- Healthcare providers looking to use mHealth in behavioral health treatment have long struggled to make the technology unobtrusive. But recent research suggests that remote monitoring of people with mental health issues could be done by analyzing their smartphone habits.
A recent report in the journal NPJ Digital Medicine suggests that a smartphone app that collects user activity in the background can help telemental healthcare providers measure that user’s daily habits and detect when those habits deviate. By combining those habits – called digital biomarkers – with other social and environmental data, providers can then identify when mental health problems occur and why they happen.
“We believe that digital biomarkers are the foundation for measurement-based mental healthcare, for which there is a massive unmet patient need,” Dr. Paul Dagum, the study’s lead author and founder and CEO of Mindstrong Health, said in a release. “To provide better mental healthcare, we need better ways to measure cognitive function and brain health that are quantitative, reproducible, continuous and objective.”
“The traditional measures of mental health are patient self-assessments or clinician-administered questionnaires,” added Mindstrong co-founder and President Dr. Tom Insel. “They have relatively low inter-rater reliability, and don’t assess patients in real world settings. Mood and cognitive function vary widely from day to day and during the day, and are subject to a range of environmental factors. Real-time, continuous, ecological measurements of the kind we are identifying are key for enabling a new outpatient care model for mental health patients.”
In Dagun’s study, 27 volunteers underwent a three-hour psychometric assessment before downloading the Mindstrong app, which passively captured the user’s smartphone activity, including swipes, taps and keystroke events. That data was used to identify more than 1,000 distinct digital biomarkers over a seven day period.
The research continues a string of projects tying mHealth devices and digital health platforms to mood measurement. Psychiatrists and other behavioral health providers see the technology as an opportunity to gain insight into a patient’s daily activities and experiences, rather than relying on the patient to recount his or her experiences later on in a doctor’s office. But too often that technology requires patient activation or input, making the patient mindful of the monitoring technology and possibly affecting daily routines.
In Dagun’s research and a few other mHealth programs, researchers are using mobile health technology that basically looks over the patient’s shoulder.
“Smartphone users touch and interact with their devices thousands of times each day, engaging in complex actions and gestures such as typing messages, scrolling or swiping through lists, and making selections - what the company calls human-computer interactions,” Mindstrong executives say in describing how their technology works. “These are complex cognitive activities whose timing and sequencing are modelled using Mindstrong technology in a way that correlates closely with the current standard measures of cognitive function - without logging or analyzing the contents of the device, user messages, geolocation or media.”
Other explorations of the technology include apps that monitor social media messages for key words and actions, smartphone apps that analyze a user’s speech patterns and word choice, and laptop and tablet applications that measure a user’s behaviors and responses to certain programs.
The technology holds promise in telemental health for both adults and children, as well as in testing and treatment for concussions and neurological disorders like Alzheimer’s and Parkinson’s disease.
Among the companies in this space are Ellipsis Health, a San Francisco-based startup whose technology analyzes conversations between patients and their doctors to spot signs of depression in patients, and Cogito, a Boston-based company whose AI-enhanced apps, long used to help call centers in customer relations, are also used to analyze a user’s smartphone habits.
Last year, researchers from the University of Illinois at Chicago, the University of Michigan and Sage Bionetworks created an app that can spot manic and depressive episodes by analyzing how someone uses a smartphone. The BiAffect app, designed for people diagnosed with bipolar disorder, tracks keystroke behaviors and looks for variations common to those experiencing manic episodes, such as typing short, quick sentences or repeatedly ignoring auto-correct suggestions.
Dubbed a “fitness tracker for the brain,” the app won top honors in the Mood Challenge for ResearchKit, a New Venture Fund program funded by the Robert Wood Johnson Foundation.
“During a manic episode, people with bipolar disorder exhibit some common behaviors, such as talking really, really fast, with diminished self-control and flight of ideas,” Alex Leow, an associate professor of psychiatry at the UIC College of Medicine, a professor of bioengineering and computer science and the winning team’s leader, said in a press release issued by UIC. “It is thus natural that they also exhibit similar abnormalities in non-verbal communications that are typed on their phones.”
“People in the midst of a manic episode commonly have reduced impulse control, so it is not surprising that our pilot data supported that they tend to blow through the spell-check alerts,” she added.
In 2016, Massachusetts General Hospital launched a two-year project, funded by a $1.8 million grant from the National Institute of Mental Health, to measure people’s moods by how they use their smartphone. They’re using Cogito’s app.
"The current practice of relying on patient-reported mood ratings via periodic surveys poses a number of limitations to the collected data," Dr. Thilo Deckersback, the study’s principal investigator and Associate Director of the Bipolar Clinic and Research Program at MGH, said in a release. "As such, it is our hope that this inclusive and comprehensive approach to tracking and understanding mood disorders will enable our clinical teams to better assess and understand the disease state and ultimately improve real-world outcomes for millions of patients."