Combining Data Sources to Identify Effect Moderation for Personalized Mental Health

This work will extend and apply methods for modeling heterogeneous treatment effects when multiple studies are available, with a particular focus on the complexities in mental health research. The methods will be illustrated in examples estimating the effects of medical treatments for major depressive disorder (duloxetine and vortioxetine) using 4 randomized controlled trials (available in the Vivli trials resource) and non-experimental data from the Duke University Health System electronic health record and the Johns Hopkins Health System electronic health record systems.

Study Overview

• Status: Completed
• Conditions: Depressive Disorder, Major
• Intervention / Treatment: Drug: Duloxetine, Vortioxetine

• Study Type: Observational
• Enrollment (Actual): 9586

Contacts and Locations

Study Locations

• United States
  • Maryland
    • Baltimore, Maryland, United States, 21205
      • Johns Hopkins Bloomberg School of Public Health

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 65 years (Adult, Older Adult)
• Accepts Healthy Volunteers: N/A

• Sampling Method: Non-Probability Sample

Study Population

Patients who received Duloxetine or Vortioxetine

Description

Duke University Health System Electronic Health Record (EHR) System and Johns Hopkins Health System EHR System

Inclusion criteria:

• All patients prescribed any Duloxetine or Vortioxetine in the health systems between 01/01/2015 - 12/31/2021.
• AND from patients with at least 1-year historical EHR data (burn-in period) without prescribing any of the two medications. During the burn-in period, patients should have at least one encounter in the system, and with no encounters during that time frame when they were prescribed Duloxetine or Vortioxetine.
• AND patients age 18-65 years old at baseline.

Exclusion criteria:

- Patients prescribed both Duloxetine and Vortioxetine at baseline would be removed.

4 randomized controlled trials (RCTs)

Inclusion criteria:

• Male or female
• Age 18 years or older
• Major depressive episode according to Diagnostic and Statistical Manual of Mental Disorders, 4th edition, Text Revision (DSM-IV-TR) classification code 296.3x
• Montgomery-Asberg Depression Rating Scale (MADRS) total score ≥ 26
• Clinical Global Impression - Severity (CGI-S) scale score ≥ 4

Study Plan

How is the study designed?

Number of groups / cohorts

6

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Duke Health System Electronic Health Record; Patients who received Duloxetine or Vortioxetine	Drug: Duloxetine, Vortioxetine; Medications for major depressive disorder
Johns Hopkins Health System Electronic Health Record; Patients who received Duloxetine or Vortioxetine	Drug: Duloxetine, Vortioxetine; Medications for major depressive disorder
Randomized Controlled Trial 1 (NCT01153009); Patients who received Duloxetine or Vortioxetine	Drug: Duloxetine, Vortioxetine; Medications for major depressive disorder
Randomized Controlled Trial 2 (NCT01140906); Patients who received Duloxetine or Vortioxetine	Drug: Duloxetine, Vortioxetine; Medications for major depressive disorder
Randomized Controlled Trial 3 (NCT00672620); Patients who received Duloxetine or Vortioxetine	Drug: Duloxetine, Vortioxetine; Medications for major depressive disorder
Randomized Controlled Trial 4 (NCT00635219); Patients who received Duloxetine or Vortioxetine	Drug: Duloxetine, Vortioxetine; Medications for major depressive disorder

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
9-question Patient Health Questionnaire (PHQ-9) or PHQ-2 scores	Higher score indicates greater severity of major depressive disorder.	6 months
9-question Patient Health Questionnaire (PHQ-9) or PHQ-2 scores	Higher score indicates greater severity of major depressive disorder.	1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Emergency room visits or inpatient admissions with diagnosis of depression or suicidality	A binary outcome equals to 'Yes' if the patient ever had any emergency department or inpatient hospital admission encounter with depression or suicidality International Classification of Diseases (ICD) codes, otherwise equals to "No".	6 months
Emergency room visits or inpatient admissions with diagnosis of depression or suicidality	A binary outcome equals to 'Yes' if the patient ever had any emergency department or inpatient hospital admission encounter with depression or suicidality International Classification of Diseases (ICD) codes, otherwise equals to "No".	1 year
Emergency room visits or inpatient admissions for any reason	A binary outcome equals to 'Yes' if the patient ever had any emergency department or inpatient hospital admission encounter after the prescription, otherwise equals to "No".	6 months
Emergency room visits or inpatient admissions for any reason	A binary outcome equals to 'Yes' if the patient ever had any emergency department or inpatient hospital admission encounter after the prescription, otherwise equals to "No".	1 year

Collaborators and Investigators

• Sponsor: Johns Hopkins Bloomberg School of Public Health
• Collaborators: Patient-Centered Outcomes Research Institute; National Institute of Mental Health (NIMH); Duke University
• Investigators: Principal Investigator: Elizabeth Stuart, PhD, Johns Hopkins Bloomberg School of Public Health; Principal Investigator: Hwanhee Hong, PhD, Department of Biostatistics & Bioinformatics, Duke University

Publications and helpful links

General Publications

• Hua K, Wojdyla D, Carnicelli A, Granger C, Wang X, Hong H. Network Meta-Analysis With Individual Participant-Level Data of Time-to-Event Outcomes Using Cox Regression. Stat Med. 2025 Feb 28;44(5):e70027. doi: 10.1002/sim.70027.
• Hua K, Wang X, Hong H. Network Meta-Analysis of Time-to-Event Endpoints With Individual Participant Data Using Restricted Mean Survival Time Regression. Biom J. 2025 Feb;67(1):e70037. doi: 10.1002/bimj.70037.
• Brantner CL, Nguyen TQ, Parikh H, Zhao C, Hong H, Stuart EA. Precision Mental Health: Predicting Heterogeneous Treatment Effects for Depression through Data Integration. J R Stat Soc Ser C Appl Stat. 2025 Dec 12:10.1093/jrsssc/qlaf068. doi: 10.1093/jrsssc/qlaf068. Online ahead of print.
• Brantner CL, Nguyen TQ, Tang T, Zhao C, Hong H, Stuart EA. Comparison of methods that combine multiple randomized trials to estimate heterogeneous treatment effects. Stat Med. 2024 Mar 30;43(7):1291-1314. doi: 10.1002/sim.9955. Epub 2024 Jan 25.
• Brantner CL, Yu W, Zhao C, Jeon K, Ringlein GV, Wang Q, Lemoto E, Nguyen TQ, Gagliardi JP, Zandi PP, Goldstein BA, Stuart EA, Hong H. The challenges of integrating diverse data sources: A case study in major depression. Health Serv Outcomes Res Methodol. 2024 Oct 23:10.1007/s10742-025-00363-8. doi: 10.1007/s10742-025-00363-8. Online ahead of print.
• Brantner CL, Chang TH, Nguyen TQ, Hong H, Stefano LD, Stuart EA. Methods for Integrating Trials and Non-experimental Data to Examine Treatment Effect Heterogeneity. Stat Sci. 2023 Nov;38(4):640-654. doi: 10.1214/23-sts890. Epub 2023 Nov 6.

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2015
• Primary Completion (Actual): December 31, 2021
• Study Completion (Actual): December 1, 2024

Study Registration Dates

• First Submitted: February 20, 2022
• First Submitted That Met QC Criteria: March 2, 2022
• First Posted (Actual): March 4, 2022

Study Record Updates

• Last Update Posted (Actual): April 13, 2026
• Last Update Submitted That Met QC Criteria: April 8, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Mental Disorders; Mood Disorders; Depressive Disorder; Depressive Disorder, Major; Sulfur Compounds; Organic Chemicals; Heterocyclic Compounds, 1-Ring; Heterocyclic Compounds; Thiophenes; Piperazines; Duloxetine Hydrochloride; Vortioxetine
• Other Study ID Numbers: IRB00017538 (JHSPH IRB)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

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