Optimizing Treatment for Depression in Primary Care Using Psychotherapy Versus Antidepressant Medication in a Low-Resource Setting: Protocol for the OptimizeD Randomized Controlled Trial

TL;DR

This study compares psychotherapy and antidepressants for depression in low-resource settings to find which treatment works best for different patients.

Contribution

The study introduces a machine learning approach to personalize depression treatment in primary care using baseline patient data.

Findings

The trial will use baseline data to develop a precision treatment rule for depression.

Cost-effectiveness of optimized treatment will be evaluated in low-resource settings.

Genetic and biological markers may improve treatment optimization.

Abstract

Background:
Psychotherapy and antidepressant medications are first-line treatments for depression, and they both have significant treatment effects on average. However, treatment response varies widely across patients, and neither approach is universally effective. Identifying the most effective treatment for each patient is critical everywhere, but particularly in low-resource settings where access to mental health care is limited. The Optimizing Depression (OptimizeD) trial aims to explore whether different patients respond differently to behavioral activation therapy versus antidepressant medication and if providing each patient with their optimal treatment improves outcomes in primary care.
Methods:
We plan to randomize 1,500 patients with moderate to severe depression (defined as a Patient Health Questionnaire [PHQ-9] score ≥10) from primary healthcare settings in Bhopal, India, with equal allocation either to a culturally adapted behavioral activation therapy delivered by trained counselors (Healthy Activity Program) or to antidepressant medication (fluoxetine). Treatment will last 3 months, with remission (defined as PHQ-9 score <5) at 3 months as the primary endpoint. Using machine learning, we will attempt to develop a precision treatment rule that leverages baseline clinical, psychological, cognitive, socioeconomic, and biological data to predict which treatment is most likely to achieve remission for each patient. Cost-effectiveness analysis will then assess whether the added costs of optimizing treatment are justified by improvements in remission, recovery, and cost savings at the health system and societal levels. Secondary and exploratory objectives include assessing the effectiveness of optimization in a range of secondary outcomes, evaluating treatment mechanisms, and exploring whether incorporating genetic and biological markers as predictors improves treatment optimization.
Discussion:
The OptimizeD trial will evaluate whether baseline information collected in routine care can inform optimal depression treatment selection and identify predictors of nonresponse to facilitate timely specialist referrals. Findings have the potential to enhance personalized depression care in primary health systems, particularly in low-resource settings, with broader implications for global public health.
Trial registration:
ClinicalTrials.gov (NCT05944926; registered July 2, 2023) and Clinical Trials Registry India (CTRI/2024/01/061932; registered January 29, 2024).