Antidepressants, Placebos, and Lithium—Some Parting Thoughts


In the case of antidepressants, a few recent articles may make it tempting to throw the baby out with the bathwater. But on close examination, that would be a mistake on many levels...

Antidepressant, placebos, lithium and more



The concept of the placebo condition has been misunderstood, and the efficacy and benefits of antidepressants and lithium have been obscured or downplayed in the media.

Imagine that you have just been diagnosed with a life-threatening form of cancer. Your doctor tells you there is no cure, but a certain type of medication produces a robust response and remission—in about 15 of 100 cases. Knowing that the medication can have some serious side effects, would you still take it? For me—and for most people I know—the answer would be, “Of course—in a New York minute!”

And yet, a recent meta-analysis finding that antidepressants produce a robust response—much stronger than placebo—in about 15% of depressed patients1 was widely depicted in the media as confirmation that antidepressants are “not much better than a sugar pill” and not worth the likely adverse effects. In my view, this is a serious misinterpretation of the data, at least with respect to individuals with severe or melancholic major depression. I’ll circle back to this issue shortly.

A recent study of lithium use and suicide rates in bipolar disorder also made headlines in the blogosphere and was widely cited by critics of psychiatric medication.2 This was another meta-analysis, analyzing randomized trials published after 2000, comparing lithium and placebo versus treatment as usual in mood disorders. The pooled suicide rate was 0.2% for people randomized to lithium, and twice that (0.4%) with placebo or treatment as usual. This result was judged as “not a statistically significant difference.” There was also no significant difference in rates of all non-fatal suicidal behavior. The authors concluded, “Evidence from randomized trials is inconclusive and does not support the idea that lithium prevents suicide or suicidal behavior.” This conclusion is roughly consistent with a recent VA study discussed in Psychiatric Times by David N. Osser, MD,3 but is contradicted by numerous studies showing that lithium use in bipolar disorder is associated with a substantial reduction in suicide rates. I will circle back to Nabi et al study as well, shortly.

The Placebo Condition is Not Just a “Sugar Pill”

In my experience, few among the general public—and not even some clinicians—fully understand how clinical studies of depression utilize placebos. To put it bluntly: There has never been, and likely never will be, a clinical study that simply compares an antidepressant to a sugar pill (ie, inert agent) and does nothing else. Quite the contrary. As Sheldon H. Preskorn, MD, has pointed out,4 the placebo condition in a typical study involves

[A]ttention, support, and patient education on a weekly basis for 6–8 weeks…[during which] the patient is typically seen weekly. Each of these visits lasts approximately 60 minutes, during which the patient’s response to the double-blind treatment is assessed and recorded. These visits involve giving the patient considerable education and support, even though ‘formal‘ psychotherapy is not provided. In other words, a considerable amount of energy is spent providing the ‘placebo’ treatment, energy that may well alter the natural outcome. Certainly, ‘placebo treatment’ in such a study is not the same as no treatment.

Preskorn righty argues that rather than calling it a “placebo response,” the effect of the placebo condition really represents a response to good clinical management. The “drug-specific response” is the difference between the total response to a drug and the response to such aforementioned good clinical management (ie, the placebo condition). So, if about 65% of patients respond well when receiving an antidepressant, and about 45% would have responded to good clinical management alone (ie, the placebo condition), the drug-specific response would be about 20%.

Preskorn has calculated that the drug-specific effect of most antidepressants in clinical trials is roughly 20-30%—not tremendous, but better than the 15% found in the Stone et al study. However, a narrow fixation on these numbers misses the big clinical picture. As Preskorn noted, “Antidepressants are tremendously valuable to the patients who specifically benefit from them and who otherwise would continue to suffer from their depressive illnesses.”4

Indeed, the ultimate value of antidepressant treatment cannot be judged based solely on the drug-specific response rate—we must also consider the seriousness of the illnesses we treat. Clinically severe major depression is a devastating illness, and it has been designated the leading cause of disability around the world.5 Severe depression is associated not only with greatly elevated rates of suicide, but also with attentional problems; memory deficits; and broad impairment in executive functions.6

Finally, it is important to point out the key differences between clinical studies and the real world of psychopharmacology. The typical clinical study examines medication effects in a large group of participants; for a short duration (6-8 weeks); and usually with administration of a single antidepressant. In clinical practice, we are concerned with a particular patient who is often treated over months or years; and often, with a variety of agents and augmentation strategies. As my colleague, George Dawson, MD, has commented to me, “We treat until the patient is better.”

Indeed, the 15% finding from the Stone et al study does not equate to a 15% chance that Mr Smith or Ms Jones will eventually get better in the hands of an experienced psychiatrist who is using all the pharmacological tools of treatment at our disposal. Furthermore, most studies show that the addition of psychotherapy enhances response to medication, and that the level of remission obtained by successful acute-phase pharmacotherapy can be increased by a subsequent psychotherapeutic treatment.7

Even if it were the case that “only” 15 in 100 depressed patients had a robust response to antidepressant treatment, the cumulative number of acutely depressed patients greatly helped by antidepressants over months and years would warrant judicious use of these medications. But again: We should be focusing not on this number, but on the soul-killing devastation of a major depressive episode. Here is one brief description from the writer William Styron (1925-2006):8

Death was now a daily presence, blowing over me in cold gusts. Mysteriously and in ways that are totally remote from normal experience, the gray drizzle of horror induced by depression takes on the quality of physical pain ...[The] despair, owing to some evil trick played upon the sick brain by the inhabiting psyche, comes to resemble the diabolical discomfort of being imprisoned in a fiercely overheated room . . . [thus] it is entirely natural that the victim begins to think ceaselessly of oblivion. . .In depression the faith in deliverance, in ultimate restoration, is absent.

Lithium and Suicide Prevention

The study by Nabi et al—which concluded that the evidence “does not support the idea that lithium prevents suicide or suicidal behavior”—should be taken with a large grain of lithium salt!

There are numerous studies contradicting that claim, including at least 2 not cited or analyzed in the Nabi et al review. For example, Tondo and Baldessarini9 meta-analyzed 12 randomized trials in 10 reports (with at least 1 suicide in either treatment arm) including participants with bipolar disorder (BD) and major depressive disorder (MDD).Their study found a “substantial reduction of risks of suicide and [suicide] attempts with long-term lithium treatment, particularly in depressive phases of BD and in MDD.” Risk of suicide fatality was lower with lithium than with placebo, and probably lower than with mood-altering anticonvulsants or antidepressants.

Another recent study not included by Nabi et al was a systematic review of more than 200,000 participants in a variety of study types.10 Most ecological studies (eg, examining lithium in drinking water) reported a reduction in suicide in the general population, mostly in males. Most uncontrolled and controlled open-label trials also showed a reduction in the risk of suicide with long-term lithium use. These authors posited “an intrinsic anti-suicidal property of lithium, independent of its efficacy as [a] mood stabilizer.”10

These 2 studies pointing to lithium’s anti-suicidal effects are hardly conclusive, but their omission ought to raise some red flags regarding the Nabi et al study. This also merits further analysis of its inclusion criteria and statistical methods. For example, my colleague, Nassir Ghaemi, MD, has pointed out that the Nabi et al meta-analysis applied an entirely arbitrary cut-off year (ie, 2000) to the studies reviewed. And yet, 3 of 4 studies showing a benefit for lithium in prevention of suicide were published before the year 2000.11

Finally, a recent review of lithium concluded, “Lithium is the most effective medication in psychiatry, because it has disease-modifying, not just symptomatic, effects. It is effective not only for bipolar illness but also for prevention of suicide, episodes of unipolar depression, mood temperaments, and possibly dementia. Its many benefits need better appreciation, while lowered dosing can reduce risks.”12


Antidepressants are far from ideal medications. Nevertheless, on average, they are clearly superior to placebo in at least 15 in 100 depressed patients, for whom antidepressants can greatly reduce suffering and incapacity. However, the ultimate value of antidepressant treatment must be judged in light of the devastating effects of severe depression.

The ultimate value of antidepressant treatment cannot be judged based solely on the drug-specific response rate—we must also consider the seriousness of the illnesses we treat.

Having treated hundreds of patients who have bipolar spectrum disorders with lithium, I have observed that it is well-tolerated by most patients; and its adverse effects are almost always manageable with careful adjustment of lithium blood levels. Until we have more randomized, placebo-controlled studies of lithium and suicide rates, I believe that Osser’s conclusion stands, ie, “the best recommendation at this time regarding lithium in patients with bipolar disorder who are suicidal would be to consider it a worthwhile option.”3

Dr Pies is Professor Emeritus of Psychiatry and Lecturer on Bioethics and Humanities, SUNY Upstate Medical University; Clinical Professor of Psychiatry, Tufts University School of Medicine; and Editor in Chief Emeritus of Psychiatric TimesTM (2007-2010). Dr Pies is the author of several books. A collection of his works can be found on Amazon.

Author’s Note and Acknowledgments

As I inch my way toward full retirement, I will be taking a few months off to work on a book project and may not be contributing as frequently on these pages.

I would like to express my deep appreciation to Dr Sheldon Preskorn for his sage guidance and many publications pertaining to this article.Thanks to Dr Dave Osser for commenting on an early draft. This article grew from many valuable discussions with my colleagues Nassir Ghaemi, MD, Alexander Lisinski, MD, Fredrik Hieronymus, PhD, George Dawson, MD, Chris Aiken, MD, Jim Phelps, MD, Marc Stone, MD, Awais Aftab, MD, Andrew Nierenberg, MD, Randall F. Moore, MD, Roger McIntyre, MD, Daniel Morehead, MD, Cathryn Lewis, PhD, John J. Miller, MD, Mark L. Ruffalo, MSW, DPsa, and many colleagues I have inadvertently left out. A special thanks also go to Mark S. Komrad, MD, Cynthia Geppert, MD, MA, MPH, MSBE, DPS, and Anne Hanson, MD, for their collegial support and contributions to Psychiatric TimesTM. Thanks as well to the diligent editorial staff of PsychiatricTimesTM.


1.Stone MB, Yaseen ZS, Miller BJ, et al. Response to acute monotherapy for major depressive disorder in randomized, placebo controlled trials submitted to the US Food and Drug Administration: individual participant data analysis. BMJ. 2022;378:e067606.

2. Nabi Z, Stansfeld J, Plöderl M, Wood L, Moncrieff J (2022). Effects of lithium on suicide and suicidal behaviour: a systematic review and meta-analysis of randomised trials. Epidemiol Psychiatr Sci. 31: e65.

3. Osser DN. Lithium and Suicide Prevention: More Thoughts on Recent VA Study. Lithium and Suicide Prevention: More Thoughts on Recent VA Study . Psychiatric Times 39(10):26.

4. Preskorn SH. What do the terms "drug-specific response/remission rate" and "placebo" really mean? J Psychiatr Pract. 2011 Nov;17(6):420-424.

5. Friedrich M. Depression is the leading cause of disability around the world. JAMA. 2017;317(15):1517.

6. Fossati P. Is major depression a cognitive disorder? Rev Neurol (Paris). 2018;174(4):212-215.

7. Guidi J, Fava GA. Sequential combination of pharmacotherapy and psychotherapy in major depressive disorder: a systematic review and meta-analysis. JAMA Psychiatry. 2021 Mar 1;78(3):261-269.

8. Styron W. Darkness Visible. Vintage, 1992.

9. Tondo L, Baldessarini RJ. Antisuicidal effects in mood disorders: are they unique to lithium? Pharmacopsychiatry. 2018 Sep;51(5):177-188.

10. Del Matto L, Muscas M, Murru A, et al. Lithium and suicide prevention in mood disorders and in the general population: A systematic review. Neurosci Biobehav Rev. 2020;116:142-153.

11. Ghaemi SN. How to mislead with meta-analysis: Lithium and suicide. Psychiatry Letter. Oct. 24, 2022. Accessed November 9, 2022.

12.Barroilhet SA, Ghaemi SN. When and how to use lithium. Acta Psychiatr Scand. 2020;142(3):161-172.

Further reading

Lisinski A. Hooked on hype? A critical examination of recent anti-antidepressant reporting. Psychiatric Times, in press.

McIntyre RS, Lophaven S, Olsen CK. A randomized, double-blind, placebo-controlled study of vortioxetine on cognitive function in depressed adults. Int J Neuropsychopharmacol. 2014 Oct;17(10):1557-1567.

Cipriani A, Hawton K, Stockton S, Geddes JR. Lithium in the prevention of suicide in mood disorders: updated systematic review and meta-analysis. BMJ. 2013;27(346):f3646.

Ruffalo ML. A Brief History of Lithium Treatment in Psychiatry. Prim Care Companion CNS Disord. 2017;19(5):17br02140.

Related Videos
brain depression
© 2024 MJH Life Sciences

All rights reserved.