Abstract
BACKGROUND: The use of antipsychotic drugs in dementia has been reported to be associated with increased risk of cerebrovascular events and mortality. There is an international drive to reduce the use of these agents in patients with dementia and to improve the safety of prescribing and monitoring in this area. OBJECTIVES: The aim of this project was to use enhanced automated regular feedback of information from electronic health records to improve the quality of antipsychotic prescribing and monitoring in people with dementia. METHODS: The South London and Maudsley NHS Foundation Trust (SLaM) incorporated antipsychotic monitoring forms into its electronic health records. The SLaM Clinical Record Interactive Search (CRIS) platform provides researcher access to de-identified health records, and natural language processing is used in CRIS to derive structured data from unstructured free text, including recorded diagnoses and medication. Algorithms were thus developed to ascertain patients with dementia receiving antipsychotic treatment and to determine whether monitoring forms had been completed. We used two improvement plan-do-study-act cycles to improve the accuracy of the algorithm for automated evaluation and provided monthly feedback on team performance. RESULTS: A steady increase in antipsychotic monitoring form completion was observed across the study period. The percentage of our sample with a completed antipsychotic monitoring form more than doubled from October 2017 (22%) to January 2019 (58%). CONCLUSION: 'Real time' monitoring and regular feedback to teams offer a time-effective approach, complementary to standard audit methods, to enhance the safer prescribing of high risk drugs.