To stay on antipsychotics or not to stay on antipsychotics? A longstanding question with an update


Although antipsychotic medications have been a first-line intervention for the treatment of schizophrenia and other psychotic disorders since their initial development in the 1950s, the question of how long antipsychotic therapy should be continued – and at what dose – continues to be debated some 70 years later.

Most randomised controlled trials (RCTs) support the long-term use of antipsychotic medications for relapse prevention. For example, based on the results of 75 RCTs comparing antipsychotic medication to placebo in patients with schizophrenia spectrum disorders conducted between 1959 and 2017, a recent meta-analysis found that compared to antipsychotic continuation, discontinuation was associated with a threefold greater risk of relapse at one year (Ceraso et al., 2021). Another meta-analysis of RCTs of up to 104 weeks duration similarly found that compared to standard antipsychotic dosing as defined by World Health Organization guidelines, low-dose (within 50-99% of the lower limit of the standard dose) and very low-dose (<50% of the lower limit) antipsychotic therapy were associated with 44% and 72% greater risks of relapse respectively for patients with schizophrenia or schizoaffective disorder (Højlund et al., 2021).

Such results stand in contrast to those of non-randomised longitudinal studies where the outcomes have been less consistent. Tiihonen et al. (2018) found that within a cohort of 7,434 patients with schizophrenia followed over 20 years, continuous antipsychotic treatment was associated with the lowest risk of rehospitalisation or death compared to antipsychotic discontinuation following different durations of treatment. Hayes et al. (2019) also indicated that relapse was least likely among those who remained on antipsychotic medication for more than 5 years. On the contrary, Harrow and colleagues (2021) concluded that those who were not on antipsychotics after 2 years had better outcomes than those who continued antipsychotic medication.

Determining the answers to several relevant questions could help disentangle these varied outcomes:

  • Can disparate study results be explained by diagnostic differences? It could be that those with a less well-defined first episode or affective psychoses may be more likely to benefit from discontinuation than those with chronic schizophrenia and schizoaffective disorder (Pierre, 2021).
  • Does antipsychotic discontinuation result in better outcomes as Harrow et al. (2021) claim or does recovery precede the justifiable tapering of antipsychotic therapy to discontinuation (Pierre et al., 2021)?
  • Does symptomatic worsening following antipsychotic discontinuation reflect clinical relapse or the effects of antipsychotic drug withdrawal (Murray et al., 2016)?

To address such questions and resolve ongoing debates about the utility of long-term antipsychotic medication for the treatment of psychotic disorders, Ostuzzi et al. (2022) completed a new study to determine whether continuing, reducing, switching, or stopping antipsychotic therapy is best at preventing relapse among clinically stable individuals with schizophrenia-spectrum disorders.

Antipsychotic therapy is a clearly effective treatment for psychotic disorders in the short-term, it is less clear whether—or for whom—long-term treatment is needed.

Antipsychotic therapy is an effective treatment for psychotic disorders in the short-term, but it is less clear whether—or for whom—long-term treatment is needed.


To compare continuation, reduction, switching, or stopping antipsychotic medications, the authors utilised a network meta-analysis to examine the results of 98 RCTs that included adults with schizophrenia spectrum disorders who were clinically stabilised on antipsychotic maintenance therapy. Included studies required at least 25 subjects with randomised comparisons of at least two of these treatment strategies where reduced dosing was defined as doses lower than those recommended by the International Consensus Study of Antipsychotic Dosing or other dosing guidelines. Studies also had to be at least 6 weeks in duration with a pre-randomisation wash-out period of up to 4 weeks. Clinical trials involving treatment-refractory patients were excluded.

The primary outcome measure was the number of subjects who relapsed by the end of the clinical trial as defined by each study. If not defined in the original study, relapse was defined in the meta-analysis based on symptomatic worsening on clinical rating scales (e.g. PANSS increase > 25%, BPRS > 30% or CGI-S > 2 points). Secondary outcomes included change scores for symptoms, quality of life, and functional status; rates of hospitalisation; and the reason for study drop-out.


The 98 RCTs included in the network meta-analysis had a mean sample size of 148.5 subjects whose mean age was 38.8 years. The proportion of men in the studies was 62% while the proportion of those of White or Caucasian race/ethnicity was 56%. Most trials involved subjects with schizophrenia (72.5%) or both schizophrenia and schizoaffective disorder (16.3%) who had multiple episodes of disease (94.9%). The majority of participants were treated with oral (59.3%) and first-generation (59.8%) antipsychotic medications.

Antipsychotic continuation was the most common (40.6%) treatment strategy in the RCTs, followed by stopping (29.0%), switching (20.8%), and dose reduction (9.6%). Double-blinding was present in 79.6% of the included trials.

For the primary outcome of relapse, continuation, switching, and reducing were all significantly more effective than stopping antipsychotic medication. Pairwise comparisons between stopping and other treatment strategies revealed that the relative risk (RR) of relapse was 0.37 for continuation, 0.44 for switching, and 0.68 for reducing. These results indicate that the risk of relapse was reduced more than two-fold by staying on antipsychotic medication. Relapse was also less likely when continuing (RR 0.55) or switching (0.65) medication compared to dose reduction. There was no significant difference between continuing or switching.

Staying on antipsychotic medication also outperformed stopping on several secondary measures. Risk of hospitalisation was significantly lower for continuation (RR 0.53) switching (0.56), and reducing (0.62) compared to stopping. All three strategies also outperformed stopping based on symptomatic improvement. The risk of discontinuing study participation due to the inefficacy of treatment was significantly less likely with continuation (RR 0.38) and switching (0.49) compared to stopping as well. Finally, continuing and reducing outperformed stopping for functional status while continuing outperformed stopping based on the quality of life.

In order to examine the potential confound of abrupt antipsychotic discontinuation leading to withdrawal symptoms, a network meta-analysis was performed with the exclusion of studies where discontinuation was “abrupt or fast” (< 4 weeks). Based on the remaining 33 studies, continuation (RR 0.32), switching (0.32); and reducing (0.56) were still all significantly more effective at preventing relapse compared to stopping.

Continuing, switching, and reducing antipsychotic medication (than stopping) was more effective to prevent relapse among people with chronic schizophrenia and schizoaffective disorder.

Continuing, switching, and reducing antipsychotics were more effective than stopping to prevent relapse in chronic schizophrenia and schizoaffective disorder.


  • This is the first published study utilising network meta-analysis to compare the risk of relapse with antipsychotic continuation, switching, dose reduction, and discontinuation among patients with schizophrenia-spectrum disorders.
  • Based on 98 RCTs involving nearly 14,000 participants, all treatment strategies that involved maintaining patients on antipsychotic medication were more effective for the primary outcome measure of preventing relapse than discontinuation.
  • Secondary outcomes that included risk of hospitalisation and quantifiable measures of symptom severity, quality of life, and functional status mirrored the primary outcome results with antipsychotic continuation outperforming discontinuation on all measures.

Based on the number needed to treat for the main study outcome, the authors concluded that…

…for every 3 individuals continuing antipsychotic treatment at standard doses, one additional individual will avoid relapse compared to stopping antipsychotic treatment, which can be regarded as a large effect magnitude according to commonly used thresholds.

The risk of relapse in schizophrenia and schizoaffective disorder is significantly lower with antipsychotic maintenance compared to discontinuation.

The risk of relapse in schizophrenia and schizoaffective disorder is significantly lower with antipsychotic maintenance compared to discontinuation.

Strengths and limitations

Network meta-analysis is a relatively recently developed statistical method that combines direct and indirect evidence to make multiple comparisons at the same time, with the strength of its conclusions dependent on heterogeneity and study quality within the network (Rouse et al., 2017). Ostruzzi et al.’s network meta-analysis of antipsychotic maintenance strategies was based on nearly 100 RCTs with comparisons showing moderate confidence levels based on the Confidence in Network Meta-Analysis (CINemA) method that examines heterogeneity and potential bias (Nikolakopoulou et al., 2020). Although CINeMA revealed “some concerns” about heterogeneity and reporting/sponsorship bias, the significant effectiveness of continuation, switching, and dose reduction compared to discontinuation remained when trials with an overall high risk of bias were excluded.

The authors acknowledge several other study limitations including a smaller number of studies that reported secondary outcome measures and that involved more gradual discontinuation of antipsychotic medications beyond 4 weeks. Despite this smaller number of studies, however, the secondary outcome findings were nonetheless consistent with the primary outcome results and the primary outcome finding remained significant when studies of abrupt discontinuation were excluded.

As the authors point out, since most RCTs compared antipsychotic treatment to placebo rather than no treatment at all, it is possible that the difference in effectiveness between maintenance antipsychotic therapy and discontinuation might be larger in real world clinical practice.

It should be noted that most of the studies included in the network meta-analysis involved adult participants with chronic schizophrenia and schizoaffective disorder who were treated with first-generation antipsychotic medications. The results may be less generalisable to patients with psychotic disorders on the milder side of the schizophrenia spectrum, such as those with affective and first-episode psychoses.

it is possible that the difference in effectiveness between maintenance antipsychotic therapy and discontinuation might be larger in “real world” clinical practice.

The difference in effectiveness between maintenance antipsychotic therapy and discontinuation might be larger in real world clinical practice than the findings.

Implications for practice

The results of clinical trials to date suggest that while maintenance antipsychotic therapy is associated with the best outcomes for some patients with chronic schizophrenia-spectrum illness, others can fare well with discontinuation. Consequently, there is no “one size fits all” answer to the question of how long antipsychotic medication should be continued for the treatment of schizophrenia-spectrum psychotic disorders. Though some studies have attempted to determine specific predictors of long-term antipsychotic benefit (Tani et al., 2018; Bogers et al. 2020), it remains unclear who can be safely tapered off antipsychotic medication. Given the often significant side effect burden of antipsychotic therapy, obtaining a better understanding of who is most likely to benefit from long-term antipsychotic treatment vs. discontinuation is crucial to optimising the care of those suffering from psychotic disorders.

Ostruzzi et al. (2022) add to a substantial body of evidence that supports long-term antipsychotic treatment for individuals with chronic schizophrenia and schizoaffective disorder. Although it is possible that the results could be explained by too rapid antipsychotic discontinuation as argued by the commentary that accompanies Ostruzzi et al.’s study (Sommer et al., 2022), the primary outcome still held true based on 33 studies involving gradual discontinuation over more than 4 weeks. Furthermore, maintaining standard antipsychotic dosing was more effective than reducing the antipsychotic dose to lower than standard doses, which could be regarded as a proxy for gradual discontinuation.

These results suggest that most patients with chronic schizophrenia or schizoaffective disorder should be maintained on antipsychotic medication in the long term. Although dose reduction should always be a consideration to minimise side effects, reducing the dose below standard dose ranges should be weighed carefully against an increased risk of relapse.

 To minimise the risk of relapse in chronic schizophrenia and schizoaffective disorder, most patients should continue rather than discontinue taking antipsychotic medications.

To minimise the risk of relapse in chronic schizophrenia and schizoaffective disorder, most patients should continue taking antipsychotic medications.

Statement of interests



Primary paper

Ostuzzi G, Vita G, Bertoloni F, Tedeschi F, De Luca B, Gastaldon C, Nosé M, Papola D, Purgato M, Del Giovane C, Correll CU, Barbui C. (2022) Continuing, reducing, switching, or stopping antipsychotics in individuals with schizophrenia-spectrum disorders who are clinically stable: a systematic review and network meta-analysis. Lancet Psychiatry 2022; 1-9

Other references

Bogers JPA, Hambarian G, Michiels M, Vermeulen J, de Haan L. Risk factors for psychotic relapse after dose reduction or discontinuation of antipsychotics with chronic schizophrenia: a systematic review and meta-analysis. Schizophr Bull Open 2020; 1:sgaa002.

Ceraso A, LIN JJ, Schneider-Thoma J, Siafis S, Tardy M, Komossa K, Heres S, Kissling W, Davis JM, Leucht S. Maintenance treatment with antipsychotic drugs for schizophrenia. Cochrane Database of Systematic Reviews 2020; 8: CD008016

Harrow M, Jobe TH, Tong L. Twenty-year effects of antipsychotics in schizophrenia and affective psychotic disorders. Psychol Med 2021; February 8: 1-11.

Hayes J, Osborn DPJ, Lundin A, Dalman C. Psychiatric hospitalization following antipsychotic medication cessation in first episode psychosis. J Psychopharmacol 2019; 33:532-534.

Højlund M, Kemp AF, Haddad PM, Neill JC, Correll CU. Standard versus reduced dose of antipsychotics for relapse prevention in multi-episode schizophrenia: A systematic review and meta-analysis of randomised controlled trials. Lancet Psychiatry 2021; 8:471-476.

Murray RM, Quatorre D, Natesan S, van Os J, Nordentoft M, Howes O, Di Forti M, Taylor D. Should psychiatrists be more cautious about the long-term prophylactic use of antipsychotics? Brit J Psychiatr 2016; 209:361-365.

Nikolakopoulou A, Higgins JPT, Papakonstantinou T, Chaimani A, Del Giovane C, Egger M, Salanti G. CINeMA: an approach for assessing confidence in the results of a network meta-analysis. PLoS Med 2020; 17:e1003082.

Pierre JM, Zito MF, Yang TS, Marder SR. Antipsychotic discontinuation and recovery: chicken or egg? Psychological Medicine 2021; 1-2

Pierre JM. Diagnostic uncertainty, antipsychotic dosing, and optimal psychosocial interventions: unanswered questions in first-episode psychosis. Schizophr Res 2021; 228:600-601.

Rouse B, Chaimani A, Li T. Network meta-analysis: an introduction for clinicians. Intern Emerg Med 2017; 12:103-111.

Sommer IE, Horowitz M, Allott K, Speyer H, Begemann MJH. Antipsychotic maintenance treatment versus dose reduction: how the story continues. Lancet Psychiatry 2022.

Tani H, Suzuki T, Fleishhhacker WW, Tomita M, Mimura M, Uchida H. Clinical characteristics of patients with schizophrenia who successfully discontinued antipsychotics. J Clin Psychopharmacol 2018: 38:582-589.

Tiihonen J, Tanskanen A, Taipale H. 20-year nationwide follow-up study on discontinuation of antipsychotic treatment in first-episode schizophrenia. Am J Psychiatry 2019; 175:765-773.

Wunderink L, Nieboer RM, Wiersma D, Sytema S, Nienhuis FJ. Recovery in remitted first-episode psychosis at 7 Years of follow-up of an early dose reduction/discontinuation or maintenance treatment strategy: Long-term follow-up of a 2-Year randomized clinical trial. JAMA Psychiatry 2013; 79:913-920.

Photo credits

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+
Mark as read
Create a personal elf note about this blog
Profile photo of Joseph Pierre

Joseph Pierre

Joe Pierre, MD is a Health Sciences Clinical Professor in the Department of Psychiatry and Behavioral Sciences at UCSF and the Unit Chief of the Langley Porter Psychiatric Hospital Adult Inpatient Unit. He has extensive clinical experience working with individuals with psychotic disorders, substance abuse, and those with “dual diagnosis” as well as research experience participating as a primary investigator and collaborator for clinical trials in schizophrenia and early intervention for young persons at high risk for psychosis. Dr. Pierre has authored over 100 papers, abstracts, and book chapters related to schizophrenia, antipsychotic medications, substance-induced psychosis, delusions and delusion-like beliefs, auditory hallucinations and voice-hearing, and a variety of other topics including the neuroscience of free will and culturally sanctioned suicide. He is also the author of the Psych Unseen blog at Psychology Today and serves as an expert witness consultant in forensic/legal cases involving schizophrenia, the intersection of psychosis and religion, and delusion-like beliefs and conspiracy theories.

More posts - Website

Follow me here –