ISQUA17-2435QUALITATIVE EVALUATION OF THE IMPLEMENTATION OF A NATIONALLY USED MEDICATION SAFETY DATA COLLECTION TOOL IN ENGLAND

Abstract

Objectives: Approximately 10% of patients are harmed by healthcare, and 15% of these harms are medication-related (1). To improve medication safety it must be measured, however research studies are time-consuming and expensive, and voluntary reporting cannot be used for measurement. Therefore, a large multidisciplinary collaborative, within England's NHS, developed the Medication Safety Thermometer to collect monthly medication safety data, that could be used for improvement (2). This study aimed to understand its implementation into routine practice and explore users' views and experiences. Methods: Fifteen in-depth interviews were conducted either faceto-face or via telephone, with pharmacists, nurses, pharmacy technicians, pre-registration pharmacists and audit clerks. Participants were selected by purposive convenience sampling from primary and secondary care settings. Interviews were transcribed verbatim and thematic analysis was based on the four constructs of Normalisation Process Theory: Coherence (if participants understand the purpose of the tool); Cognitive participation (engagement with the tool and implementation); Collective action (work undertaken that drives the intervention forwards); Reflexive monitoring (assessment of the impact of the tool). Relevant ethics and governance approvals were obtained. Results: Coherence existed in secondary care, where users understood that the purpose of the tool was to measure medication safety and related improvement. However, other uses were reported, such as pinpointing individuals displaying poor practice. Confusion about the purpose existed in primary care, despite further training, suggesting the tool to be unsuitable in primary care. Cognitive participation depended on ownership of medication safety, as staff more responsible for harm from medication demonstrated increased engagement with further improvement work and related meetings. However, participants were often “one-man bands” with subjective support from senior levels, which had a knock-on effect to frontline teams, causing lack of engagement overall. Collective action work was undertaken to drive scale-up of the use of the tool, for example, through securing additional funding, despite uncertainty about how to use the data. Successful improvement was often at ward-level and not recognised wider within organisations. Reflexive monitoring showed mixed feedback from healthcare staff about the value of the tool, often due to a perceived lack of “capacity”. However, participants displayed interest in learning how to use their data, and unexpected uses of data were reported, such as judicial use, use for freedom of information requests, and to change guidelines for improving patient care. Conclusion: The tool can aid improvement of medication safety, however, many organisations have not seen improvement, often because data is not analysed, understood or used. The national normalisation of the tool has allowed more rapid shared learning, however, a considerable number of difficulties exist, particularly in primary care settings, where a separate tool may be required. Issues with using the data for improvement must be addressed by organisations, before scaling up.