Dr. Howland is Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania.
Dr. Howland acknowledges grant support from the National Institute of Mental Health.
Address correspondence to Robert H. Howland, MD, Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, 3811 O’Hara Street, Pittsburgh, PA 15213; e-mail: HowlandRH@upmc.edu.
The goal of antidepressant therapy should be the absence of significant depressive symptoms, an outcome referred to as full remission (Rush, Kraemer, et al., 2006). On the basis of typical results of most controlled antidepressant trials, approximately 50% of patients will have a significant treatment response (defined as a 50% or greater decrease in depressive symptoms) with an initial medication. However, in such controlled trials, only 25% to 33% will have attained full remission. A significant proportion of depressed patients are left with residual symptoms, despite an adequate initial antidepressant drug therapy. What is the best next step treatment for these patients (Howland & Thase, 1999)? The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study was conducted to answer this question. In this article, I will describe the overall design of the STAR*D study. Next month, I will review the main outcome findings from STAR*D.
Design of the STAR*D Study
The STAR*D study, funded by the National Institute of Mental Health, is the largest prospective, randomized antidepressant treatment trial ever conducted in patients with major depressive disorder, involving more than 4,000 adult patients from real-world clinical settings (Rush et al., 2004). The treatment goal in STAR*D was to achieve clinical remission. The main objective of the study was to determine which treatments are most effective after nonremission or intolerance to an initial serotonin reuptake inhibitor (SRI) antidepressant medication (Level 1 treatment with citalopram [Celexa®]). This objective was accomplished by sequentially investigating a series of three subsequent levels of randomized treatments to achieve remission.
STAR*D was conducted across the United States at 18 primary care and 23 psychiatric settings that serve public and private sector patients. Clinical research coordinators (CRCs) at each site assisted patients and clinicians in protocol implementation and the collection of clinical measures. A central pool of research outcome assessors (ROAs), not located at any clinical site, conducted telephone interviews to obtain primary outcomes data.
STAR*D enrolled treatment-seeking patients who visited these clinics rather than soliciting symptomatic volunteers through advertising or other typical research recruitment procedures. To maximize the generalizability of the results, broad inclusion and minimal exclusion criteria were used. Patients with a baseline score of 14 or greater on the 17-item Hamilton Rating Scale for Depression (HAM-D) were enrolled if their treating clinicians determined that outpatient treatment with an antidepressant medication was appropriate.
Exclusion criteria included bipolar disorder; psychotic symptoms or disorders; current primary diagnosis of obsessive-compulsive disorder, anorexia nervosa, or bulimia; general medical conditions contraindicating medications used in the first two treatment levels of the study; substance dependence requiring inpatient treatment; or a clear history of nonresponse or intolerance (during the current depressive episode) to any medication used in the first two treatment levels of the study. Patients who were pregnant, breast-feeding, or intending to conceive during the 9 months subsequent to study entry were also excluded. Thus, the study permitted nonpsychotic, nonbipolar outpatients with most kinds of comorbid psychiatric and general medical conditions, including active substance abuse or suicidality, to be enrolled as long as outpatient treatment was considered clinically appropriate.
The success of translating clinical trials research findings to real-world practice has been limited, primarily because strategies effective in clinical trials have not been sufficiently studied in routine clinical care among patients with psychiatric and general medical comorbidities. The STAR*D study required that the dosage and duration of antidepressant medication treatment be optimal, while accommodating the flexibility necessary to ensure patient safety, given the wide range of comorbid general medical and psychiatric disorders allowed in the trial.
The study used a measurement-based care (MBC) approach and an automated feedback system to ensure adequate and safe antidepressant treatment delivery suitable for both clinical research and routine practice (Trivedi et al., 2007). Ratings of depressive symptom severity and side effect frequency, intensity, and burden were obtained at each treatment visit by CRCs using the MBC system, which guided medication dosage adjustments and treatment duration, documented clinician adherence to treatment recommendations, and provided prompt feedback to clinicians to enhance appropriate treatment decisions. Physician adherence to protocol-specific treatment was monitored based on measured symptoms and side effect burden, as well as the dosage and duration of antidepressant agents at each critical decision point during the acute phase treatment.
Feedback was provided at the point of care by the CRCs, assisted by Web-based reports following each treatment visit. During Level 1 treatment with citalopram, for example, more than 85% of treatment encounters were consistent with study recommendations. Most deviations from treatment recommendations occurred late in treatment and were often justifiable. The MBC system was found to be feasible and effective in busy primary care and psychiatric settings.
Four Treatment Levels
Throughout the four levels of STAR*D, a diverse range of antidepressant treatments were chosen for comparison of their efficacy and tolerability (Rush et al., 2004). At each of the four levels, the highest tolerated medication dosage (within the usual clinical dosage range for each drug) necessary to achieve remission was used, guided by the MBC system. Patients were treated for up to 12 weeks at each level (with an optional week 14 visit, if needed). All patients started with citalopram at Level 1 (Trivedi, Rush, et al., 2006). After Level 1, the three subsequent levels of treatment options were randomized and open label (i.e., not blinded or placebo controlled). Similar to clinical practice, however, patients could accept or decline particular treatments as long as sufficient options were left that allowed a randomization between at least two different options (Wisniewski et al., 2007).
Patients who did not remit or were intolerant to citalopram could enter Level 2. This level was the most clinically and scientifically important phase of STAR*D, because patients were randomly switched from citalopram to one of four alternatives (Rush, Trivedi, et al., 2006), or they continued to take citalopram and were augmented with one of three additional treatments (Trivedi, Fava, et al., 2006). The four switch options were to bupropion SR (Wellbutrin SR®), sertraline (Zoloft®), venlafaxine XR (Effexor XR®), or cognitive psychotherapy. The three augmentation options were adding bupropion SR, buspirone (Buspar®), or cognitive psychotherapy. For those patients having inadequate benefit from switching to or augmenting with cognitive psychotherapy (Thase et al., 2007), the next step (Level 2A) was a switch to bupropion SR or venlafaxine XR. Level 2A ensured that all patients entering Level 3 had received two different pharmacotherapy trials.
Patients without adequate benefit from Level 2 or 2A treatment were advanced to Level 3, where they were randomly switched from their Level 2/2A medication (Fava et al., 2006), or their Level 2/2A medication was augmented (Nierenberg et al., 2006). The two Level 3 switch options were to mirtazapine (Remeron®) or nortriptyline (Pamelor®). The two Level 3 augmentation options were adding lithium (Eskalith®) or thyroid hormone (triiodothyronine, T3 [Cytomel®]). Finally, at Level 4, patients were randomly switched to tranylcypromine (Parnate®) or to the combination of venlafaxine XR plus mirtazapine (McGrath et al., 2006).
The HAM-D scale was the primary research outcome obtained by ROAs using telephone-based structured interviews in English or Spanish. Secondary outcomes were based on the 16-item Quick Inventory of Depressive Symptomatology, Self-Report (QIDS-SR) scale, obtained at baseline and at each treatment visit by the CRCs. Remission was defined as an exit score of 7 or less on the HAM-D (primary outcome measure) or a score of 5 or less on the QIDS-SR (secondary outcome measure). Response was defined as a reduction of 50% or greater on the QIDS-SR score from baseline to exit. Side effects were assessed at each visit using the self-report Frequency, Intensity, and Burden of Side Effects Ratings (FIBSER) scale.
Through the MBC system, QIDS-SR and FIBSER data from each visit were used to guide treatment decisions. For any level of treatment, patients who reached full remission, and those with a satisfactory response, could continue the same treatment during 1-year naturalistic follow up. Patients who did not reach full remission were encouraged to enter subsequent levels. Patients could discontinue treatment before 12 weeks if intolerable side effects warranted a medication change, if an optimal dosage increase was not possible (because of side effects or by patient choice), or if significant depressive symptoms were still present after 9 weeks despite maximum tolerated dosages. These patients were eligible to enter subsequent levels.
For patients not responding adequately to an initial SRI antidepressant medication, the STAR*D study was designed to investigate the effectiveness of a sequential series of next-step treatments guided by an MBC system. Nurses should be familiar with the design of STAR*D because the study compared a diverse assortment of antidepressant therapies in real-world clinical settings. The main outcome findings from STAR*D will be reviewed in next month’s article.
- Fava, M, Rush, AJ, Wisniewski, SR, Nierenberg, AA, Alpert, JE & McGrath, PJ et al. 2006. A comparison of mirtazapine and nortriptyline following two consecutive failed medication treatments for depressed outpatients: A STAR*D report. American Journal of Psychiatry, 163, 1161–1172. doi:10.1176/appi.ajp.163.7.1161 [CrossRef]
- Howland, RH & Thase, ME1999. What to do with SSRI nonresponders?Journal of Practical Psychiatry and Behavioral Health, 5, 216–223.
- McGrath, PJ, Stewart, JW, Fava, M, Trivedi, MH, Wisniewski, SR & Nierenberg, AA et al. 2006. Tranylcypromine versus venlafaxine plus mirtazapine following three failed antidepressant medication trials for depression: A STAR*D report. American Journal of Psychiatry, 163, 1531–1541. doi:10.1176/appi.ajp.163.9.1531 [CrossRef]
- Nierenberg, AA, Fava, M, Trivedi, MH, Wisniewski, SR, Thase, ME & McGrath, PJ et al. 2006. A comparison of lithium and T(3) augmentation following two failed medication treatments for depression: A STAR*D report. American Journal of Psychiatry, 163, 1519–1530. doi:10.1176/appi.ajp.163.9.1519 [CrossRef]
- Rush, AJ, Fava, M, Wisniewski, SR, Lavori, PW, Trivedi, MH & Sackeim, HA et al. 2004. Sequenced Treatment Alternatives to Relieve Depression (STAR*D): Rationale and design. Controlled Clinical Trials, 25, 119–142. doi:10.1016/S0197-2456(03)00112-0 [CrossRef]
- Rush, AJ, Kraemer, HC, Sackeim, HA, Fava, M, Trivedi, MH & Frank, E et al. 2006. Report by the ACNP task force on response and remission in major depressive disorder. Neuropsychopharmacology, 31, 1841–1853. doi:10.1038/sj.npp.1301131 [CrossRef]
- Rush, AJ, Trivedi, MH, Wisniewski, SR, Stewart, JW, Nierenberg, AA & Thase, ME et al. 2006. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. New England Journal of Medicine, 354, 1231–1242. doi:10.1056/NEJMoa052963 [CrossRef]
- Thase, ME, Friedman, ES, Biggs, MM, Wisniewski, SR, Trivedi, MH & Luther, JF et al. 2007. Cognitive therapy versus medication in augmentation and switch strategies as second-step treatments: A STAR*D report. American Journal of Psychiatry, 164, 739–752. doi:10.1176/appi.ajp.164.5.739 [CrossRef]
- Trivedi, MH, Fava, M, Wisniewski, SR, Thase, ME, Quitkin, F & Warden, D et al. 2006. Medication augmentation after the failure of SSRIs for depression. New England Journal of Medicine, 354, 1243–1252. doi:10.1056/NEJMoa052964 [CrossRef]
- Trivedi, MH, Rush, AJ, Gaynes, BN, Stewart, JW, Wisniewski, SR & Warden, D et al. 2007. Maximizing the adequacy of medication treatment in controlled trials and clinical practice: STAR*D measurement-based care. Neuropsychopharmacology, 32, 2479–2489. doi:10.1038/sj.npp.1301390 [CrossRef]
- Trivedi, MH, Rush, AJ, Wisniewski, SR, Nierenberg, A, Warden, D & Ritz, L et al. 2006. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: Implications for clinical practice. American Journal of Psychiatry, 163, 28–40. doi:10.1176/appi.ajp.163.1.28 [CrossRef]
- Wisniewski, SR, Fava, M, Trivedi, MH, Thase, ME, Warden, D & Niederehe, G et al. 2007. Acceptability of second-step treatments to depressed outpatients: A STAR*D report. American Journal of Psychiatry, 164, 753–760. doi:10.1176/appi.ajp.164.5.753 [CrossRef]