Your smartphone data can diagnose depression with 87% accuracy

Depression is a huge societal problem, but for many it’s also an invisible one. People can fake being “OK” to the degree that even friends and family may be unaware of their misery, but, according to new research from Northwestern University, your smartphone could be more in tune with your emotions than the people around you.

The researchers studied location and activity data taken from a group of participants, of which half had depression. They discovered that by correlating the amount of time spent interacting with the phone alongside GPS data, they could determine which were depressed with an 87% accuracy. That’s a better strike-rate than the pen-and-paper daily PHQ-9 survey routinely used in mental health assessment, which the researchers also administered to the group.

“The significance of this is we can detect if a person has depressive symptoms, and the severity of those symptoms, without asking them any questions,” explained the research paper’s senior author David Mohr. “We now have an objective measure of behavior related to depression – and we’re detecting it passively. Phones can provide data unobtrusively and with no effort on the part of the user.”

The researchers found that on average, participants without depression would spend just 17 minutes on their phone on a daily basis, compared to 68 minutes for the depressed subjects.

They were also able to correlate this against location data, knowing that people with depression on average visit fewer locations, leave the house less frequently and keep more irregular hours.

The next step for the researchers is to see whether nudging the depressed subjects’ behaviour to mirror the non-depressed ones’ routines would improve their mood. “We will see if we can reduce symptoms of depression by encouraging people to visit more locations throughout the day, have a more regular routine, spend more time in a variety of places or reduce mobile phone use,” explained Sohrob Saeb, a lead author on the paper.

While the research presents some interesting possibilities about the role of smartphones in passively diagnosing mental illness, it’s important to note that this was a pretty small study: just 28 participants were involved, of which 14 were depressed.

There were also 20 women compared to just eight men, with an average age of 29 – and gender and age are just two variables that could lead to very different conclusions. Speaking as someone who routinely spends two hours of every 24 commuting, 67 minutes of smartphone usage sounds a touch low to me – and I’m, on balance, as happy as Larry. So let’s file this under “interesting, but needs more study” for now.

Images: Andy Rennie and Esther Vargas used under Creative Commons