Ginger.io Behavioral Health Study

Procedures involved in the Human Research Overview: The investigators will use a
quasi-experimental study with a pre-post intervention design to evaluate the usability,
feasibility, acceptability, and potential utility of a smartphone app to improve
patient-behavioral health provider communication and care for patients with mood disorders
(e.g. depression, anxiety, bipolar). The smartphone app will collect patient smartphone
interaction data and survey data from patients on their health and functioning including
mood, depression, and anxiety levels. At various points during the study, patients will be
surveyed about their experience using the smartphone application, Ginger.io.

All behavioral health providers at Community Health Center, Inc. (CHCI) who are using the
app with their patients will be granted access to a secured, password-protected, web-based
dashboard that will collect and display information related to patients' health and
behavior. The research assistant and behavioral health providers will use the web-based
dashboard to monitor patient activation and survey completion, and the behavioral health
provider will use it to proactively reach out to patients and aid in discussions with the
patient about the care he/she receives at CHCI. At various points during the study,
behavioral health providers will be surveyed about their experience using the secured,
password-protected web-based provider dashboard.

Behavioral health providers at multiple CHCI sites will be using this app with their
patients as part of their care. From the pool of Ginger.io existing users, about 500
patients will sign written informed consent and join the research study. These patients will
provide survey information that will be used in the research study. Comparison data will
also be gathered from CHCI's electronic health record and via the completion of paper
surveys for approximately 100 patients that sign written informed consent and who are
receiving routine behavioral health care at CHCI but are not using the Ginger.io app.
Baseline data on the patient's depression, bipolar and anxiety will be collected the first
time using the Ginger.io app or filling out paper forms in control clinics.

Intervention patients will be asked to use the Ginger.io smartphone app for a minimum of 6
months, to answer depression, anxiety and mood questions. The Ginger.io app may also be used
by patients on a less frequent (e.g. weekly, biweekly, or monthly) basis to answer more
detailed surveys like the self-report hospitalization or Patient Health Questionnaire
(PHQ-9), which collects data on depression. These daily and weekly assessments of patient
mood, depression and anxiety, including data on the patient's mobility and communication,
will be collected by the Ginger.io smartphone application, and available 24/7 in the
dashboard, so that the dashboard monitor can take appropriate action (or should they choose,
the behavioral health clinician could view as well at monthly visits). Most importantly,
behavioral health providers will be performing proactive outreach to patients that are
identified by alerts as being at higher likelihood of mood disorder irregularities. These
alerts are powered by patient survey responses and smartphone passive behavioral changes.

Intervention Description:

Patient Data Collected Using the Ginger.io Smartphone Application:

Patients who are referred to the study by a participating behavioral health provider or are
recruited via phone, email, etc. provide data for inclusion in the research study will be
current users of the Ginger.io smartphone application that is available free of charge and
will receive instruction on how to download and use the app if applicable. Once the app is
installed on their phone, patients will receive prompts on their smartphone to complete
assessments. The Ginger.io application will also collect data passively on the patient's
levels of communication and mobility while using their smartphone. The patient will not be
required to complete an assessment of communication and mobility - once the application is
installed on the patient's smartphone, it will collect this data automatically. Patients
will have the ability and option to turn these features off if that is their preference.
Patients will be asked to use the Ginger.io application for at least six consecutive months.
Communication and mobility data will also be collected by the application for the duration
of the patient's participation in the study. Patients will also be asked to complete brief
patient surveys about their experience with using the Ginger.io app. The data collection
period will begin on the day of app download, but this data will not be examined for the
purposes of this study until a patient has signed written informed consent documentation.
When downloading the Ginger.io app, patients indicate their consent (or not) to allow the
app to capture passive motion, mobility, and communication data. If the patient has allowed
the Ginger.io app to capture this data, passive motion and mobility data will be used by the
patient's behavioral health provider to supplement between-visit information provided by the
patient while completing surveys. The data will also be used by CHCI and Ginger.io to refine
and improve the threshold that triggers an alert that is generated by the Ginger.io app and
sent to the CHCI behavioral health provider when it is suspected that the patient is
experiencing exacerbated mental health symptoms. As mentioned, these features can be turned
off by the patient. This passive data collection does not collect the content of calls or
text messages and does not track the exact location of the patient (i.e. the patient went to
a specific restaurant or store). It simply measures volumes of contact and amount of
movement to attempt to identify deviations from normal (i.e. a patient no longer calls or
texts anyone, a patient doesn't leave home for several days, a patient is using their phone
in the middle of the night) to detect exacerbations of their mood disorder.

Alerts Triggered by Patient Responses to Ginger.io Survey Question Assessments:

When a patient's mood scores and activity exceed thresholds set by a predetermined Ginger.io
algorithm, the Ginger.io app creates an alert. Alerts generated by the Ginger.io system will
be relayed to behavioral health providers via the provider dashboard. The behavioral health
provider will decide how best to respond, e.g. by contacting the patient or having a nurse
or medical assistant contact the patient. Behavioral health providers will be asked to check
for alerts at least once each business day. They will not be required to check on weekends
or days off. A research assistant will have administrative access to all dashboards to
ensure that behavioral health providers are checking and resolving alerts as they are
received and that no alert goes unaddressed.

If a patient is completing surveys in a manner than indicates an emergency, the app will
deploy a screen that informs that patient that they should seek help and presents them with
contact information for the Suicide Hotline. Additionally, the written informed consent
document that the patients will have a copy of will direct them for how to get after-hours
help through CHCI and to call the Suicide Hotline or 911 in case of an emergency.

Patient and Behavioral Health Provider Review of Patients' Ginger.io Data:

While the main intervention is alert-enabled outreach, participating patients and their
behavioral health providers will have the opportunity to review the patient's data should
they desire, presented in graphical form in the dashboard. This summary will provide a
longitudinal view of the patient's responses to questions about their health (including
depression, bipolar and anxiety). Patients and behavioral health providers will be able to
review these results together at behavioral health office visits and use them to inform
their discussion about the patient's care experience, the efficacy of the patient's
treatment regimen, and whether to modify the patient's treatment plan. Monthly during the
study, behavioral health providers will be surveyed using Survey Monkey about their
experience using the secured, password-protected web-based provider dashboard.

Periodic Delivery of Health Tips and Coping Strategies for Depression & Anxiety & Bipolar In
addition to surveys, the Ginger.io app delivers periodic health tips. The tips are
evidence-based, created by professional psychologists, and offer light-weight practical
means for patients to cope with bipolar, depression & anxiety.

Data management Electronic Health Record Data: CHCI has an electronic health record system
which includes all patient health records. The present study will use data elements from the
system, including primary care performance measures, chart review data, pharmacy prescribing
data, lab data and utilization data (outpatient visits, urgent visits). For each patient, an
index date will be defined - for patients in the intervention group, the date will be the
app activation date; for patients in the control group it will be a comparable visit date.
Electronic health record data will be extracted for 6-months before and after the index date
for each patient.

In addition to these data elements, the investigators will also analyze the survey data
collected via the Ginger.io app.

The following are measures to be obtained by the research team:

1. Behavioral health provider measures:

a. Monthly behavioral health provider surveys

2. Patient Measures:

1. Bidaily depression, mood & anxiety

2. Substance abuse questions

3. Weekly treatment adherence question

4. Patient Health Questionnaire (PHQ-9) and Generalized Anxiety Disorder Screen
(GAD-7) biweekly, alternating

5. A drug/alcohol screen:Single Question Alcohol Screen (SQAS), Single Question Drug
Screen (SQDS), Alcohol Use Disorder Identification Test (AUDIT-C), Drug Abuse
Screening Test (DAST-10)

6. Hospitalization surveys

7. Monthly app satisfaction surveys

8. Comprehensive data will be collected from eClinical Works (ECW) for participating
patients, including demographic data, co-morbidity, behavioral health diagnoses,
clinic utilization, and medication prescribed.

Withdrawal of subjects Patient subjects will be able to withdraw from the study at any time
without penalty. Patients may be withdrawn from the study by research personnel if they lose
access to their smartphones for an extended period of time during the study period or if
they uninstall the Ginger.io smartphone application without re-installing it. Enrolled
patient participants who are not completing daily and weekly surveys on a consistent basis
may be contacted by a Research Assistant or Ginger.io and offered technical assistance with
using the application, if needed. Patients who accept this assistance may meet in-person
with a Research Assistant to relearn how to use the app, or to re-install it, if necessary.
Patients will be eligible to receive additional technical assistance from Research
Assistants at any point during their enrollment in the study. All patient data collected
using Ginger.io between enrollment in the study and withdrawal from the study will be
retained and analyzed by the research team at CHCI and by Ginger.io.

CHCI behavioral health provider subjects may withdraw from the study at any time without
penalty. Behavioral health providers who are found to not be following protocol will be
contacted by research staff to inquire why. When a provider withdraws from the study, his or
her enrolled patients will be given the option to stop providing study data via the
Ginger.io smartphone application.

Inclusion Criteria:

- Age 18 or older

- Experiencing mood disorders (e.g. depression, anxiety, bipolar) defined as a visit
with a specified DSM-V code (295.7 Schizoaffective Disorder (Depressive type), 295.7
Schizoaffective Disorder (Bipolar type), 296.41 Bipolar I Current or most recent
episode manic Mild, 296.42 Bipolar I Current or most recent episode manic mod, 296.43
Bipolar I Current or most recent episode manic sev, 296.44 Bipolar I Current or most
recent episode manic w psychotic, 296.45 Bipolar I Current or most recent episode
manic in partial remission, 296.46 Bipolar I Current or most recent episode manic in
full remission, 296.40 Bipolar I Current or most recent episode unspecified, 296.40
Bipolar I Current or most recent episode hypomanic, 296.45 Bipolar I Current or most
recent episode hypomanic in partial remission, 296.46 Bipolar I Current or most
recent episode hypomanic in full remission, 296.40 Bipolar I Current or most recent
episode hypomanic unspecified, 296.51Bipolar 1 Current or most recent episode
depressed mild, 296.52 Bipolar 1 Current or most recent episode depressed mod, 296.53
Bipolar 1 Current or most recent episode depressed sev, 296.54 Bipolar 1 Current or
most recent episode depressed w psychotic, 296.55 Bipolar 1 Current or most recent
episode depressed in partial remission, 296.56 Bipolar 1 Current or most recent
episode depressed in full remission, 296.50 Bipolar 1 Current or most recent episode
depressed unspecified, 296.7 Bipolar 1 current or most recent episode unspecified,
296.89 Bipolar II, 301.13 Cyclothymic, 296.83 Bipolar due to Medical Condition,
296.89 Other Specified Bipolar, 296.80 Unspecified Bipolar, 296.99 Disruptive Mood
Dysregulation disorder, 296.21 Major Depressive Disorder Single Episode Mild, 296.22
Major Depressive Disorder Single Episode mod, 296.23 Major Depressive Disorder Single
Episode sev, 296.24 Major Depressive Disorder Single Episode w psychotic, 296.25
Major Depressive Disorder Single Episode in partial remission, 296.26 Major
Depressive Disorder Single Episode in full remission, 296.20 Major Depressive
Disorder Single Episode unspecified, 296.31 Major Depressive Disorder Recurrent
Episode mild, 296.32 Major Depressive Disorder Recurrent Episode moderate, 296.33
Major Depressive Disorder Recurrent Episode severe, 296.34 Major Depressive Disorder
Recurrent Episode w psychotic, 296.35 Major Depressive Disorder Recurrent Episode in
partial remission, 296.36 Major Depressive Disorder Recurrent Episode in full
remission, 296.30 unspecified, 300.4 Persistent Depressive Disorder (Dysthymia),
625.4 Premenstrual Dysphoric Disorder, 293.83 Depressive Disorder due to another
medical condition, 311 Other specified depressive disorder, 311 Unspecified
Depressive disorder, 309.21 Separation anxiety disorder, 300.29 Specific Phobia,
300.23 Social anxiety do, 300.01 Panic Disorder, 300.22 Agoraphobia, 300.02 GAD,
293.84 Anxiety do due to another med condition, 300.09 other specified anxiety do,
300.00 Uspecified Anxiety do, 309.0 Adjustment do with depressed mood, 309.24
Adjustment do with anxiety, 309.28 Adjustment do with mixed, 309.4 Adjustment do with
w disturbance of emotions and conduct, 309.9 Adjustment do unspecified, 300.7 Illness
anxiety disorder, 293.83 Mood do NOS, 296.9 Mood do unspecified, 291.89 alcohol
inducted mood do, 292.84 amphetamine inducted mood do 292.84 cocaine inducted mood
do, unspecified inducted mood do, 296.66 episodic mood dis unspecified, 296.64 Severe
mixed bipolar do)

- Possessing a smartphone with either an Android or iOS platform

- Able to read and write in English

- Have a designated primary behavioral health care provider who is either using the app
with their patients (intervention) or is issuing the surveys on paper in their clinic
(control)

Exclusion Criteria: