Researchers: “Antidepressants Should Be Avoided in Bipolar Depression”

New research finds that antidepressants are not effective for bipolar disorder and can worsen symptoms of mania.


A new study in the Journal of Clinical Psychiatry demonstrated that antidepressants were no better than placebo for treating either bipolar I or bipolar II disorder. In fact, taking the antidepressant actually worsened mania after one year. The researchers, led by the psychiatrist S. Nassir Ghaemi from Tufts Medical Center, write:

“Citalopram, added to standard mood stabilizers, did not have clinically meaningful benefit versus placebo for either acute or maintenance treatment of bipolar depression. Acute mania did not worsen with citalopram, but maintenance treatment led to worsened manic symptoms, especially in subjects with a rapid-cycling course.”

According to the study authors, antidepressants are the most commonly used drug to treat bipolar disorder, despite a recent meta-analysis that found them to be ineffective. Antidepressants have also been demonstrated to induce manic symptoms. Finally, “maintenance treatment”—using antidepressants long-term—has been found to increase the risk of mania.

“Clinicians and patients often choose antidepressants, especially serotonin reuptake inhibitors (SRIs), to treat bipolar depression, but the evidence for benefit and safety of these medications has been poorly proven or controversial,” Ghaemi and his co-authors explain. 

However, these findings are difficult to interpret because very few randomized, controlled trials (RCTs—considered the gold standard of objective research) have been conducted to test these findings, especially around maintenance treatment.

In fact, according to the authors of the current study, their trial was “the first placebo-controlled RCT of any SRI in maintenance prevention of depressive episodes in bipolar illness at 1-year follow-up.”

In their study, 119 participants diagnosed with bipolar I or bipolar II disorder were randomly assigned to take either citalopram (an SSRI, brand name Celexa) or a placebo. All the participants were also already taking a “mood stabilizer” such as lithium, which they continued taking throughout the study.

The primary outcome was to compare depression and mania scores between the placebo and citalopram groups at six weeks—using the Montgomery-Asberg Depression Rating Scale (MADRS) and the Mania Rating Scale of the Schedule for Affective Disorders and Schizophrenia (MRS-SADS). The secondary outcome was to compare the same scores after one year of long-term “maintenance treatment.”

The researchers found that the difference between the groups was neither statistically nor clinically significant on any of the measures—meaning that citalopram was no better than placebo at relieving depression or mania. This was true whether people had the bipolar I or bipolar II diagnosis.

There was one difference—the people who were randomly assigned to citalopram for a year of “maintenance treatment” had worse mania scores at the one-year follow-up than those who continued taking a placebo. That is, the antidepressant worsened their mania over the course of a year.

“Maintenance treatment led to worsened manic symptoms,” the researchers write.

Although this was true for all participants on average, it appeared to be even worse in those who had a “rapid-cycling” version of bipolar disorder—but this analysis was underpowered and needed further validation from future studies.

The researchers noted that when untreated, bipolar disorder episodes usually resolve naturally, “typically within six months or less.” They note that in clinical practice, since most patients will receive an antidepressant at the beginning of their episode, this natural improvement “will be attributed to antidepressant use, producing the clinical impression of drug efficacy.”

Thus, doctors will see the natural improvement that would have occurred without treatment and believe that the drug they prescribed is responsible for that improvement. This is why RCTs like this one are necessary, as they provide a comparison group that naturally improves without the use of an antidepressant. The researchers write:

“SSRIs like citalopram are not helpful to treat bipolar depression or to prevent it, and they may worsen manic symptoms if used long-term, especially in patients with a rapid-cycling course.”
“Antidepressants should be avoided in bipolar depression.”



Ghaemi SN, Whitham EA, Vohringer PA, et al. Citalopram for Acute and Preventive Efficacy in Bipolar Depression (CAPE-BD): A randomized, double-blind, placebo-controlled trial. J Clin Psychiatry. 2021;82(1):19m13136. (Link)


  1. “Antidepressants should be avoided in bipolar depression,” and “mania.” So should the antipsychotics, since combining the antidepressants and/or antipsychotics can result in anticholinergic toxidrome, which looks like “bipolar” or “schizophrenia” to the DSM deluded.

    Probably they should be avoided for all those stigmatized by the “mental health professionals” with any of the DSM disorders, since all their DSM “bible” disorders have been confessed to be “invalid.”

    “The researchers noted that when untreated, bipolar disorder episodes usually resolve naturally, ‘typically within six months or less.'”

    Then why use drugs at all for the “invalid” “bipolar” stigmatization? And why do DSM deluded “mental health” workers believe “bipolar” is a “lifelong, incurable, genetic mental illness,” with zero scientific proof?

    I’m quite certain all the non-medically trained, DSM deluded, “mental health” workers, who know nothing about the common adverse and withdrawal effects of the psych drugs, should learn to live and let live, and leave other people alone.

    Because some of us, have done our homework, and can embarrass you. By pointing out your industry’s, and your religion’s, systemic child rape covering up crimes.

    Report comment

  2. This is the sort of thing that happens when shrinks don’t explore the perceptual status of their patients, but rely on “diagnostic jive” to guide their prescribing, ignoring the simple reality that individuals with numerous perceptual distortions shouldn’t get stimulants, no matter what the DSM “diagnosis” states is proper.

    Report comment

  3. How about avoid antidepressants altogether? There are reports that antidepressants might cause bipolar disorder or rather the manic part of the disorder. Therefore, this may be a circular logic discussion. Or even how about avoiding the alleged diagnosis of bipolar disorder, formerly manic depression altogether? What do you think? Less stress and damage to both brain and body. Thank you.

    Report comment

    • I agree. The title should read, “Antidepressants can cause manic behavior in many who take them.” Honestly, from talking to a number of people who have taken them, I kind of think that’s what they do. For people who are really withdrawn or super anxious, it probably feels good as long as it doesn’t go too far. But I’ve never taken them myself, so I really don’t know. In all likelihood, different people have totally different experiences. Which is really not a good thing if everyone who “has depression” is encouraged/forced to take them.

      Report comment

    • Usually, if I have a “client”, it’s due to some professional’s failed treatment, so I have to find out what’s going on in a symptomatic individual, in order to treat it with the correct nutrients, if I can. I use various combinations of vitamins (and sometimes minerals and amino acids) if I can’t find something out of my league, such as thyroid problems, where I’d recommend a proper GP. I couldn’t prescribe anything, even if I wanted to.

      Report comment

  4. It’s really time to stop saying anti-depressants or anti-psychotics or one of my perennial favorites, mood stableizers. (I can’t spell that word!) and all the rest medications. These are nothing less than drugs, drugs like any street drug pusher would try to sell you. The only difference is these are legal and it’s people with alleged authoritative degrees who sell them to you. Like they stop the madness, but the madness is not in us it’s in those who sell it to us and profit off of us and try to control us with it. “Just Say No” has never sounded so good and is a powerful tool in its own right. Thank you.

    Report comment

    • Or you can just set aside the fact that mainstream psychiatric researchers use terms that alarm you and read it as “More evidence for antidepressant harm in people who get diagnosed with ‘bipolar disorder'”.

      Seems to me there’s a lot of energy spent on language policing on MiA that might be better used somewhere else.

      Report comment

        • I think we’ve all drawn attention to it plenty of times, including me.

          Maybe we should stop obsessing over the branding and pay more attention to the product. As history tells us, when the labels get dirty the shrinks just make new ones.

          When the idea was to pack more distressed people into loony bins neuroleptics were ‘major tranquilisers’ to keep things calm. When the bins had to be emptied and their contents dumped onto the streets they became ‘antipsychotics’ to ‘cure’ those we still designate as human trash, regardless of the currently fashionable pseudo-scientific euphemism we’re papering over the attitude with.

          Abolish ‘depression’ and ‘antidepressants’ will get a new label to be inflicted onto the same people as something else. And we’re back to square one.

          What Ghaemi is telling us here is that even within the epistemological and procedural framework of mainstream psychiatric research there is no justification for current clinical practice. We can use that. If it means less people are having the pills inflicted upon them then I don’t really care what the labels on the bottle or the person say.

          Report comment

        • sam plover, I think you make an interesting point. However, sometimes, maybe I just need a little clarification. How do you distinguish between “psych terms” and “psych language?” Also, in this distinguishing, will we be helping or hurting more those already damaged by psychiatry, etc. ? I think someone said something about “language policing.” Would this be considered “language policing?” Yes, “words do matter” but getting caught up too much in the words can sometimes do more harm than good. Thank you.

          Report comment

          • Because in the end, they promote that the diagnosis is real, and so it also means that further studies will be needed to create new drugs. And what will we call them? Anti-bipolars?
            How does this label affect people when they go to ER with leg pain?
            It is not the “drugs” that affect people in being seen as credible humans.
            How does a label affect a 17 year old who sees themselves as “having” an “illness”?

            If they are not real illnesses, the diagnosis should be a choice and nothing that will damage you in receiving unbiased care or view of you should be written down as if they are facts, nor should biased information about you ever be shared.

            Being seen as bi-polar by your neighbour is not the same as a doctor reading your chart.

            So yes, words can indeed do harm. It is not policing, to expose the BS in the DSM.
            And in turn, I won’t be suggesting to someone that they are too caught up in labels/terms/language, when they make note of the BS.

            Right now there is someone being held against their will in a psych hospital because their physical damage is being referred to as being “bipolar”. By psych and by MD’s.
            It is not for any other reason than the “diagnosis”

            Report comment

  5. Yeah, sure, the whole notion of mental illness and diagnosis is f*ed up, abusive and oppressive. I think we all get that. And it’s something that needs addressing. But we’re not going to see it fixed tomorrow.

    In the meantime there is research like this being published from within the f*ed up paradigm of mental illness that can be used to argue against coercive and abusive drugging of people today, if we don’t just hold up our “no such thing as mental illness” crucifix and try to banish it. The organisation I work for is always looking for stuff like this to use in the cases we take before the NSW Mental Health Review Tribunal and the courts in attempting to get forced medication Community Treatment Orders lifted and get people released from psychiatric detention. We’re doing that here and now. Not in some future utopia where everyone understands mental illness is a myth and always uses the ‘correct’ terms when referring to abuse and suffering.

    Sure it would give me a smug sense of righteousness in speaking truth to power to stand before the MHRT panel and say “This person shouldn’t be forced to take drugs for her mental illness because there is no such thing as mental illness.”, but I don’t think that’s going to help anyone.

    If you want to argue against medical authorities about coerced psychiatric treatment – either on your own behalf or on behalf of someone else – you’d best adapt your communication to their epistemological framework, because you can be pretty damned sure they aren’t going to adjust theirs to engage with you.

    Report comment

    • This is my experience also. Practical considerations sometimes requires using the language of the oppressor to get one’s point across and have the desired effect. Real advocacy requires a focus on getting the product, however it has to be gotten. The finer points of philosophy can be addressed in a different forum.

      Report comment

    • What you are doing, is real work on the frontlines. Trying to get people to safety. It is really hard work, and why should it be so hard. And I thank you for every person you have helped.
      But this work will be never ending and people get tired. The APA has no intention of making it easier to get innocent people out of their grasp.
      The question remains why? Why do people have to be protected from the very system that was supposed to help them?

      It is okay to have many different approaches. There has never been only one approach that works to change something.

      Please sir, may I have my freedom? Sure, go ahead. We set it up so that you cannot compete and succeed.
      Psychiatry is only carrying out what our governments set up.

      Report comment

        • I think we’re in furious agreement about the root of the problem. As with so many issues of dignity and survival, the corruption and inhuman rapaciousness of our own institutions are what we’re really fighting here.

          I guess my point was mainly against letting ideological purity stop us from finding allies and ammo where we can. There’ll be a lot more battles before the end of the war is in sight.

          Ghaemi has long fought to end some of the most harmful prescribing practices those with bipolar diagnoses are subjected to. I don’t have to agree with him about the validity of the diagnosis to know I’m on his side.

          Report comment

          • I totally agree! I’ll talk with anyone who wants to change things, no matter what they think the “new solution” is. Even those who are “reformers” are potential recruits for a more radical analysis. Heck, I was a “reformer” once upon a time! And we can work to get agreement with “reformers” on certain goals that will help in either purpose, like getting Miranda rights read to people receiving an “evaluation.” We need to generate power, and ideological purity isn’t a requirement to get things done.

            Report comment

  6. BTW, there’s an element of circularity in this. Sort of psychiatry eating its own tail.

    Bipolar II was posited largely in response to the fact that ‘depressed’ people prescribed ‘antidepressants’ often developed manic or psychotic symptoms. Of course it couldn’t be that the medicines were causing a disease, so the shrinks explained that these weren’t real depression patients at all. They were in fact bipolar patients who were quite happy with their upswings but seeking treatment for their downswings. In treating the depressive symptoms the pills were unmasking the untreated manic symptoms. They weren’t causing bipolar, they were revealing it. Any initial misdiagnosis was the patient’s fault for not speaking up about their hypomania.

    But even with the new diagnosis psychiatrists just couldn’t resist prescribing antidepressants to their bipolar II patients. The drug reps and key opinion leaders say such nice things about them …

    So people like Ghaemi, who take the diagnoses seriously and believe in ‘evidence based’ therapies for them, find themselves in opposition to the mainstream practices of their profession.

    Oh what a tangled web we weave …

    Report comment