Dr. Howland is affiliated with the Department of Psychiatry, Western Psychiatric Institute and Clinic, 3811 O'Hara St., Pittsburgh, PA 15213 (e-mail: howlandrh@upmc.edu).

Recently, an outpatient whom I have known a long time came for a routine medication visit. He has schizophrenia and has been clinically stable for many years on a well-tolerated dosage of fluphenazine. Unexpectedly, he asked me why I was giving him an "archaic" drug. He had just seen his primary care physician, Dr. R, for a routine medical visit. Dr. R had referred to fluphenazine as an "archaic 19th century drug" and thought that it was "time for a change." My patient did not understand what the problem was with his medication and was quite disturbed by these comments. We talked about his long-term experience with fluphenazine and discussed newer alternatives. Ultimately, we agreed that there was no need to make a change.

In this issue of Psychiatric Services, Arbuckle and colleagues (1) report on a survey of psychiatrists regarding their opinion of second-generation antipsychotics. This topic is timely, given increasing concerns about the costs, efficacy, and safety of these agents compared with first-generation antipsychotics. The survey findings are predictable. Only 3% of respondents preferred first-generation agents. If psychiatrists do not prefer these drugs, and perhaps are not as familiar with them, why would they choose to use them or feel optimistic about their use? The vast majority (88%) was optimistic about the benefits of second-generation agents. If psychiatrists were not optimistic about these medications, would they still predict that they would be effective for patients who do not respond to first-generation antipsychotics? Moreover, why would a psychiatrist choose to not use a medication that he or she optimistically believes will offer an improvement?

Even though less than a quarter of respondents (22%) cited drug representatives or advertisements as being an influential source of information, it is notable that having weekly visits from drug representatives was one of two factors significantly associated with optimism about second-generation antipsychotics (the other factor being familiarity with practice guidelines). This finding is not necessarily surprising. Would many psychiatrists admit to prescribing drugs on the basis of marketing influences rather than for strict clinical or scientific reasons? What occurs during these office visits? Brand-name drug samples are given. Branded pens, notepads, and educational materials are left behind. Copies of published studies about the product are presented, sometimes with published treatment guidelines if they are favorable or are not unfavorable to the product. By contrast, use of first-generation antipsychotics is not promoted through office visits, drug samples, educational materials, or other advertising efforts.

The small number and narrow focus of the survey questions in this study, as well as the small unrepresentative sample of psychiatrists surveyed, limit the overall value and interpretation of the results. Despite these limitations, the authors know more about this group of psychiatrists than I know about Dr. R. I know nothing about Dr. R, except that he is not a psychiatrist and that he apparently believes that my prescribing fluphenazine is behind the times. Second-generation antipsychotics have been actively investigated for conditions other than schizophrenia, and some have gained approval for treatment of mood disorders. Consequently, they are being marketed more widely to the general public as well as to nonpsychiatrists. Although I do not know what sources of information influence Dr. R's opinion about antipsychotics for schizophrenia, it would surprise me if he, as a primary care physician, reads relevant research reports or practice guidelines. However, I think it is likely that drug representatives visit his office. If they do visit Dr. R, is he less likely to be optimistic toward the second-generation antipsychotics than the psychiatrists surveyed by Arbuckle and colleagues?

With the advent of the second-generation agents, are patients with schizophrenia better off than they were 15 years ago? Schizophrenia is a devastating and difficult-to-treat illness, and it is always good to have treatment options. But the choice of drug treatment for an individual patient should be based as much as possible on the best unbiased clinical and scientific information available, together with sensitivity to practical cost considerations.