A study from Washington University in St. Louis reveals that sleep variability—captured via wearable devices—may help predict preterm birth risk. Researchers analyzed actigraphy data from 665 pregnant individuals, focusing on sleep duration, timing, and consistency during the first two trimesters. They found that irregular sleep patterns were more predictive of preterm birth than average sleep quality or duration.
Using machine learning, the team combined wearable data with patient surveys to build predictive models. The results suggest that consistent sleep schedules may be more protective than simply getting more sleep. This insight opens the door to early, non-invasive interventions that could reduce the risk of preterm delivery, which remains a leading cause of infant mortality worldwide.
The researchers plan to validate their findings across broader populations and explore how behavioral interventions might stabilize sleep patterns during pregnancy. This work exemplifies how everyday wearables can become powerful tools in maternal-fetal medicine, offering personalized insights long before symptoms arise.
Article from WashU: Sleep data from wearable device may help predict preterm birth
Abstract from npj Women’s Health: Validation of sleep-based actigraphy machine learning models for prediction of preterm birth
