Mobile Self-Tracking

Serious mental illnesses require years of monitoring and adjustments in treatment. Stress, substance abuse or reduced medication adherence cause rapid worsening of symptoms, with consequences that include job loss, homelessness, suicide, incarceration, and hospitalization. Treatment visits can be infrequent. Illness exacerbations usually occur with no clinician awareness, leaving little opportunity to make treatment adjustments. Tools are needed that quickly detect illness worsening. At least two thirds of Veterans with serious mental illness use a smart phone. These phones generate data that characterize sociability, activity and sleep. Changes in these are warning signs for relapse. Members of this project developed an app that monitors and transmits these mobile data. This project studies passive mobile sensing that allows Veterans to self-track their activities, sociability and sleep; and studies whether this can be used to track symptoms. The project intends to produce a mobile platform that monitors the clinical status of patients, identifies risk for relapse, and allows early intervention.

Background: Serious mental illnesses are common, disabling, challenging to treat, and require years of monitoring with adjustments in treatments. Stress or reduced medication adherence can lead to rapid worsening in symptoms and functioning with consequences that include relapse, job loss, homelessness, incarceration, hospitalization and suicide. In usual care, clinician visits are infrequent, with intervals ranging from monthly to yearly. Communication between patients and clinicians between visits is challenging and often nonexistent. Patient illness exacerbations and relapses generally occur with little or no clinician awareness in real time, leaving little opportunity to adjust treatments. Significance/Impact: For the large population of Veterans with serious mental illness, tools are needed that passively monitor their mental health status, allowing them to self-track their behaviors, quickly detect worsening of mental health, and support prompt assessment and intervention. At least 60% of Veterans with serious mental illness use a smart phone. These generate data that characterize sociability, activity, and sleep. Changes in these behaviors are warning signs of relapse. Passive self-tracking could be used to identify and predict worsening of illness in real time. Innovation: Passive mobile sensing is a novel approach to illness self-tracking and monitoring. There has been relatively little research on passive self-tracking in serious mental illness, with limited analytics development in this area, and none in VA. Specific Aims: This project studies passive mobile sensing with Veterans in treatment for serious mental illness. Data are used for self-tracking of behaviors and symptoms. While passive mobile sensing has been feasible, acceptable and safe in patients with serious mental illness, these are studied for the first time in VA. Analytics are developed that use passive data to predict behaviors and symptoms. This project responds to the HSR&D priority areas of Mental Health and Healthcare Informatics. The project has these objectives: Conduct user-centered design of passive mobile self-tracking to support Veterans' management of their mental health. Study the feasibility, acceptability and safety of passive self-tracking of mental health that includes feedback of mental health status to the Veteran. Use mobile sensor and phone utilization data to develop individualized estimates of sociability, activities, and sleep as measured by weekly interviews. Study the predictive value of using data on sociability, activities, and sleep to identify exacerbations of psychiatric symptoms. Methodology: Activities can be assessed with data on movement, location, and habits. Sociability can be assessed with data on communication and public interactions. Sleep can be assessed using data on light, sound, movement, and phone use. Investigators on this project developed a functional mobile app that monitors and transmits mobile sensor and utilization data. Focus groups and in-lab usability testing inform further app and intervention development. Mixed methods research study deployment in Veterans who passively self-track their behaviors and psychiatric symptoms. If this project meets intended goals, the VA will have a mobile analytics platform that continuously monitors behaviors and symptoms of patients with serious mental illness.

Mental health care, Informatics, Predictive Modeling, Bipolar disorder, Psychotic Disorders, Machine learning

VetThrive is a mobile smartphone application that monitors and transmits mobile sensor and utilization data. This app is deployed in Veteran patients who passively self-track their behaviors and psychiatric symptoms.

Feasibility of passive self-tracking of mental health assessed using qualitative interview data based on the Technology Assessment Model.

Study the predictive value of using data on sociability, activities, and sleep to identify exacerbations of psychiatric symptoms.

Acceptability of passive self-tracking of mental health assessed using qualitative interview data based on the Technology Assessment Model.

Safety of passive self-tracking of mental health assessed using the occurrence of serious adverse events.

Inclusion Criteria: Veteran patient at the Greater Los Angeles Veterans Healthcare Center with a chart diagnosis of serious mental illness, defined as a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder Risk for symptoms based on having had, during the past year, psychiatric hospitalization, psychiatric emergency care, lived at a crisis program, or more than 6 outpatient visits; and, Ownership of a smartphone with a data plan Exclusion Criteria: Under age 18 Has a conservator/legally authorized representative

Young AS, Choi A, Cannedy S, Hoffmann L, Levine L, Liang LJ, Medich M, Oberman R, Olmos-Ochoa TT. Passive Mobile Self-tracking of Mental Health by Veterans With Serious Mental Illness: Protocol for a User-Centered Design and Prospective Cohort Study. JMIR Res Protoc. 2022 Aug 5;11(8):e39010. doi: 10.2196/39010.