Background Early integration of oncology and palliative care has been recommended as a major strategy to improve patient outcomes at the end of life. A standardised palliative care pathway(PCP) may be useful to support such integration. We studied whether implementation of a PCP had an effect on place of death, hospitalisations, and documentation of advance care planning.

Methods In a single centre pre- and post-intervention study, data were collected for adult patients with cancer who had been treated at inpatient or outpatient clinics and died between February 2014 and February 2015 (pre-PCP period) or between November 2015 and November 2016 (post-PCP period).

Results 852 patients were included, 426 in each study period. The PCP was used for 237 patients (56%) after implementation. The PCP was started 33 days (IQR 12–74days) before death. 76% and 77% of the patients died outside the hospital in the pre- and post-PCP period, respectively (p=0.57). No differences were found for hospitalisations in the last 3 months of life. In the pre-PCP period, bad-news conversations (75% and 62% respectively, p<0.001) and preferred place of death (47% and 32% respectively, p< 0.001) were documented more often, whereas a DNR code was more often documented during the post-PCP period (79% and 89% respectively, p<0.001).

Conclusions A high percentage of patients died outside the hospital in both groups, not increased by implementation of the PCP. Start of a PCP in the last two months of life may be too late to facilitate ACP.