[Skip to Content]
[Skip to Content Landing]
Figure 1.
Flow of Patients Through the VAST-D Trial of Antidepressant Switching vs Augmentation
Flow of Patients Through the VAST-D Trial of Antidepressant Switching vs Augmentation

DSM-IV indicates Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; MDD, major depressive disorder; QIDS-C16, 16-item Quick Inventory of Depressive Symptomatology-Clinician Rated; SNRI, serotonin norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor.

Patients followed up to week 12 completed the acute treatment phase. All 1522 patients were included in the primary analysis, following the intention-to-treat principle.

aOnly the most frequent reasons for not enrolling screened individuals are shown. Individuals could have more than 1 reason for exclusion.

bOther includes patients who moved away or withdrew for other illness.

cLost to follow-up includes patients who did not return to the clinic and could not be located for contact.

Figure 2.
Cumulative Probability of Remission and Response Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment Phase
Cumulative Probability of Remission and Response Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment Phase

HR, hazard ratio. Remission was defined as a 16-item Quick Inventory of Depressive Symptomatology-Clinician Rated (QIDS-C16) score (range, 0-27; 0 indicates better symptoms, 27 indicates worse symptoms) of 5 or less for 2 consecutive weeks after baseline during the acute treatment phase. Response was defined as reduction in QIDS-C16 score of 50% or more from baseline at any scheduled visit after baseline through week 12 or improvement in Clinical Global Impression Severity score (range, 1-7; 1 indicates less severe, 7 indicates more severe) of “much improved” or “very much improved” at any scheduled visit after baseline through week 12. Time to first remission or first response was determined by 16-item Quick Inventory of Depressive Symptomatology-Clinician Rated (QIDS-C16) assessments at scheduled visits. In this analysis, follow-up was censored if the patient withdrew from the study. A, HRs from stratified life-table regressions: augment-aripiprazole group vs switch group, 1.28 (95% CI, 1.00-1.64); augment-bupropion group vs switch group, 1.15 (95% CI, 0.89-1.48); augment-aripiprazole group vs augment-bupropion group, 1.06 (95% CI, 0.84-1.35). Median follow-up time to remission or censoring was 81 days (interquartile range [IQR], 41-90) for the switch group, 83 days (IQR, 42-90) for both the augment-bupropion group and augment-aripiprazole group. Plot is truncated at 91 days; 32 observations for week 12 occurred after 91 days. B, HRs from stratified life-table regressions: augment-aripiprazole group vs switch group, 1.32 (95% CI, 1.13-1.54); augment-bupropion group vs switch group, 1.05 (95% CI, 0.89-1.23); augment-aripiprazole group vs augment-bupropion group, 1.23 (95% CI, 1.05-1.43). Median follow-up time to response or censoring was 30 days (IQR, 14-70) for the switch group, 30 days (IQR, 14-69) for the augment-bupropion group and 28 days (IQR, 12-65) for the augment-aripiprazole group. Plot is truncated at 91 days; 25 observations for week 12 occurred after 91 days.

Figure 3.
Cumulative Probability of Relapse After Remission Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment Phase
Cumulative Probability of Relapse After Remission Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment Phase

HR, hazard ratio. Time to first relapse was determined by first occurrence of QIDS-C16 score of 11 or more after remission in the acute treatment phase through the week 36 visit. Follow-up time 0 indicates the day that remission was determined. The proportions relapsing within each treatment group were 26 of 114 patients (22.8%) for the switch group, 35 of 136 patients (25.7%) for the augment-bupropion group, and 37 of 146 patients (25.3%) for the augment-aripiprazole group. Follow-up time was censored in this analysis if the patient withdrew from the study. HRs for stratified life-table regressions for relapse: augment-bupropion group vs switch group, 1.36 (95% CI, 0.78-2.39); augment-aripiprazole group vs switch group, 1.12 (95% CI, 0.65-1.94); augment-aripiprazole group vs augment-bupropion group, 0.96 (95% CI, 0.58-1.59). Median follow-up time to relapse or censoring was 163 days (interquartile range [IQR], 70-201) for the switch group, 163 days (IQR, 68.5-197) for the augment-bupropion group and 160.5 days (IQR, 59-203) for the augment-aripiprazole group. Maximum follow-up time was 245 days.

Table 1.  
Baseline Characteristics of Randomized Patients With Antidepressant-Resistant Major Depressive Disorder by Treatment Groupa
Baseline Characteristics of Randomized Patients With Antidepressant-Resistant Major Depressive Disorder by Treatment Groupa
Table 2.  
Patients With Antidepressant-Resistant Major Depressive Disorder Achieving Cumulative Remission and Response, Acute Treatment Phase
Patients With Antidepressant-Resistant Major Depressive Disorder Achieving Cumulative Remission and Response, Acute Treatment Phase
Table 3.  
Treatment Comparisons for Remission and Response at Week 12 Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment Phasea
Treatment Comparisons for Remission and Response at Week 12 Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment Phasea
Table 4.  
Adverse Events and Serious Adverse Events Reported by Treatment Group Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment and Continuation Phases
Adverse Events and Serious Adverse Events Reported by Treatment Group Among Patients With Antidepressant-Resistant Major Depressive Disorder, Acute Treatment and Continuation Phases
1.
Murray  CJ, Atkinson  C, Bhalla  K,  et al; US Burden of Disease Collaborators.  The state of US health, 1990-2010: burden of diseases, injuries, and risk factors.  JAMA. 2013;310(6):591-608.PubMedGoogle ScholarCrossref
2.
Center for Behavioral Health Statistics and Quality.  Key substance use and mental health indicators in the United States: results from the 2015 National Survey on Drug Use and Health. https://www.samhsa.gov/data/sites/default/files/NSDUH-FFR1-2015/NSDUH-FFR1-2015/NSDUH-FFR1-2015.pdf. Accessed June 19, 2017.
3.
Cleare  A, Pariante  CM, Young  AH,  et al; Members of the Consensus Meeting.  Evidence-based guidelines for treating depressive disorders with antidepressants: a revision of the 2008 British Association for Psychopharmacology guidelines.  J Psychopharmacol. 2015;29(5):459-525.PubMedGoogle ScholarCrossref
4.
Kolovos  S, van Tulder  MW, Cuijpers  P,  et al.  The effect of treatment as usual on major depressive disorder: a meta-analysis.  J Affect Disord. 2017;210:72-81.PubMedGoogle ScholarCrossref
5.
American Psychiatric Association.  Practice Guideline for the Treatment of Patients With Major Depressive Disorder. 3rd ed. Arlington, VA: American Psychiatric Association; 2010.
6.
Management of MDD Working Group.  Clinical practice guideline: management of major depressive disorder (MDD). https://www.healthquality.va.gov/mdd/mdd_full09_c.pdf. Accessed June 19, 2017.
7.
Kennedy  SH, Lam  RW, McIntyre  RS,  et al; CANMAT Depression Work Group.  Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 clinical guidelines for the management of adults with major depressive disorder: section 3: pharmacological treatments.  Can J Psychiatry. 2016;61(9):540-560.PubMedGoogle ScholarCrossref
8.
Goldberg  JF, Freeman  MP, Balon  R,  et al.  The American Society of Clinical Psychopharmacology survey of psychopharmacologists’ practice patterns for the treatment of mood disorders.  Depress Anxiety. 2015;32(8):605-613.PubMedGoogle ScholarCrossref
9.
Rush  AJ, Trivedi  MH, Wisniewski  SR,  et al; STAR*D Study Team.  Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression.  N Engl J Med. 2006;354(12):1231-1242.PubMedGoogle ScholarCrossref
10.
Trivedi  MH, Fava  M, Wisniewski  SR,  et al; STAR*D Study Team.  Medication augmentation after the failure of SSRIs for depression.  N Engl J Med. 2006;354(12):1243-1252.PubMedGoogle ScholarCrossref
11.
Rush  AJ, Warden  D, Wisniewski  SR,  et al.  STAR*D: revising conventional wisdom.  CNS Drugs. 2009;23(8):627-647.PubMedGoogle Scholar
12.
Mohamed  S, Leslie  DL, Rosenheck  RA.  Use of antipsychotics in the treatment of major depressive disorder in the US Department of Veterans Affairs.  J Clin Psychiatry. 2009;70(6):906-912.PubMedGoogle ScholarCrossref
13.
Arbaizar  B, Dierssen-Sotos  T, Gómez-Acebo  I, Llorca  J.  Aripiprazole in major depression and mania: meta-analyses of randomized placebo-controlled trials.  Gen Hosp Psychiatry. 2009;31(5):478-483.PubMedGoogle ScholarCrossref
14.
Han  C, Wang  S-M, Kwak  K-P,  et al.  Aripiprazole augmentation versus antidepressant switching for patients with major depressive disorder: a 6-week, randomized, rater-blinded, prospective study.  J Psychiatr Res. 2015;66-67:84-94.PubMedGoogle ScholarCrossref
15.
Mohamed  S, Johnson  GR, Vertrees  JE,  et al.  The VA augmentation and switching treatments for improving depression outcomes (VAST-D) study: rationale and design considerations.  Psychiatry Res. 2015;229(3):760-770.PubMedGoogle ScholarCrossref
16.
Kroenke  K, Spitzer  RL, Williams  JB.  The PHQ-9: validity of a brief depression severity measure.  J Gen Intern Med. 2001;16(9):606-613.PubMedGoogle ScholarCrossref
17.
Rush  AJ, Trivedi  MH, Ibrahim  HM,  et al.  The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression.  Biol Psychiatry. 2003;54(5):573-583.PubMedGoogle ScholarCrossref
18.
Trivedi  MH.  Tools and strategies for ongoing assessment of depression: a measurement-based approach to remission.  J Clin Psychiatry. 2009;70(suppl 6):26-31.PubMedGoogle ScholarCrossref
19.
Wisniewski  SR, Rush  AJ, Balasubramani  GK, Trivedi  MH, Nierenberg  AA; STARD Investigators.  Self-rated global measure of the frequency, intensity, and burden of side effects.  J Psychiatr Pract. 2006;12(2):71-79.PubMedGoogle ScholarCrossref
20.
Israel  JA.  Remission in depression: definition and initial treatment approaches.  J Psychopharmacol. 2006;20(3)(suppl):5-10.PubMedGoogle ScholarCrossref
21.
Guy  W.  ECDEU Assessment Manual for Psychopharmacology-Revised. Rockville, MD: Dept of Health, Education, and Welfare; 1976.
22.
Levine  J, Schooler  NR.  SAFTEE: a technique for the systematic assessment of side effects in clinical trials.  Psychopharmacol Bull. 1986;22(2):343-381.PubMedGoogle Scholar
23.
Lieberman  JA, Stroup  TS, McEvoy  JP,  et al; Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) Investigators.  Effectiveness of antipsychotic drugs in patients with chronic schizophrenia.  N Engl J Med. 2005;353(12):1209-1223.PubMedGoogle ScholarCrossref
24.
Hochberg  Y.  A sharper Bonferroni procedure for multiple tests of significance.  Biometrika. 1988;75(4):800-802.Google ScholarCrossref
25.
Zisook  S, Tal  I, Weingart  K,  et al.  Characteristics of US veteran patients with major depressive disorder who require “next-step” treatments: a VAST-D report.  J Affect Disord. 2016;206:232-240.PubMedGoogle ScholarCrossref
26.
Fava  M, Rush  AJ, Alpert  JE,  et al.  Difference in treatment outcome in outpatients with anxious versus nonanxious depression: a STAR*D report.  Am J Psychiatry. 2008;165(3):342-351.PubMedGoogle ScholarCrossref
27.
Sansone  RA, Sansone  LA.  Antidepressant adherence: are patients taking their medications?  Innov Clin Neurosci. 2012;9(5-6):41-46.PubMedGoogle Scholar
28.
Adams  JR, Drake  RE.  Shared decision-making and evidence-based practice.  Community Ment Health J. 2006;42(1):87-105.PubMedGoogle ScholarCrossref
29.
American Diabetes Association; American Psychiatric Association; American Association of Clinical Endocrinologists; North American Association for the Study of Obesity.  Consensus development conference on antipsychotic drugs and obesity and diabetes.  J Clin Psychiatry. 2004;65(2):267-272.PubMedGoogle ScholarCrossref
30.
Rush  AJ, Trivedi  MH, Wisniewski  SR,  et al.  Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report.  Am J Psychiatry. 2006;163(11):1905-1917.PubMedGoogle ScholarCrossref
Original Investigation
July 11, 2017

Effect of Antidepressant Switching vs Augmentation on Remission Among Patients With Major Depressive Disorder Unresponsive to Antidepressant Treatment: The VAST-D Randomized Clinical Trial

Author Affiliations
  • 1Veterans Affairs (VA) New England Mental Illness Research, Education, and Clinical Center, VA Connecticut Healthcare System, West Haven
  • 2Yale University School of Medicine, West Haven, Connecticut
  • 3Cooperative Studies Program Coordinating Center, VA Connecticut Healthcare System, West Haven
  • 4Louis Stokes VA Medical Center and Case Western Reserve University School of Medicine, Cleveland, Ohio
  • 5Central Texas Veterans Healthcare System and Department of Psychiatry and Behavioral Science, Texas A&M Health Science Center College of Medicine, Temple
  • 6Central Texas Veterans Healthcare System, Temple
  • 7Tuscaloosa VA Medical Center, Tuscaloosa, Alabama
  • 8University of Alabama School of Medicine, Birmingham
  • 9Health Economics Resource Center, VA Palo Alto, Menlo Park, California
  • 10Department of Veterans Affairs, Office of Research and Development, Washington, DC
  • 11Cooperative Studies Program Clinical Research Pharmacy Coordinating Center, Albuquerque, New Mexico
  • 12VA San Diego Healthcare System, San Diego, California
  • 13University of California, San Diego
  • 14Philadelphia VA Medical Center, Philadelphia, Pennsylvania
  • 15Cooperative Studies Program Central Office, Department of Veterans Affairs Office of Research and Development, Washington, DC
JAMA. 2017;318(2):132-145. doi:10.1001/jama.2017.8036
Key Points

Question  Is there a difference among pharmacotherapeutic approaches for treating patients with depression unresponsive to an antidepressant course?

Findings  In a 12-week follow-up of a randomized clinical trial of 1522 patients with major depressive disorder (85% men) unresponsive to previous antidepressant treatment, 29% achieved remission after augmenting their antidepressant with the antipsychotic aripiprazole vs 22% who switched to the antidepressant bupropion. Other remission comparisons were not significant.

Meaning  Augmentation with aripiprazole resulted in a statistically significant, but only modestly increased, likelihood of remission during 12 weeks of treatment compared with switching to bupropion alone.

Abstract

Importance  Less than one-third of patients with major depressive disorder (MDD) achieve remission with their first antidepressant.

Objective  To determine the relative effectiveness and safety of 3 common alternate treatments for MDD.

Design, Setting, and Participants  From December 2012 to May 2015, 1522 patients at 35 US Veterans Health Administration medical centers who were diagnosed with nonpsychotic MDD, unresponsive to at least 1 antidepressant course meeting minimal standards for treatment dose and duration, participated in the study. Patients were randomly assigned (1:1:1) to 1 of 3 treatments and evaluated for up to 36 weeks.

Interventions  Switch to a different antidepressant, bupropion (switch group, n = 511); augment current treatment with bupropion (augment-bupropion group, n = 506); or augment with an atypical antipsychotic, aripiprazole (augment-aripiprazole group, n = 505) for 12 weeks (acute treatment phase) and up to 36 weeks for longer-term follow-up (continuation phase).

Main Outcomes and Measures  The primary outcome was remission during the acute treatment phase (16-item Quick Inventory of Depressive Symptomatology-Clinician Rated [QIDS-C16] score ≤5 at 2 consecutive visits). Secondary outcomes included response (≥50% reduction in QIDS-C16 score or improvement on the Clinical Global Impression Improvement scale), relapse, and adverse effects.

Results  Among 1522 randomized patients (mean age, 54.4 years; men, 1296 [85.2%]), 1137 (74.7%) completed the acute treatment phase. Remission rates at 12 weeks were 22.3% (n = 114) for the switch group, 26.9% (n = 136)for the augment-bupropion group, and 28.9% (n = 146) for the augment-aripiprazole group. The augment-aripiprazole group exceeded the switch group in remission (relative risk [RR], 1.30 [95% CI, 1.05-1.60]; P = .02), but other remission comparisons were not significant. Response was greater for the augment-aripiprazole group (74.3%) than for either the switch group (62.4%; RR, 1.19 [95% CI, 1.09-1.29]) or the augment-bupropion group (65.6%; RR, 1.13 [95% CI, 1.04-1.23]). No significant treatment differences were observed for relapse. Anxiety was more frequent in the 2 bupropion groups (24.3% in the switch group [n = 124] vs 16.6% in the augment-aripiprazole group [n = 84]; and 22.5% in augment-bupropion group [n = 114]). Adverse effects more frequent in the augment-aripiprazole group included somnolence, akathisia, and weight gain.

Conclusions and Relevance  Among a predominantly male population with major depressive disorder unresponsive to antidepressant treatment, augmentation with aripiprazole resulted in a statistically significant but only modestly increased likelihood of remission during 12 weeks of treatment compared with switching to bupropion monotherapy. Given the small effect size and adverse effects associated with aripiprazole, further analysis including cost-effectiveness is needed to understand the net utility of this approach.

Trial Registration  clinicaltrials.gov Identifier: NCT01421342

Introduction

Quiz Ref IDMajor depressive disorder (MDD) is a chronic, debilitating disorder1 that affected an estimated 16.1 million adults in the United States in 2015.2 Given that less than one-third of patients achieve remission with their first course of antidepressant pharmacotherapy,3,4 an estimated 10.8 million US residents may benefit from an alternative treatment each year. For these patients, most treatment guidelines recommend either switching to another antidepressant or adjunctive use of another antidepressant or nonantidepressant agent.5-7 Among the most commonly used of these treatment strategies are switching to bupropion, a norepinephrine-dopamine reuptake inhibitor, and adjunctive use of either bupropion or aripiprazole, a second-generation antipsychotic that is a partial dopamine agonist.8

The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial showed that bupropion was at least as effective as other switching9 and augmenting agents.10 However, STAR*D was not powered to compare bupropion switching and augmentation strategies,11 and atypical antipsychotics, frequently used as adjunctive agents for treatment-resistant MDD even prior to US Food and Drug Administration approval,12 were not included. Several studies have shown efficacy of aripiprazole as an antidepressant augmentation strategy13 and a recent study suggested aripiprazole augmentation is more beneficial than antidepressant switching and has comparable tolerability.14 However, adequately powered and well-controlled clinical trials have yet to compare the effectiveness of these 2 treatments or compare them with augmentation with a second antidepressant.

The primary objective of this randomized clinical trial was to compare the effectiveness and adverse effect profiles of 3 commonly used alternative MDD treatment strategies: switch to the antidepressant bupropion sustained release; augment current treatment with bupropion sustained release; or augment current treatment with the antipsychotic aripiprazole.15

Methods

The Veterans Affairs (VA) Office of Research and Development and VA Central Institutional Review Board approved the study, the National Institutes of Health provided a certificate of confidentiality, the VA Central Institutional Review Board conducted annual continuing review, and a data and safety monitoring committee reviewed the study biannually. All patients provided written informed consent and privacy authorization. The full study protocol can be found in Supplement 1.

Study Design

VA Augmentation and Switching Treatments for Improving Depression Outcomes (VAST-D) was a multisite randomized, single-blind, parallel-assignment trial including US Veterans Health Administration (VHA) patients whose condition was unresponsive to at least 1 course of antidepressant treatment meeting minimal standards for dose and duration.

Patient Selection

Participants were VHA patients, 18 years or older, with an MDD diagnosis, who were referred by their VA clinicians. Diagnostic eligibility was further established by research staff using criteria from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR), supplemented with the 9-Item Patient Health Questionnaire (PHQ-9; range, 0-27; 0 indicates better health, 27 indicates worse health).16 Study clinicians determined final diagnoses. Race and ethnicity were determined by self-report using open-ended questions and were recorded to document the level of inclusion of minority populations.

Patients with a suboptimal response to a treatment course with a selective-serotonin reuptake inhibitor, serotonin and norepinephrine reuptake inhibitor, or mirtazapine that met or exceeded minimal standards for dose and duration of treatment were eligible (Supplement 2).15 Suboptimal response was defined as a score of 16 or more (indicating severe depression) on the 16-Item Quick Inventory of Depressive Symptomatology-Clinician Rated (QIDS-C16) questionnaire17 after at least 6 weeks of treatment or a score of 11 or more (indicating moderately severe depression) after at least 8 weeks of treatment with the 3 most recent weeks at a stable “optimal” dose (Supplement 2).

Quiz Ref IDPatients were excluded if they were receiving current treatment with bupropion or any antipsychotic; had a lifetime history of bipolar disorder, schizophrenia, schizoaffective disorder, or other psychosis; had current dementia, an eating disorder, or a seizure disorder; suicidal ideation requiring inpatient treatment; had an unstable medical condition; had need of immediate psychiatric hospitalization; had substance dependence requiring detoxification in the past 30 days; were pregnant, lactating, or planning to become pregnant; or were unable or unwilling to provide informed consent or declined to participate prior to randomization.

Randomization

Patients at 35 VA medical centers were randomized to 1 of 3 treatments: switch to another antidepressant, bupropion sustained release (switch group); augment current treatment with bupropion sustained release (augment-bupropion group); or augment current treatment with an antipsychotic, aripiprazole (augment-aripiprazole group) (Figure 1). They were randomized using a stratified randomization scheme balanced (1:1:1) within each medical center using a random permuted-block scheme with variable block sizes (3 or 6) and random number generation in SAS Proc Plan (SAS Institute) prepared by VA Cooperative Studies Program Coordinating Center. After patient eligibility was confirmed, randomization was completed by site personnel using a web-based application. Outcomes were assessed by independent evaluators blind to treatment assignment.

Interventions

Treatments included titration (cross-titration for the switch group) from standard starting doses of 150 mg of bupropion sustained release to 300 mg or 400 mg daily; or from 2 mg of aripiprazole with titration to 5, 10, or 15 mg daily, until depressive symptoms remitted or adverse effects were intolerable. Dose adjustments were guided by “measurement-based care”18 using a patient-rated symptom measure (PHQ-9)16 and a global adverse effects measure (Frequency, Intensity, and Burden of Side Effects Rating [FIBSER]; range 0-18, 0 indicates no adverse effects, 18 indicates severe adverse effects; scores >8 suggest doses should not be increased due to adverse effects.)19 at each visit. Acute treatment visits occurred at baseline and at weeks 1, 2, 4, 6, 8, 10, and 12. The acute treatment phase was up to 12 weeks and up to 36 weeks for longer-term follow-up.

Patients who tolerated acute treatment and achieved adequate benefit were eligible to enter a 24-week continuation phase to evaluate relapse and other outcomes. Adequate benefit was defined as either QIDS-C16 score of 8 or less at 12 weeks or QIDS-C16 score of 9 or 10 with clinician judgment of adequate benefit. This article addresses relapse among those who achieved remission during the acute treatment phase with follow-up for relapse of symptoms for up to 36 weeks. Assigned treatment was sustained after 12 weeks and follow-up visits occurred every 4 weeks.

Outcome Measures

The primary outcome was remission (close to asymptomatic status), defined as a QIDS-C16 score (range, 0-27 with higher scores indicating more severe symptoms) of 5 or less at 2 consecutive scheduled follow-up visits during the acute treatment phase. The QIDS-C16 questionnaire was administered every visit. Secondary outcomes included 2 measures of response: reduction in QIDS-C16 score from baseline by 50% or more20 and a Clinical Global Impression (CGI)21 Improvement scale rating (range, 0-7; 0 indicates not assessed, 1 indicates very much improved since the initiation of treatment, 7 indicates very much worse since the initiation of treatment) of 2 (much improved) or 1 (very much improved) at any scheduled visit through week 12; and 1 measure of relapse: QIDS-C16 score of 11 or more during the continuation phase after achieving remission in the acute treatment phase.

Other secondary measures that were assessed but not reported in this article include suicidal ideation and behaviors, anxiety, global improvement, quality of life, health-related costs and cost-effectiveness, and medication adherence and satisfaction.

At each visit, treatment-emergent adverse effects were queried with a checklist based on the Systematic Assessment for Treatment of Emergent Events—Specific Inquiry (SAFETEE-SI)22 and vital signs were obtained. A 7% weight gain was considered clinically important.23 Every 12 weeks, laboratory assessments and metabolic indicators were obtained for safety monitoring. Serious adverse events, reported to the sponsor as required, were tabulated for 30 days after withdrawal from the study.

Statistical Methods

A clinically significant difference in remission of 10% was chosen for estimating sample size and power based on results of previous trials.9,10 A target sample size of 1518 patients (506 per treatment group) was chosen to give 90% power to detect a 10% increase in remission for an augmentation group compared with the switch group (hypothetically 35% vs 25% remission, odds ratio [OR], 1.62) at a P value of less than .05, and 84% power to detect a 10% increase in remission for the other co-primary hypothesis at a P value of less than .025. The secondary hypothesis comparing the augment-aripiprazole group and augment-bupropion groups would then have 80% power to detect a 9% absolute difference (30% vs 39%; OR, 1.49) in remission at a P value of less than .05.

To compare the proportion of patients achieving remission in each augmentation group relative to the proportion of patients achieving remission in the switch group, the intention-to-treat analysis for these co-primary hypotheses used logistic regression models stratified by participating medical center. The 2 co-primary hypotheses were tested using the Wald test statistic and the sequentially rejective procedure of Hochberg24 with a familywise type I error rate of .05 (2-sided) for the largest P value, and, if not significant, a test at the .025 level (2-sided) for the smallest P value. Using a gatekeeping approach, the treatment effect for the augment-aripiprazole group vs augment-bupropion group was evaluated at .05 as the second family of comparisons only if at least 1 of the co-primary hypotheses tests were statistically significant. To correct for overestimation of the risk ratio by the odds ratio, log-binomial regression was used for analysis of remission and the analysis of the secondary response outcomes; relative risk (RR) ratios with 95% CIs are reported (SAS PROC GENMOD [SAS Institute]). A supportive mixed-model analysis including participating medical centers as random effects was also conducted (SAS PROC GLIMMIX [SAS Institute]).

Time-to-event analysis of relapse among patients achieving remission in the acute treatment phase was based on time from remission to relapse within the 36-week study period. The Kaplan-Meier method was used to calculate cumulative time-to-event curves for remission, response, and relapse. Treatment differences were tested using the Cox proportional-hazards regression models stratified by site. For relapse, treatment comparisons yielding a P value of less than .025 were considered significant, except for the augment-aripiprazole group vs augment-bupropion group for which a P value of less than .05 was considered significant.

Results for the primary outcome, remission in the acute phase, and secondary outcomes including response and relapse among those who achieved remission in the acute treatment phase as well as adverse events occurring throughout follow-up are presented in this article. Missing data due to missed assessments were treated as missing and were not imputed in the primary analysis. Overall, 178 scheduled visits (1.7%) in the acute treatment phase were missed (eTable 1 in Supplement 2). For data missing due to withdrawal before week 12, the status of the patient at the last completed assessment was retained. There was no imputation of QIDS-C16 scores for missing assessments.

Multiple imputation methods (SAS Procedures MI and MIANALYZE [both from SAS Institute]) were used in sensitivity analyses to impute values (0 or 1) for protocol remission and protocol response for patients who withdrew prior to week 12. Patients who withdrew due to lack of treatment effect were considered to be unresponsive to treatment. Baseline QIDS-C16 score, age, sex, days of participation, and treatment group were used to impute missing values for 19.4% of the patients.

The proportions of patients who developed adverse effects were compared across treatment groups using χ2 tests for the difference in proportions at the .05 significance level, and subgroupings of serious adverse events (SAEs) or nonserious adverse events at the .01 significance level. Pairwise comparisons were performed for the composite akathisia symptom and extrapyramidal adverse effects count, and a P value of less than .025 was considered significant, except for the augment-aripiprazole group vs augment-bupropion group for which a P value of less than .05 was considered significant.

For other secondary (supportive) analyses, a P value of less than .025 was considered significant, except for the augment-aripiprazole group vs augment-bupropion group for which a P value of less than .05 was considered significant.

SAS software (SAS Institute), version 9.3, was used to complete these analyses.

One interim analysis using conditional power calculations was conducted for futility (conditional power lower boundary of 25%) or sample size reestimation (25% < conditional power < 80%) after approximately 50% enrollment. No sample size increase was indicated and the study was continued.

Results
Study Population

Of the 7234 VHA patients screened, 1861 patients (25.7%) consented to participate and, of these, 1522 (81.8%) were randomly assigned to the switch group (n = 511), augment-bupropion group (n = 506), and augment-aripiprazole group (n = 505) (Figure 1). Baseline characteristics, described previously,25 were similar across treatment groups (Table 1).

Treatment and Retention

Index antidepressants are presented in Table 1; the bupropion or aripiprazole dose prescribed at each visit is presented in eTable 1 in Supplement 2. In the switch group, the modal dose of bupropion reached a maximum of 200 mg twice daily by 6 weeks and remained at that level through 12 weeks. The augment-bupropion group also reached the maximum modal dose by 6 weeks. In the augment-aripiprazole group, the modal dose was 5 mg daily from weeks 2 through 10 and reached 10 mg daily at week 12. Doses of study medications were initiated at randomization per protocol for all but 3 patients.

Retention of patients through the 12-week acute treatment phase was greatest for the augment-aripiprazole group (80.4%), lower for the augment-bupropion group (74.7%) and lowest for the switch group (69.1%) (Figure 1). Withdrawal for lack of response was lowest for the augment-aripiprazole group (2.6%), followed by the augment-bupropion group (5.5%) and the switch group (9.6%). Withdrawal for medication adverse effects was also lower for the augment-aripiprazole group (5.3%) than either the augment-bupropion group (7.3%) or switch group (10.0%) (Figure 1). The major reason for withdrawal at week 12 was lack of response per protocol (Figure 1). Those who achieved remission (n = 396) were followed up for relapse for up to 36 weeks after randomization.

Outcomes

The primary outcome of remission occurring through week 12 was significantly, albeit modestly, higher for the augment-aripiprazole group (28.9%) compared with the switch group (22.3%; RR, 1.30 [95% CI, 1.05-1.60]; P = .02) but not compared with the augment-bupropion group (26.9%; RR, 1.08 [95% CI, 0.88-1.31]; P = .47) (Table 2 and Table 3). Remission with the augment-bupropion group was not significantly different than the switch group (RR, 1.20 [95% CI, 0.97-1.50]; P = .09). Cox regression analysis of time to remission showed no significant differences in cumulative remission for the augment-aripiprazole group vs the switch group (hazard ratio [HR], 1.28 [95% CI, 1.00-1.64]; P = .05), the augment-bupropion group vs the switch group (HR, 1.15 [95% CI, 0.89-1.48]; P = .27), or the augment-aripiprazole group vs the augment-bupropion group (HR, 1.06 [95% CI, 0.84-1.35]; P = .61) (Figure 2A). The supportive mixed-model analysis including participating medical centers as random effects showed essentially the same results as the stratified primary analysis (Supplement 2).

Response, a secondary outcome based on 50% reduction in QIDS-C16 score, was significantly higher for the augment-aripiprazole group (74.3%) than for both the switch group (62.4%; RR, 1.19 [95% CI, 1.09-1.29]; P < .001) and the augment-bupropion group (65.6%; RR, 1.13 [95% CI, 1.04-1.23]; P = .003), with no significant difference between the augment-bupropion group and the switch group (RR, 1.05 [95% CI, 0.96-1.15]; P = .29) (Table 2 and Table 3). Response measured by CGI improvement similarly favored the augment-aripiprazole group (79%) compared with both the switch group (70%; RR, 1.14 [95% CI, 1.06-1.22]; P < .001) and the augment-bupropion group (74%; RR, 1.07 [95% CI, 1.00-1.14]; P = .07) (Table 2 and Table 3).

In the Cox regression analysis of response based on the QIDS-C16 score, cumulative response for the augment-aripiprazole group was greater than both the switch group (HR, 1.32 [95% CI, 1.13-1.54]; P < .001) and augment-bupropion group (HR, 1.23 [95% CI, 1.05-1.43]; P = .007) (Figure 2B). Cumulative response did not differ significantly between the augment-bupropion group and switch group (HR, 1.05 [95% CI, 0.89-1.23]; P = .56).

The results of the multiple-imputation sensitivity analysis for protocol remission at week 12 showed comparison of the augment-aripiprazole group vs the switch group to be statistically significant (RR, 1.33 [95% CI, 1.07-1.66]; P = .01), but not the augment-bupropion group vs the switch group (RR, 1.16 [95% CI, 0.93-1.46]; P = .19). For response as defined by the protocol, there was a statistically significant effect for the augment-aripiprazole group vs the switch group (RR, 1.26 [95% CI, 1.13-1.39]; P < .001) but not for the augment-bupropion group vs the switch group (RR, 1.10 [95% CI, 0.99-1.23]; P = .08).

Among the 396 patients achieving remission in the acute treatment phase, there were no significant differences in the secondary outcome of cumulative relapse: augment-bupropion group vs switch group (HR, 1.36 [95% CI, 0.78-2.39]; P = .70); augment-aripiprazole group vs switch group (HR, 1.12 [95% CI, 0.65-1.94]; P = .68); or augment-bupropion group vs augment-aripiprazole group (HR, 0.96 [95% CI, 0.58-1.59]; P = .87) (Figure 3).

Adverse Events

Among 165 patients (10.8%), a total of 207 SAEs occurred (Table 4). Eight deaths occurred among study patients during the safety reporting period: 3 during follow-up; and 5 in the 30 days after withdrawal or completion of follow-up, with 1 completed suicide. Causes of death included completed suicide, sudden unexpected death, chronic obstructive pulmonary disease, bilateral pulmonary embolisms, arteriosclerosis, acute myocardial infarction, lung cancer, and rhabdomyolysis. The investigators and the data and safety monitoring committee agreed that the deaths were not related to study medication.

There were 4356 nonserious adverse events recorded (eTable 2 in Supplement 2). Specific nonserious adverse events occurred in a greater proportion of patients in the switch group than in the other groups and included nausea, irritability, and hypomania (Table 4). Nonserious adverse events occurring in greater proportions of both the switch and augment-bupropion groups than among patients in the augment-aripiprazole group included anxiety, decreased appetite, dry mouth, and increased blood pressure. Nonserious adverse events occurring in a greater proportion of patients in the augment-aripiprazole group compared with the other 2 groups included fatigue, increased appetite, increased weight, akathisia, somnolence, and abnormal values for several laboratory tests. At week 12, weight gain of 7% or greater was more frequent for the augment-aripiprazole group (38 of 399 patients [9.5%]) compared with the switch group (8 of 347 patients [2.3%]) and the augment-bupropion group (7 of 369 patients [1.9%]). For the subset of patients continuing through week 36, the proportion with weight gain 7% or greater was also greater for the augment-aripiprazole group (53 of 210 patients [25.2%]) compared with the augment-bupropion group (10 of 193 patients [5.2%]) and the switch group (8 of 153 patients [5.2%]) (Table 4).

Discussion

Quiz Ref IDIn this randomized clinical trial in a predominantly male population with MDD who were unresponsive to antidepressant treatment, augmentation with aripiprazole resulted in a statistically significant but only modestly increased likelihood of remission during 12 weeks of treatment compared with switching to bupropion monotherapy. On 2 measures of response, the secondary outcome, the augment-aripiprazole group exceeded both the switch group and the augment-bupropion group. The lowest discontinuation rates were also observed for the augment-aripiprazole group. There were no significant differences in relapse rates between treatment groups among acute treatment phase remitters during the continuation phase. On adverse effects, the results were mixed with treatment-emergent anxiety favoring the augment-aripiprazole group and akathisia, somnolence, and weight gain favoring the switch group and the augment-bupropion group.

Quiz Ref IDThe most clinically meaningful adverse event occurring in at least 5% of patients in the switch and augment-bupropion groups was increased anxiety (which also included reports of nervousness). Anxiety is known to be a negative prognostic factor in patients with MDD associated with poor response, increased relapse and recurrence, and suicide risk.26 On the other hand, the augment-aripiprazole group reported more somnolence, extrapyramidal effects (including akathisia), and weight gain than the 2 bupropion groups. Although each of these adverse effects has been associated with dropping out of treatment,27 dropout rates in this study were smallest for the augment-aripiprazole group, suggesting limited effect of these adverse effects on treatment adherence. Quiz Ref IDAlthough weight gain did not lead to medication discontinuation, it is related to metabolic syndrome and could lead to serious health concerns in the long-term. There is a need to further investigate the benefit-risk ratio of long-term risks of weight gain and other adverse effects compared with the benefit of modestly reducing depression. A major asset of the VAST-D study is the comparative evaluation of the longer-term benefits and risks of this second-generation antipsychotic, the first approved for this indication.

Although use of the augment-aripiprazole strategy as an alternative treatment may offer an increased likelihood of remission, patients should be alerted to the risk of weight gain in a process of shared decision making28 and baseline screening and follow-up monitoring of metabolic indicators should be provided.29

Limitations

This study had several limitations. First, only 1 antidepressant (bupropion sustained release) and 1 antipsychotic (aripiprazole) were evaluated, and the generalizability of the results to other medications is unknown.

Second, only 1137 patients (74.7%) completed the 12-week acute treatment phase and differences in outcomes between groups were small in magnitude. Multiple imputation was used to address missing data due to withdrawals, and the results of imputing 12-week outcomes were consistent with the direction and magnitude of effects reported in the primary analysis of protocol defined remission and response. A limitation of the imputation analysis was that withdrawals may be associated with outcome or treatment assignment so that missing data could not be assumed to be missing at random.

Third, it is possible that discontinuation in the switch group was increased by withdrawal symptoms from their previous treatment. Fourth, the study was conducted with VA patients with many exclusion criteria. As a result, the generalizability of results from this older, predominantly male population is also unknown. However, with a mean QIDS-C16 score of 16.6 to 16.9 across groups (SD, 3.2-3.3), this sample was in the severe depression range (score, 16-20),17 similar to the STAR*D study (mean score, 16.2 [SD, 4.1]) although the duration of the current episode in the VAST*D study was a mean of 87 months, which is considerably longer, and suggests less potential for treatment responsiveness than the mean of 28 months in the second step of the STAR*D study.30 Other important secondary outcomes, including cost-effectiveness, quality of life, and suicidal ideation that may clarify the balance between clinical benefits and adverse effects still need to be addressed.

Conclusions

Among a predominantly male population with major depressive disorder unresponsive to antidepressant treatment, augmentation with aripiprazole resulted in a statistically significant but only modestly increased likelihood of remission during 12 weeks of treatment compared with switching to bupropion monotherapy. Given the small effect size and adverse effects associated with aripiprazole, further analysis including cost-effectiveness is needed to understand the net utility of this approach.

Back to top
Article Information

Corresponding Author: Somaia Mohamed, MD, PhD, the VA Connecticut Healthcare System, 950 Campbell Ave, Mailstop (182), West Haven, CT 06516 (somaia.mohamed@va.gov).

Accepted for Publication: June 13, 2017.

Authors/VAST-D Investigators: Sanjai D. Rao, MD; Patricia D. Pilkinton, MD; James A. Wilcox, MD, PhD; Ali Iranmanesh, MD; Mamta Sapra, MD; George Jurjus, MD; James P. Michalets, MD; Muhammed Aslam, MD; Thomas Beresford, MD; Keith D. Anderson, PharmD; Ronald Fernando, MD; Sriram Ramaswamy, MD; John Kasckow, MD, PhD; Joseph Westermeyer, MD, PhD; Gihyun Yoon, MD; D. Cyril D’Souza, MD; Gunnar Larson, MD; William G. Anderson, MD; Mary Klatt, MD; Ayman Fareed, MD; Shabnam I. Thompson, DO; Carlos J. Carrera, MD; Solomon S. Williams, MD; Timothy M. Juergens, MD; Lawrence J. Albers, MD; Clifford S. Nasdahl, PharmD; Gerardo Villarreal, MD; Julia L. Winston, MD; Cristobal A. Nogues, MD; K. Ryan Connolly, MD; Andre Tapp, MD; Kari A. Jones, MD; Gauri Khatkhate, MD; Sheetal Marri, MD; Trisha Suppes, MD, PhD; Joseph LaMotte, PharmD; Robin Hurley, MD; Aimee R. Mayeda, MD; Alexander B. Niculescu III, MD, PhD; Bernard A. Fischer, MD; David J. Loreck, MD; Nicholas Rosenlicht, MD; Steven Lieske, MD, PhD; Mitchell S. Finkel, MD; John T. Little, MD.

Affiliations of Authors/VAST-D Investigators: Veterans Affairs (VA) New England Mental Illness Research, Education, and Clinical Center, VA Connecticut Healthcare System, West Haven (D’Souza); Louis Stokes VA Medical Center and Case Western Reserve University School of Medicine, Cleveland, Ohio (Jurjus); Central Texas Veterans Healthcare System, Temple (Williams); Tuscaloosa VA Medical Center, Tuscaloosa, Alabama (Pilkinton); VA San Diego Healthcare System, San Diego, California (Rao); University of California, San Diego (Rao); Philadelphia VA Medical Center, Philadelphia, Pennsylvania (Connolly); Southern Arizona VA Healthcare System, Tucson (Wilcox); Salem VA Medical Center, Salem, Virginia (Iranmanesh, Sapra); Charles George VA Medical Center, Asheville, North Carolina (Michalets); Cincinnati VA Medical Center, Cincinnati, Ohio (Aslam); VA Eastern Colorado Healthcare System, Denver (Beresford); Kansas City VA Medical Center, Kansas City, Missouri (K. D. Anderson); VA Loma Linda Healthcare System, Loma Linda, California (Fernando); VA Nebraska Western Iowa Healthcare System, Omaha (Ramaswamy); VA Pittsburgh Healthcare System, Pittsburgh, Pennsylvania (Kasckow); Minneapolis VA Health Care System, Minneapolis, Minnesota (Westermeyer, Yoon); Clement J. Zablocki VA Medical Center, Milwaukee, Wisconsin (Larson, W. G. Anderson, Klatt); Atlanta VA Medical Center, Atlanta, Georgia (Fareed); Phoenix VA Health Care System, Phoenix, Arizona (Thompson, Carrera); William S. Middleton Veterans Hospital, Madison, Wisconsin (Juergens); Long Beach VA Healthcare System, Long Beach, California (Albers); Memphis VA Medical Center, Memphis, Tennessee (Nasdahl); New Mexico VA Healthcare System, Albuquerque (Villarreal); James A Haley VA Hospital, Tampa, Florida (Winston); Bruce W. Carter VA Medical Center, Miami, Florida (Nogues); VA Puget Sound Health Care System, American Lake/Tacoma, Washington (Tapp, Jones); Edward Hines Jr VA Hospital, Hines, Illinois (Khatkhate, Marri); VA Palo Alto Healthcare System, Palo Alto, California (Suppes); W.G. Hefner VA Medical Center, Salisbury, North Carolina (LaMotte, Hurley); Richard L Roudebush VA Medical Center, Indianapolis, Indiana (Mayeda, Niculescu); VA Maryland Healthcare System, Baltimore (Fischer, Loreck); San Francisco VA Health Care System, San Francisco, California (Rosenlicht, Lieske); Louis A. Johnson VA Medical Center, Clarksburg, West Virginia (Finkel); Washington DC VA Medical Center (Little).

Author Contributions: Mr Johnson and Ms Planeta had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Mohamed, Johnson, Chen, Hicks, J. Yoon, Gleason, Vertrees, Tal, Wilcox, Williams, Juergens, Lawrence, Huang, Zisook.

Acquisition, analysis, or interpretation of data: Mohamed, Johnson, Chen, Hicks, Davis, J. Yoon, Vertrees, Weingart, Tal, Rao, Pilkinton, Wilcox, Iranmanesh, Sapra, Jurjus, Michalets, Aslam, Beresford, K. Anderson, Fernando, Ramaswamy, Kasckow, Westermeyer, G. Yoon, D'Souza, Larson, W. Anderson, Klatt, Fareed, Thompson, Carrera, Albers, Nasdahl, Villarreal, Winston, Nogues, Connolly, Thase, Tapp, Jones, Khatkhate, Marri, Suppes, LaMotte, Hurley, Mayeda, Niculescu, Fischer, Loreck, Rosenlicht, Lieske, Finkel, Little, Scrymgeour, Lawrence, Planeta, Huang, Zisook.

Drafting of the manuscript: Mohamed, Johnson, Chen, Hicks, Davis, Gleason, Vertrees, Weingart, Tal, Wilcox, Sapra, Michalets, Thompson, Winston, Nogues, Finkel, Scrymgeour, Lawrence, Planeta, Zisook.

Critical revision of the manuscript for important intellectual content: Mohamed, Johnson, Chen, Hicks, Davis, J. Yoon, Gleason, Vertrees, Weingart, Tal, Rao, Pilkinton, Wilcox, Iranmanesh, Sapra, Jurjus, Aslam, Beresford, K. Anderson, Fernando, Ramaswamy, Kasckow, Westermeyer, G. Yoon, D'Souza, Larson, W. Anderson, Klatt, Fareed, Carrera, Williams, Juergens, Albers, Nasdahl, Villarreal, Connolly, Thase, Tapp, Jones, Khatkhate, Marri, Suppes, LaMotte, Hurley, Mayeda, Niculescu, Fischer, Loreck, Rosenlicht, Lieske, Little, Scrymgeour, Planeta, Huang, Zisook.

Statistical analysis: Mohamed, Johnson, Thase, Lawrence, Planeta, Zisook.

Obtained funding: Mohamed, Johnson, Vertrees, Huang, Zisook.

Administrative, technical, or material support: Mohamed, Johnson, Chen, Hicks, Davis, J. Yoon, Gleason, Vertrees, Weingart, Tal, Pilkinton, Wilcox, Sapra, Jurjus, Beresford, G. Yoon, W. Anderson, Fareed, Thompson, Williams, Juergens, Villarreal, Nogues,, Tapp, Jones, Khatkhate, LaMotte, Hurley, Mayeda, Niculescu, Fischer, Loreck, Finkel, Little, Scrymgeour, Huang, Zisook.

Supervision: Mohamed, Johnson, Chen, Davis, Vertrees, Tal, Iranmanesh, Fernando, Kasckow, Westermeyer, D'Souza, Fareed, Nasdahl, Jones, Suppes, Loreck, Rosenlicht, Lieske, Finkel, Little, Huang, Zisook.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported. Mr Johnson reports his spouse being an employee of and owning stock in Bristol-Myers Squibb. Dr Davis reports receiving grant funding from Tonix and personal fees from Otsuka, Tonix, Lundbeck, and Bracket. Dr Rao reports being in the speaker’s bureau for Janssen, Otsuka, and Alkermes. Dr Thase reports consulting for Alkermes, AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Forest Laboratories, Gerson Lehman Group, GlaxoSmithKline, Guidepoint Global, Lundbeck, MedAvante, Merck, Neuronetics, Ortho-McNeil Pharmaceuticals, Otsuka, Pfizer, Roche, Shire US, Sunovion, Takeda, American Psychiatric Foundation, Guilford Publications, Herald House, W.W. Norton & Company, Peloton Advantage, Cerecor, Moksha8, Pamlab, Allergan, Trius Therapeutical, Fabre-Kramer Pharmaceuticals; and receiving grant funding from Eli Lilly, Forest Laboratories, Otsuka. Dr Suppes reports receiving grant funding from Merck, Sunovion, Stanley Medical Research Institute, Palo Alto Health Sciences Services and Elan Pharma International Limited; personal fees from AstraZeneca, Lundbeck, Merck, Sunovion, Global Medication Education, CMEology, and Medscape Education; and royalties from Jones and Bartlett and UpToDate. No other disclosures were reported.

Funding/Support: This study was supported by the Veterans Affairs Cooperative Studies Program of the Department of Veterans Affairs, Office of Research and Development, which participated in the design and oversaw the conduct of the study including data collection and management and analysis. Bristol-Myers Squibb provided aripiprazole for use in this study.

Role of the Funder/Sponsor: The Veteran Affairs Cooperative Studies Program was involved in the design and conduct of the study and the collection, management, analysis and interpretation of the data, and preparation, review and approval of the manuscript, however, had no role in the decision to submit the manuscript for publication.

Group Information: A complete list of collaborators for the VAST-D study is provided in the eAppendix in Supplement 3.

Disclaimer: Opinions herein are those of the individual authors and the contents do not represent views of the Department of Veterans Affairs or the US government.

References
1.
Murray  CJ, Atkinson  C, Bhalla  K,  et al; US Burden of Disease Collaborators.  The state of US health, 1990-2010: burden of diseases, injuries, and risk factors.  JAMA. 2013;310(6):591-608.PubMedGoogle ScholarCrossref
2.
Center for Behavioral Health Statistics and Quality.  Key substance use and mental health indicators in the United States: results from the 2015 National Survey on Drug Use and Health. https://www.samhsa.gov/data/sites/default/files/NSDUH-FFR1-2015/NSDUH-FFR1-2015/NSDUH-FFR1-2015.pdf. Accessed June 19, 2017.
3.
Cleare  A, Pariante  CM, Young  AH,  et al; Members of the Consensus Meeting.  Evidence-based guidelines for treating depressive disorders with antidepressants: a revision of the 2008 British Association for Psychopharmacology guidelines.  J Psychopharmacol. 2015;29(5):459-525.PubMedGoogle ScholarCrossref
4.
Kolovos  S, van Tulder  MW, Cuijpers  P,  et al.  The effect of treatment as usual on major depressive disorder: a meta-analysis.  J Affect Disord. 2017;210:72-81.PubMedGoogle ScholarCrossref
5.
American Psychiatric Association.  Practice Guideline for the Treatment of Patients With Major Depressive Disorder. 3rd ed. Arlington, VA: American Psychiatric Association; 2010.
6.
Management of MDD Working Group.  Clinical practice guideline: management of major depressive disorder (MDD). https://www.healthquality.va.gov/mdd/mdd_full09_c.pdf. Accessed June 19, 2017.
7.
Kennedy  SH, Lam  RW, McIntyre  RS,  et al; CANMAT Depression Work Group.  Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 clinical guidelines for the management of adults with major depressive disorder: section 3: pharmacological treatments.  Can J Psychiatry. 2016;61(9):540-560.PubMedGoogle ScholarCrossref
8.
Goldberg  JF, Freeman  MP, Balon  R,  et al.  The American Society of Clinical Psychopharmacology survey of psychopharmacologists’ practice patterns for the treatment of mood disorders.  Depress Anxiety. 2015;32(8):605-613.PubMedGoogle ScholarCrossref
9.
Rush  AJ, Trivedi  MH, Wisniewski  SR,  et al; STAR*D Study Team.  Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression.  N Engl J Med. 2006;354(12):1231-1242.PubMedGoogle ScholarCrossref
10.
Trivedi  MH, Fava  M, Wisniewski  SR,  et al; STAR*D Study Team.  Medication augmentation after the failure of SSRIs for depression.  N Engl J Med. 2006;354(12):1243-1252.PubMedGoogle ScholarCrossref
11.
Rush  AJ, Warden  D, Wisniewski  SR,  et al.  STAR*D: revising conventional wisdom.  CNS Drugs. 2009;23(8):627-647.PubMedGoogle Scholar
12.
Mohamed  S, Leslie  DL, Rosenheck  RA.  Use of antipsychotics in the treatment of major depressive disorder in the US Department of Veterans Affairs.  J Clin Psychiatry. 2009;70(6):906-912.PubMedGoogle ScholarCrossref
13.
Arbaizar  B, Dierssen-Sotos  T, Gómez-Acebo  I, Llorca  J.  Aripiprazole in major depression and mania: meta-analyses of randomized placebo-controlled trials.  Gen Hosp Psychiatry. 2009;31(5):478-483.PubMedGoogle ScholarCrossref
14.
Han  C, Wang  S-M, Kwak  K-P,  et al.  Aripiprazole augmentation versus antidepressant switching for patients with major depressive disorder: a 6-week, randomized, rater-blinded, prospective study.  J Psychiatr Res. 2015;66-67:84-94.PubMedGoogle ScholarCrossref
15.
Mohamed  S, Johnson  GR, Vertrees  JE,  et al.  The VA augmentation and switching treatments for improving depression outcomes (VAST-D) study: rationale and design considerations.  Psychiatry Res. 2015;229(3):760-770.PubMedGoogle ScholarCrossref
16.
Kroenke  K, Spitzer  RL, Williams  JB.  The PHQ-9: validity of a brief depression severity measure.  J Gen Intern Med. 2001;16(9):606-613.PubMedGoogle ScholarCrossref
17.
Rush  AJ, Trivedi  MH, Ibrahim  HM,  et al.  The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression.  Biol Psychiatry. 2003;54(5):573-583.PubMedGoogle ScholarCrossref
18.
Trivedi  MH.  Tools and strategies for ongoing assessment of depression: a measurement-based approach to remission.  J Clin Psychiatry. 2009;70(suppl 6):26-31.PubMedGoogle ScholarCrossref
19.
Wisniewski  SR, Rush  AJ, Balasubramani  GK, Trivedi  MH, Nierenberg  AA; STARD Investigators.  Self-rated global measure of the frequency, intensity, and burden of side effects.  J Psychiatr Pract. 2006;12(2):71-79.PubMedGoogle ScholarCrossref
20.
Israel  JA.  Remission in depression: definition and initial treatment approaches.  J Psychopharmacol. 2006;20(3)(suppl):5-10.PubMedGoogle ScholarCrossref
21.
Guy  W.  ECDEU Assessment Manual for Psychopharmacology-Revised. Rockville, MD: Dept of Health, Education, and Welfare; 1976.
22.
Levine  J, Schooler  NR.  SAFTEE: a technique for the systematic assessment of side effects in clinical trials.  Psychopharmacol Bull. 1986;22(2):343-381.PubMedGoogle Scholar
23.
Lieberman  JA, Stroup  TS, McEvoy  JP,  et al; Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) Investigators.  Effectiveness of antipsychotic drugs in patients with chronic schizophrenia.  N Engl J Med. 2005;353(12):1209-1223.PubMedGoogle ScholarCrossref
24.
Hochberg  Y.  A sharper Bonferroni procedure for multiple tests of significance.  Biometrika. 1988;75(4):800-802.Google ScholarCrossref
25.
Zisook  S, Tal  I, Weingart  K,  et al.  Characteristics of US veteran patients with major depressive disorder who require “next-step” treatments: a VAST-D report.  J Affect Disord. 2016;206:232-240.PubMedGoogle ScholarCrossref
26.
Fava  M, Rush  AJ, Alpert  JE,  et al.  Difference in treatment outcome in outpatients with anxious versus nonanxious depression: a STAR*D report.  Am J Psychiatry. 2008;165(3):342-351.PubMedGoogle ScholarCrossref
27.
Sansone  RA, Sansone  LA.  Antidepressant adherence: are patients taking their medications?  Innov Clin Neurosci. 2012;9(5-6):41-46.PubMedGoogle Scholar
28.
Adams  JR, Drake  RE.  Shared decision-making and evidence-based practice.  Community Ment Health J. 2006;42(1):87-105.PubMedGoogle ScholarCrossref
29.
American Diabetes Association; American Psychiatric Association; American Association of Clinical Endocrinologists; North American Association for the Study of Obesity.  Consensus development conference on antipsychotic drugs and obesity and diabetes.  J Clin Psychiatry. 2004;65(2):267-272.PubMedGoogle ScholarCrossref
30.
Rush  AJ, Trivedi  MH, Wisniewski  SR,  et al.  Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report.  Am J Psychiatry. 2006;163(11):1905-1917.PubMedGoogle ScholarCrossref
×