Study shows link between mobility and depression – update

Smartphones could potentially be used to determine the likelihood of depressive symptoms in people. That’s what a recent study from the American Northwestern University suggests. The researchers linked people’s locations to their moods.

The scientists’ researchers asked 40 adults to share GPS data for several weeks. The data came from the open source Purple Robot app for Android, which the scientists developed especially for the study. In the end, 28 adults were able to share the data, because the other participants could not pass on their data correctly. Of the 28 participants, half were ‘healthy’ and the other half had complaints of depression, ranging from mild to severe complaints.

Purple Robot can read out various data from Android phone. For example, he can register where someone comes, how many Facebook friends someone has and whether the smartphone screen is on or off. For the study, only the GPS data and the extent of telephone use were recorded. Purple Robot registered the location every five minutes.

The researchers first looked at which stationary states there were per person. For example, they wanted to know which locations belonged to the work and which to the house. As soon as someone went for a walk and went above a certain threshold value, the steady state shifted. The scientists clustered the locations with k-means, a widely used quantization technique for data research. That way they could determine the exact number of states.

The research showed that someone who exhibits depressive symptoms, compared to a ‘healthy’ person, is more likely to stay in one place. It also seems that he or she uses a telephone more often and for a longer period of time, but follow-up research with more participants should confirm this in order to draw further conclusions.

The researchers published their findings Thursday in the Journal of Medical Internet Research, but the results are also available online. They conclude their research by saying that although the study was on a limited scale, the results could open new doors for healthcare. For example, software can give a warning to a healthcare agency as soon as the location data shows that someone is in danger of becoming depressed.

Update, Saturday 9am: description of the participants clarified.