Introduction and aims The Edmonton Symptom Assessment System (ESAS) has been validated for patient-completed, caregiver-assisted and caregiver-rated scoring of patients' symptoms, enabling symptom recording to continue as patients deteriorate. However, it has not been used to assess symptoms at the end of life specifically. We aimed to assess the utility of ESAS in scoring symptoms in the dying patient.

Methods 70 patients were recruited; 40 before and 30 following the introduction of a simple end-of-life (EOL) care tool. Nursing staff were asked to complete 12 hourly ESAS scores from the time the ‘diagnosis of dying’ was made until death.

Results 249 of a required 418 (59%) ESAS were completed in the pre- and 140 of 302 (46%) in the post-implementation group. 29/40 (72%) and 25/30 (83%) patients had an ESAS completed in the last 24 h in pre- and post- groups respectively. Comparison of scores for EOL symptoms before and after the introduction of the tool showed less symptom burden in the group where the tool was used. The differences in scores with 95% CIs (0–100 scales) were: pain 12.4 (−2.3 to 27.1); shortness of breath 11.7 (−6.2 to 29.6); anxiety 7.3 (−7.8 to 22.5); nausea 5.9 (−2.6 to 14.5) and secretions 12.3 (−6.7 to 31.3).

Conclusion The amount of missing data suggests 12-hourly ESAS scoring was onerous for ward staff and would not therefore capture longitudinal symptom changes. However, ESAS was useful for assessing symptoms in the last 24 h of life, enabling comparison between the two groups.