Introduction: The commercial introduction of atypical antipsychotics, called second-generation antipsychotics (SGAs), a few years ago, has led to a world-wide reappraisal of the established treatment strategies for people with psychotic or bipolar disorders. They permitted improvements in the pharmacologic management of psychiatric diseases. As compared to conventional neuroleptics or first-generation antipsychotics (FGAs), they promised better efficacy especially on negative symptoms and cognitive impairments of psychiatric diseases and, at the same time, better tolerance on neurological side effects. Now, they have shown other side effects and they have a higher acquisition cost than FGAs.

Objectives and methods: The aim of this paper is to describe and analyse the prescribing practices of antipsychotic drugs in French psychiatric hospitals for adult inpatients and to compare them with other surveys and guidelines. In June 2004, we conducted a one-day, cross-sectional, observational and naturalistic study in 13 hospitals, members of the PIC network.

Results: Two thousand one hundred and ninety-two prescriptions with antipsychotic treatment were collected. One thousand one hundred and fifty-four prescriptions (52.6%) included a SGA, but the FGAs were the most prescribed (65.8%; n=2259), principally cyamemazine (24.7%). There was one antipsychotic in 50.7% of prescriptions, two antipsychotics in 42.2%, but the second neuroleptic used was a sedative (82.6%), principally cyamemazine. Multiple antipsychotics were present in 1081 prescriptions (49.3%), with an average number of 1.57 antipsychotics. A mood stabiliser, an antidepressant, an anxiolytic and a hypnotic were coprescribed in respectively 37, 30.5, 65.1 and 41.6%. There were 2.48 psychotropic drugs associated with the principal antipsychotic; in total, with correctors of side-effects of the antipsychotics, there were 3.38 drugs per prescription. The SGAs aimed more often for psychotic (F20-F29) patients (61.9% versus 43.3% with FGAs), who were males (61.4% versus 68%), younger (42.6 years versus 44.1 years; p<0.02), with higher average daily doses, more associated with other neuroleptics (p<0.0004) and less associated with anticholinergic antiparkinsonian agents (p<10(-4)) than FGAs. Compared to other surveys, these results showed that the SGAs have become the first-line treatment for psychiatric disorders. The highest average daily doses corresponded to treatments of psychotic patients and, hence, the values might largely exceed the authorized maximum doses. Furthermore, in more than half of the cases, an FGA, generally a sedative, was associated with an SGA that did not comply with the principle of monotherapy established by the national and international guidelines; that also annulled the expected benefit of the SGAs on the awakening, cognition and the neurological tolerability of the treatment. The coprescriptions of the other psychotropic drugs to neuroleptics also remained the rule in psychiatry, showing all the complexity of pharmacological psychiatric medications. Prescriptions also included treatments for side effects of antipsychotics; even on the prescriptions including the SGAs, there was the coprescription of anticholinergic antiparkinsonian drugs, the deleterious character of which one knows on cognition. This resulted in a difficulty of understanding the prescription for the patient, associated with reduced compliance and increased risks of pharmacological side effects. The heterogeneity of the situations of crisis in psychiatric hospitals could make the strict application of guidelines' recommendations difficult. Nevertheless, the educational interventions in psychopharmacology for patients and the training campaigns for psychiatrists and nurses are necessary to improve the therapeutic management of the patient and ensure him/her optimal quality of life.

Conclusion: This kind of survey, far too rare, was very important because it showed the routine clinical settings in which these new drugs were really used. The results showed that SGAs appeared to take the place of the FGAs used in the treatment of psychoses, particularly schizophrenia, but also in the treatment of mood disorders and they reflected actual clinical practices. Other surveys must be conducted to see whether our study confirms the general trend concerning the use of these drugs and, therefore, to reassess these prescribing practices.