Predicting postpartum depression

3 September 2020

Alkistis Skalkidou, professor at the Department of Women’s and Children’s Health and researcher of postpartum depression.

Hi Alkistis Skalkidou, professor at the Department of Women’s and Children’s Health and researcher of postpartum depression. Your research team has developed the research app Mom2B for expecting mothers and recent mothers. What is the purpose of this app?

“We want to test a completely new way of collecting data for studies on how women feel in association with their pregnancy and childbirth. We collect different types of data, both what the mothers’ report and passive measurements on how active they are, how much they use their mobile phones, the sound of their voice... All this data is analysed with artificial intelligences to predict which ones will develop symptoms of depression.

How are you using artificial intelligence?
“The data models identify women who will later develop depression based on what is different compared to other women who remain healthy. There are studies that show that a change in your voice and changes in how you move are evident much earlier than depression. The women may begin with a certain pattern of behaviour and then that pattern changes, which indicates that they are developing depression before they are aware of it.”

How common is postpartum depression?
“The numbers vary since it depends on what country you are in and how you measure, but it is around 12 per cent and it is about the same during and after the pregnancy. Depression is one of the most common complications. Often, we think of such complications as diabetes, pre-eclampsia, bleeding, but they are extremely rare compared to this.”

And it has severe consequences?
“Yes, since the mother is in the middle of a period when she needs to care for and has intensive contact with the child. They are often young women in the beginning of a relationship and career. To have a depression at this critical juncture impacts the child’s emotional development, the relationship with the partner, and the woman herself. Once you have a first depression, which is often the case with these women, then you can more easily have additional episodes, so it can have a huge impact.”

Why is it important to be able to predict these depressions?
“We are not very good at identifying high-risk groups within the health care system. We ask if they have had a depression previously, and it is naturally a major risk factor, but for many women this is their first episode.

If we identify a risk group before they become ill, we can focus the health care system’s limited resources on this group as a preventive measure. It can be simple things like a few extra phone calls from the midwife and taking part in discussion groups with others. These are measures that have proven to be effective for high-risk groups.”

How do the expecting mothers benefit from using the app?
“Those who use the app can register important information about their pregnancy and receive information about it and how the child is growing. They also can see statistics about how much they move each day, weight gain, how much they use their mobile phone, and their own assessments of how they feel. If they fill in the surveys and their responses indicate a potential depression, we note this and contact them.

More than 1,000 mothers have registered to use the app, of which 500 have used it in the last week. We want to reach many more and are aiming for 20,000 users throughout the country to allow us to build the mathematical models.”