News

New study will combine patients’ self-reports with data from mobile apps, wearable devices, genetic samples and clinical records to improve the success of first-time mental health treatment.

People with depression, anxiety, and other mental health concerns face many options to try and ease their symptoms, from talk therapy and medication to movement, sleep and mindfulness.

However, finding an option or combination that works can take months or years for each person because no one combination has proven best across the board. And it can take weeks to tell if a mental health treatment is helping or not.

Plus, searching for an effective combination only starts after someone finds a mental health care provider who has appointments available and accepts their insurance.

All of this means lost time, money and – in the most serious cases – loss of life.

A new $17.9M grant to University of Michigan researchers aims to change that.

The team will enroll thousands of patient volunteers by invitation in the period before they start mental health care at U-M Health. The new study is called COMPASS, for Comprehensive Mobile Precision Approach for Scalable Solutions in Mental Health Treatment.

The COMPASS team will harness massive amounts of mobile technology, genomic, behavioral and electronic health record data to create decision aids that they hope will make it easier to predict which approaches will work best for an individual.

Their goal: the same kind of precision for mental health care that patients and providers already have for the treatment of many types of cancer, heart disease, diabetes and more.

The new grant comes from the National Institute of Mental Health, part of the National Institutes of Health. It recognizes the promise shown in previous U-M research involving a smaller group of people.

How the study will work

Soon, thousands of patients seeking mental health care at a wide range of U-M Health outpatient clinics and U-M student-focused clinics will be offered the opportunity to take part in the COMPASS study. This includes those using a collaborative care model in which psychiatrists support U-M Health primary care providers in treating patients with mild to moderate mental illnesses.

Both patients who have never received mental health care and those who have previously received mental health care from non-U-M providers will be eligible to enroll in the new study in the weeks before their first appointment with a U-M provider. The researchers will ask for permission to access information about any mental health medications they’ve taken in the past.

If patients agree to enroll in COMPASS, they’ll get free access to mobile mental health apps while awaiting their first appointment with their U-M provider. They will also receive wearable trackers and take surveys and genetic tests over the course of a year.

Rolling this information together with patients’ response to treatments, the researchers hope, will pave the way to use data science and machine learning tools to search for patterns and predictions.

More about the study

The COMPASS team is led by researchers from U-M Medical School and the School of Public Health, who are members of the U-M Eisenberg Family Depression Center, U-M Precision Health, the U-M Opioid Research Institute, the Michigan Neuroscience Institute and the U-M Institute for Healthcare Policy and Innovation. They received the grant as part of NIMH’s IMPACT Mental Health effort.

“We need better ways of predicting which patients are likely to get better with digital solutions and lifestyle changes, which ones need specific medications or specific types of therapy, and which might need more involved treatments such as ketamine, rTMS or ECT,” said Srijan Sen, M.D., Ph.D., co-leader of the new study, director of the Eisenberg Family Depression Center, and a psychiatrist and neuroscientist at the Medical School. Both rTMS, which stands for repetitive transcranial magnetic stimulation, and ECT, which stands for electroconvulsive therapy, are approved for severe depression and other conditions that don’t respond to other treatments.

The team secured the funding based on the promising results of previous initiatives, such as the PROviding Mental health Precision Treatment (PROMPT) study and the Mental Health Biobank, which were funded by the Eisenberg Family Depression Center, Precision Health and other sources.

The first data from the PROMPT study were presented in early June at the national Society for Ambulatory Assessment meeting at U-M and are in the process of being published in a peer-reviewed journal.

The research team also hopes that the patient volunteers recruited and data collected through COMPASS can serve as a foundation for other U-M researchers interested in studying specific patient populations or data types included in the study.

The U-M team will collaborate with teams at other universities receiving IMPACT-MH funding from NIMH to learn from one another’s findings and validate new digital tools.

Weaving genetics with other information

Sen notes that the genetic component of the study seeks to go beyond the current commercial genetic tests that use metabolism-related variations in DNA to try to predict which mental health medications might cause more or fewer side effects in an individual.

Instead, the team will use a broader assessment of each volunteer’s genome to help predict response to treatments.

Project co-leader Lars Fritsche, Ph.D., associate research scientist in the U-M Department of Biostatistics in the School of Public Health, is an expert in statistical genetics and will bring that knowledge to the study.

“By integrating genetics, especially pharmacogenetic and polygenic predictors, with electronic health records and behavioral data, this innovative and comprehensive approach will improve predictions for treatment responses and outcomes and ultimately advance personalized mental health care,” Fritsche said.

“Our goal is to combine information to build tools that will help providers choose the right treatment and support program for the right person at the right time,” said Amy Bohnert, Ph.D., M.H.S., co-leader of the project and a professor in the Department of Anesthesiology at the Medical School. “We want to help patients get to a point of recovery faster.”

Filling the gaps in mental health care

The importance of reducing time to recovery can solve fundamental problems in access to mental health services.

Currently, high demand for mental health care from U-M psychiatrists, psychologists, mental health nurse practitioners and clinical social workers means multiple weeks or even months can elapse between the time a person calls for an appointment and the day they’re seen.

In fact, researchers say this gap between contact and care is an especially important time to use other tools to support patients. It’s common nationwide to have such gaps because of the rise in people seeking mental health care. 

Mental health smartphone apps have the potential to help fill this gap.

But while hundreds of mobile apps claim to impact mental health, there is little actual evidence showing which ones are most effective and which are most likely to work for which person.

Sen said the COMPASS team will assess mental health apps and hopes to facilitate the use of these tools to help meet the demand for mental health interventions.

He and his colleagues also envision a clinician dashboard to help providers see composite data from their patients’ smartphone apps and wearable devices in real time. This dashboard can help monitor when patients are doing well and when their mental health may be declining, signaling that they need more immediate help.

Such dashboards have already become helpful in the ongoing management of other chronic conditions, such as high blood pressure, diabetes and asthma.

Another goal is to help patients understand how to use their personal devices in ways that will benefit them the most, such as tracking and setting goals for sleep, exercise, natural light exposure and screen time.

The COMPASS study is supported by grant U01MH136025 from the National Institute of Mental Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.