Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
The aim of this study is to evaluate the efficacy of using a reinforcement learning algorithm to determine the optimal content of a mobile health intervention (message delivered via smartphone) for improving the mood, physical activity, and sleep of medical interns.
Due to their high workloads, less sleep and physical activity and other stressors, medical interns suffer from depression at higher rates than the general population. The goal of this study is to evaluate the efficacy of a mobile health intervention intending to help prevent the degradation of health behaviors and the development of depression. The intervention sends mobile phone notifications which aim to help interns improve their mood, maintain physical activity, and obtain adequate sleep during their internship year. A reinforcement learning algorithm will use prior survey, daily mood, and wearable data to decide each day whether to send a message or not.
Not provided
Not provided
Not provided
Not provided
Not provided
| Label | Type | Description | Intervention Names |
|---|---|---|---|
| Within-participant micro-randomization | Experimental | Each week a policy outcome is chosen at random with â…“ mood, â…“ activity, â…“ sleep - this determines which category of message a participant will receive. Each day in the study, a reinforcement learning algorithm will determine if a participant will receive a notification that day or no notification that day. If assigned to receive a notification, 1 core message set will be randomly selected from a pool of 358 core message sets. Each core message set will be comprised of 4 messages containing comparable content, however they will be tailored based on the participant's wearable (steps, sleep) or mood data for the specified time interval (7 days, 30 days, since the start of internship) as follows: 1) no data, 2) low, 3) moderate, or 4) high. |
|
| Name | Type | Description | Arm Group Labels | Other Names |
|---|---|---|---|---|
| Intern Health Study behavioral change mobile notification | Behavioral | The study's mobile app will be used to deliver push notifications. The notifications appear on the participant's phone lock screen. The notifications include 3 categories: mood notifications, activity notifications, sleep notifications. Mood notifications aim to increase the participant's mood. Activity notifications aim to increase the participant's physical activity. Sleep notifications aim to increase the participant's sleep duration. All notifications are categorized as one of five therapeutic approaches: 1) CBT-Behavioral, 2) CBT-Cognitive, 3) Distanced Self-Talk, 4) Mindfulness, 5) Motivational Interviewing. |
| Measure | Description | Time Frame |
|---|---|---|
| Average daily mood | Through the mobile app, participants enter a mood score (scale 1 - 10) every day of the study. 1 corresponds to lowest mood and 10 corresponds to highest mood. | Daily, through study completion at the end of intern year (1 year) |
| Average daily step count | Participant's daily step counts are recorded through a fitness tracker. High step counts are considered a positive outcome as it indicates more physical activity. | Daily, through study completion at the end of intern year (1 year) |
| Average nightly sleep duration | Participant's nightly sleep duration (in minutes) is recorded through a fitness tracker. High sleep duration is considered a positive outcome. | Daily, through study completion at the end of intern year (1 year) |
| Patient Health Questionnaire-9 (PHQ-9) | Prior to the start of the intervention and at quarterly intervals throughout internship year, all participants complete the Patient Health Questionnaire 9 (0-27). Higher scores on the PHQ-9 correspond to a larger number of depressive symptoms (worse outcome). | Quarterly (every 3 months for 1 year) |
Not provided
Not provided
Inclusion Criteria:
Exclusion Criteria:
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
| Facility | Status | City | State | ZIP | Country | Contacts |
|---|---|---|---|---|---|---|
| University of Michigan | Ann Arbor | Michigan | 48375 | United States |
Not provided
| Label | URL |
|---|---|
| Sen Lab Website | View source |
| Intern Health Study Participant Website | View source |
Not provided
De-identified survey data (baseline survey, plus quarterly survey which contains the PHQ-9) will be shared via the ICPSR repository (https://www.openicpsr.org/openicpsr/project/129225/version/V1/view).
Not provided
Data will be made available 12 months after the end of the study It will be made available indefinitely after that date.
Deidentified data will be publicly available via ICPSR https://www.openicpsr.org/openicpsr/project/129225/version/V1/view
Not provided
| ID | Term |
|---|---|
| D003863 | Depression |
| D003865 | Depressive Disorder, Major |
| D009043 | Motor Activity |
| ID | Term |
|---|---|
| D001526 | Behavioral Symptoms |
| D001519 | Behavior |
| D003866 | Depressive Disorder |
| D019964 | Mood Disorders |
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
|
| D001523 |
| Mental Disorders |