Variation and Drivers of Costs for Emergency Department Visits Among Children in 8 States.

OBJECTIVE: To describe variation in costs for emergency department (ED) visits among children and to assess hospital and regional factors associated with costs. METHODS: Cross-sectional study of all ED encounters among children under 18 years in 8 states from 2014 to 2018. The primary outcome was each hospital's mean inflation-adjusted ED costs. We evaluated variability in costs between hospitals and determined factors associated with costs using hierarchical linear models at the state, region, and hospital levels. Models adjusted for pediatric case mix, regional wages, Medicaid share, trauma status, critical access status, ownership, and market competitiveness. RESULTS: We analyzed 22.9 million ED encounters across 713 hospitals. The median ED-level cost was $269 (range 99-1863). There was a 5.1-fold difference in median ED-level costs between the lowest- and highest-cost regions (range 119-605). ED-level costs were associated with case mix index (+38% per 10% increase, 95% confidence interval [CI] 30 to 47); wages [+7% per 10% increase, 95% CI 5 to 9]); critical access (adjusted costs, +24%, 95% CI 13 to 35); for profit status (-20%, 95% CI -26 to -14) compared with nonprofit, lowest trauma designation (+17%, 95% CI 5 to 30); teaching hospital status (+7%, 95% CI 1 to 14); highest number of inpatient beds (+13%, 95% CI 4 to 23); and Medicaid share versus quarter (Q)1 (Q2: -12%, 95% CI -18 to -7; Q3: -13%, 95% CI -19 to -7; Q4: -11%, 95% CI -17 to -4). CONCLUSIONS: Our results suggest nonclinical factors are important drivers of pediatric health care costs.

Characteristics and Outcomes of Culture-Positive and Culture-Negative Pediatric Sepsis.

OBJECTIVES: To compare the outcomes of pediatric severe sepsis and septic shock among patients with culture-positive and culture-negative sepsis and to determine if there are differentiating markers of disease severity between these 2 populations during their initial presentation and emergency department (ED) stay. STUDY DESIGN: Retrospective cohort study of patients ≤21 years of age who presented to the ED of a single children's hospital with severe sepsis or septic shock from June 1, 2017 to June 5, 2019. RESULTS: There were 235 patients who met criteria for severe sepsis or septic shock. Of these, 139 (59.1%) had culture-negative sepsis and 96 (40.9%) had culture-positive sepsis. In the adjusted multivariable model, children with culture-negative sepsis had more intensive care unit (ICU)-free days than those with culture-positive sepsis (27.3 vs 24.1; adjusted median differences [aMD] -2.6 [-4.4, -0.8]). There were no differences in mortality or hospital-free days. On initial presentation, there were no differences in fever, hypothermia, tachycardia, tachypnea, or hypotension between the 2 groups. There were no differences in proportion of patients receiving the following interventions: intravenous (IV) antibiotics, IV fluids, vasoactive medications, CPR, intubation, or time from arrival to provision of these interventions. CONCLUSIONS: Culture-negative sepsis constitutes a substantial proportion of pediatric severe sepsis and septic shock. In this study, patients with culture-negative and culture-positive sepsis presented similarly on arrival to the ED and received similar treatments while there. Patients with culture-negative sepsis had more ICU-free days than those with culture-positive sepsis, although differences in hospital length of stay (LOS) and mortality were not observed.

Changes Made to Orders Placed by Overnight Admitting Residents on Teaching Rounds the Next Day.

OBJECTIVES: Increased focus on health care quality and safety has generally led to additional resident supervision by attending physicians. At our children's hospital, residents place orders overnight that are not explicitly reviewed by attending physicians until morning rounds. We aimed to categorize the types of orders that are added or discontinued on morning rounds the morning after admission to a resident team and to understand the rationale for these order additions and discontinuations. METHODS: We used our hospital's data warehouse to generate a report of orders placed by residents overnight that were discontinued the next morning and orders that were added on rounds the morning after admission to a resident team from July 1, 2017 to June 29, 2018. Retrospective chart review was performed on included orders to determine the reason for order changes. RESULTS: Our report identified 5927 orders; 538 were included for analysis after exclusion of duplicate orders, administrative orders, and orders for patients admitted to non-Pediatric Hospital Medicine services. The reason for order discontinuation or addition was medical decision-making (n = 357, 66.4%), change in patient trajectory (n = 151, 28.1%), and medical error (n = 30, 5.6%). Medical errors were most commonly related to medications (n = 24, 80%) and errors of omission (n = 19, 63%). CONCLUSIONS: New or discontinued orders commonly resulted from evolving patient management decisions or changes in patient trajectory; medical errors represented a small subset of identified orders. Medical errors were often errors of omission, suggesting an area to direct future safety initiatives.

Utility of specific laboratory biomarkers to predict severe sepsis in pediatric patients with SIRS.

OBJECTIVE: To identify the association between readily available laboratory biomarkers and the development of severe sepsis in children presenting to the emergency department (ED) with systemic inflammatory response syndrome (SIRS). METHODS: In this retrospective cohort study, ED patient encounters from June 2018 to June 2019 that triggered an automated sepsis alert based on SIRS criteria were analyzed. Encounters were included if the patient had any of the following laboratory tests sent within 6 h of ED arrival: C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), lactic acid, and procalcitonin. For each of the biomarkers, a receiver operating characteristic (ROC) curve was created for our primary outcome, severe sepsis within 24 h of ED disposition, and our secondary outcome, severe sepsis with a positive bacterial culture. For each ROC curve, we calculated the area under the curve (AUC) with 95% confidence intervals (95% CI) and created cutoff points to achieve 90% sensitivity and 90% sensitivity for the primary and secondary outcomes. RESULTS: During the study period, 4349/61,195 (7.1%) encounters triggered an automated sepsis alert. Of those, 1207/4349 (27.8%) had one of the candidate biomarkers sent within 6 h of ED arrival and were included in the study. A total of 100/1207 (8.3%) met criteria for severe sepsis within 24 h of arrival, and 41/100 severe sepsis cases (41%) were deemed culture-positive. Procalcitonin had the highest AUC for identifying severe sepsis [0.62 (95% CI 0.52-0.73)] while ESR and CRP had the highest AUC for culture-positive sepsis [0.68 (95% CI 0.47-0.89) and 0.67 (95% CI 0.53-0.81), respectively]. At 90% sensitivity for detecting severe sepsis, all of the biomarker threshold values fell within that laboratory test's normal range. At 90% specificity for severe sepsis, threshold values were as follows: procalcitonin 2.72 ng/mL, CRP 16.79 mg/dL, ESR 79.5 mm/h and lactic acid 3.6 mmol/L. CONCLUSION: Our data indicate that CRP, ESR, lactic acid, and procalcitonin elevations were all specific, but not sensitive, in identifying children in the ED with SIRS who go on to develop severe sepsis.

Outcomes of Patients with Sepsis in a Pediatric Emergency Department after Automated Sepsis Screening.

OBJECTIVE: To determine the effect of an automated sepsis screening tool on treatment and outcomes of severe sepsis in a pediatric emergency department (ED). STUDY DESIGN: Retrospective cohort study of encounters of patients with severe sepsis in a pediatric ED with a high volume of pediatric sepsis cases over a 2-year period. The automated sepsis screening algorithm replaced a manual screen 1 year into the study. The primary outcome was the proportion of patients treated for sepsis while in the ED. Secondary outcomes were time from ED arrival to first intravenous (IV) antibiotic and first IV fluid bolus, volume of fluid administered in the ED, 30-day mortality, intensive care unit-free days, and hospital-free days. RESULTS: In year 1 of the study, 8910 of 61 026 (14.6%) of encounters had a manual sepsis screen; 137 patients met criteria for severe sepsis. In year 2, 100% of 61 195 encounters had an automated sepsis screen and there were 136 cases of severe sepsis. There was a higher proportion of patients with severe sepsis who had an active malignancy and indwelling central venous catheter in year 2. There were no differences in the proportion of patients treated for sepsis in the ED, time to first IV antibiotic or first IV fluid bolus, fluid volume delivered in the ED, hospital-free days, intensive care unit-free days, or 30-day mortality after implementation of the automated screening algorithm. CONCLUSIONS: An automated sepsis screening algorithm introduced into an academic pediatric ED with a high volume of sepsis cases did not lead to improvements in treatment or outcomes of severe sepsis in this study.

Comparison of Manual and Automated Sepsis Screening Tools in a Pediatric Emergency Department.

OBJECTIVES: To compare the performance and test characteristics of an automated sepsis screening tool with that of a manual sepsis screen in patients presenting to a pediatric emergency department (ED). METHODS: We conducted a retrospective cohort study of encounters in a pediatric ED over a 2-year period. The automated sepsis screening algorithm replaced the manual sepsis screen 1 year into the study. A positive case was defined as development of severe sepsis or septic shock within 24 hours of disposition from the ED. We calculated the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and positive and negative likelihood ratios with 95% confidence intervals (CIs) for each. RESULTS: There were 122 221 ED encounters during the study period and 273 cases of severe sepsis. During year 1 of the study, the manual screen was performed in 8910 of 61 026 (14.6%) encounters, resulting in the following test characteristics: sensitivity of 64.6% (95% CI 54.2%-74.1%), specificity of 91.1% (95% CI 90.5%-91.7%), PPV of 7.3% (95% CI 6.3%-8.5%), and NPV of 99.6% (95% CI 99.5%-99.7%). During year 2 of the study, the automated screen was performed in 100% of 61 195 encounters, resulting in the following test characteristics: sensitivity of 84.6% (95% CI 77.4%-90.2%), specificity of 95.1% (95% CI 94.9%-95.2%), PPV of 3.7% (95% CI 3.4%-4%), and NPV of 99.9% (95% CI 99.9%-100%). CONCLUSIONS: An automated sepsis screening algorithm had higher sensitivity and specificity than a widely used manual sepsis screen and was performed on 100% of patients in the ED, ensuring continuous sepsis surveillance throughout the ED stay.

Clinical Effects of Pediatric Clonidine Exposure: A Retrospective Cohort Study at a Single Tertiary Care Center.

BACKGROUND: Pediatric clonidine ingestions frequently result in emergency department visits and admission for cardiac monitoring. Detailed information on the clinical course and specifically time of vital sign abnormalities of these patients is lacking. OBJECTIVE: The objective of this study was to provide descriptive analysis of the rates and times to vital sign abnormalities, treatment, disposition, and outcomes in a single-center cohort of pediatric patients with report of clonidine poisoning. METHODS: We performed a retrospective cohort study of patients younger than 21 years who presented to a large, urban, tertiary care center with a report of single substance clonidine exposure between January 2004 and November 2017. Patients were dichotomized into younger (≤9 years or younger) and older (10-21 years) groups based on the expected physiologic and psychologic differences between older and younger children. RESULTS: Eighty-eight patients met our inclusion criteria. Younger patients (≤9 years or younger; n = 47) were more likely to be exposed to someone else's medication (53%) and older patients (10-21 years; n = 41) overwhelmingly (85%) were exposed to their own medication. Thirty-nine (45%) became bradycardic, 27 (32%) became bradypneic, and 38 (44%) became hypotensive. Eighty percent of patients had depressed mental status. Thirty-three (38%) patients received at least one dose of naloxone (median 0.07 mg/kg; interquartile range 0.03-0.11 mg/kg). Of those who received naloxone, 50% had a documented clinical response. CONCLUSIONS: In this study of patients at a pediatric tertiary referral center, pediatric patients with report of clonidine exposures were likely to exhibit altered mental status and frequently develop vital sign abnormalities. Naloxone exhibited some effectiveness; given its wide safety margin, high-dose naloxone should be used in critically poisoned non-opioid-dependent patients. Because adolescents are much more likely to ingest their own clonidine medication, counseling with parents and other caregivers regarding safe medication storage is paramount.

Applying Cognitive Learning Strategies to Enhance Learning and Retention in Clinical Teaching Settings.

Introduction: Cognitive learning strategies are strategies that improve a learner's ability to process information more deeply, transfer and apply information to new situations, and result in enhanced and better-retained learning. Methods: We developed an interactive workshop for a national conference of pediatric educators to teach five cognitive learning strategies. The specific strategies were (1) spaced retrieval practice, (2) interleaving, (3) elaboration, (4) generation, and (5) reflection. Each strategy was taught using an active learning exercise. We evaluated the effectiveness of the workshop through a commitment-to-change exercise in which we asked participants to commit to making a change in their teaching as it related to the workshop and then queried them 6 weeks later about their implementation successes and barriers. Results: Of the 161 participants registered for the workshop, 52 completed the voluntary workshop evaluation. All 52 participants committed to making a change in their teaching as a result of the workshop. Of those 52 participants, 24 completed the 6-week follow-up survey. Eighty-two percent of those respondents (n = 18) reported implementing a change based on the workshop, with 77% of respondents implementing a change that they had committed to directly after the workshop and 55% implementing a change that they had not originally committed to at the end of the workshop. Discussion: This workshop successfully led to behavioral change in the teaching of cognitive learning strategies. We anticipate that this will lead to improved learning among the trainees whom participants teach.

Identification and characterization of T reg-like cells in zebrafish.

Regulatory T (T reg) cells are a specialized sublineage of T lymphocytes that suppress autoreactive T cells. Functional studies of T reg cells in vitro have defined multiple suppression mechanisms, and studies of T reg-deficient humans and mice have made clear the important role that these cells play in preventing autoimmunity. However, many questions remain about how T reg cells act in vivo. Specifically, it is not clear which suppression mechanisms are most important, where T reg cells act, and how they get there. To begin to address these issues, we sought to identify T reg cells in zebrafish, a model system that provides unparalleled advantages in live-cell imaging and high-throughput genetic analyses. Using a FOXP3 orthologue as a marker, we identified CD4-enriched, mature T lymphocytes with properties of T reg cells. Zebrafish mutant for foxp3a displayed excess T lymphocytes, splenomegaly, and a profound inflammatory phenotype that was suppressed by genetic ablation of lymphocytes. This study identifies T reg-like cells in zebrafish, providing both a model to study the normal functions of these cells in vivo and mutants to explore the consequences of their loss.