Predicting postpartum depression with the Mom2B app: a large-scale Swedish study using artificial intelligence to improve mothers´ mental health

Peripartum depression (PPD) is a common, serious and potentially life-threatening disorder with high societal costs.

Preventive efforts targeting PPD are cost-effective only among high-risk groups, but our ability to predict PPD is poor.

This project uses a unique constellation of researchers, resources and methodologies available at Uppsala University and proposes a transdisciplinary project to design, develop and evaluate effective and user-centered methods to predict depression around childbirth.

The accuracy of machine learning methods in predicting the development of PPD in the postpartum period, using data collected during pregnancy and early postpartum will be assessed.

We will draw data from self-reports, geographical movement patters, general smartphone activity, internet usage, social media activity and voice recordings from 20,000 pregnant women who are being recruited to the Mom2B smartphone application study (>450 already recruited, analyses planned for 2023-3024).

In a sub-study, we will use qualitative methods to explore the experiences, attitudes and concerns of 60 participants towards using the Mom2B application (2021-2022).

Identifying an appropriate high-risk group for PPD preventive interventions will lead to better prevention outcomes and cost effectiveness. Results could be implemented into the national health care system to inform clinical decisions. Further, the results are expected to be informative also for depression beyond the peripartum period.