How Many Antipsychotics Before Clozapine? When to Start

Clozapine is indicated for schizophrenia that hasn't responded adequately to standard antipsychotic treatment.¹ Consensus criteria define that as at least two different antipsychotic trials, each at an adequate dose and duration, with adherence checked.² In practice the wait is much longer. In one clinic study of 149 patients, the mean theoretical delay before clozapine was started was 47.7 months, with wide variation between patients.³

During that delay, 36.2% of patients had been given two antipsychotics at once and 34.2% had been given high doses, despite treatment guidelines.³ Clozapine gets called a last resort. The consensus criteria point to two adequate trials, then clozapine.

What counts as treatment-resistant schizophrenia

The Treatment Response and Resistance in Psychosis (TRRIP) working group reviewed 42 trials and found that half gave no operational definition of treatment resistance. The group's 2017 consensus set out the elements a definition needs:²

• Current symptoms of a minimum duration and severity, measured with a standardized rating scale
• Moderate or worse functional impairment
• At least two different antipsychotic trials, each of a minimum duration and dose
• Adherence that has been checked systematically
• Ideally, at least one trial that was followed prospectively

A medication that wasn't taken hasn't failed. Each trial has to be long enough, dosed adequately, and actually taken, and substance use or another diagnosis has to be considered as the explanation.

What the evidence shows

The trial that established clozapine enrolled 268 patients who had not responded to at least three antipsychotics and then failed a prospective six-week trial of haloperidol. After six weeks, 30% of those assigned to clozapine met response criteria, compared with 4% assigned to chlorpromazine.⁴

A 2016 meta-analysis covering 21 papers and 25 comparisons found a number needed to treat of 9 and superiority for positive symptoms in both the short and long term.⁵ Pooled response data from the same review put the response rate at 40.1%, with a mean reduction of 22 points on the Positive and Negative Syndrome Scale.⁶ In the CATIE study, patients who had stopped a newer antipsychotic stayed on clozapine for a median of 10.5 months, compared with 3.3 months on quetiapine and 2.8 months on risperidone. The difference from olanzapine was not statistically significant for that outcome.⁷

The trial evidence is less settled than the guidelines suggest

A 2025 individual patient data meta-analysis in Lancet Psychiatry pooled 19 blinded trials with 1,599 participants. It found a mean difference of 0.64 points on the PANSS in favor of other second-generation antipsychotics, mainly olanzapine and risperidone, with a credible interval that included no difference. The authors graded confidence in the evidence as very low.⁸

While short blinded trials haven't shown clear superiority over olanzapine or risperidone, the real-world data point the other way. A meta-analysis of 63 cohort studies covering 109,341 patients found that clozapine was associated with a lower risk of hospitalization (risk ratio 0.82) and of stopping treatment for any reason (risk ratio 0.73), even though the patients on clozapine were more severely ill.⁹ Cohort studies can't establish causation. The trials measured change at six to eight weeks.⁸

The same cohort data carry the cost. Clozapine was associated with 1.7 kg more weight gain and a higher risk of type 2 diabetes than other second-generation antipsychotics.⁹ The side effects page covers what to monitor.

How long a clozapine trial should last

Response in the original trial was measured at six weeks.⁴ The 2016 meta-analysis concluded that if there is no response by six months, medications with fewer adverse effects should be considered.⁵ A fair trial runs for months at an adequate dose.

When clozapine doesn't work well enough

About 40% of people with treatment-resistant schizophrenia meet response criteria on clozapine.⁶ For the rest, the evidence on add-on treatments is mixed. In a single-blind randomized trial of 39 patients, without a sham comparison, 50% of those who received electroconvulsive therapy added to clozapine met response criteria, compared with none who continued clozapine alone.¹⁰ In Finnish and Swedish national registers, adding medium-dose aripiprazole to high-dose clozapine was associated with a lower risk of hospitalization for psychosis than clozapine alone, with an adjusted hazard ratio of 0.68.¹¹ The register finding is observational. The later and less favorable trials are covered on the page about what to do when clozapine isn't working.

Why clozapine is started late or never

In a Veterans Affairs cohort that my colleagues and I studied, 246 of 14,620 patients with schizophrenia received clozapine, or 1.7%.¹² Clozapine and long-acting injectable antipsychotics have been described as among the most effective treatments in schizophrenia and as markedly underused.¹³ Across 75 countries, clozapine use tracked the number of psychiatrists and was unrelated to how strict the blood monitoring rules were.¹⁴ A 2024 expert review named the lack of physician training as one of the main obstacles.¹⁵

The blood test requirement itself changed in 2025. I've written separately about clozapine blood tests after the REMS ended.

What to ask the psychiatrist

• How many antipsychotics have been tried at an adequate dose and for long enough?
• Was adherence confirmed for each one, and was a long-acting injectable considered?
• Has the diagnosis been reviewed, including substance use and mood symptoms?
• Is there a reason clozapine has not been offered?
• If the current prescriber does not start clozapine, who nearby does?

Clozapine also has a second FDA-approved use that does not require treatment resistance: reducing the risk of recurrent suicidal behavior in schizophrenia or schizoaffective disorder.¹ That evidence is on the clozapine and suicide risk page.

Limitations of the evidence

This evidence has limitations. First, the original trial compared clozapine with chlorpromazine, and newer blinded trials against second-generation drugs are less clear. Second, the cohort studies are observational, and who gets selected for clozapine affects the result. Nevertheless, the indication is clear, and the delay before a trial is measured in years.

In a crisis: call or text 988, or go to the nearest emergency department. A first episode of psychosis belongs in a hospital-based or coordinated specialty care program.

I start and manage clozapine in an outpatient practice in Chelsea, Manhattan. Details are on the schizophrenia care in NYC page. Clozapine should not be stopped, skipped, or restarted without the prescriber. This page is general information and does not replace an evaluation.

Related: Schizophrenia and psychosis treatment guide · Clozapine blood tests after the REMS ended · Clozapine side effects · Missed clozapine doses · Clozapine and suicide risk · Cannabis and psychosis · Cobenfy vs clozapine · Injections vs pills · Stimulants and psychosis · Drug-induced psychosis · Clozapine levels · Clozapine myocarditis · Clozapine constipation · Clozapine drooling · Clozapine weight gain · Who can prescribe clozapine · Clozapine in teenagers · When clozapine isn't enough · Benign ethnic neutropenia · Starting clozapine · Clozapine and pneumonia · Clozapine seizures · Clozapine in pregnancy · Clozapine rechallenge · Key clozapine studies