Predictors of and trends in high-intensity end-of-life care among children with cancer: A population-based study using health services data

Abstract

Background: Children with cancer often receive high intensity (HI) medical care at the end-of-life (EOL), possibly increasing already high symptom burdens. Previous EOL studies are limited to single centers or lack detailed clinical data. We determined predictors of and trends in HI-EOL care by linking population-based clinical and health services databases. Methods: A retrospective cohort of all children with cancer who died 2000-2012 in Ontario, Canada was assembled using the provincial childhood cancer registry POGONIS, and linked to population-based healthcare data capturing all inpatient, outpatient, emergency (ER) and home care visits. The primary composite outcome, HI-EOL care, comprised any of: IV chemotherapy LT 14 days from death; GT 1 ER visit, GT 1 hospitalization, or any intensive care unit (ICU) admission LT 30 days from death. Secondary outcomes included individual indicators and measures of highest intensity EOL care [e.g. mechanical ventilation (MV) LT 14 days of death]. We determined study outcome predictors with regression models. In sensitivity analyses, we excluded cases of treatment related mortality (TRM). Results: 815 patients met inclusion criteria. 331 (40.6%) experienced HI-EOL care, with patients with hematologic malignancies at significantly higher risk [odds ratio (OR) 2.5, 95th confidence interval 1.8-3.6; p LT 0.001)]. Patients with hematologic cancers, living close to a treatment center, and who died in 2005-2008 and 2009-2012 were significantly more likely to experience indicators of highest intensity care (ICU, MV; ORs ranging from 2.2-4.9). Excluding cases of TRM did not substantively change the results. Conclusions: Ontario children with cancer continue to experience HI care at the EOL. Patients with hematologic malignancies are at highest risk even when excluding those with TRM. Worryingly, rates of highest intensity EOL care have increased over time despite increased palliative care access. Linkage of health services and clinical data offers a method of monitoring population trends in EOL care and identifying high-risk populations targetable by future interventions.