Tracking Depression Symptoms With a Health Chatbot

Not Applicable

Completed

Brief Summary: The goal of this research is to bridge a significant "effectiveness" gap in the treatment of depression. The investigators have developed a chatbot which will assist in performing measurement-based care (MBC) via Facebook Messenger. Participants will be randomized to either Usual Care or Usual Care with additional Chatbot Care.

Detailed Description: The goal of this research is to bridge a significant "effectiveness" gap in the treatment of depression. The investigators will be looking specifically at depressed women soon after they give birth, otherwise known as known as perinatal depression. Measurement-based care (MBC) which involves the use of quantitative assessments for depression screening, diagnosis, and symptom monitoring is recognized as one of the keys to depression management. MBC can be difficult to implement within the clinical setting due to the administrative overhead and close follow-up required. It is reasonable to hypothesize that low rates of MBC adoption are responsible for low rates of adequate depression care.

The investigators have developed an automated conversational agent or "chatbot" (CB) for delivering MBC via Facebook Messenger. The investigators hypothesize that the use of CB-MBC, which combines a conversational interface with state-of-the-art quantitative assessments, will improve depression symptom severity for patients with perinatal depression. If successful, this project will provide primary care physicians with a new and inexpensive method for delivering better and safer care to their patients with depression. This study works to determine the impact of the chatbot in improving depression outcomes for patients under treatment for perinatal depression.

Recruitment & Eligibility

Inclusion Criteria

Provider Subject Cohort: residents and attendings from University of Chicago Medical Center OBGYN and NorthShore Hospital OBGYN clinics.
Patient Subject Cohort:
1. Age ≥ 18 years and able to and demonstrate English reading literacy of at least 8th-grade level (REALM-R ≥ 6)
2. Postnatal women newly diagnosed or receiving treatment for non-psychotic unipolar depression with a severity-based entry criterion of moderate to severe (EPDS ≥ 12)
3. Willing to participate and able to give written informed consent
4. Must own a smart phone with a data plan
5. Must have a Facebook account and Facebook Messaging app on phone (or willing to create account/download app today)
6. Sufficient cognitive ability to provide self-report data on a computer touchscreen/ standard computer with minimal assistance

Exclusion Criteria

Provider Subject Cohort: None
Patient Subject Cohort:
1. Subjects with documented dysthymia or Axis II diagnoses
2. Subjects with self-reported or documented history of: anorexia or bulimia, obsessive compulsive disorder or prior hospitalization for suicidal ideation
3. Active suicidality as determined by clinician
4. Non-English speakers

Arm && Interventions

Group	Intervention	Description
Chabot Care Group	Chatbot Intervention	Subjects will undergo screening including REALM-R and EPDS and a few additional questions. Eligible subjects randomized to chatbot care will receive monthly reminder phone calls from a research coordinator. Subjects will receive an email with a link to surveys for the purpose of collecting information on depression severity, medication adherence, self-efficacy, side-effect burden, and maternal functioning. In addition to the above procedures, eligible subjects randomized to chatbot care will receive weekly messages from the chatbot asking them to complete a depression severity measure and side-effect burden assessments. Within week 1 subjects will receive a check-in call from a study coordinator to answer any questions regarding use of the chatbot. All participants will be asked to participate in a semi-structured debrief interview upon study completion.

Primary Outcome Measures

Name	Time	Method
Longitudinal Change in Depression Severity	3 months	Measured by the Edinburgh Postnatal Depression Scale (EPDS) Range 0-30 (dimensionless) Risk of depression goes up with an increasing score

Secondary Outcome Measures

Name	Time	Method
Change in Maternal Confidence	3 months	Measured by Maternal Confidence Questionnaire
Medication Adherence	3 months	Measured by 1-item medication adherence questionnaire
Change in Depression Severity	3 months	Measured by the Computerized Adaptive Test-Depression Inventory (CAT-DI) and calculated change in Edinburgh Postnatal Depression Scale (EPDS) between Baseline and 3 months. Edinburgh Postnatal Depression Scale (EPDS) Range 0-30 (dimensionless) Risk of depression goes up with an increasing score
Change in Side-effect Burden	3 months	Measured by the Frequency, Intensity, and Burden of Side Effects Rating (FIBSER) questionnaire
Change in Maternal Function	3 months	Measured by the Barkin Index of Maternal Functioning