The STAR*D Trial: Landmark Evidence Reshaping Depression Treatment
The STAR*D (Sequenced Treatment Alternatives to Relieve Depression) trial fundamentally demonstrated that no single antidepressant or switching/augmentation strategy is superior to another, while revealing the sobering reality that only 28-37% of patients achieve remission with initial SSRI treatment and cumulative remission rates plateau at approximately 67% even after four sequential treatment attempts. 1, 2, 3
Revolutionary Study Design and Scale
STAR*D enrolled 4,041 adult outpatients (ages 18-75) with nonpsychotic major depressive disorder across 41 representative primary care and specialty care settings, making it the largest pragmatic, real-world effectiveness trial ever conducted for depression treatment. 4, 2, 3
The study cost $35 million over 6 years and was conducted without pharmaceutical company support, eliminating industry bias that plagued prior antidepressant research. 5, 3
Unlike traditional efficacy trials with highly selected patients, STAR*D included patients with substantial chronic and recurrent depression, multiple psychiatric comorbidities, and general medical conditions—reflecting actual clinical practice populations. 2, 6
Core Findings That Changed Clinical Practice
Level 1: Initial Treatment Reality Check
Only one-third (28-37%) of participants achieved remission with first-line citalopram treatment, with 50% of these remissions occurring within 6 weeks. 1, 2
Higher baseline depression severity was paradoxically associated with lower likelihood of achieving remission with medication—contradicting the common belief that more severe depression responds better to pharmacotherapy. 1
Patients most likely to remit were white women who were employed, had higher education/income levels, and shorter episode duration. 1
Level 2: No Winner Among Switch or Augmentation Strategies
Switching strategies showed equivalent remission rates: sertraline (27%), bupropion-SR (26%), venlafaxine-XR (25%), and cognitive therapy (31%)—no statistically significant differences between within-class, out-of-class, or dual-action agents. 1, 2
Augmentation strategies also showed similar efficacy: bupropion-SR (39%), buspirone (33%), and cognitive therapy (31%) augmentation of citalopram produced comparable remission rates. 1, 2
Bupropion-SR augmentation had significantly better tolerability than buspirone (12.5% vs. 20.6% discontinuation due to adverse events, P < 0.001). 1
Cognitive therapy augmentation required longer time to remission than medication augmentation (55 vs. 40 days). 2
Level 3 and 4: Diminishing Returns
Remission rates dropped dramatically: mirtazapine (8%), nortriptyline (12%), lithium augmentation (13%), T3 augmentation (25%). 2, 3
Level 4 treatments (tranylcypromine 14% vs. venlafaxine-XR plus mirtazapine 16%) showed no significant differences. 2
Patients requiring more than two well-delivered treatments should be characterized as treatment-resistant, given substantially lower remission rates beyond that point. 2
Critical Clinical Implications
The Remission Imperative
Relapse rates during naturalistic follow-up were dramatically lower for patients achieving remission versus response-without-remission: Level 1 (34% vs. 59%), Level 2 (47% vs. 68%), Level 3 (42% vs. 76%), Level 4 (50% vs. 83%). 2
This finding established remission—not just response—as the mandatory treatment target. 2, 3
Predictors of Poor Outcome
Minority status, socioeconomic disadvantage, more axis I and III comorbid disorders, lower baseline function, and anxious or melancholic features all predicted worse outcomes. 2, 3
Longer index episodes and more concurrent psychiatric/medical disorders reduced likelihood of remission. 3
Augmentation May Trump Switching
Although direct comparison was not possible due to study design, the numerically higher remission rates at Level 2 with medication augmentation (39% bupropion, 33% buspirone) versus switching (25-27%) suggest augmentation may be preferable for partial responders. 1, 2
The American College of Physicians' 2023 systematic review incorporating STAR*D data found no significant differences between switching and augmentation strategies overall. 1
Limitations and Controversies
The Missing Placebo Problem
STAR*D did not include placebo arms, making it impossible to determine how many patients would have remitted without active treatment or with alternative/no therapy. 1
This design flaw prevents definitive conclusions about absolute effectiveness versus natural course or placebo response. 1
The Low Bar for Success
The 28% initial remission rate and 67% cumulative remission rate after four treatment attempts reveal that current antidepressant strategies fail the majority of patients. 5
These sobering results suggest "the bar for antidepressants has been set far too low" and highlight the urgent need for novel treatment approaches. 5
Generalizability Strengths
No differences in outcomes between primary care versus specialist settings validate the applicability of findings across practice types. 3
No significant differences between clinician-rated versus self-reported depression scores support the validity of both assessment methods. 3
Lasting Impact on Guidelines
The 2023 American College of Physicians guidelines explicitly cite STAR*D findings to support recommendations that various switch and augmentation strategies show similar efficacy, allowing clinician and patient preference to guide selection. 1
STAR*D demonstrated that comorbid anxiety and depression severity do not significantly affect comparative efficacy of switch or augmentation strategies, simplifying treatment algorithms. 1
The trial established that measurement-based care using itemized symptom and adverse effect assessments at each visit is feasible in real-world settings and results in adequate dosing and treatment duration. 2, 6