We read with interest the paper by Harrington et al (Psychiatric Bulletin, November 2002, 26, 414-418). It mirrored the results of a recent audit we did on atypical, high dose and co-prescribing of antipsychotics, as well as concurrent anticholinergic prescribing. Drug card information on all general adult inpatients, aged 18-65, prescribed an antipsychotic on a regular basis was collected during a single visit in April 2000. Re-audit occurred in June 2001, and on this occasion adult day hospital patients were included.

Overall, 231 inpatients received a regular prescription for an antipsychotic drug in 2000 and 321 patients in 2001 (213 in-patients and 99 day hospital patients). There were high levels of atypical antipsychotic prescribing (49%, 95% confidence interval [CI] 42-55 in 2000 and 52%, 95% CI 46-57 in 2001), co-prescribing (20%, 95% CI 15-26 in 2000 and 25, 95% CI 20-30 in 2001) and above British National Formulary (2002) limits prescribing (15%, 95% CI 11-20 in 2000 and 17%, 95% CI 12-21 in 2001) when compared to total antipsychotic prescribing. Furthermore, there were high levels of anticholinergic prescribing with atypical antipsychotics alone (21, 95% CI 13-31 in 2000 and 18%, 95% CI 12-26 in 2001).

These prescribing patterns increase the risk of side-effects and negate the cost benefits of atypical antipsychotics. They also run counter to 1993 Royal College of Psychiatrists guidelines on antipsychotic prescribing and World Health Organization anticholinergic prescribing guidelines (World Health Organization, 1990; Reference BarnesBarnes, 1990). This audit appears to indicate problems in dissemination and a lack of widespread knowledge of current guidelines.