High-Intensity End-of-Life Care Common Among Pediatric Cancer Patients

Children with cancer frequently receive high-intensity end-of-life care, despite advances in palliative medicine, according to the results of a retrospective study published in Journal of Clinical Oncology.

-----
Related Content
Bloodstream Infections Increase Costs, Length of Stay for Pediatric Cancer and Transplant Patients
Clinical Pathways Improve Outcomes and Decrease Costs in Pediatric Care
-----

The effect is most pronounced in hematologic cancers, results showed.

Roughly 20% of children with cancer will die from their diseases, despite treatment advances across pediatric oncology platforms. Research has shown significant end-of-life (EOL) symptom burdens in this patient population, often associated with advanced disease and high-intensity (HI) medical care.

“HI medical care at the EOL is variably defined in the literature but includes measures during the last month of life such as in-hospital deaths, intensive care unit (ICU) admissions, emergency department (ED) visits, and intravenous (IV) chemotherapy administration,” wrote Sumit Gupta, MD, PhD, staff oncologist and clinician investigator at The Hospital for Sick Children (Toronto, ON), and colleagues. “Each of these events can be appropriate in individual cases; however, in the aggregate, they may indicate a tendency to focus on anticancer treatment at the expense of palliative care.”

Using linked clinical and health services databases, Dr Gupta and colleagues identified all children diagnosed with cancer prior to age 15 years in Ontario, Canada, who subsequently died between 2000 and 2012 aged 18 years or younger.

The researchers measured HI-EOL care through a composite that included the receipt of IV chemotherapy in the 14-day period preceding death; more than one ED visit or hospitalization in the last 30 days of life; or any ICU admission in the last 30 days of life. They further examined potential predictive variables for HI-EOL receipt, including patient age (0-4 years, 5-9 years, 10-14 years, or 15-18 years), patient sex, and cancer type (hematologic malignancies, solid tumors, or central nervous system tumors). Chart abstraction was used to determine whether deaths were due to disease or treatment-related mortality (TRM), such as acute toxicities.

After excluding 85 patients who died of external causes or within 30 days of diagnosis, the study included data from 815 children (56.6% male; 31.2% aged 5-9 years). In total, 40.6% (n = 331) of the cohort received HI-EOL care as defined by the composite measure, and 43.4% (n = 354) died while hospitalized.

A multivariate analysis found that only a diagnosis of hematologic cancer served as an independent predictive variable for experiencing HI-EOL care (odds ratio [OR] = 2.5; 95% CI, 1.8-3.6; P < .001), although female patients were more likely to experience an ICU admission in the last 30 days of life (OR = 1.6; 95% CI, 1.1-2.2), be placed on mechanical ventilation (MV) in the 14-day period preceding death (OR = 1.6; 95% CI, 1.1-2.4), or die while hospitalized (OR = 1.5; 95% CI, 1.1-2).

Patients treated in the middle (2005-2008) or late (2009-2012) study period were more likely to receive HI-EOL measures than patients treated between 2000 and 2004. “Our data suggest that the most intensive measures of EOL care, such as ICU admissions and MV are, in fact, increasing over time,” Dr Gupta and colleagues wrote. “These results seem incongruous with national and international efforts over the last decade to improve pediatric palliative care.”

A total of 110 children experienced TRM. Patients with hematologic cancers were more likely to experience TRM (73.6% vs 26.1%; P < .001); alternately, children treated in the latest study period were less likely to experience TRM (21.8% vs 30.2%; P < .02). An analysis excluding children who died from TRM showed that patients with hematologic cancers continued to experience a higher risk for HI-EOL care than patients with solid malignancies (OR = 2; 95% CI, 1.5-3; P = .004).  

The researchers acknowledged study limitations, including their inability to ascertain patient and family preferences regarding EOL care, as well as their lack of knowledge regarding receipt of palliative care counselling. “Future studies should focus on exploring the relationship between high-quality palliative care and HI-EOL care, particularly in vulnerable populations,” the researchers concluded.