To the Editor: The recently published APA Practice Guideline for schizophrenia (1) completes a trilogy of guidelines for schizophrenia; the other two are the Texas Medication Algorithm Project (2) and the Expert Consensus Guideline Series (3). We can use these guidelines as suggestions, or we can use these them as strict rules. In Philadelphia, where I work, the administrators of AmeriChoice of Pennsylvania, a managed care company, insist that we use these guidelines as strict rules.

According to AmeriChoice’s interpretation of these guidelines, we "shalt not" combine atypical antipsychotics until we first use clozapine. I do not believe that this commandment is rational or therapeutic for a certain group of patients. These are the patients who have a significant but partial response, with minimal or no side effects, to the maximum dose of an atypical antipsychotic. For example, before medication, these patients might have intrusive auditory hallucinations constantly or almost every day. After taking the maximum dose of an atypical antipsychotic, they might have less intrusive auditory hallucinations that occur only 1 or 2 days a week.

These patients usually want and need a fine-tuning of their medication, not a complete overhaul. Substituting clozapine would require laboratory tests every 2 weeks—possibly for the rest of their lives—and would increase the risks of weight gain, lethargy, seizures, and agranulocytosis. Why would these patients want to switch to clozapine if they are already taking a medication that provides significant although partial relief from their symptoms, that requires relatively infrequent laboratory monitoring, and that has few or minimal side effects?

In my clinical experience, adding an atypical antipsychotic (other than clozapine) to the original atypical antipsychotic might reduce or eliminate symptoms for these patients. The additional atypical antipsychotic could be safely and easily withdrawn if it were ineffective or had adverse side effects. An additional atypical antipsychotic would be preferable to an additional typical antipsychotic because of its lower risks for extrapyramidal symptoms and tardive dyskinesia.

This particular clinical situation that I am describing—of patients having a significant but partial response, with minimal or no side effects, to the maximum dose of an atypical antipsychotic—is frequent among chronically psychotic patients. By refusing to pay for an additional atypical antipsychotic before clozapine is tried, AmeriChoice is encouraging us to choose clozapine, which is often inappropriate and impractical in this situation. The potential benefits of clozapine often do not outweigh the labor and the potential risks necessary for its use in this situation. For all practical purposes, AmeriChoice is using these guidelines to deny patients a medication regimen, combining atypical antipsychotics that might reduce or eliminate symptoms and that is often more appropriate and more practical than substituting clozapine in this situation.