Background Benefits of early palliative care (PC) referral for oncology patients are well-documented. The development of PC referral ‘Triggers’ may help identify which patients should be referred. A number of ‘Triggers’ have been proposed, but are currently not commonly used in clinical practice. This project aimed to evaluate a number of published referral Triggers through a retrospective note review of PC team involvement with patients who died during a non-elective hospital admission.

Method All patients admitted to the oncology hospital non-electively during one year who died during that admission were selected, excluding patients admitted to CCU (N=159). The timing of referral to PC, and whether patients met any published PC referral triggers before and during their terminal admission was assessed.

Results 88% (140/159) of patients would have been eligible to ‘trigger’ a PC referral during their terminal admission, no matter which Triggers tool was used. Information relating to disease and symptom-related ‘Triggers’ are most commonly documented in medical notes. Psychosocial factors were less well-documented.

46.5% of patients were known to PC prior to their final admission. All but 3 of the remaining patients were referred during admission, with a median time (range) between admission and referral of 4 days (0–58 days), and between referral and death of 9 days (0–90 days).

Conclusions Currently, many patients are not known to palliative care until their last hospital admission, and therefore until their last days of life. In this patient cohort, a palliative referral Triggers Tool would have proactively identified most or all patients who died during their non-elective hospital admission. The inclusion of disease and symptom-related parameters in a PC referral Trigger tool appears feasible as these are often documented in oncology records, whereas oncologists may not routinely identify psychosocial factors.