End-of-life decisions and palliative care in a children's hospital

Objective: To investigate clinical and demographic factors affecting the nature of end-of-life decisions and pediatric palliative care. Design: Charts of 236 expired children were retrospectively reviewed for presence of end-of-life care (EOLC) discussions and spiritual support, the nature of EOLC decisions, and the degree of opioid analgesics (OA) and sedatives (SDT) administration. Results: Approximately 60% of patients had EOLC discussion, of whom 87.4% obtained an EOLC decision, mostly opting for withholding therapy (68.8%). Presence of EOLC discussion was associated with a longer hospital stay (univariate analyses: odds ratio [OR] = 1.9; p < 0.029), higher number of failed organs (OR = 2.5; p < 0.003), chronic illnesses (OR = 2.4; p < 0.002), spiritual support (OR = 1.8; p < 0.028) and respiratory diseases (OR = 3.1; p < 0.0006). Younger patients and those with higher number of failed organs were more likely to have withdrawal of therapy (OR = 10.9 and 6.0; p < 0.0001 and <0.002, respectively), whereas patients with chronic illness opted for withholding of therapy (OR 3.1; p < 0.006). Spiritual support was associated with higher use of both OA and SDT (OR 1.9 and 2.3; p < 0.014 and p < 0.005, respectively). Younger patients received less OA and SDT (OR = 0.2 and 0.4, respectively; p < 0.0001). Multivariate analyses showed that EOLC discussion is associated with higher use of OA and SDT (OR = 4.4 and 4.2; p < 0.00001 and p < 0.0001, respectively), whereas younger age is associated with withdrawal of therapy (OR = 8.3; p < 0.0005) and lower use of SDT (OR = 0.23; p < 0.0001). Conclusions: Patterns of care at the end of life vary in children with differing clinical and demographic characteristics. Because EOLC discussions are associated with greater focus on palliative care, strategies to enhance EOLC communications for pediatric patients should be further evaluated.