In 2010, Temel et al demonstrated early referral to palliative care (PC) improved quality of life and overall survival (OS) with less aggressive end of life (EOL) care in patients with newly diagnosed metastatic NSCLC. We examined the timing of PC referral for patients with metastatic NSCLC treated in Southwest Sydney and its impact on OS and aggressiveness of EOL care.
We retrospectively reviewed electronic medical records for patients with newly diagnosed metastatic NSCLC who were treated in Southwest Sydney, Australia. Patients were analysed pre- (1/1/2008-31/12/2009) and post-Temel study (1/1/2011-31/12/2012), respectively, to identify changes in referral patterns. Patients diagnosed in 2010 were excluded (same year as Temel's publication). Early referral to PC was defined as within 8 weeks of diagnosis. Patients receiving aggressive EOL care were defined as those receiving chemotherapy within 14 days of death and/or death in a hospital. Cox regression was used to analyse OS in the early and late groups.
266 patients were included: 134 from 2008-2009 and 132 from 2011-2012. In total, 249 (94%) were referred to PC. 156 (59%) were referred early and 110 (41%) were referred late or not referred. There was no change in PC referral patterns pre- or post- Temel study. Median OS was shorter in the early group (3.1 months vs 9.8 months; HR = 2.51, 95% CI = 1.93- 2.97; p < 0.0001). More patients in the late group were ECOG performance status 0-1(85% vs 46%) and received ≥ 2 lines of systemic therapy (70% vs 31%). Patients in the late group were also less likely to die in hospital (26% vs 35%) but more likely to receive chemotherapy within 14 days of death (5% vs 1%). However, the overall proportions of patients receiving aggressive EOL care were similar (early 35% vs late 32%).
Among patients with metastatic NSCLC treated in Southwest Sydney, early referral to PC was not associated with longer OS or less aggressive EOL care. This likely reflects that referrals to palliative care services are tailored to patient clinical factors and performance status. More resources into PC services are required to reduce aggressive EOL care and deaths in acute hospitals.
Clinical trial identification
All authors have declared no conflicts of interest.