Variation in Pediatric Emergency Department Care of Sickle Cell Disease and Fever

Abstract

Objectives: The objective was to study the variation in pediatric emergency department (PED) practice patterns for evaluation and management of children with sickle cell disease (SCD) and fever in U.S. children's hospitals. Methods:A cross-sectional study of visits by children 3 months to 18 years of age with SCD and fever evaluated in 36 U.S. children's hospital PEDs within the 2010 Pediatric Health Information System database. The main outcome measures were the proportions of SCD visits that received evaluation (laboratory testing and chest radiographs [CXRs]) and treatment (parenteral administration of antibiotics) and were admitted for fever.Results: Of the 4,853 PED visits for SCD and fever, 91.7% had complete blood counts (CBCs), 93.8% had reticulocyte counts, 93% had blood cultures obtained, 68.5% had CXRs, and 91.7% received antibiotics. Most (81.4%) patients received the recommended National Heart, Lung and Blood Institute evaluation (CBC, reticulocyte count, and blood culture) and treatment (parenteral antibiotics). In multivariate regression modeling controlling for hospital-and patient-level effects, age groups ≥1 to <5 years (odds ratio [OR] = 0.32, 95% confidence interval [CI] = 0.25 to 0.40) and ≥5 to <13 years (OR = 0.40, 95% CI = 0.32 to 0.50), and those visits that did not have CXRs had lower odds of hospital admission. After adjusting for age, payor status, receipt of laboratory testing, antibiotics, and CXRs, admission rates varied by sevenfold across U.S. children's hospitals (p < 0.001).Conclusions: Standardization of practice exists across children's hospitals regarding obtaining laboratory studies and administering antibiotics for patients with SCD and fever. However, admission rates vary significantly. Evaluating the causes and consequences of such significant variation needs further exploration to improve the quality of care for patients with SCD.ACADEMIC EMERGENCY MEDICINE 2015;22:423-430

Which Febrile Children With Sickle Cell Disease Need a Chest X‐Ray?

“…Our data suggest that there is strong evidence to add chest pain to these high‐risk criteria, while a high WBC count and history of ACS can further identify patients at risk for ACS. Given the previously documented variability in practice, it is clear that data‐driven evidence is needed to further determine which patients require a CXR . While neither the NHLBI guideline nor our model identified all instances of ACS, our model provides additional evidence of high‐risk groups that might be more likely to benefit from obtaining a CXR.…”

“…The variables abstracted from the patient chart and these composites were compared between those with ACS and those without ACS using a chi‐square test for categorical variables (and Fisher's exact test where indicated due to small sample sizes) and a Wilcoxon rank‐sum test for nonparametric continuous variables with a Type I error rate of <0.05. Similar to previous studies, all visits were treated as independent occurrences for analysis . We performed binary recursive partitioning using classification and regression tree (CART) analysis to determine predictive factors for a diagnosis of ACS (CART PRO 6.0, Salford Systems) .…”

“…Due to these conflicting recommendations in the literature, significant variation exists in clinical practice. A recent analysis of the 2010 Pediatric Health Information System demonstrated that the percentage of febrile children with SCD who received a CXR in the emergency department (ED) ranged from less than 40% to more than 90% across 36 tertiary care children's hospitals in the United States . This variation in care suggests that there is an absence of data‐driven evidence supporting the clinical guidelines for CXR utilization in the ED.…”

Variation in hospital admission of sickle cell patients from the emergency department using the Pediatric Health Information System

“…In particular, we saw that visit‐level factors such as weekday versus weekend did not have a significant effect on fever admission variation. Similar to the admission variation seen in pain, other nonvisit‐level factors are more likely to be influencing this variation, as mentioned in previous studies . For example, social and environmental factors such as distance from a hospital with experience caring for patients with SCD, reliable transportation, and compliance may all play a role in determining disposition.…”

“…Fever, the next most common reason for a child with SCD to visit the ED showed much greater variation than overall or pain in admission percentages between hospitals, despite having established recommendations for hospitalization of children with fever by the NHLBI and the American Academy of Pediatrics (AAP) Section of Hematology/Oncology . Ellison et al using 2010 PHIS data demonstrated significant practice variation in ED care of fever in children with SCD even when controlling for patient‐level characteristics . A recent study also showed significant provider and hospital practice variation in SCD fever management in a retrospective review of three pediatric EDs .…”