Palliative care allocation among critically ill children is highly variable in the United States

Palliative care allocation among critically ill children is highly variable in the United States

O'Keefe S; Maddux A; Bennett K; Youngwerth J; Czaja A

Critical Care Medicine

2019

artificial ventilation; child; cohort analysis; conference abstract; controlled study; critically ill patient; female; hospitalization; household income; human; infant; insurance; length of stay; major clinical study; male; morbidity; mortality; multicenter study; neonatal intensive care unit; newborn; palliative therapy; race; resource allocation; retrospective study; suburban area; United States

Learning Objectives: Palliative Care (PC) is a scarce resource. Little is known about its allocation among critically ill children. Previously proposed criteria may help identify children who may benefit from PC. Method(s): This is a retrospective cohort study of patients aged <40 years requiring ICU admission (excluding neonatal ICU) among 51 children's hospitals from 2010-2017. Hospitalizations were categorized into 3 mutually exclusive groups: 1. PICU based criteria adapted from previously published criteria (PC-ICU) 2. Additional Criteria (AC = chronic complex condition not in PCICU) 3. No Criteria (NC). Characteristics, outcomes and PC use (based on ICD administrative charges) were compared using chi2 or kruskal-wallis. We also explored PC patterns over time. Result(s): 93499 subjects with 114510 hospitalizations had >= 1 ICU admission. 15% subjects had >1 hospitalization. Median age at admission was 19 months (IQR 2-117). 56%, 39% and 5% of hospitalizations met PC-ICU, AC and NC criteria respectively. PC-ICU admissions had higher severity of illness, number of procedures, need for mechanical ventilation, CPR, mechanical ventilation days, length of stay, charges and mortality (p<=0.001). PC consult was present in 4.5% of hospitalizations, higher among PC-ICU than AC and NC groups (5.8%, 2.9% and 0.3% respectively, p<=0.001). Median age of children receiving PC was younger in PC-ICU (34 months v 52 and 59 months in AC and NC groups respectively). PC use was slightly higher among patients living in urban/suburban areas (4.5% v 4%, p<0.004), with governmentbased insurance (4.7% v 4%, p<=0.0001) and of non-white race (4.6% v 4.3%, p<0.04). Household income did not differ between PC and no-PC (p= 0.6). PC use increased steadily from 2010-2017 from 1% to 6.7%, with greatest change observed in PC-ICU group (1.4% to 9%), followed by the AC group (1% to 4.7%) and with little change in NC group (0 to 0.5%). PC use varied among institutions ranging from 0-44% in PC-ICU, 0-12% in AC and 0-2% in NC groups respectively. Conclusion(s): National PC use remain low but has increased over time, especially among those meeting PC-ICU criteria who are at high risk of morbidity and mortality. However, PC use varies substantially across ICUs and across certain demographic groups. Future research to understand the impact of PC for critically ill children could help optimize resource allocation.

Article information provided for research and reference use only. PedPalASCNET does not hold any rights over the resource listed here. All rights are retained by the journal listed under publisher and/or the creator(s).

December 2019 List