Data from wearable devices link changes in typical sleep and activity patterns during pregnancy to a risk of premature delivery, according to an NHLBI-funded study.
In the study, published in npj Digital Medicine, researchers gathered data from wearable devices, similar to Fitbits or Apple Watches, from more than 1,000 women during pregnancy. Feeding the data into a new artificial intelligence algorithm, the researchers were able to detect even minute changes in sleep and physical activity during pregnancy. The algorithm also used the data to build a “clock” that was then able to predict how far along a pregnancy was.
In a typical pregnancy, as the women advanced towards later stage pregnancy, they became less physically active and their sleep more disrupted. However, some women’s sleep and activity patterns changed on a faster timeline relative to how far along they were in their pregnancies, sleeping less and becoming more sedentary earlier in their pregnancies. The study found that these individuals were more likely to have a preterm birth.
The results suggest that continuous monitoring of sleep and activity throughout pregnancy may enable doctors to better support their patients who could be at higher risk of preterm birth.