Administering injectable medications prescribed in the anticipation of the end of life in the community: A mixed-methods observational study

Abstract Background The prescription of injectable anticipatory medications ahead of possible need for last-days-of-life symptom relief is established community practice internationally. Healthcare teams and policy makers view anticipatory medication as having a key role in optimising effective and timely symptom control. However, how these medications are subsequently administered (used) is unclear and warrants detailed investigation to inform interdisciplinary practice and guidance. Objective To identify the frequency, timing and recorded circumstances of the administration of injectable end-of-life anticipatory medications prescribed for patients living at home and in residential care. Design A retrospective mixed-methods observational study using general practitioner (family doctor) and community nursing held clinical records. Setting(s) Community-based care in two English counties. Participants 167 deceased adult patients (aged 18 +) registered with eleven general practitioner practices and two associated community nursing services. These were patients prescribed anticipatory medications, identified from the 30 most recent deaths per practice. Patients died between 1 March 2017 and 25 September 2019, from any cause except trauma, sudden death or suicide. Methods Patient characteristics, anticipatory medication discussions, recorded administration contexts and decision-making, medication details, recorded symptom control and comfort at death were collected from clinical records. Data analysis combined quantitative and qualitative analyses in a mixed methods approach. Results Anticipatory medications were administered to 59.9 % (100/167) patients, commenced between 0 and 586 days before death (median 3 days). Their usage was similar for patients who died from cancer and non-cancer conditions. Anticipatory medications were almost universally started and titrated by visiting nurses. Eleven patients had medications started between 59 days and 586 days before death for recorded reversible non-end-of-life care conditions. Only 5 % (5/100) of patient records contained detailed accounts of patient participation in decisions to start medications: four were recorded as being reluctant to commence medications but agreed to trial injections to relieve symptoms. Crucially, there was recurrent under-recording of the effectiveness of injectable medications and patient comfort. Conclusions Prescribed medications were commonly administered by visiting community nurses to help manage last-days-of-life symptoms. However, patient records infrequently referred to the effectiveness of administered medication and perceived patient comfort. Most recorded references to patient and family preferences for involvement in anticipatory medication decision-making and their experiences of care were brief and perfunctory. More detailed information should be routinely recorded in clinical records to enable assessment of the appropriate and effective use of anticipatory medicines and how inter-professional collaboration and services could be developed to provide adequate twenty-four-hour cover.