Abstract
Objective:
To assess whether children with sepsis in general Emergency Departments (EDs) with low versus high pediatric sepsis volumes are less likely to receive care concordant with Surviving Sepsis pediatric guidelines.
Methods:
A retrospective cohort study of children <18 years with sepsis presenting to 29 general EDs. ED and hospital data were abstracted from the medical record of two large health care systems, including all hospitals to which children were transferred. Pediatric sepsis yearly case volume was categorized as low (0-10), medium (>10 - 100) or high (>100). Guideline concordant care was defined as intravenous antibiotics within 3 hours, intravenous fluid bolus within 3 hours, lactate measured.
Results:
Between 1/1/2015 and 9/30/2021, 1527 ED visits were included. 301 (19%) occurred in 25 low-volume EDs, 466 (31%) in 3 medium-volume EDs, 760 (50%) in a high-volume ED. Care was concordant in 627 (41.1%) visits. The relative risk of concordant care did not differ between low versus high pediatric sepsis volume EDs (RR=0.95, 95% CI: 0.81-1.12), or medium- vs. high-volume EDs (RR=0.91, 95% CI 0.79-1.05). Care concordance improved from 23.1% in 2015 to 52.8% in 2021.
Conclusions:
Children had equal risk of receiving guideline-concordant sepsis care at low versus high pediatric sepsis volume general EDs. Most children in this study did not receive concordant care, although this improved over time. Pediatric sepsis volume has been postulated to drive care quality, however identifying other drivers and leveraging existing improvement is important for the majority of US children first treated in general EDs.
Introduction
Background:
Pediatric sepsis is a leading cause of childhood mortality, in which early, evidence-based emergency care can prevent death and improve outcomes.1 In the United States, >70,000 children are hospitalized with sepsis yearly with 5-10% case fatality rates, and 35% with long-term morbidities.2-4 Recent research and quality improvement have improved care and outcomes of pediatric sepsis in high-volume, pediatric hospitals.5 However, the quality of care for children who initially present to general Emergency Departments (EDs) has not been well-described. The few studies addressing pediatric sepsis in general EDs suggest that guideline-concordant care is lower and mortality is higher in EDs with low pediatric sepsis volume.6,7 These studies have been limited by non-standardized case definitions, lack of data about ED treatments and timing, and lack of linking data in patients who transfer between hospitals, which is common for critically ill children who first present to a hospital without pediatric intensive care. Most studies of children with sepsis treated in general hospitals have used administrative data from the hospital of admission, limiting insight into the initial care of children in general EDs who were transferred.7-9
Importance:
The majority of US children receive emergency care in general EDs.10 At any individual general ED, pediatric sepsis is a high-risk, low-frequency event, but in aggregate, approximately 65% of children with sepsis first receive emergency care in general EDs.7 Global efforts to improve pediatric sepsis outcomes require more data to understand the factors associated with care quality in EDs that infrequently treat pediatric sepsis. The National Pediatric Readiness Project identified implementation of pediatric guidelines in general EDs as a priority area facing ongoing barriers; sepsis is an important exemplar condition with pediatric-specific guidelines in which there is little data about guideline-concordant care in general EDs.11 Improved understanding of the numbers of children with sepsis initially treated in low pediatric volume EDs, and their care and outcomes would identify factors that improve care, and conversely, focus efforts on EDs that most need support in pediatric guideline implementation.
Goals of this Investigation:
We sought to assess pediatric sepsis care quality in general EDs, addressing prior data limitations by applying a standardized sepsis definition to ascertain cases of pediatric sepsis directly from the Electronic Health Record (EHR) in general EDs, and assessing subsequent hospital course. In particular, we sought to understand the relationship between pediatric sepsis volume at the presenting ED and care concordant with Surviving Sepsis pediatric guidelines and outcomes.
Methods
Study Design and setting:
This retrospective cohort study was set in 29 Colorado and Wyoming general EDs, with combined annual total encounters of approximately 487,000, of which approximately 53,000 were children <18 years old. The study period was January 1, 2015 through September 30, 2021, beginning in 2015 when the Epic (Epic Systems Corporation, Verona, WI) EHR in use was linked into a unified data instance. This facilitated patient-level data extraction from all EDs for this study, but the clinician-facing EHR had differences across sites. The EDs were part of a large, university-affiliated extended health system, which included five ED provider practice groups. The EDs were spread over 200 miles, included trauma (levels 1, 2, 3, 4) and non-trauma centers, hospital based and free-standing EDs, some with pediatric inpatient beds. All had <25% of yearly encounters by children <18 years and all ED providers treated all ages of patients. General/adult sepsis quality improvement occurred during this study period, but not dedicated pediatric sepsis quality improvement. Hospitalization data was obtained from the general health care system and a separate children’s hospital system, which included 2 tertiary hospitals with Pediatric Intensive Care Units (ICUs), and 2 secondary hospitals without PICUs. Colorado and Wyoming did not have laws relating to pediatric sepsis. The Colorado Multiple Institutional Review Board determined this study to be exempt research. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.
Objective:
To assess whether ED-level yearly pediatric sepsis volume was associated with care concordant with the Surviving Sepsis Campaign pediatric care bundle in pediatric sepsis encounters at general EDs.
Population:
Inclusion criteria were presentation to an included general ED, age 30 days-18 years, inpatient hospitalization within 72 hours of presentation, and any one of three criteria: 1) ED diagnostic code of sepsis (R65.20 or R65.21), 2) ED sepsis order set used, 3) blood culture in the first 12 hours with ED intravenous fluid bolus and ED intravenous antibiotic administration. These criteria were modified from the Improving Pediatric Sepsis Outcomes (IPSO) definition that was developed for quality improvement to identify cases in EHR data.12 IPSO criteria were designed to identify patients receiving ED and inpatient treatment at the same hospital and require at least two fluid boluses; this study modified criteria to include encounters with one bolus in the presenting ED because many expeditious transfers were expected. A subgroup analysis of patients meeting original IPSO criteria was performed.12
Exclusion criteria were incomplete encounters, transfers out of the two hospital systems (transfers among the 29 general EDs and hospitals and 4 children’s hospitals were included). We excluded encounters identified on chart review to have an alternate, non-sepsis ED condition. Encounters with an ED diagnosis of acute appendicitis were added as an exclusion after initial chart reviews demonstrated that this was a common reason patients met EHR inclusion (transfer to a children’s hospital, intravenous antibiotics and fluids) without evidence of organ dysfunction or clinical concern for sepsis from the treating team.
Data Collection:
Encounters meeting inclusion criteria were initially identified from the general hospital system’s EHR. ED and hospital course data were electronically extracted from the general hospital’s EHR into a secure datafile. Additional standardized chart review was performed in all encounters that included any of the following: transfer to a children’s hospital, vasopressors, airway interventions or ventilation.
Chart review data were collected and managed using REDCap electronic data capture (Appendix 1).13,14 Chart review was used to ascertain vasoactive and ventilator duration. In patients transferred to the children’s system, chart review was used to ascertain the outcomes of hospitalization, length of stay, in-hospital mortality, which were not in the electronic extraction from the general hospital system. Three physician investigators (SB, EM, HS) conducted chart review, without knowledge of the predictor. Additionally, the reviewers identified and excluded encounters identified by electronic data capture in whom chart review demonstrated that a primary non-sepsis etiology was known and treated in the ED (Appendix 2). To ensure inter-rater reliability, reviewers classified ten charts independently, achieving a kappa of 0.737. These charts were reviewed, discussed and criteria for classification were reconciled. Subsequent independent review of ten charts led to perfect agreement (kappa =1.0). Ongoing audit of 10% of charts which were double-coded yielded a kappa >0.8 for all outcomes. REDCap and EHR-derived datasets were then merged in SAS 9.4 (SAS, Cary, NC) for analysis.
Where available, ED pediatric readiness scores and nationally-benchmarked quartiles were obtained from the 2020 National Pediatric Readiness Project assessment, a weighted score based on answers to 55 questions assessed in United States EDs since 2013.11,15 Readiness scores were not assessed for free-standing EDs.
Variables and Outcomes:
The primary predictor was annual ED volume of pediatric sepsis, assessed as a continuous variable. The primary outcome was concordance with three elements of initial sepsis resuscitation in pediatric Surviving Sepsis Campaign (SSC) guidelines: intravenous antibiotics within 3 hours, intravenous fluid bolus within 3 hours, lactate measured.1 A pre-planned subgroup analysis in children with systolic hypotension in the ED assessed concordance with antibiotics and bolus within 1 hour, and lactate measured, following SSC guidelines for shock. Changes in case volume and care concordance over time were evaluated. The secondary outcome was hospital length of stay, which has been used previously as an outcome measure in pediatric sepsis studies, and was chosen a priori.16,17 All available encounters were included without an a priori sample size calculation, because of an absence of existing literature from which to generate estimated case numbers or effect size; however, we did not expect to have sufficient power to detect differences in mortality or organ dysfunction.18
To enhance understanding of the included hospitals, ED sites were grouped by size, described and compared. EDs were classified as low, medium, and high-volume for pediatric sepsis. There is not consensus on definitions of low pediatric sepsis volume, or low pediatric ED volume.6,7,19 We assessed the number of yearly pediatric sepsis encounter per site, and grouped them by orders of magnitude, after initially assessing data for encounter numbers and prior to assessing outcomes. Free-standing EDs were compared to hospital-based EDs in a secondary analysis planned a priori. A post-hoc exploratory analysis compared highest to lowest two quartiles of pediatric readiness scores.15
Statistical Analysis:
Demographic and hospital data were summarized using frequencies and percentages for categorical variables, and median and interquartile range (IQR) for non-normal continuous variables. The association between predictors and outcomes were assessed using relative risks (RR) with 95% confidence intervals (CI). Trends in concordance over time were evaluated with the Cochran-Armitage test of trend.
Multivariable Analysis:
To assess the association between ED pediatric sepsis volume and concordant care, we created a generalized estimating equation, with detailed methods described in Appendix 3. The clustering variable was provider group, rather than site, because providers, nurses and pharmacists frequently worked at multiple sites within a provider group. This analysis was planned to adjust for covariates chosen clinically and because they were established in the literature as sepsis severity indicators. We considered the following covariates: blood pressure hypotensive for age (Pediatric Advanced Life Support thresholds),20 age in years, arrival mode, year, pulse oximeter value, temperature, heart rate above 95th centile for age, respiratory rate above 95th percentile for age.21 After our initial analyses demonstrated that there was only one site with >100 pediatric sepsis visits yearly, in post-hoc analysis, additional models were created with that site removed, to evaluate if the relationship between case number and concordant care was different among the lower volume sites.
Pediatric readiness scores were not included in multivariable analysis, as they were available for a minority of EDs (12 of 27).
Results
Characteristics:
1527 ED encounters were included in 1402 distinct patients [Figure 1]. Population and hospital course characteristics are shown in Table 1. Transfers to children’s hospitals comprised 448 (29.3%) encounters. Among the 29 EDs, 25 had ≤10 annual pediatric sepsis encounters, 3 had 11 to 100 pediatric sepsis encounters, and 1 had >100 pediatric sepsis encounters; 12 were free standing and 17 were hospital-based [Table 2].
Figure 1:
Study flow diagram.
Table 1:
Population characteristics, ED care and hospital course of study population. 1527 is the denominator for all percentages in the table unless otherwise listed.
| Median [Interquartile Range] or n (% of study population) |
|
|---|---|
| Characteristics | |
| Age, years | 5.5 [1.6, 13.6] |
| Arrival mode | |
| Walk-in | 1313 (86.0%) |
| Ambulance | 201 (13.2%) |
| Other | 9 (0.6%) |
| Unknown | 4 (0.3%) |
| First ED lactate level, mmol/L (lactate measured, n=847) | 1.6 [1.1, 2.4] |
| Systolic hypotension on ED arrival (blood pressure recorded, n=1368) | 30 (2.2%) |
| Systolic hypotension in presenting ED (blood pressure recorded, n=1368) | 104 (7.6%) |
| Course at Presenting (General) ED | |
| ED sepsis order set used | 283 (18.5%) |
| Sepsis diagnostic code associated with ED visit | 72 (4.7%) |
| ED respiratory intervention | 182 (11.9%) |
| ED intubation | 50 (3.3%) |
| ED intravenous fluid volume/weight, mL/kg (in those receiving fluid with weight recorded, n=1494) | 20 [20-29.9] |
| ED vasoactive agent administered | 36 (2.4%) |
| ED length of stay, hours | 4.5 [3.3, 5.8] |
| ED disposition | |
| Admitted | 949 (62.2%) |
| Transferred to another facility | 561 (36.7%) |
| Transferred to a Children’s Hospital site | 448 (29.3%) |
| Transferred to a general hospital system site | 113 (7.4%) |
| Discharge with revisit with admission in 72 hours | 17 (1.1%) |
| Hospital Course Following ED Care | |
| Hospital length of stay in calendar days | 3 [2, 5] |
| Admission to ICU during hospitalization | 190 (12.4%) |
| Positive pressure calendar days (ventilated, n=90) | 4.5 [2.0, 8.0] |
| Vasoactive calendar days (vasoactive patients, n=36) | 3.0 [2.0, 4.0] |
| ICU length of stay, days (ICU patients, n=190) | 4.0 [2.0, 7.0] |
| Extra-corporeal membrane oxygenation | 3 (0.2%) |
| 30-day mortality | 5 (0.3%) |
Abbreviations: ED – Emergency Department, ICU – Intensive Care Unit
Table 2.
Guideline concordant-care and outcome by site characteristics. Sites grouped by a) volume of pediatric sepsis encounters per year; b) free-standing vs. hospital-based ED; c) National quartile of readiness,15 as measured by ED Pediatric Readiness Score (PRS), in a limited analysis including only EDs with PRS available.
| a) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ED Pediatric Sepsis Volume Category |
Number of sites |
Number of visits |
Visit/year Range |
Number of sites with PRS |
PRS, median [IQR] |
Proportion of visits with SSC- concordant care, n (%) |
30-day mortality, n (%) |
HLOS, calendar days, median [IQR] |
Positive- pressure ventilation n (%) |
Vasoactive agent use n (%) |
| Lowest Volumes: 0-10 cases/year | 25 | 301 | 0.3-8.3 | 8 | 74 [52-76] |
122 (40.5%) | 1 (0.3%) |
3 [2-4] |
41 (13.6%) |
10 (3.3%) |
| Middle Volumes: >10-100 cases/year | 3 | 466 | 15.4-32.9 | 3 | 74 [63-74] |
181 (38.8%) |
1 (0.2%) |
3 [2-5] |
56 (12.0%) |
10 (2.1%) |
| Highest Volumes: >100 cases/year | 1 | 760 | 108.6 | 1 | 91 | 324 (42.6%) |
3 (0.4%) |
4 [3-5] |
78 (10.3%) |
16 (2.1%) |
| b) | ||||||||
|---|---|---|---|---|---|---|---|---|
| ED Type | Number of sites |
Number of visits |
Visit/year Range |
Proportion of visits with SSC-concordant care, n (%) |
30-day mortality, n (%) |
HLOS, calendar days, median [IQR] |
Positive- pressure ventilation n (%) |
Vasoactive agent use n (%) |
| Free-standing | 12 | 63 | 0.3-4.8 | 27 (42.9%) |
1 (1.6%) |
2 [1-4] |
6 (9.5%) |
2 (2.2%) |
| Hospital-based | 17 | 1464 | 0.5-108.6 | 600 (41.0%) |
4 (0.3%) |
3 [2-5] |
169 (11.5%) |
34 (2.3%) |
| c) | ||
|---|---|---|
| ED PRS Quartile | Number of Visits to EDs in these Quartiles | Number of visits with SSC-concordant care |
| Quartile 1 and 2 | 444 | 185 (41.7%) |
| Quartile 3 and 4 | 951 | 388 (40.8%) |
Abbreviations: ED – Emergency Department, PRS – Pediatric Readiness Score, SSC – Surviving Sepsis Campaign, HLOS – Hospital Length of Stay, IQR – inter-quartile range
Main Results:
Care was concordant with the SSC guidelines in 627 (41.1%) encounters; lactate measurement was the element missed most often, in 680 (44.5%) encounters [Table 3]. The association between yearly pediatric sepsis encounters and the probability of concordant care was minimal, with an odds ratio of 1.002 (1.001-1.003) in a general estimating equation including covariates reported in Appendix 4. When the largest site was removed from the general estimating equation, there was no association between the probability of receiving concordant care and yearly pediatric sepsis (odds ratio 0.998, 0.987-1.01).
Table 3:
Proportion of study encounters with care concordant with SSC guidelines for pediatric sepsis
| Encounters with SSC-Concordant Care, n (%) n = 1527 |
|
|---|---|
| SSC Care Bundle Elements | |
| ED blood culture obtained | 1516 (99.3%) |
| Intravenous fluid administered within 180 minutes of arrival | 1355 (88.7%) |
| Intravenous antibiotic administered within 180 minutes of arrival | 1006 (65.9%) |
| Lactate measured in ED | 847 (55.5%) |
| Concordant with all above | 627 (41.1%) |
Abbreviations: SSC – Surviving Sepsis Campaign, ED – Emergency Department
Minimal association was seen between hospital length of stay and ED pediatric sepsis volume (incident rate ratio 1.002, 95% confidence interval 1.002-1.003 with all sites included; 1.004, 95% confidence interval 0.999-1.01 with the largest site removedThe ≤10% or fewer encounters with missing variables were excluded from multivariable analysis (Appendix 4). The risk of receiving concordant care was similar in free-standing compared to hospital-based EDs (Table 2). Among encounters in 12 EDs with readiness scores, the risk of concordant care was similar in the lowest two score quartiles compared to the highest two score quartiles (Table 2).
In the entire study population, care concordance significantly improved over time from 23.1% in 2015 to 52.8% in 2021 (Cochran-Armitage trend p< 0.001) [Fig 2, Appendix 5]; improvement was steady despite fluctuating pediatric sepsis volume and the Covid-19 pandemic.
Figure 2:
Proportion of sepsis patients across all sites who received SSC concordant care with 95% confidence intervals (dot and whiskers) and yearly number of pediatric sepsis patients treated.
Subgroup Analyses
In a pre-planned subgroup analysis of 120 encounters with hypotension or vasopressor use in the ED, concordance with the 3-hour bundle was higher, 59%, but only 13.3% met SSC 1-hour goals for septic shock. The subgroup of 372 encounters meeting IPSO Sepsis criteria in the presenting ED are described in Appendix 6. Care was concordant in 201 (51%) of these encounters.
Limitations
This study was conducted in two health systems in Colorado and Wyoming, which together included approximately 80% of all pediatric critical care hospitalizations in Colorado. Findings may not be generalizable, particularly to general hospitals that are not part of large health systems. While the multivariable analysis controlled for differences in known potential confounders, there are likely additional confounders not measured in this retrospective data that may have influenced the association between ED size and length of stay. The sepsis definitions rely on treatment for sepsis, which may vary based on pediatric case volume, and this study could not assess the association between pediatric volume and likelihood of initiating sepsis treatment. There was only one ED with high pediatric sepsis volume, limiting generalizability to general EDs with large case numbers. There was no a priori power calculation; due to limited prior literature assessing care quality of pediatric sepsis in general EDs. Results were not adjusted for within-patient correlations over time. This study relied on retrospective EHR data which is subject to misclassification of predictors and outcomes if the data were incorrect in the EHR.
Discussion
Children initially treated for sepsis in general EDs had similar odds of receiving guideline-concordant care regardless of the ED’s annual volume of pediatric sepsis encounters. Similarly, there was no difference in the risk of receiving guideline-concordant care between free-standing and hospital-based EDs. These findings are in contrast to prior studies that have suggested that EDs with low pediatric sepsis volume are least likely to deliver a pediatric sepsis care bundle, but these prior studies have either not identified sepsis cases in a similar manner across sites, or they have not linked transferred patient data from the presenting ED to their inpatient hospitalization.6,7
The majority of children in this study received their initial care in EDs that care for pediatric sepsis infrequently. In this study, 51% received care at 28 different EDs that saw <35 pediatric sepsis encounters annually. While hospitals that see pediatric sepsis frequently might be the most efficient targets for quality improvement, sites that rarely treat pediatric sepsis individually, treat substantial numbers of cases in aggregate.
It was an unexpected finding that the proportion of encounters with concordant care was so similar across ED types. Whether EDs were grouped by pediatric sepsis volume category, free-standing vs. hospital-based or pediatric readiness score, 40% of encounters had concordant care [Table 2]. Identifying that pediatric sepsis volume and ED type alone do not drive the quality of care can help design quality improvement and policy interventions. It was surprising that readiness quartile, which has been associated with improved outcomes in other conditions, was not associated with care concordance in this study, although this analysis was limited only to sites with available scores.15,22 The National Pediatric Readiness Program, which has driven improvements in pediatric equipment and personnel preparedness, has identified guideline implementation as an ongoing priority.11 Of note, the National Pediatric Readiness Program separately assesses the presence of quality and performance improvement programs that include children but does not address specific conditions such as sepsis. Our finding that guideline concordance was not associated with readiness score suggests that additional factors drive guideline implementation success; local and state readiness programs should consider adding sepsis quality improvement to verification programs, and future research should focus on identifying implementation strategies that are associated with guideline concordance.
The severity of illness in this cohort was lower than has been reported in children’s hospitals-based ED sepsis studies. Vasoactive medications were used in 2.4% and ventilation in 4.5% of children with sepsis in this study, compared to 1-5% and 5-10% of children’s hospital ED encounters.12,23,24 12% of were admitted to ICU, lower than reported in pediatric ED studies. These results may reflect a tendency to bypass general EDs to bring more severely ill or complex children to children’s hospitals when possible. The proportion of children experiencing ECMO and death in this study was 0.3%, lower than has been seen in children’s hospital ED studies, where mortality is 4%.5,25 Given the much larger numbers of children who initially receive care in general hospitals, these smaller proportions nonetheless represent large numbers of children with critical illness.
Particularly concerning was the finding that children with shock did not receive the recommended one-hour bundle of septic shock care, however many of these children did not have shock on ED arrival. Interestingly, the proportion of children receiving care concordant with sepsis guidelines increased over time, although pediatric sepsis-specific quality interventions or EHR changes were not implemented in this system. This trend persisted through pediatric volume increases and decreases associated with the Covid-19 pandemic. This study began January 1, 2015, and the CMS core measure for sepsis in adults was launched October 1 of that year. It is possible that sepsis quality improvement work in adults influenced sepsis care quality in children as well, but further studies are needed to test whether this trend was similar nationally.
Annual numbers of ED pediatric sepsis encounters did not have a strong effect on concordant care, nor did it strongly affect hospital length of stay. Identifying factors beyond pediatric encounter numbers that influence sepsis care and outcomes is important.. For example, in our study, one-third were transferred to another facility;, in prior studies, clinical leaders in general EDs identified the process of transfer to pediatric hospitals as a driver or barrier to pediatric care quality.26,27 Understanding factors such as transfer that may drive pediatric sepsis outcomes beyond the initial bundled measures should be an ongoing focus of research.
Our study highlights challenges to studying pediatric sepsis in general EDs: <5% of cases in this study would have been identified with an approach based on the initial ED diagnostic code. In prior qualitative interviews, general ED clinicians have stated that they are reluctant to apply the code of sepsis to an ill child whom they are quickly transferring out, before a diagnosis has been established, making internal audit of sepsis quality difficult.26 Additionally, 20% of the cases occurred in 25 different EDs, seeing <10 pediatric sepsis cases yearly, representing a substantial but difficult-to-study population. Clinical improvement work needs to account for the lack of repetition that individual clinicians at such sites would have, with attention to data capture.
A strength of this study was the very small proportion of cases lost to transfer outside of the included systems (<1%), an important feature because 30% of encounters were transferred out of the initial ED. Studies that do not link encounters from initial ED presentation to inpatient hospitalization have been limited in their ability to assess outcomes related to initial care. Identifying replicable and efficient approaches to link medical record data across general and pediatric systems is essential to ongoing research and clinical quality efforts focused on the emergency care of children across all emergency settings.
In conclusion, this study directly identified pediatric sepsis cases initially presenting to general EDs using a standardized definition. Pediatric sepsis care improved over time, but opportunities for improvement remained, especially in septic shock. Half of children received care at EDs with low or medium volumes of pediatric sepsis, and annual pediatric sepsis volume was not associated with the probability of receiving of concordant care. Future pediatric sepsis research in general EDs should include very low pediatric volume settings, which treat large numbers of children in aggregate, and seek to understand and leverage factors driving existing improvement.
Supplementary Material
Funding Source:
Dr. Scott's institution received grant support for this research from the Agency for Healthcare Research and Quality (K08HS025696), and the use REDCap for this research was funded by the NIH/NCATS Colorado CTSA (UL1 TR002535)
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest: None
Meetings: A preliminary version of this study was presented in abstract form at the Pediatric Academic Societies Meeting in 2022 in Denver, Colorado.
Data Sharing:
The data dictionary and analytic code for this investigation are available upon request. Partial or complete Limited Data Sets for this study will be shared with investigators who establish a Data Use Agreement, by contacting Halden Scott at halden.scott@cuanschutz.edu, after article publication.
References
- 1.Weiss SL, Peters MJ, Alhazzani W, et al. Surviving Sepsis Campaign International Guidelines for the Management of Septic Shock and Sepsis-Associated Organ Dysfunction in Children. Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. 2020;21(2):e52–e106. [DOI] [PubMed] [Google Scholar]
- 2.Tan B, Wong JJ, Sultana R, et al. Global Case-Fatality Rates in Pediatric Severe Sepsis and Septic Shock: A Systematic Review and Meta-analysis. JAMA pediatrics. 2019;173(4):352–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zimmerman JJ, Banks R, Berg RA, et al. Trajectory of Mortality and Health-Related Quality of Life Morbidity Following Community-Acquired Pediatric Septic Shock. Critical care medicine. 2020;48(3):329–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hartman ME, Linde-Zwirble WT, Angus DC, Watson RS. Trends in the epidemiology of pediatric severe sepsis. Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. 2013;14(7):686–693. [DOI] [PubMed] [Google Scholar]
- 5.Paul R, Melendez E, Wathen B, et al. A Quality Improvement Collaborative for Pediatric Sepsis: Lessons Learned. Pediatr Qual Saf. 2018;3(1):e051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Evans IVR, Phillips GS, Alpern ER, et al. Association Between the New York Sepsis Care Mandate and In-Hospital Mortality for Pediatric Sepsis. JAMA : the journal of the American Medical Association. 2018;320(4):358–367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Baker AH, Monuteaux MC, Eisenberg MA, Hudgins JD. Pediatric sepsis survival in pediatric and general emergency departments. The American journal of emergency medicine. 2022;51:53–57. [DOI] [PubMed] [Google Scholar]
- 8.Ames SG, Davis BS, Angus DC, Carcillo JA, Kahn JM. Hospital Variation in Risk-Adjusted Pediatric Sepsis Mortality. Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. 2018;19(5):390–396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gigli KH, Davis BS, Yabes JG, et al. Pediatric Outcomes After Regulatory Mandates for Sepsis Care. Pediatrics. 2020;146(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McDermott KW (IBM Watson Health) EAA, Sun R (AHRQ). Trends in Hospital Inpatient Stays in the United States, 2005-2014. HCUP Statistical Brief #225. June 2017. Agency for Healthcare Research and Quality, Rockville, MD. www.hcup-us.ahrq.gov/reports/statbriefs/sb225-Inpatient-US-Stays-Trends.pdf. . [Google Scholar]
- 11.Gausche-Hill M, Ely M, Schmuhl P, et al. A national assessment of pediatric readiness of emergency departments. JAMA pediatrics. 2015;169(6):527–534. [DOI] [PubMed] [Google Scholar]
- 12.Scott HF, Brilli RJ, Paul R, et al. Evaluating Pediatric Sepsis Definitions Designed for Electronic Health Record Extraction and Multicenter Quality Improvement. Critical care medicine. 2020;48(10):e916–e926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Harris PA, Taylor R, Minor BL, et al. The REDCap consortium: Building an international community of software platform partners. Journal of biomedical informatics. 2019;95:103208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. Journal of biomedical informatics. 2009;42(2):377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ames SG, Davis BS, Marin JR, et al. Emergency Department Pediatric Readiness and Mortality in Critically Ill Children. Pediatrics. 2019;144(3). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Paul R, Neuman MI, Monuteaux MC, Melendez E. Adherence to PALS Sepsis Guidelines and Hospital Length of Stay. Pediatrics. 2012;130(2):e273–280. [DOI] [PubMed] [Google Scholar]
- 17.Depinet H, Macias CG, Balamuth F, et al. Pediatric Septic Shock Collaborative Improves Emergency Department Sepsis Care in Children. Pediatrics. 2022;149(3). [DOI] [PubMed] [Google Scholar]
- 18.Weiss SL. Clinical Trials in Pediatric Sepsis: What's the (End) Point? Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. 2017;18(3):296–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Samuels-Kalow ME, Sullivan AF, Boggs KM, Gao J, Alpern ER, Camargo CA. Comparing definitions of a pediatric emergency department. J Am Coll Emerg Physicians Open. 2021;2(3):e12478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hagedoorn NN, Zachariasse JM, Moll HA. A comparison of clinical paediatric guidelines for hypotension with population-based lower centiles: a systematic review. Critical care. 2019;23(1):380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bonafide CP, Brady PW, Keren R, Conway PH, Marsolo K, Daymont C. Development of heart and respiratory rate percentile curves for hospitalized children. Pediatrics. 2013;131(4):e1150–1157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Newgard CD, Lin A, Goldhaber-Fiebert JD, et al. Association of Emergency Department Pediatric Readiness With Mortality to 1 Year Among Injured Children Treated at Trauma Centers. JAMA Surg. 2022;157(4):e217419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Balamuth F, Scott HF, Weiss SL, et al. Validation of the Pediatric Sequential Organ Failure Assessment Score and Evaluation of Third International Consensus Definitions for Sepsis and Septic Shock Definitions in the Pediatric Emergency Department. JAMA pediatrics. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lloyd J, Depinet H, Zhang Y, Semenova O, Meinzen-Derr J, Babcock L. Comparison of children receiving emergent sepsis care by mode of arrival. The American journal of emergency medicine. 2021;47:217–222. [DOI] [PubMed] [Google Scholar]
- 25.Scott HF, Brou L, Deakyne SJ, Kempe A, Fairclough DL, Bajaj L. Association Between Early Lactate Levels and 30-Day Mortality in Clinically Suspected Sepsis in Children. JAMA pediatrics. 2017;171(3):249–255. [DOI] [PubMed] [Google Scholar]
- 26.Scott HF, Kempe A, Bajaj L, Lindberg DM, Dafoe A, Dorsey Holliman B. "These Are Our Kids": Qualitative Interviews With Clinical Leaders in General Emergency Departments on Motivations, Processes, and Guidelines in Pediatric Sepsis Care. Annals of emergency medicine. 2022;80(4):347–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Goldman MP, Wong AH, Bhatnagar A, Emerson BL, Brown LL, Auerbach MA. Providers' Perceptions of Caring for Pediatric Patients in Community Hospital Emergency Departments: A Mixed-methods Analysis. Academic emergency medicine : official journal of the Society for Academic Emergency Medicine. 2018;25(12):1385–1395. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data dictionary and analytic code for this investigation are available upon request. Partial or complete Limited Data Sets for this study will be shared with investigators who establish a Data Use Agreement, by contacting Halden Scott at halden.scott@cuanschutz.edu, after article publication.