Major Depression in the Real World – The STAR*D Trial

Major depression is a very common and debilitating. It is characterized by low mood, changes in sleeping patterns, changes in appetite, lack of energy and a very substantial loss of quality of life. Depression may not improve for long periods of time, often years, and that someone who has experienced one episode of major depression is likely to experience more. Frequently, depression is accompanied by other medical conditions, like for example cancer or heart disease.

Since depression is so frequent it is not surprising that there are a wide range of treatments available. Options include antidepressant medication, cognitive behavioral therapy, and elctroconvulsive therapy in relatively rare and severe cases. Now, how treatable is depression? The answer to that is complex. There appears to be no one best treatment for major depression, and not all treatments work for everybody. Moreover, the treatments do not work quickly. So if a depressed patient sees their physician today, the prescription they get may not really take effect for several weeks, if at all. So, now what?

For the patient who did not get better, this means there are other treatments available. Should the next treatment be from the same class drug class or a different one? Should the first medication continue, or should a second one be added? Maybe cognitive behavioral therapy would help? Questions like these were explored in the largest prospective randomized treatment study to date, the Sequenced Treatments to Relieve Depression Trial (STAR*D). Unlike most previous trials, STAR*D allowed patients with comorbidities, and was conducted under real world conditions. It followed an initial sample of over 4000 patients over four treatment steps. A new treatment was tried whenever the previous treatment did not lead to remission of the patient’s depressive symptoms. Everyone started on an SSRI in step 1. In step 2, several other drugs and cognitive behavioral therapy became available. Step 3 added more drugs, among them the tricyclic antidepressants. And finally in step 4 electroconvulsive therapy and yet another drug class, monoamine oxidase inhibitors were tried.

And what was learned? Some patients reached remission at every step, meaning they had no depressive symptoms. At the first step, it was 37%, at the second step, 31%, 14% at the third step and finally 13% at the last step. At the end of the trial over 70% of patients had remitted. So, mostly depression is treatable, although it can take time, each of the treatment trials could take 12 weeks, and four of them might be needed.

But the knowledge gained does not stop there. STAR*D yielded a huge set of raw data that will be valuable for a long time to come. Let’s look at two examples.

First, the placebo effect has long been known and describes the sum of all treatment effects that are not directly attributable to the medication under study. In a placebo controlled trial, a patient in the placebo group will be treated exactly like one on active, medication. That is, they will see the physician, the nurses, will receive some positive attention, and all that may help with their illness. The effects of the medication are over and above the placebo effect. We know from placebo controlled phase III registration trials of antidepressants that large placebo effects are common. If applied to the first steap in the STAR*D trial, this consideration of the placebo effect then tells us this: 37% of the patients remitted from the combined effects of citalopram and all the non-medication effects of treatments. So not only did 63% of the patients not remit, but a lot of the remissions may not have been due to the medication at all. So, the contribution of citalopram may to the patient’s recovery may have been relatively small. In a recent column in the Journal of Psychiatric Practice, Dr. Sheldon Preskorn estimated that the drug specific response rate for modern antidepressants is about one in four patients.

Another example of how knowledge from STAR*D remains useful after the trial comes from a study assessing the interaction of anti inflammatory drugs and antidepressants. Antidepressants raise the level of certain cytokines in the cortex, anti inflammatory drugs lower them. Is this interaction clinically relevant? Looking back to the data from STAR*D confirms that in clinical practice under real world conditions antidepressants and anti inflammatory drugs have antagonistic effects. In this way STAR*D is a a large repository of information that will remain useful into the future.


Kessler, R. (2003). The Epidemiology of Major Depressive Disorder: Results From the National Comorbidity Survey Replication (NCS-R) JAMA: The Journal of the American Medical Association, 289 (23), 3095-3105 DOI: 10.1001/jama.289.23.3095

Stafford, R., Ausiello, J., Misra, B., & Saglam, D. (2000). National Patterns of Depression Treatment in Primary Care The Primary Care Companion to The Journal of Clinical Psychiatry, 02 (06), 211-216 DOI: 10.4088/PCC.v02n0603

Warden, D., Rush, A., Trivedi, M., Fava, M., & Wisniewski, S. (2007). The STAR*D project results: A comprehensive review of findings Current Psychiatry Reports, 9 (6), 449-459 DOI: 10.1007/s11920-007-0061-3

Preskorn SH (2011). What Do the Terms “Drug-Specific Response/Remission Rate” and “Placebo” Really Mean? Journal of psychiatric practice, 17 (6), 420-424 PMID: 22108399

Warner-Schmidt, J., Vanover, K., Chen, E., Marshall, J., & Greengard, P. (2011). From the Cover: Antidepressant effects of selective serotonin reuptake inhibitors (SSRIs) are attenuated by antiinflammatory drugs in mice and humans Proceedings of the National Academy of Sciences, 108 (22), 9262-9267 DOI: 10.1073/pnas.1104836108

Image via Nomad_Soul / Shutterstock.

Alex Kadner, PhD

Alex Kadner, PhD is a scientist and medical writer with 10 years of clinical experience as a psychiatric nurse's aide, over 10 years of academic and 2 years of agency experience. His strengths are in writing and editing peer reviewed manuscripts and grant applications, developing power point slide sets and scripts for video recordings. He has written in both English and German for a broad range of audiences, ranging from scientists and clinicians to the lay public. His peer-reviewed papers have appeared in Neuroscience and the Journal of Neurophysiology among others.
See All Posts By The Author