Effectiveness of a cognitive behavioural workbook for changing beliefs about antipsychotic polypharmacy: analysis from a cluster randomized controlled trial

Abstract Rationale, aims and objectives Educational workbooks have been used in psychiatry to influence patient but not clinician behaviour. Targeted education interventions to change prescribing practice in other areas of medicine have only looked at changes in prescribing and not attitudes or beliefs related to the prescribing. We aimed to examine whether clinicians' beliefs about a common prescribing issue in psychiatry (antipsychotic polypharmacy prescription) changed alongside behaviour as a result of a complex intervention. Method Medical and nursing staff were recruited from 19 general adult psychiatry units in the south‐west of the UK as part of a cluster randomized controlled trial. A questionnaire was used to assess beliefs on the prescribing of antipsychotic polypharmacy as a secondary outcome before and after completion of a cognitive behavioural ‘self‐help’ style workbook (one part of a complex intervention). A factor analysis suggested three dimensions of the questionnaire that corresponded to predetermined themes. The data were analysed using a random‐effects regression model (adjusting for clustering) controlling for possible confounders. Results There was a significant change in beliefs on both of the factors: antipsychotic polypharmacy (coefficient = −0.89, P < 0.01) and rapid tranquilization (coefficient = −0.68, P = 0.01) specifically targeted by the workbook. There was a modest but statistically significant change in antipsychotic polypharmacy prescribing (odds ratio 0.43, 95% confidence intervals 0.21–0.90). Conclusions The workbook appeared to change staff beliefs about antipsychotic polypharmacy, but achieving substantial changes in clinician behaviour may require further exploration of other factors important in complex prescribing issues.