Abstract
Introduction Adverse drug events (ADEs) constitute a significant problem in hospitals worldwide. However, little is known about their burden in mental health hospitals. Objective The objective of this study was to determine the prevalence, nature, severity and preventability of ADEs across three mental health trusts in England. Methods Trained clinical pharmacists retrospectively screened randomly sampled medical records to identify ADEs. An expert panel assessed all suspected ADEs to determine their causality, preventability and severity. Multivariable regression analysis (adjusted for clustering between hospitals) examined risk factors associated with ADEs. Results In total, 227 patient admissions comprising 10,164 patient-days of follow-up were included in the study. The adjusted rate of confirmed ADEs was 12.6 (95% confidence interval [CI] 5.6-26.0) per 100 admissions and 2.6 (95% CI 1.0-6.9) per 1000 patient-days, with almost a fifth of these ADEs judged as preventable 19.1% (n = 9/47). The majority of ADEs were of at least moderate clinical severity (29/47; 61.7%), and medicines from the central nervous system class were most commonly implicated in ADEs (45/47; 95.7%) including antipsychotics (31/45; 68.8%) and antidepressants (7/45; 15.5%). Patients with a hospital stay of more than 30 days (odds ratio 16.58, 95% CI 3.77-72.85) and patients with a stay of 8-30 days (odds ratio 5.32, 95% CI 1.22-23.07) were more likely to experience an ADE compared with patients with a stay of 1-7 days. Conclusions Adverse drug events in National Health Service mental health hospitals pose an important threat to patient safety. Targets for remedial interventions have been suggested for further exploration to improve patient safety in this setting.