The quest for the correct dose | Behavioral Healthcare Executive Skip to content Skip to navigation

The quest for the correct dose

September 1, 2010
by William M. Glazer, MD
| Reprints
Recent efforts shed light on antipsychotic dosing, but find much work is left to be done

Doctors have a lot of antipsychotic agents to choose from. With a complex foundation of evidence and experience, prescribers need to know optimal dosing strategies for a wide range of eligible patients. Recent efforts to characterize differences in effectiveness among antipsychotic medications have been inconclusive while, at the same time, FDA indications for this class of medications are expanding beyond schizophrenia and bipolar disorder.1,2

According to a group of investigators from Dalhousie University in Nova Scotia and Harvard Medical School, the proper dosage for an antipsychotic medication should be determined from “randomized, … head-to-head, comparisons of a range of antipsychotics at several fixed doses, and across various diagnostic, illness severity, comorbidity, and demographic variables.”3 But such informative evidence is badly lacking, and it is widely assumed that uncertainty about clinically equivalent doses of antipsychotics contributes to variability in outcomes in clinical practice and to inconsistent findings among experimental clinical trials.

An international consensus

The investigators decided to ask 43 experts from around the world, “What are the best dosing practices for first and second generation antipsychotic medications?”

Their study, entitled International Consensus Study of Antipsychotic Dosing, employed the “Delphi method”-which draws anonymous opinions from a broad range of experts, and then guides them to reach a consensus on dosing of these agents for patients in various clinical scenarios. By keeping the experts’ opinions anonymous, it ensures that a famous or well-regarded expert's opinion does not supersede the lesser-known ones.

Table: Clinical dosing equivalencies of selected first and second generation antipsychotics available in the U.S.


Milligram equivalent to 20 mg of olanzapine (Zyprexa)


Key: H = high confidence level; M = moderate confidence level.

Not included at the time of this survey: Asenapine (Saphris), Iloperidone (Fanapt). Adapted from Gardner et al. Am J Psychiatry 2010; 167:6, p. 687.

Second generation antipsychotics

Aripirazole (Abilify)



Clozapine (Clozaril)



Paliperidone (Invega)



Quetiapine (Seroquel)



Risperidone (Risperdal)



First generation antipsychotics

Haloperidol (Haldol)



Perphenazine (Trilafon)



Trifluoperazine (Stelazine)



The experts were asked to reach a consensus on clinically equivalent doses among 59 antipsychotic agents: 36 oral, 13 short-acting, and 10 long-acting, injectable medications. Respondents were asked:

“What dose do you consider to be clinically equivalent to 20 milligrams per day of olanzapine in treatment of the reference case, a moderately symptomatic adult man with DSM-IV schizophrenia with ≥2 years of antipsychotic treatment and not considered treatment refractory?”

The clinicians rated the confidence level for their opinion as:

  • Low (L): limited experience or frequent exceptions due to clinical circumstances;

  • Moderate (M): some experience, probably applies to all patients; or

  • High (H): based on extensive experience, applies to most patients.

The experts were instructed to give priority to efficacy over tolerability in their dosing estimates.

The table displays the clinically equivalent antipsychotic dosing recommendations for oral agents. Not included here are dose equivalents for short-acting and long-acting injectables. Overall, the authors reported reasonably strong conformity on clinically equivalent dosing estimates.

Though not exhibited in the table, the experts agreed that the overall mean median starting dose for oral antipsychotics in olanzapine equivalents was 4.8 milligrams per day, with a target range of 10.2 to 25.5 milligrams per day; and the overall mean median maximum dose for oral antipsychotics in olanzapine equivalents was 30.9 milligrams per day.

Several patient factors affected their dosing recommendations. Experts recommended the use of lower doses in:

  • Younger and elderly patients;

  • Patients with certain medical comorbidities, such as diabetes, heart disease, hepatic or renal impairment, or organic brain syndrome;

  • East Asian versus Caucasian patients;

  • Adult women; and

  • Underweight patients.

Most of these recommendations were consistent with current labeling and clinical practice.

The authors felt that the effectiveness of the two-stage Delphi method was supported by observed increases in agreement on dose estimates between the two survey stages, and by the achievement of a 76 percent final consensus rate.

Alternative perspectives

On the other hand, Dr. Gary Remington of the Schizophrenia Program at the Centre for Addiction and Mental Health in Toronto wrote in an editorial that he felt the 76 percent consensus rate in Gardner et al's study underscored “the lack of agreement, even among experts.”4