Artificial Intelligence in Depression - Medication Enhancement (AID-ME)

Artificial Intelligence in Depression - Medication Enhancement: A Randomized, Patient and Rater Blinded, Active-Controlled Trial of a Hybrid-Classic/Machine-Learning Enabled Clinical Decision Aid for Personalized and Individualized Pharmacological Depression Treatment Selection

This study will determine the safety and potential effectiveness of a digital health platform aimed at improving treatment outcomes for patients with depression by assisting physicians with clinical decision making about depression treatment.

Study Overview

• Status: Completed
• Conditions: Depression
• Intervention / Treatment: Device: Clinical Decision Support System

• Study Type: Interventional
• Enrollment (Actual): 94
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Canada
  • Ontario
    • Toronto, Ontario, Canada, M6J1H4
      • CAMH: The Centre for Addiction and Mental Health
  • Quebec
    • Montréal, Quebec, Canada, H1N3V2
      • CIUSSS de l'Est-de-l'Île-de-Montréal
    • Montréal, Quebec, Canada, H3T1E2
      • CIUSSS De Centre Ouest De L'île de Montréal
    • Montréal, Quebec, Canada, H4A3J1
      • McGill University Health Care Centre (MUHC)
    • Montréal, Quebec, Canada, H4H1R3
      • CIUSSS De L'Ouest de L'île de Montréal
    • Verdun, Quebec, Canada, H4H 1R3
      • Douglas Mental Health University Institute
• United States
  • Connecticut
    • West Haven, Connecticut, United States, 06516
      • VA New Haven, VA Connecticut Healthcare System
  • Florida
    • Miami, Florida, United States, 33125
      • South Florida Veterans Affairs
  • Georgia
    • Atlanta, Georgia, United States, 30322
      • Emory University
  • Michigan
    • Ann Arbor, Michigan, United States, 48109
      • University of Michigan, Michigan Medicine
  • Virginia
    • Salem, Virginia, United States, 24153
      • Salem Vamc

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No

Description

Patient Inclusion Criteria:

• diagnosed with major depressive disorder by a physician using DSM-V criteria
• able to provide informed consent
• patients must confirm that they are comfortable being treated for depression by their physician, who may propose a range of treatment options, such as medications or psychotherapies, consistent with best practice guidelines for depression which are included in the application. Physicians will be required, as in usual practice, to explain treatments to patients and patients will be able to give and withdraw consent for treatment in general or for specific treatments as in usual practice.

Patient Exclusion Criteria:

• bipolar disorder of any type
• inability or unwillingness of the individual to give informed consent
• inability to manage patient in an outpatient setting (i.e. imminent suicidality)
• active major depression is not the main condition being treated (i.e. the patient has depressive symptoms in the context of severe substance abuse or a psychotic disorder, but a primary diagnosis of major depressive episode (MDE) cannot be made or would result in inappropriate care).
• inability to use the tool (i.e. patient cannot interface with a mobile phone or computer due to delirium, or another medical condition)* *Note that for patients who do not have access to mobile or desktop devices but are able to use them or to be trained to use them, these will be provided to them at no cost.

Physician Inclusion Criteria:

• any family doctor/primary care physician or psychiatrist accredited in Canada or the USA who treats patients with depression on at least a monthly basis, as well as residents from these specialities supervised by a participating physician
• able to provide informed consent
• comfortable prescribing the range of potential treatments which could have probabilities provided for them by the CDA

No Physician Exclusion Criteria

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Double

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Active Intervention; Intervention delivered to patients by digital health platform.	Device: Clinical Decision Support System; Clinical Decision Support System using Measurement-Based Care and Digital Decision Support Platform
Active Comparator: Active Control; Intervention delivered to patients by digital health platform.	Device: Clinical Decision Support System; Clinical Decision Support System using Measurement-Based Care and Digital Decision Support Platform

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Rate of Safety - Adverse Events	Adverse and Serious Adverse Events	3 months
Effectiveness in Reduction of Depression Symptoms	This is approved as a physician partially-blinded study and, as such, physicians are blinded to the primary outcome.	3 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Time to Remission	To evaluate time to remission between groups. The hypothesis is that time to remission will be shorter in the active intervention group compared to the active control group.	3 months
Response Rate	To evaluate response rate between groups. The hypothesis is that the response rate (defined as 50% improvement in symptoms) and time to response, will be higher and lower, respectively, in the active intervention group compared to the active control group.	3 months
Patient Disability with WHODAS rating scale	To evaluate between-group differences in patient disability outcomes. The hypothesis is that the physicians in the active intervention arm will produce superior patient outcomes to those physicians using usual guideline-informed practice, in terms of patient function as measured by a rating scale (WHODAS).	3 months

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Number of ER visits, admissions, and re-admissions	We will examine impact of health resource utilization by comparing rates of visits, admissions, and re-admissions between groups.	3 months
Medication Adherence Rates	We will asses medication adherence rates, as measured by the Brief Adherence Rating Scale, after each visit for the duration of the study and compare between groups.	3 months
Patient Questionnaire Response Rate	We will assess the frequency at which patients complete the questionnaires in the digital health platform in both groups.	3 months

Collaborators and Investigators

• Sponsor: Aifred Health
• Collaborators: McGill University

Study record dates

Study Major Dates

• Study Start (Actual): June 16, 2022
• Primary Completion (Actual): December 31, 2023
• Study Completion (Actual): December 31, 2023

Study Registration Dates

• First Submitted: November 23, 2020
• First Submitted That Met QC Criteria: November 30, 2020
• First Posted (Actual): December 7, 2020

Study Record Updates

• Last Update Posted (Actual): January 30, 2024
• Last Update Submitted That Met QC Criteria: January 29, 2024
• Last Verified: January 1, 2024

More Information

Terms related to this study

• Keywords: Depression; Decision Support; Artificial Intelligence (AI); Treatment Selection; Machine Learning (ML)
• Additional Relevant MeSH Terms: Behavioral Symptoms; Mental Disorders; Mood Disorders; Depression; Depressive Disorder
• Other Study ID Numbers: IUSMD 18-04

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: With participating site and on request. Further sharing will be determined at a later date.
• IPD Sharing Time Frame: To be determined.
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP; ICF; CSR

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: Yes
• product manufactured in and exported from the U.S.: No