Transition Points for the Routine Integration of Palliative Care in Patients With Advanced Cancer

Anna Collins, Vijaya Sundararajan, Jodie Burchell, Jeremy Millar, Sue Anne McLachlan, Meinir Krishnasamy, Brian H. Le, Linda Mileshkin, Peter Hudson, Jennifer Philip

Research output: Contribution to journalArticleResearchpeer-review

33 Citations (Scopus)


Context: Increasing emphases are being placed on early integration of palliative care for patients with advanced cancers, yet barriers to implementation in clinical practice remain. Criteria to standardize referral have been endorsed, but their application is yet to be tested at the population level. Objectives: This study sought to establish the need for standardized referral by examining current end-of-life care outcomes of decedents with cancer and define transition points within a cancer illness course, which are associated with poor prognosis, whereby palliative care should be routinely introduced to augment clinician-based decision making. Methods: Population cohort study of admitted patients with advanced cancer diagnosed with non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), prostate or breast cancer between 2000 and 2010 in Victoria, Australia, identified from routinely collected, linked, hospital discharge, emergency department, and death registration data. Descriptive statistics described quality indicators for end-of-life care outcomes for decedents. Kaplan-Meier analyses were used to test the predefined transition point that mostly accurately predicted survival of six months or lesser. Results: About 46,700 cases (56% females) were admitted with metastatic NSCLC (n = 14,759; 31.6%), SCLC (n = 2932; 6%), prostate (n = 9445; 20.2%), and breast cancer (n = 19,564; 41.9%). Of the 29,680 decedents, most (80%) died in hospital, had suboptimal end-of-life care outcomes (83%), and 59% received a palliative approach to care, a median of 27 days before death. Transition points in the cancer illness course of all cases were identified as first admission with any metastatic disease (NSCLC: 3.8 months [interquartile range {IQR} 1.1, 16.0]; n = 14,666; and SCLC: 4.2 months [IQR 1.0, 10.6]; n = 2914); first multiday admission with any metastatic disease (prostate: 6.0 months [IQR 1.3, 26.4]; n = 7174); and first multiday admission with at least one visceral metastatic site (breast: 6.0 months [IQR 1.2, 29.8]; n = 7120). Conclusion: Despite calls for integrated palliative care, this occurs late or not at all for many patients with cancer. Our findings demonstrate the application of targeted cancer-specific transition points to trigger integration of palliative care as a standard part of quality oncological care and augment clinician-based referral in routine clinical practice.

Original languageEnglish
Pages (from-to)185-194
Number of pages10
JournalJournal of Pain and Symptom Management
Issue number2
Publication statusPublished - 1 Aug 2018
Externally publishedYes


  • cancer
  • health services research
  • integration
  • Palliative care

Cite this