Background Continuous Subcutaneous Infusion (CSCI) prescription practice is an area of high clinical importance across all palliative care settings. Frequent evaluation of CSCI prescribing is essential to ensure appropriate and justified prescriptions are made.

Methods Across three palliative care settings, a hospice and two hospitals (A and B) within a trust, prescriptions of anticipatory (PRN) medications, opioids and continuous subcutaneous infusions (CSCI) of 60 patients who died in April 2019 were reviewed retrospectively.

Results In the hospice anticipatory CSCIs were prescribed for all patients expected to deteriorate with a range given for each medication. This practice was not observed in the hospital setting where it is not recommended to prescribe anticipatory CSCIs. 70% of the hospice patients had a CSCI in use at the end of life compared to 25% of the hospital patients. All hospital CSCIs contained midazolam (vs 71% in the hospice), however the total number of patients receiving midazolam CSCI was higher in the hospice. The numbers of patients dying with a CSCI containing an anti-secretory agent was low in all settings (20% at hospice, 5% A, 10% B). No hospital patients were receiving anti-emetic medication CSCI compared to 57% of hospice patients. In the hospice, patients died with between 0–5 medications administered CSCI, compared to 0–3 in the hospital setting.

Conclusion There were variations in the prescription of CSCIs for patients dying in an acute hospital compared with a hospice. Not only will this likely reflect the varying complexity of patients within the different settings but also staff confidence and competence in recognising and managing symptoms of patients who are at the end of their lives. Patient comfort at the end of life was not evaluated in this audit, review of clinical notes would guide further work.