The use of wearables to monitor health over the course of a patient’s day has grown sharply in recent years, and a new federally backed initiative aims to test mHealth monitoring in behavioral health care.
The initiative, run by Massachusetts General Hospital and its patient-driven research network, the MoodNetwork, will test the use of technology that monitors patients' voices, texts and call patterns to predict the recurrence of depression and mania. The National Institute of Mental Health funded the work with a $1.8 million grant, and comes amid a larger movement by the federal government and organizations like the American Hospital Association to improve behavioral health care.
The approach could circumvent a problem with traditional mood-tracking apps that don’t always apply to approaches that track such things as heart rate or blood pressure. “People are not very good at monitoring their mood,” says Thilo Deckersbach, associate director of Mass General’s Bipolar Clinic and Research Program and site principal investigator of the study. “They’re not going to do it on a regular basis,” Deckersbach says.
The way this approach is structured, patients have to do essentially nothing but be themselves. The predictive algorithm embedded in their smartphones will gather data points to identify patterns that predict when depression or a manic episode is likely to be coming.
The key to the algorithm, which is produced by behavioral analytics company Cogito Corp., is to figure out when certain behaviors monitored through the phone escalate from being normal activity to being a predictive action. “It’s like your fire alarm. You don’t want your fire alarm going off three times a day because you will stop paying attention to it,” Deckersbach says. “The models that ultimately do the mathematics need to be very sensitive and they need to be reliable,” he says.
According to Cogito, its predictive behavioral models rely on data that have been validated through work with NIMH, the Defense Advanced Research Projects Agency and the Veterans Affairs department.
Also key to the initiative is the role of the nationally focused MoodNetwork, because the 1,000 test patients will come from its ranks.
This will be the first NIMH study to rely on the national network of behavioral health patients, which itself is a by-product of the Patient-Centered Outcomes Research Institute, Deckersbach says.
Members of the network not only will be monitored — with full knowledge of what is happening — but they helped to create the test, he says. The mission of the network is to bring together people with mood disorders, physicians and researchers.
“When we design a study, it’s not just the researchers anymore,” he says. “That’s what’s really exciting about this study.”