Gradual Tapering is Most Successful for Withdrawal from Antipsychotics

Mixed-Methods study explores the experiences of antipsychotic discontinuation among service users


A mixed-methods published in the journal Psychiatry Research surveyed individuals who had attempted, at least once, to discontinue taking antipsychotic medications. Fifty-five percent of participants reported successfully stopping, and 50% reported no current use. In the findings, a gradual approach to withdrawal was positively associated with successful discontinuation and was negatively associated with relapse, suggesting that gradual withdrawal (more than one month) is a more effective method of discontinuation.

The authors write that, “When considered in conjunction with the longitudinal research and studies exploring psychiatric medication withdrawal in general . . . discontinuation is a legitimate choice that requires and justifies appropriate support.”

Photo Credit: Flickr

Persons prescribed antipsychotics often make adjustments to their medications independent of their prescriber or attempt to discontinue the medication on their own. As the authors point out, 60-80% of individuals with schizophrenia spectrum diagnoses report discontinuation.

Additionally, in other studies, as much as 55% of patients who were on APs stopped taking their medication without consulting their prescriber, and 41% withdrew abruptly. Of those patients that discontinued the drug, 78% experienced withdrawal effects, and 21% completely stopped taking APs. Studies of AP use have found that those who discontinued AP treatment had better long-term functional outcomes and lower rates of relapse than those with continuous use. While discontinuation can result in better long-term outcomes, the process of discontinuation can come with numerous withdrawal effects including, somatic, emotional, and cognitive effects as well as psychotic or manic relapse.

The authors of this study sought to elucidate the association between gradual withdrawal methods, withdrawal effects, and successful outcomes across different diagnoses. The authors also aimed to explore how people manage discontinuation. To do this, a mixed-methods (qualitative and quantitative) study was conducted via an online survey of adults (n=105) who were taking or had taken antipsychotics for at least three months and had at least one attempt to stop taking antipsychotics.

Among other demographic data, participants were asked to identify the primary symptoms they were experiencing when they began taking antipsychotics, medication history, treatment history, and the number of attempts to discontinue and age at those attempts. Participants were asked: “What was the outcome of your most recent attempt to stop taking antipsychotics?” and “are you still taking oral antipsychotic medication?” Participants were also asked about their method of discontinuation and (slow reduction over time or abruptly stopping) and the withdrawal effects they experienced.

Thirty-five percent of the participants had received a bipolar diagnosis, 18% schizophrenia, 28% other (unipolar depression, anxiety, obsessive-compulsive disorder, post-traumatic stress disorder, personality disorder). Only 48% of participants reported speaking with a doctor about their most recent attempt to discontinue, 55% successfully stopped taking APs, and 50% reported no current use.

Participants who identified as successfully stopping described experiences of improved psychosocial well-being, as captured in the quotes below:

“I experience no psychosis and only occasional anxiety […] I am (reasonably) physically healthy, extremely mentally healthy, working and enjoying life”

“I manage my mental health well. I have occasionally visited [a] counselor since stopping the medication, but mostly use my support network for help now.”

Continued difficulties after successfully stopping were also reported such as ongoing mental health problems and unresolved adverse effects of antipsychotic medication. For those the participants who resumed the drug, they described hospitalization or a compulsory treatment order (CTO) or described returning to prevent hospitalization and mandatory treatment.

Others described changes to dosage, medication, or clinician, e.g. “changed psychiatrist, correct diagnosis and prescribed different drugs.” Additionally, some participants who resumed taking antipsychotic medication described accepting having to take antipsychotics, saying, “I have to stay on it for life I can’t cope without it.” Lastly, a subset of participants described having adverse events that affected their mental health, including losing a job, a relationship ending, or a growing sense of failure.

Half of the participants reported gradual withdrawal, while the other half reduced their medication abruptly (one month or less). Those who gradually withdrew did not report relapse, described successfully stopping, and were not currently using antipsychotics.

Gradual withdrawal over one month was positively associated with successful discontinuation and no current use. There was also a significant positive association between following a gradual withdrawal method and consulting a doctor. However, 47% of those who asked a doctor withdrew abruptly, while 52% withdrew gradually. Unsurprisingly, relapse was negatively associated with discontinuation and no current use.

Withdrawal Effects

Two-thirds reported unwanted withdrawal effects, 18% reported no withdrawal effects, and 13% reported positive effects. Some of the adverse emotional withdrawal effects included: anxiety and fear, low mood, sadness and depression, irritability and agitation, suicidality, and mood swings. On the physical side, withdrawal effects included: nausea, diarrhea, vomiting, headaches, unpleasant bodily sensations, appetite, and rapid weight loss, insomnia or disturbed sleep, shaking, sweating, and one person reported seizures.


Twenty-seven percent of the sample reported relapse of psychosis or mania during withdrawal. Thirteen percent of those who discontinued gradually reported relapse while 34% of those who stopped abruptly reported relapse.

The authors conclude that “it is possible to successfully discontinue antipsychotic medication, relapse during withdrawal presents a major obstacle to successfully stopping an antipsychotic medication, and people who withdraw gradually across more than one month may be more likely to stop and to avoid relapse during withdrawal.”



Larsen-Barr, M., Seymour, F., Read, J., & Gibson, K. (2018). Attempting to discontinue antipsychotic medication: withdrawal methods, relapse, and success. Psychiatry research. (Link)

Previous articleEmbracing Human Complexity Over Labels
Next articleDid You Hear That? On Destigmatizing Hallucinations
Bernalyn Ruiz
MIA Research News Team: Bernalyn Ruiz-Yu is a Postdoctoral Fellow in the Department of Psychiatry and Biobehavioral Sciences at the University of California, Los Angeles. She completed her Ph.D. in Counseling Psychology from the University of Massachusetts Boston. Dr. Ruiz-Yu has diverse clinical expertise working with individuals, families, children, and groups with a special focus on youth at risk for psychosis. Her research focuses on adolescent serious mental illness, psychosis, stigma, and the use of sport and physical activity in our mental health treatments.


  1. “Unsurprisingly, relapse was negatively associated with discontinuation and no current use.”

    There is a more powerful way to express exactly the same thing:

    “Unsurprisingly, relapse was positively associated with continuation and current use.”

    Report comment

      • Are we talking about Relapse or Rebound?

        “..A Depot Neuroleptic Withdrawal Study..”

        :…A controlled study of the clinical effects of the withdrawal of depot fluphenazine decanoate and depot flupenthixol decanoate in chronic schizophrenic patients…”*

        This study quotes a withdrawal FAILURE Rate of 97%. But I believe neuroleptic drug Withdrawal Syndrome overides any original problems.

        I believe the “Chronic Schizophrenics” were trapped.

        My target was to come off Depot Injection medication because this drug was genuinely Disabling (extrapyramidal).

        I was given permission by my Psychiatrist to stop Depot (abruptly) but I FAILED and ended up in hospital several times as a result, until I went on oral medication and tapered very carefully.

        After this I consumed very very low oral doses for many years until the drugs eventually disappeared of their own accord.

        For me It was possible to come off the Depot Injection Medication but only very carefully – and this was something that the Doctor didn’t know.

        I believe a 97% Depot Withdrawal SUCCESS Rate could be Projected with – A Careful Withdrawal Approach.

        *The above study is dated, but I think it’s appropriate as an example.

        Report comment

        • Interesting study, but his results are very bad. The results of the meta-analysis of Vigera are much better: only 30% relapse in 24 months after progressive weaning.

          Some remarks on this study:

          1) The definition of relapse is very extensive: it is a CPRS score.

          If we use a more restrictive measure, the rehospitalization rate, relapse rates are as follows (in 6 months):

          neuroleptic: 0%
          weaning: 31%

          In addition, one patient from the neuroleptic group had a portion of the injection-depot removed surgically and did not relapse.

          2) The use of emergency neuroleptic tablets was prohibited. The use of emergency neuroleptic tablets is absolutely necessary, because the concentration of neuroleptic in the blood does not decrease perfectly regularly with the depot injections. The depot injection is a progressive weaning, but it is not perfectly regular. The author himself notes this problem in the “Discussion” chapter.

          3) The average duration of the “disease” was greater than 10 years: they was thus very addict “chronic” subjects.

          The author emphasizes that other studies have shown that the success of weaning depends on many factors, including the duration of exposure and the dose of neuroleptic. In the studies of Engelhardt et al. (1960) for example, the duration of exposure was short and the dose was low, and there was only 25% relapse in the weaning group after 12 and >18 months. This is a very encouraging result.

          4) Finally, the author points out that despite relapses, “This study was supported by the interest of patients who were enthusiastic and positive in their participation and visited our team at almost every rating.”

          Relapse rate is therefore not necessarily a relevant criterion for measuring the success of withdrawal: researchers should instead focus on improving quality of life, social relationships and other personal and social parameters.

          Report comment

          • Sylvain,

            Once I got off the injection I was no longer disabled, and could become independent of Psychiatry.

            But I found that I didn’t return with the same “brain” as I started with, and levelling myself off was a big job. However, once I moved onto the taper I never “relapsed” again, or had another hospitalisation.

            So, coming off the strong medication returned me to longterm wellness!

            Report comment

      • What really concerns me is, why are we calling it relapse? How can we assume it is definitely relapse? Relapse implies a return to an original state. What if this is not relapse, but an artifact of withdrawal?

        We know withdrawal can cause psychosis, akathisia, mania, suicidality, etc. So when a patient experiences this, why do we only label it a relapse, and not consider that it might just be withdrawal? Especially in light of the evidence that this “relapse” risk decreases with a slow taper instead of cold turkey.

        I say this as someone who experienced the full aspects of “relapse” while withdrawing. The kicker is I never had those problems prior to drugs, so I couldn’t really be relapsing, could I?

        Report comment

        • You are right, of course. Relapse implies a specific condition from which one is recovering, and none of the “mental illnesses” can meet that criterion. Perhaps addiction, but there you at least have the measurable question of whether one has returned to the use of substances or whatever one is addicted to. “Relapse” is pretty meaningless when applied to “depression” or the like, particularly when withdrawal or other drug effects are in play. It all starts with the mistaken concept that saying someone “has depression” or whatever has any scientific meaning, which it does not.

          Report comment

          • We need to address the life and death question of serious iatrogenic harm. I wound up with the deadly condition of frontotemporal lobar degeneration because of the antipsychotic Seroquel that was given to me for sleep and not for any psychiatric condition. People with this condition lose their minds and everyone takes advantage of whatever they can get from them and then they die in short order. The younger they are, the faster the disease progresses to death. This is not a relapse but something unspeakably awful

            Report comment

  2. thanks for this. i don’t think this could have been more timely and encouraging for me, not to mention the many, many others out there who want our selves and lives on our own terms, free from neuroleptics and everything else the psychiatrists have thrown our way.

    ive noticed that my shrinks rarely tapered most drugs. part of it was that i was low status and didn’t “know my place,” but I had good insurance (this equals: controlled substances, 0 informed consent, 0 real monitoring, and insurance fraud). the other issue, i think, is that psychiatrists are so gung ho about selling us the myths that these are good medicines that help heal sickness, etc. etc. etc., that…with all their knowledge, they refuse to treat the drugs as drugs, even though they’re obviously drugs. oh, that and…i suspect many shrinks play with the drugs to punish and torment the person/patient.

    thanks again for this. i think you just made my day, big time. 🙂

    Report comment

  3. Personally I think you should be recommending withdrawal over much more than merely one month.

    “Studies of AP use have found that those who discontinued AP treatment had better long-term functional outcomes and lower rates of relapse than those with continuous use.” So the reality is our society needs to get the doctors to stop forcing the antipsychotics on people altogether.

    I’d say, especially since the antipsychotics can create both the negative and positive symptoms of schizophrenia, which isn’t beneficial to the patients, and a highly unethical thing for the psychiatrists to do. The positive symptoms are created via antipsychotic induced anticholinergic toxidrome and the negative symptoms are created via neuroleptic induced deficit syndrome.

    “While discontinuation can result in better long-term outcomes, the process of discontinuation can come with numerous withdrawal effects including, somatic, emotional, and cognitive effects as well as psychotic or manic relapse.” True, but these are always misdiagnosed as a “return of illness” instead of properly being diagnosed as withdrawal symptoms.

    Report comment

  4. I feel an incredible sadness about antipsychotics. Many of them are of marginal efficacy in RCTs, and the disabling effects are all too evident, people tolerating metabolic deterioration believing their abnormal flawed brains are only kept in check by a pill. I went in search of tapering guidelines from NICE, patient leaflets or RCPsych and found none. Please correct me if any exist.

    What there is, are studies showing that once you’ve been on them for a while your chances of staying alive or out of hospital are better if you stay on them. I doubt this, because how you fare must depend on how you are tapered and other support you receive, but there does least seem to be a red flag that the longer you are on them the more careful the taper needs to be.

    The same Finnish study I’m thinking of also would indicate that since the hospitalisation risk on discontinuation continues to rise on anti psychotics, it’s better to come off earlier rather than leave it too late.

    We really don’t know how badly these drugs damage people and cause dependence, whilst all the time stopping people functioning and amazingly actually shrinking their brains.

    This paper says that half the people didn’t even ask their doctor before stopping, presumably because they knew the answer – “don’t do it”. It’s a mess.

    I hope the UK parliamentary group on prescribed drug dependence has the time to look at antipsychotics as well, because although the numbers are smaller the disablement is huge.

    Report comment

    • From your comment I can tell you have never been subject to them. I can inform you they are far far worse than you could imagine. It’s a major human rights crime to force these drugs. Olanzapine is nothing short of torture. They directly contribute to acts of terrible violence by removing your ability to care about anything. Now add that to getting akathisia and you have people turned into potential and actual killers.

      Report comment

  5. Taking Olanzapine as an example (being a notoriously diabolical concoction):

    Seratonin 5HT2A (4), 5HT2C (11) and 5HT6 (5).
    Histamine H1 (7).
    Dopamine D1-D4 (11 – 31)
    Adrenergic α1 (19)

    So, theoretically it blocks (in order) 5HT2A, 5HT6, H1, 5HT2C, before dopamine and α1.

    I read somewhere that on average, 5HT2A blocking is max at about 5mg. At 2,5mg dopamine blocking is negligible.

    It seems to me there are quite a few obstacles to overcome in order to taper from this drug. Withdrawal from different chemical effects at different dosages. Above 5 mg the dopamine blocking effect, below seratonin and histamine.

    Report comment

  6. Many thanks for reviewing these results and helping share them further, Bernalyn. Really appreciate all the comments above.

    For those wanting to know more about the finer points of the paper, the fulltext is available free to anyone using this link until the 25th of November:

    Withdrawal is not something we can realistically put a specified timeframe around and say that’s long enough for everyone. The associations I found with gradual withdrawal were small, and as a number of people point out above, method is only one small part of the picture, and definitely relapse should not be the main outcome being looked at. I was more focused on looking at what helps people succeed than relapse. But relapse did get in the way of success, so looking at ways to assist people to manage that does seem important.

    In another paper I show that having support for the attempt also shows association with success and avoiding relapse during the attempt. And in another yet to be published study habitual use of avoidant coping partially predicted quality of life among those who continued and those who stopped (available on research gate in my thesis), and I think this would likely also be at play during withdrawal, and certainly all the qualitative studies highlight the role of coping too.

    It really is a very complicated process, and everyone caught in the loop of it has all my empathy (and my commitment to keep doing my little bit to help shift things in a direction that might make it easier).

    Report comment

    • I hate to say it, but I do wonder if the whole language of relapse is a mistake. It implies a deterioration after a period of improvement, or a return to a previous worse state – as judged by one person with a clear observational bias, the psychiatrist.

      The misnamed “anti-psychotics” are really broad spectrum brain suppressants, that, at the very best, suppress psychotic symptoms that may well return at a worse level upon discontinuation. Its worth noting that even this “beneficial” effect is really very weak in most of the drugs. In the process they damage your body, make you dopey and apathetic, and, I would argue, depressed, fatalistic, defeated and psychologically dependent. They therefore put you in an uncomfortable and unsustainable limbo, not a relapsed state of good health.

      Relapse implies they got better. They didn’t. The drugs just stuck a lid on things and I think recovery can only begin when you start to come off them.

      Report comment

      • Concerned Carer

        I was happy to “decline” neuroleptics to begin with – but after several years of neuroleptic consumption I was incapable of surviving without them. So as far as I can see the drugs are causing the problems.

        ( the end of the Day any problems I had responded to a non drug approach).

        Report comment

        • Fiachra: I’m interested in what you think about this idea that people tend to “relapse” when they stop the meds, and it’s just a question of how many. Do you think for example , the drugs led to “remission” of the problems you had, or did they just suppress a subset of the symptoms for a while, at the same time giving you some more.

          It used to really irk me when we were told that the drugs “got you better”, or even more ludicrous “got you back into life” or bizarrely “shifted these ideas” once you got to “therapeutic dose”. It all looks like a load of cods now.

          I can’t put my finger on what exactly the drugs do , but I don’t think they get you better , so I’m not sure it’s right to say that people relapse. I know this is an extreme view, basically saying that they are useless and just fool the doctors into thinking the person is “better” when in fact they are just coshed.

          Report comment

          • Thanks Concerned Carer,

            I had hospitalization after hospitalization while I consumed the neuroleptic injection (1980 to 1984), and my records reflect a non recovery situation. My last hospitalization was in 1984 with a very poor Prognosis – and then full Recovery.

            I did suffer from what Robert Whitaker describes (very accurately ) as (neuroleptic withdrawal) “High Anxiety” BUT this responded successfully to practical inexpensive psychotherapy.

            The practical Psychotherapy was the key.

            Report comment

      • Most of today’s “mentally ill” have real physical problems related to involuntary drug use for years on end. TBI is a legitimate brain disorder created by the shrinks pretending to fix a problem that never existed till they rushed in.

        Where angels feared to tread. 😛

        Report comment

    • In my experience at SA (and in the real world), tapering antipsychotics is extremely difficult, because the drugs destroy the frontal lobes – centres of concentration, discipline and reasoning required to come off the drug slowly.

      This is why so many impulsively get off – and in many ways it’s the only way for some people. Is to just stop destroying their mind (and tolerate the disabling intrusive symptoms for as long as it takes to get better).

      This is not “relapse,” withdrawal symptoms can strike 6-18 months out. “one month” is hardly a test.

      Report comment

      • I struggle with poor concentration–though making myself read short stories, essays, novellas is helping.

        What’s worse is BAD short term memory, non-existent organizational skills. Seriously, I can’t organize a dresser drawer. My coordination is poor so I keep knocking stuff over. Mom keeps yelling at me for my carelessness, telling me I’ll need to go to a home when she dies. 45 is too late she says.

        Report comment

  7. I have been diagnosed with behavioral variant frontotemporal dementia, known as young onset dementia in the UK. This is due to taking Seroquel for sleep over the span of 9 years. I stopped Seroquel this April and did it over the span of a week and still got horrendous shrinkage of the frontal and temporal lobes

    Report comment

    • How much Seroquel were you taking?

      I took very low dose Seroquel about 6 mg per night for years. But when I stopped my sleeping went haywire. It took a number of years to come back to normal.

      I also had liver problems for quite a while when I stopped, I think this was possibly due to the absence of chemicals passing through my system.

      Report comment

  8. I was taking 400 mg per night for the last two of the 9 years. How were you able to get 6mg? I started with 25. I withdrew over the course of a week in April and started having tardive dyskinesia, tardive dystonia and very strange symptoms like the mouth opening very wide and needing to open it even wider, tongue protruding, face contorting into strange grimaces, the neck twisting around, one shoulder hunching and not the other, legs freezing into a strange gait where I would drag one leg. I still have those symptoms and constant spasms in my right foot and toes. I can’t drive anymore because of this. A friend thought it was funny to say that I should be part of a circus freak show. I don’t know what to do now. I have been severely damaged forever. I hide from people. I can’t think, I can’t function, and I can’t take showers, just like people with old age dementia. I am only 45

    Report comment