April 5, 2022
The Electronic Framingham Heart Study (eFHS) Was Designed To Leverage Commercial Wearables and Digital Medical Devices
September 30, 2021
Recruiting since May 2020, the study on mobile technology in mental health care tracks the effects of wearable and mobile technology to both reduce mental health symptoms and predict response to clinic-based treatments.
For the University of Michigan’s PROviding Mental health Precision Treatment (PROMPT) study, Drs. Srijan Sen and Amy Bohnert configured the RKStudio digital clinical research platform to incorporate eConsent, wearable device data (sleep, steps, HR data), and PROs (daily mood, PHQ9, GAD7) to understand patient factors and phenotypes that influence depression and anxiety occurence and test the effectiveness of new digital treatments (e.g. eCBT, meditation app) that may have an impact on a participant’s mental health.
Participants downloaded MyDataHelps from the app store and received surveys at baseline, 6 weeks, 18 weeks, and 12 months in addition to a mood survey they were prompted to complete each day. To date, 92% of enrolled patients submitted six-week follow up surveys, and 81% completed 18-week surveys. The average completion rate of daily mood score surveys during the first six weeks has been 82.6%—a high level of engagement for a daily study.
With the University of Michigan’s PROviding Mental Health Precision Treatment (PROMPT) study, Co-Investigators Amy Bohnert (PhD, Professor of Anesthesiology, Psychiatry and Epidemiology) and Srijan Sen (MD, PhD, Director of the Frances and Kenneth Eisenberg and Family Depression Center) designed the study to address two fundamental problems: the long waits for patients to receive mental health care and standardized traditional treatment approaches that lead to patients cycling through treatments before finding one that works.
Mobile technology has the potential to address the dual problems of limited clinical capacity and inadequate and untimely data, but with little being known about how to most effectively use this technology, Dr. Bohnert and Dr. Sen are working to discover patient-specific treatments to help each individual patient derive the greatest benefit.
Daily Mood Survey Completion Rate
Participants to Date
Participant Age Range
The study was designed to collect a rich array of biologic, behavioral, social, and symptom data from participants, including:
By including genetic samples and measuring sleep & activity, the investigators hope to understand which phenotypes and genotypes are predictive of mental health, so they can create better predictions for which treatments will be most effective for future individual patients and develop deeper knowledge of how mobile interventions and therapeutics can treat patients more quickly and accurately.
We hope to more effectively bring a precision approach to mental health and be able to predict beforehand which treatment will be most effective for a specific patient.
Srijan Sen, MD, Ph.D.
Associate Vice President for Research – Health Sciences
Research Professor, Michigan Neuroscience Institute
Participants were divided into three groups:
Participants randomized to the groups receiving app-based interventions downloaded additional apps to their smartphones designed to address mental health symptoms. Participants were computer-assigned to one of the two app-based interventions (mindfulness via the Headspace app or cognitive behavioral therapy via the SilverCloud app).
There’s suggestive evidence that mobile mental health treatments work, but mostly as short-term trials outside of existing care systems. We want to test mobile interventions in a way that’s integrated into the flow of patient care under real-world conditions. We’re looking at what [mobile interventions] can add for patients who are also going to get therapy or also get medications, depending on what their clinical team decides.
Amy Bohnert, Ph.D.
Professor of Anesthesiology, Psychiatry and Epidemiology,
University of Michigan
Standard Feedback (SF)
All participants received the feedback that is standard with the Apple Watch or Fitbit devices. Both sensors provide feedback on activity level, heart rate, and progress toward daily activity goals on the device itself. For the Apple Watch, this also includes feedback from the iPhone’s “Activity” and “Watch” apps. For Fitbit users, this includes feedback in the “Fitbit” app.
Enhanced Feedback (EF)
Participants received this feedback from the MyDataHelps study app. This includes varying types of text and visual feedback based on data collected through the app. Feedback was displayed to participants on a dashboard in the app or delivered via pop-up notifications. Participants will receive a combination of text and visual feedback.
Examples of potential text feedback:
Participants downloaded the study app, MyDataHelps, on their smartphone. This app serves as a vehicle for eConsent, data collection, delivery of all study questionnaires, automated notifications, and reminders.
The app also received data from the health app and study devices. Participants were encouraged to allow push notifications and were asked to complete daily mental health symptom assessments via the app throughout the duration of their enrollment in the study. The app also presented enhanced feedback on participant progress via a personalized “dashboard.”
Participants were given an Apple Watch or Fitbit, which they wore throughout the entire study—allowing daily collection of activity, sleep & heart rate. Participants were also asked for permission to collect passive data, such as screen time, from their smartphones.
Sen and Bohnert are planning to expand their research into mental health treatments in primary care, and to develop and test a reinforcement learning algorithm to match patients to treatments. Among PROMPT participants to date, there has been substantial decreases in symptoms over six weeks of mobile intervention use, with different patterns of sleep, cardiac, and physical activity changes with different mobile treatment types.