retrospective data of 588,622 pregnancies in Israel for which comprehensive electronic health records were available. Our models predict GDM with high accuracy even at pregnancy initiation (area under the receiver operating curve (auROC) = 0.85), substantially outperforming a baseline risk score (auROC = 0.68). We validated our results on both a future validation set and a geographical validation set from the most populated city in Israel, Jerusalem, thereby emulating real-world performance. Interrogating our model, we uncovered previously unreported risk factors, including results of previous pregnancy glucose challenge tests. Finally, we devised a simpler model based on just nine questions that a patient could answer, with only a modest reduction in accuracy (auROC = 0.80). Overall, our models may allow early-stage intervention in high-risk women, as well as a cost-effective screening approach that could avoid the need for glucose tolerance tests by identifying low-risk women. Future prospective studies and studies on additional populations are needed to assess the real-world clinical utility of the model. Leveraging the availability of nationwide electronic health records from over 500,000 pregnancies in Israel, a machine-learning approach offers an alternative means of predicting gestational diabetes at high accuracy in the early stages of pregnancy.