Skip to main content

Transforming the understanding
and treatment of mental illnesses.

Celebrating 75 Years! Learn More >>

Questions and Answers about the NIMH Sequenced Treatment Alternatives to Relieve Depression (STAR*D) Study — Background

January 2006

1. What is the STAR*D study?

A. The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) Study, funded by the NIH National Institute of Mental Health, is a nationwide public health clinical trial. The purpose of the trial is to determine the effectiveness of different treatments for people with Major Depressive Disorder (MDD) who have not responded to initial treatment with an antidepressant. This is the largest and longest study ever done to evaluate depression treatment. Over a seven-year period, the study enrolled more than 4,000 outpatients, aged 18-75 years. The participants were people who came to their doctors for care and who had other psychiatric and medical conditions like those regularly seen in typical clinical practices (see Ref 1). Patients from both mental health (specialty care) and non-mental health (primary medical care) practices in diverse racial, ethnic, and socioeconomic populations were enrolled in the study.

See Questions and Answers about Level 1 Results.

2. Why is the STAR*D study important?

A. Each year, 9.5 percent of the population, or about 20.9 million American adults, will struggle with depressive illness. MDD is a recurring and chronic illness, frequently returning for two or more episodes, with episodes that often last two years or more. Depression is currently the fourth most disabling illness worldwide, and according to the World Health Organization, it will be the second leading cause of disability by the year 2020. About 10 percent of men and up to 25 percent of women will experience depression in their lifetime. People with depression fare even more poorly at work, socially, and with their families than do people with a variety of general medical conditions. Depression is responsible for up to 70 percent of psychiatric hospitalizations and about 40 percent of suicides. The cost of depression in the United States in the year 2000 was estimated to be $83 billion, including both $26 billion in costs of treatment and $57 billion in losses such as absenteeism, reduced productivity at work, and the value of lifetime earnings lost due to suicide-related deaths.

There is only limited information about how to successfully treat people with depression, especially how to treat people with depression who have not gotten better with the first treatment tried. Depression is considered “treatment-resistant” when at least one adequately delivered treatment does not lead a person to reach a “remission” of their depressive symptoms, that is, to become symptom-free. About half the participants who were treated with a first medication or talk therapy in prior studies showed a reduction in their symptoms to at least half the level of intensity. Of those, about two-thirds actually became symptom-free.

3. How will the results of STAR*D affect the care of depression?

A. There are nearly 20 different antidepressant medications and several well-established forms of talk therapy available to treat depression. Many people get better on the first treatment they receive for depression. Many, however, will need a second, third, or even fourth treatment. The main goal of the STAR*D study was to identify the best next steps for those people with depression who need to try more than one treatment. That is, which treatment strategies are the most effective after people do not become symptom-free after one or more treatments?

The results of STAR*D will provide data needed to inform the practice guidelines that doctors and other practitioners use for treating people who have depression. The results will provide descriptive information about specific treatment strategies, such as, at what dose a medication should be prescribed and how long a treatment should be tried before deciding that it is ineffective and moving on to another treatment strategy. Results also will provide much needed information on the long-term course of depression in terms of the frequency, nature, and timing of relapses (return of depressive symptoms less than 6 months after a remission) and will provide information about the cost effectiveness of various treatments. Furthermore, the results will help determine whether there are characteristics of patients, their depressive symptoms and/or their responses to different treatments that might help doctors and other practitioners prescribe the most effective treatment for a particular patient at the outset.

4. How is the STAR*D study different from other studies of depression?

A. STAR*D has a unique study design (see Ref 2) that mimics clinical practice and ensures high levels of patient participation, making the results directly applicable to practitioners and their patients. This study was conducted by psychiatrists and primary care physicians in both private practice and public clinics to reflect the treatments patients typically receive in community settings. The study reflects what is done in clinical practice by allowing the study participants to choose what treatments would be acceptable to them, and then limiting randomization of each participant only to his/her range of acceptable treatment strategies. No prior studies have evaluated the acceptability of different treatment strategies in broadly defined participant groups treated in diverse care settings.

STAR*D aimed for the first time to provide solid scientific evidence regarding which next treatment steps are best for treatment-resistant depression. The results should produce clinical, administrative and economic benefits.

5. How did the researchers measure how well the medications worked?

A. Each participant was evaluated upon entry into the study and frequently throughout the study to assess symptoms, side effects, work, social and family function, as well as costs of treatment.

At each participant visit, STAR*D investigators measured symptoms and side effects to determine when and how much to increase medication doses or change to other treatments. Consistently providing this quality treatment ensured that participants had the best possible chance of benefiting from the treatments. Use of these standardized measurements is different than what is commonly done in regular community treatment practices, but was necessary in a research study to ensure that all participants were treated similarly and that decisions about each participant’s medication treatment were based on the same kinds of information. However, these measurements are simple and easy to use and could easily be adopted by the larger clinical community.

To ensure that there would be no bias in assessing how well each treatment worked, the information that was used for measuring the outcome results of the study was collected both by an expert clinician over the phone who had no knowledge of what treatment the participants were receiving and by a novel computer-based interactive voice response system.

6. What treatments were used in STAR*D and how was it decided which medication to give to which participants?

A. STAR*D treatments were selected with attention to what practitioners commonly use to treat clinical depression. The treatments selected were among the safest, easiest to take, and most frequently used. There were no placebo treatments used in this study. Four levels of treatment were used in STAR*D and treatment at each level could continue for up to 14 weeks.

All participants began at Level 1, and were treated with the antidepressant citalopram, a selective serotonin reuptake inhibitor (SSRI). If a participant did not become symptom-free or could not tolerate side effects from the citalopram, he/she was encouraged to progress to the next level (Level 2), where new treatment choices would be available.

Level 2 of the study offered seven different treatments; four options “switched” the citalopram participants to a new medication or talk therapy, and three options “augmented” citalopram treatment by adding a new medication or talk therapy to the citalopram they were already receiving. If participants progressed to Level 2 treatment, they were agreeing to have a treatment selected randomly (like by tossing a coin) from among all or some of the available treatment approaches that were personally acceptable.

There were two additional levels of treatment (Level 3 and Level 4) that were available for participants who did not become symptom-free on prior level treatments and who wanted to continue to try other treatments. Each of these levels also included random selection of treatment from among individually acceptable treatment options.

After each level of treatment, those participants whose depression did not respond to the treatment were encouraged to go on to the next level of treatment. Those participants who got somewhat better, but did not become symptom-free were also were encouraged to go on to the next level of treatment; however, if those participants chose to stay with their present treatment, they could enter a 12-month follow-up period during which their long-term outcomes were monitored. Participants who fully recovered (reached remission of their depressive symptoms) began the 12-month follow-up process at that point.

STAR*D Study Medications

  • Citalopram (Celexa)
  • Sertraline (Zoloft)
  • Bupropion SR (Wellbutrin SR)
  • Venlafaxine XR (Effexor XR)
  • Buspirone (BuSpar)
  • Mirtazapine (Remeron)
  • Triiodothyronine (T3) (Cytomel)
  • Nortriptyline (Pamelor, Aventyl)
  • Tranylcypromine (Parnate)
  • Lithium (Eskalith, Lithobid)

STAR*D Talk Therapy

  • Cognitive Therapy

7. Who participated in STAR*D and how were they chosen?

A. Overall, 4,041 participants were enrolled into Level 1 of the STAR*D study. The average age was 41 years, with about 13 years of education; 64 percent were female, 36 percent were male. These participants represent the broad range of ethnic and socioeconomic populations in the U.S. as indicated by the similarity to U. S. Census figures: about 18 percent were African American; 76 percent were Caucasian and 6 percent, other races. Approximatley 13 percent were Hispanic.

Practitioners in the clinical sites identified people with depression who were either already in treatment or were about to be treated at the clinics and offered them the opportunity to participate in the study. Media advertising was not used to attract participants to the STAR*D study. The participants in STAR*D had moderate to severe depressive symptoms, with 75 percent having two or more episodes of depression in their lifetime and 25 percent reporting that their current episode of depression had continued for at least two years. More than one third of the participants were under age 18 when they first experienced depression and participants had an average of three or more general medical conditions.

Of the 4041 that entered Level 1, 1438 entered the Level 2 randomized portion of the trial, 376 entered Level 3, and 108 entered Level 4. Participants who did not move on to another level had either achieved remission or a satisfactory response at that level, or left the STAR*D study completely at their own discretion.

8. Who conducted STAR*D?

A. After a competitive, peer-reviewed process, NIMH selected the University of Texas Southwestern Medical Center in Dallas to implement the STAR*D trial. The trial was led by Dr. John Rush of the University of Texas Southwestern Medical Center, with the help of Dr. Madhukar Trivedi of the University of Texas Southwestern Medical Center and Dr. Maurizio Fava from Massachusetts General Hospital, as well as Dr. Stephen Wisniewski from the Epidemiology Data Center at the University of Pittsburgh and Dr. Andrew Nierenberg from Massachusetts General Hospital. Other key scientific leadership included Drs. Michael Thase and Edward Friedman from the University of Pittsburgh, Dr. Fred Quitkin from Columbia University and Dr. Jonathan Alpert from Massachusetts General Hospital. A National Coordinating Center at the University of Texas Southwestern Medical Center that included Drs. Diane Warden, Melanie Biggs and Kathy Shores-Wilson, and Diane Stegman, RN oversaw the execution and overall operation of the study. The Epidemiology Data Center at the University of Pittsburgh, led by Dr. Stephen Wisniewski, collected and analyzed the study data.

There were 14 Regional Centers that managed the implementation of the study at 41 clinical sites where participants were enrolled and treated:

Regional Center Regional Center Director(s)
Massachusetts General Hospital Maurizio Fava, M.D.
Andrew Nierenberg, M.D.
Univ. of North Carolina at Chapel Hill Robert Golden, M.D.
Brad Gaynes, M.D.
The University of Texas Southwestern Medical Center at Dallas Mustafa Husain, M.D.
Clinical Research Institute-Kansas Sheldon Preskorn, M.D.
University of Michigan Elizabeth Young, M.D.
Northwestern University William McKinney, M.D.
William Gilmer, M.D.
New York State Psychiatric Institute Fred Quitkin, M.D.
Patrick McGrath, M.D.
University of Pittsburgh Michael Thase, M.D.
University of California at San Diego Sidney Zisook, M.D.
Laureate Psychiatric Hospital & Clinic Jeff Mitchell, M.D.
Oklahoma Tuscaloosa VA Medical Center Lori Davis, M.D.
University of California at Los Angeles Andrew Leuchter, M.D.
Vanderbilt University Steven Hollon, Ph.D.
Richard Shelton, M.D.
Virginia Commonwealth University Susan Kornstein, M.D.

The overall cost of the study was $35 million over 6 years.

9. What role did the pharmaceutical companies have in STAR*D?

A. All of the study medications were provided free of charge by the pharmaceutical companies that manufactured them. Aside from providing feedback about the dose proposed by STAR*D investigators for each company’s medication, the pharmaceutical companies had no other input into the design or implementation of the study, nor did they have involvement in planning or conducting the data analyses, and they did not participate in preparing manuscripts for publication.

The Level 1 antidepressant Celexa (citalopram) was provided by Forest Laboratories. Level 2 drugs were Zoloft (sertraline), provided by Pfizer; Wellbutrin SR (bupropion) from GlaxoSmithKline; Effexor (venlafaxine XR) from Wyeth; and BuSpar (buspirone) from Bristol-Myers-Squibb. Level 3 medications provided were Remeron SolTabs (mirtazepine) from Organon, and Cytomel (T3) triiodothyronine) from King Pharmaceuticals. Other medications such as lithium, nortriptyline and tranylcypromine were purchased in generic form and were paid for out of study funds. All study medications were dispensed at no charge to study participants from any pharmacy that they chose to access with their prescriptions and the STAR*D pharmacy card. Pfizer provided Viagra (sildenafil) at no cost to study participants for the management of sexual dysfunction side effects from the antidepressant medication.

10. How will the results of STAR*D be made known to practitioners, patients, and the public?

A. The results of STAR*D will be made available in a variety of ways to reach a wide audience of patients, mental health professionals, medical providers and the public. A series of publications about the development of STAR*D and the characteristics of the participants have already appeared in many scientific and medical journals. See STAR*D on .

Results of Levels 1 and 2 are being disseminated in scientific journal articles to appear in early 2006. A newsletter will be available to the participants of STAR*D, informing them of the results and thanking them for their participation in this groundbreaking study. The STAR*D research clinicians and research staff will make the results available to their clinical sites, and professional associations through newsletters, teleconferences and presentations at psychiatric meetings and advocacy organizations. The NIMH will disseminate the results of STAR*D through a defined outreach to practitioners, the media, and the public. As results become available from further detailed analyses of Levels 1 and 2, and from Levels 3 and 4, they too will be disseminated using similar strategies.


1. Gaynes BN, Rush AJ, Trivedi M, Wisniewski SR, Balasubramani GK, Spencer DC, Petersen T, Klinkman M, Warden D, Schneider RK, Castro DB, Golden RN. A direct comparison of presenting characteristics of depressed outpatients from primary vs. specialty care settings: preliminary findings from the STAR*D clinical trial. Gen Hosp Psychiatry. 2005 Mar-Apr;27(2):87-96.

2. Rush AJ, Fava M, Wisniewski SR, Lavori PW, Trivedi MH, Sackeim HA, Thase ME, Nierenberg AA, Quitkin FM, Kashner TM, Kupfer DJ, Rosenbaum JF, Alpert J, Stewart JW, McGrath PJ, Biggs MM, Shores-Wilson K, Lebowitz BD, Ritz L, Niederehe G; STAR*D Investigators Group. Sequenced treatment alternatives to relieve depression (STAR*D): rationale and design. Control Clin Trials. 2004 Feb;25(1):119-42.