Interrater reliability of the Antipsychotic Non-Neurological Side-Effects Rating Scale measured in patients treated with clozapine

Abstract

The Antipsychotic Non-Neurological Side-Effects Rating Scale (ANNSERS) was developed to provide a comprehensive measure for rating non-neurological adverse drug reactions (ADRs) to antipsychotics. Although there were already available measures that adequately rated specific non-neurological ADRs, such as sexual side effects, a need was identified for a scale that comprehensively rated the full range of non-neurological ADRs commonly seen across the spectrum of first and second generation antipsychotic drugs, including metabolic and autonomic ADRs. This article reports on work to establish the interrater reliability of an early version and a later, more comprehensive version of the ANNSERS (versions 1 and 2, v1 and v2, respectively). The measures were administered in London centres to patients treated with clozapine. Trained clinicians rated the patients simultaneously and independently. Interrater reliability on the scores was calculated using the kappa coefficient method. The results (mean kappa coefficients of 0.77 and 0.72, respectively) indicate that substantial interrater reliability was achieved for both versions. Items for which the main basis for rating was laboratory investigations rather than patient interview were largely excluded from this study, and kappas were also not calculated for items with a low frequency (less than 10%) of endorsement. Samples of patients on other antipsychotics would be required to reliably calculate kappa coefficients for these items. In conclusion, the ANNSERS represents a clinically applicable research innovation, with good interrater reliability on clinician judged items, which is now available for the comprehensive assessment of non-neurological ADRs to antipsychotics, to aid the processes of clinical audit, research and drug discovery.