Siri, the iPhone's smart guide, can tell you what's on your calendar today. But not if the event is an episode of depression, nor can it tell you what to do about it.
A new program for smartphones in the pockets of people with major depression can do both. The software, developed by researchers at Northwestern University, combines data gathered by phone sensors with reports of mood to predict events likely to trigger an emotional plunge. Then the phone suggests ways to ward off the blues. In a small preliminary trial, patients who scored high on a test for major depression didn't even meet the criteria for the disorder after two months.
"I don't want to make huge claims, but we're encouraged," says David C. Mohr, the psychologist who headed the team that developed the program, Mobilyze. In a few weeks he will begin testing an enhanced version of the program, with a larger trial planned for this summer, backed by part of a $5-million grant from the National Institutes of Health. He and others think the program could fill a major health-services gap.
"Health data show that 60 million people in this country have problems that should be treated by a mental-health specialist. Obviously they aren't," because of insurance problems, lack of counselors, and other barriers to access, says Mr. Mohr, a professor of preventive medicine at Northwestern's medical school. "So this technology has the potential of extending treatment to people who are underserved and undertreated."
His is just one of several new attempts to expand treatment for mental illness using smartphones. Others target bipolar disorder and post-traumatic stress disorder. But there are cautions. "You need long-term adherence to these things," says Colin A. Depp, an assistant professor of psychology at the University of California at San Diego who did the work with bipolar patients. "We saw a decrease in symptoms, but none of us have done long trials, or used comparison groups to see if similar patients would get better anyway."
"I am, though, very excited about the potential here," he says. The work at Northwestern, he adds, "is a new frontier because of the way it uses the phone sensors. And ultimately it is automatic and doesn't require the patient to enter any information in order to get help."
Mobilyze starts by asking a little bit of the patient, who keeps a mood diary on the phone, noting when he or she feels high and low. At the same time, the smartphone's 40-odd sensors—GPS and wireless network finders to determine location, accelerometers to detect motion, light sensors, Bluetooth detection of other devices, and more—collect a lot of information. And then they connect the data to the mood ratings, and come up with a predictive model so the patient doesn't have to enter any more information.
"If the system sees that you've been sitting at home for four hours on Saturday morning, it knows that, historically, this leads to depressed feelings later in the day," Mr. Mohr says. "Then it will send you a text message with that observation, and perhaps suggest that you call a friend."
The key points here are that the interventions are customized for each patient, and drawn from a recognized effective therapy technique called behavioral activation. They may sound simple or hokey, but in the initial trial of seven patients, which lasted for eight weeks, scores on several standard depression surveys dropped by more than 50 percent, as did anxiety scores. The computer predictions of mood triggers, like location or social contacts, also proved to be pretty accurate.
The program isn't totally automated. Patients got a weekly call from a counselor to discuss issues. "We're not trying to replace therapists, but augment them," Mr. Mohr says. Mr. Depp adds that "if you embed the program, like the reminders, with language that people wrote themselves, it's a lot more personal."
Mia L. A. Lustria, an associate professor of communication and informatics at Florida State University who studies Internet-based health treatments, also thinks this work shows a lot of promise. But she points to some potential downsides. "If you provide seriously depressed patients with these tools, are they in a position to use them? Do they have the energy and motivation?" Mr. Mohr admits that since he recruited the people in his study, they were likely to be more curious and motivated than the typical depressed patient is.
Mr. Depp also cautions that short-term trials might not reveal device fatigue: People get excited when they are given a new phone with a new program, but after a few weeks the novelty wears off and they may stop using it.
He also notes some advantages, though. "We heard from our participants," he says, "that getting advice from a phone is more benign than, say, getting advice from their mother."
