Depression monitoring — there might be an app for that

Advertisement

Someday, a smartphone application that asks what you’re feeling 10 times a day may tell you if you’re edging closer to depression — and recommend that you seek therapy or drugs.

Scientists have discovered that how quickly someone bounces back from negative feelings, over hours or days, can predict whether that person is at risk of an episode of major depressive disorder.

“The holy grail of depression epidemiology is that we want to intervene early to prevent people from having depressive episodes,” says social scientist Stephen Gilman of Harvard University, not involved in the study. “Where this work is headed is making an advance in that direction, toward early detection and therefore early intervention.”

Researchers asked more than 600 people — some healthy and some with a diagnosis of depression — to track their emotions for five or six days. Ten times a day, at random intervals, a watch would beep and the subjects would record how strongly they identified with each of four emotions: cheerful, content, sad, and anxious.

Six to eight weeks later, participants filled out a more detailed questionnaire that rated their levels of clinical depression.

By the end of the follow-up period, about 13 percent of the subjects had experienced a shift toward being more depressed, a number consistent with what would be expected in the general population.

Trends in the daily mood records, the team discovered, could predict whether a previously healthy person would make that shift toward depression.

Mathematically, it turns out, the shift from a healthy state to a depressed state resembles other so-called tipping points — moments of critical mass where a system, such as changes to Earth’s climate or a social trend — shift rapidly from one state to another. Theories on tipping points suggest that as a system nears a tipping point, it becomes less resilient.

“In any system, if you push the system a little bit out of equilibrium, then the closer it is to the tipping point, the longer it takes to return to equilibrium after that perturbation,” explains Ingrid van de Leemput, an ecologist at Wageningen University and Research Centre in the Netherlands who led the new work.

Indeed, the longer a patient took to recover from feelings of sadness and anxiety, the more likely they were to be more depressed by the end of the study, suggesting that they were closer to a tipping point between health and depression, her team reported online Monday in the Proceedings of the National Academy of Sciences.

The results matched with a mathematical model that the researchers had previously created to represent how emotional swings could signal an impending tipping point.

“If a healthy person has an unpleasant call with their employer, they will be unhappy about it and dwell on it for 10 or 20 minutes but be done with it fairly quickly,” says psychiatry researcher Angelique Cramer of the University of Amsterdam, who collaborated with Van de Leemput. “What you see in people who are about to become depressed is that the next day, they are still sad about a phone call the day before.”

Over time, she says, various symptoms of depression — negative mood, fatigue and concentration problems, for example — can create a negative feedback loop that causes full-blown disease. Cramer says the new research could lead to new ways for psychiatrists to track their patients’ well-being.

“I think this could open up new avenues of research in many ways,” Gilman says. He’d like to see the work expanded to include more variables that are already known to increase depression risk, such as family history, previous episodes of major depression, and social factors.

“Really what we want to know is where on the distribution of sadness and mood is the dividing line between a serious depressive episode and nondepression,” he said. “And are there factors that can push people further from or toward that dividing line?”

Comments

Use the comment form below to begin a discussion about this content.

Sign in to comment

Click here to sign in