More than 21% of adults in the United States experience depression, which significantly impacts their quality of life. Dr. Jyoti Mishra, associate professor of psychiatry at the University of California, San Diego School of Medicine, says many people with mild to moderate depression can improve their symptoms by making adjustments to their daily habits, including sleep, exercise, diet and social interactions. However, depression varies greatly from person to person, so a one-size-fits-all lifestyle approach is not very effective.
In a first-of-its-kind study, Mishra and her team developed a machine learning-powered lifestyle coaching program based on data about participants’ moods and daily habits collected via personal devices. They found that participants who implemented the program had significantly reduced symptoms of depression after six weeks. The results of this study provide a promising approach for remotely delivering personalized depression treatment tailored to each individual’s circumstances. This research NPP – Digital Psychiatry and Neuroscience.
For two weeks, 50 adults with mild to moderate depression wore smartwatches that tracked their heart rates and exercise levels. They also tracked their mood and answered up to four short questions per day about their sleep quality, diet, activity level, and how often they talked to friends and family.
The team developed a machine learning model specific to each participant based on this data to discover which lifestyle factors best predicted an individual’s depressed mood. Each participant then worked with a health coach to implement an Individualized Mood Enhancement Plan (iMAP).
“Our goal was to understand the main lifestyle factors that cause depression, which vary from person to person, and to see whether targeting those factors through personalized coaching could actually make people feel better,” said Mishra, co-director of the Neural Engineering and Translation Laboratory (NEATLabs) at the University of California, San Diego.
Over the next 6 weeks, participants worked with their coach to implement iMAP.
Each person in the trial received a different behavioral treatment, which has already been established in the literature, depending on their top predictors. Some were working on cognitive behavioral therapy programs for insomnia, while others were implementing food-based interventions to maximize the physical activity they already did in their daily lives, strengthen social connections, and feel healthy. ”
Dr. Jyoti Mishra, Associate Professor of Psychiatry, University of California, San Diego School of Medicine
After working with a coach through short video calls for six weeks, participants:
- It has been reported that symptoms of depression have been significantly reduced. Fifty-five percent of participants no longer suffered from depression after treatment, as measured by the Patient Health Questionnaire-9 (PHQ-9), a standardized depression screening test.
- They reported a 36% reduction in anxiety symptoms as measured by the Generalized Anxiety Disorder-7 (GAD-7) screening test.
- They reported a significant improvement in their quality of life.
- They scored higher on simple memory and attention tests.
Additionally, the researchers found that treatment effects persisted for three months when participants continued to be followed after the intervention ended.
“Clinical trials show that most current interventions are only about 30% effective on average in remitting depression. Here, we see that effectiveness nearly doubled by targeting key lifestyle predictors with personalized, data-driven coaching,” said Mishra.
Mishra believes this intervention may be more effective because it departs from common behavioral health recommendations.
“We all know we need to eat healthier, get eight hours of sleep, and exercise 150 minutes a week,” she says. “But I think personalized insights can be more empowering than these general guidelines because they’re less overwhelming. When a person is in a state of depression, it’s impossible to change everything in your life. You’re just trying to survive and function day to day.”
Although small, this study provided the first evidence that digital monitoring, insights gained from machine learning, and brief, personalized weekly coaching delivered remotely may be a promising integrated approach for addressing mild to moderate depression in large populations. Larger controlled studies of this individualized treatment approach are needed to validate the results.
sauce:
University of California, San Diego
Reference magazines:
Yin, J. Others. (2026). A personalized machine learning intervention to optimize lifestyle behaviors in patients with depression: A pilot study. NPP—Digital Psychiatry and Neuroscience. DOI: 10.1038/s44277-026-00062-3. https://www.nature.com/articles/s44277-026-00062-3
