Are antipsychotics needed for the long-term treatment of schizophrenia?

‘Do we need antipsychotics for the long-term treatment of schizophrenia?’ was the title of this important debate at EPA2020. Both speakers agreed on the effectiveness of antipsychotic treatment in the acute phase1, but there was discussion around whether and how drug treatment could be reduced or stopped in the longer term.

Relapse rates are high if maintenance treatment is stopped

Relapse rates are low once patients respond and are stabilized, but as high as 97% at 36 months after treatment discontinuation

Robin Emsley (Stellenbosch University, Cape Town, South Africa) presented the need for ongoing antipsychotic maintenance treatment in schizophrenia, focusing on the consequences of stopping treatment once established.

Relapse rates are low once patients respond and are stabilized 2, but as high as 97% at 36 months after treatment discontinuation3.  One study had an 18-month relapse rate of 43%, but treatment could not be discontinued in 50% of cases4. Relapses often occur shortly after treatment discontinuation5, and a longer treatment duration does not reduce the risk of relapse3,6. There are no reliable early warning signs of impending relapse6.

Consequences of relapse may be severe, including psychological and biological effects. Recurrent episodes of relapse may lead to treatment refractoriness, with treatment not effective in 1 in 6 remitted first-episode patients after a second episode7.

Some studies show patients can remain symptom-free without medication, due to the biphasic nature of schizophrenia.  Antipsychotics are important in the active phase of deterioration, with more potential to wean medication in the chronic plateau phase.

The risk-benefit decision, concluded Prof Emsley, is between high risk of relapse and associated consequences if antipsychotics are discontinued, versus tolerability and safety concerns of long-term treatment. He advocated continuing antipsychotic maintenance therapy with lowest effective dose.

Consequences of relapse may be severe including psychological and biological effects

Long-term treatment is not necessary for all

Robin Murray (King’s College, London) took the opposing view, that antipsychotics are not necessary for long-term prophylactic treatment of all patients with schizophrenia.

Although antipsychotics are important in the acute phase, long-term prophylaxis is less clear cut. Patients are reluctant to take long-term treatment when they are feeling well, especially with a side effect burden including obesity.

Prof Murray suggested one-fifth of patients could stop antipsychotics after their first episode, and more may be able to reduce doses. In a 10-year follow-up study8, 19% of those with schizophrenia had no psychotic symptoms and were off antipsychotics. In another study, outcomes were better at 18 months in the continuation arm, but by 7 years those in the decrease/stop arm were functioning better4

Prof Murray concluded that antipsychotics should not be abandoned, but with long-term treatment to use the minimum possible dose for shortest possible time, aiming to stop in some patients.

The future

When patients initiate stopping treatment the psychiatrist’s role is to support their decision-making

Overall there was more agreement than disagreement, with both clinicians acknowledging the pros and cons of prophylactic treatment, and suggesting dose reduction is considered. They stressed the need for research to identify which patients can stop, when and how. Often the patient initiates stopping treatment, and the psychiatrist’s role is to support their decision-making, with best available evidence.

Our correspondent’s highlights from the symposium are meant as a fair representation of the scientific content presented. The views and opinions expressed on this page do not necessarily reflect those of Lundbeck.


  1. Robinson DG et al. Am J Psychiatry 1999;156:544-9
  2. Emsley R, et al. J Clin Psychopharmacol 2008;28:210-3
  3. Emsley R, et al. J Clin Psychiatry 2012;73:e541-7
  4. Wunderink L, et al. J Clin Psychiatry 2007;68:654-61
  5. Emsley R, et al. J Clin Psychiatry 2012;73:e541-7
  6. Gitlin M, et al. Am J Psychiatry 2001;158:1835-42
  7. Emsley R, et al. J Clin Psychopharmacol 2013;33:80-3
  8. Morgan C, et al. Psychol Med 2014;44:2713-26