RESEARCH ARTICLE Illness Severity of Children Admitted to the PICU From Referring Emergency Departments Jacqueline M. Evans, MD, PhD, Parul Dayal, MS, Douglas L. Hallam, BS, JoAnne E. Natale, MD, PhD, Pranav Kodali, Hadley S. Sauers-Ford, MPH, James P. Marcin, MD, MPH

ABSTRACT OBJECTIVES: To compare patient factors and outcomes among children admitted to PICUs from referring versus children’s hospital emergency departments (EDs). METHODS: Pediatric patients (,19 years old) admitted to PICUs from referring and children’s hospital EDs from July 1, 2011 to June 30, 2013. We compared demographic and clinical factors, including severity of illness as measured by a recalibrated Pediatric Index of Mortality, version 2 score. RESULTS: Of 80 045 children from 109 PICUs, 35.6% were admitted from referring EDs and 64.4% were admitted from children’s hospital EDs. Children from referring EDs had higher illness severity (Pediatric Index of Mortality, version 2–predicted risk of mortality, 3.1% vs 2.2%, P , .001), were more likely to be mechanically ventilated within their first hour in the PICU (28.4% vs 23.4%, P , .001), and had higher observed mortality (3.3% vs 2.1%, P , .001). Once adjusted for illness severity and other confounders in a multivariable logistic regression model, there was no difference in the odds of mortality between children from referring and children’s hospital EDs (odds ratio: 0.90; 95% confidence interval: 0.79 to 1.02, P 5 .09) CONCLUSIONS: Children transferred to PICUs from referring EDs had higher illness severity on arrival compared with children admitted from children’s hospital EDs. Variations in patient selection for transfer or pretransfer treatment at referring EDs may contribute to the greater illness severity of transferred children. Referring hospitals may benefit from leveraging existing resources to improve patient stabilization before transfer.

www.hospitalpediatrics.org DOI:https://doi.org/10.1542/hpeds.2017-0201 Copyright © 2018 by the American Academy of Pediatrics Address correspondence to James P. Marcin, MD, MPH, Department of Pediatrics, University of California, Davis, 2516 Stockton Blvd, Sacramento, CA 95817. E-mail: [email protected] HOSPITAL PEDIATRICS (ISSN Numbers: Print, 2154-1663; Online, 2154-1671). FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose. FUNDING: No external funding. POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose. Dr Evans conceptualized and designed the study and drafted the initial manuscript; Ms Dayal conducted the data analyses, reviewed, Department of Pediatrics, and revised the manuscript; Mr Hallam designed the data collection instruments; Dr Natale conceptualized and designed the study and University of California critically reviewed the manuscript; Mr Kodali conducted the initial data collection and analysis; Ms Sauers-Ford critically reviewed Davis Children’s Hospital, and revised the manuscript; Dr Marcin conceptualized and designed the study, supervised the data analysis, and critically reviewed Sacramento, California and revised the manuscript; and all authors approved the final manuscript as submitted.

HOSPITAL PEDIATRICS Volume 8, Issue 7, July 2018 1

Downloaded from www.aappublications.org/news by guest on September 25, 2021 In 2012, 8.8 million US infants, children, and METHODS PIM2 score and the PIM2-predicted risk of adolescents were evaluated and treated in Data Source and Patient Data mortality (ROM).18 1 an emergency department (ED). The This was a retrospective cohort study Risk Adjustment majority of these children received care in using data from the VPS database (http:// EDs within general or community hospitals www.myvps.org). This Web-based registry The ROM predicted by the PIM2 score was that treat both adults and children but do includes data from 109 US PICUs in free used to compare severity of illness between not have PICUs.2,3 Thus, community EDs cohorts. The PIM2 score is estimated by standing and nonfree-standing children’s need to transfer critically ill patients using patient-level clinical variables that are hospitals and contains .600 000 needing higher levels of care to PICUs collected at the time of admission to a PICU. admission records to date. The database located within children’shospitals.On The original model includes the following contains granular patient-specificdata average, community EDs (ie, “referring” 10 covariates: (1) systolic blood pressure; obtained from chart abstraction including EDs) are hampered by lower “pediatric (2) pupillary reaction to bright light; (3) demographic, physiologic, and laboratory readiness”4 as compared with EDs in ratio of fraction inspired oxygen and partial information, as well as the validated children’s hospitals, including but not pressure oxygen in arterial blood; (4) base pediatric illness severity score, PIM2. limited to exposure to relatively smaller excess in arterial or capillary blood; (5) VPS data elements are collected volumes of pediatric patients,5 deficiencies need for mechanical ventilation during the prospectively and deidentified by in pediatric-specific training of staff and first hour in the ICU; (6) whether admission participating PICUs. VPS data have been providers,6 variable access to pediatric was elective; (7) whether patient was in used extensively for quality improvement subspecialists,2,7 insufficient mechanisms recovery from surgery or another purposes, benchmarking, and outcomes for identifying high-risk pediatric procedure; (8) whether the patient was research.17 In this study, VPS data from a patients,8,9 inadequate resuscitation and admitted after cardiopulmonary bypass; (9) 2-year period were used to compare stabilization,1,10,11 limited pediatric-specific whether the patient presented with selected demographic and clinical characteristics resources and equipment,2,5 and a lack of high-risk diagnoses including, cardiac adherence to published pediatric-specific of pediatric patients admitted to PICUs arrest preceding ICU admission, severe ’ guidelines.12 from referring and childrenshospital combined immune deficiency, leukemia or EDs. fi Deviations from published pediatric lymphoma after rst induction, spontaneous emergency medicine guidelines in The original data set included all cerebral hemorrhage, cardiomyopathy or referring EDs2,13 are likely to exert a nonscheduled PICU admissions of children myocarditis, hypoplastic left heart , substantive impact on the outcomes of 19 years of age from July 1, 2011, to June syndrome, HIV, liver failure, and pediatric patients who require transfer 30, 2013. Patients with a physical length of neurodegenerative disorders; and , and admission to a children’shospital stay of 2 hours and patients who came to (10) whether the patient presented with PICU. Existing evidence suggests that the PICU from either an operating room or a selected low-risk diagnoses including children admitted to PICUs from referring cardiac catheterization laboratory were asthma, bronchiolitis, croup, obstructive 18 hospital EDs are sicker, more likely to be excluded, as were children admitted to the sleep apnea, and diabetic ketoacidosis. intubated, and have poorer outcomes than PICU from a pediatric ward or from a PICU The designation of high- and low-risk children admitted to PICUs from EDs in another children’s hospital. The primary diagnoses was created within the VPS located within the same hospital14–16; variable of interest was the source of PICU database. All other covariates for the PIM2 however, previous comparisons have been admission, dichotomized into referring estimation model were also available in the focusedonsinglediagnoses16 or involved hospital EDs and children’s hospital EDs. database. much smaller numbers of patients and Referring hospital EDs were defined as In this study, we recalibrated the PICUs.15 In this study, we used a large, general and community EDs that provide coefficients of the original PIM2 estimation national cohort of PICU admissions care to both adults and children and are model on our population. The reference collected by the Virtual PICU Performance located within a hospital that does not have population for the original model was System (VPS) database and a validated a PICU. Children’s hospital EDs are EDs 20 787 children from Australia, New Zealand, physiologic adjustment tool, the Pediatric located within children’s hospitals that also and the United Kingdom from 1997 to 1999.18 Index of Mortality, version 2 (PIM2) scoring have a PICU. Other variables in our study In comparison, our study included system, to generate a more robust included demographics including age, sex, 80 045 children admitted to PICUs located in comparison of demographic race, and clinical variables including the the United States. Because our study was characteristics and clinical factors, use of mechanical ventilation during the focused on patients transferred from including severity of illness and risk- first hour in the PICU, whether a trauma referring EDs, we excluded children adjusted mortality between children activation was associated with the ICU admitted to the PICU on an elective basis, admitted to PICUs from referring EDs and encounter (trauma activation, yes or no), children admitted after surgery, and those EDs located within children’s hospitals that primary diagnosis, PICU mortality, and the admitted after cardiopulmonary bypass. have a PICU. severity of illness as measured by the These 3 variables were included in the

2 EVANS et al

Downloaded from www.aappublications.org/news by guest on September 25, 2021 original model but had to be excluded from 2 cohorts from referring EDs and children’s RESULTS our recalibrated model. Thus, the hospital EDs. These ratios were compared In total, 80 045 PICU admissions met our recalibrated model included only 7 of the by using a Poisson regression approach in inclusion criteria. Of these, 35.6% were 10 independent variables that were included which the observed deaths were modeled directly admitted to a PICU after transfer in the original model (Supplemental Table as a count variable and expected ROM as from a referring hospital ED, whereas 64.4% 4). The area under the receiver operating the exposure variable. P values of were admitted to a PICU from an ED within a characteristic curve for the recalibrated ,.05 were considered to be statistically children’s hospital. As shown in Table 1, model was 0.93, demonstrating the excellent significant. children admitted from referring hospital model discrimination. The model also had Institutional Review Board EDs had higher illness severity as measured good calibration across risk quintiles by the PIM2-predicted ROM (3.1% vs 2.2%, (Hosmer-Lemeshow goodness-of-fit x2 test, The human subjects review board at , P 5 .20). The PIM2-predicted ROM is University of California Davis Health System P .001), and had higher PICU mortality , estimated from the PIM2 score and ranges approved this study. (3.3% vs 2.1%, P .001) compared with – between 0 and 1 (or 0% 100%). TABLE 1 Demographic and Clinical Characteristics of Children Admitted From Referring and Recalibration of the PIM2 estimation model Children’s Hospital EDs allowed us to more accurately estimate the Parameter Referring Hospital EDs, n (%) Children’s Hospital EDs, n (%) P PIM2-predicted ROM using this more 28 503 (35.6) 51 542 (64.4) contemporary, US-based patient population. Age in y, mean (SD) 7.1 (6.2) 6.8 (5.9) ,.001 Diagnoses Male, n (%) 15 690 (55.1) 29 191 (56.6) ,.001 The VPS primary diagnostic category Race, n (%) ,.001 scheme was used to compare diagnoses White 10 361 (36.4) 16 326 (31.7) between the 2 cohorts of patients. Results African American 3700 (13.0) 9788 (19.0) included data on the 6 most common Hispanic 3298 (11.6) 5926 (11.5) diagnostic categories: respiratory illnesses; Asian or Pacific Islander 579 (2.0) 925 (1.8) injury, poisoning or adverse drug-related Native American 270 (1.0) 204 (0.4) events; neurologic disorders; endocrine Other or multiracial 1026 (3.6) 1437 (2.8) abnormalities; cardiovascular processes; Unknown 9269 (32.5) 16 936 (32.9) and infectious diseases. Disposition, n (%) ,.001 Statistical Analysis Did not survive until discharge 930 (3.3) 1073 (2.1) Baseline characteristics were compared by Hospital ward 16 962 (59.5) 32 156 (62.4) 2 using the x test for categorical variables, Home 7191 (25.2) 12 399 (24.1) ’ the Students t test for continuous variables, Step-down or telemetry units 1750 (6.1) 3474 (6.7) and the Wilcoxon rank-sum test for Rehabilitation facility 896 (3.1) 603 (1.2) variables having skewed distributions. The Another hospital 326 (1.1) 529 (1.0) xtlogit procedure in Stata/SE 14.1 (Stata Another ICU in the same hospital 171 (0.6) 489 (1.0) Corp, College Station, TX) was used to conduct a hierarchical multivariable logistic Transitional care facilities 130 (0.5) 526 (1.0) regression analysis to compare the odds of Other 128 (0.5) 245 (0.5) mortality between children admitted to Hospice 10 (0.04) 39 (0.08) PICUs from referring EDs and those Previous PICU admission, n (%) 44 (0.2) 30 (0.1) ,.001 admitted from children’s hospital EDs. This Mechanical ventilation, n (%) 8090 (28.4) 12 074 (23.4) ,.001 model was adjusted for severity of illness as Trauma activation, n (%) 2148 (7.5) 7296 (14.2) ,.001 measured by PIM2-predicted ROM, patient PIM2 score age, provision of mechanical ventilation Mean (SD) 25.24 (1.8) 25.41 (1.6) ,.001 fi during the rst hour in the PICU, and Median (IQR) 25.59 (2.2) 25.60 (1.8) ,.001 clustering at the hospital level. We chose PIM2 ROM, % these variables using a combination of a Mean (SD) 3.1 (12.2) 2.2 (9.0) ,.001 priori standards and/or hypotheses (eg, Median (IQR) 0.371 (1.7) 0.367 (0.5) ,.001 including patient age) as well as statistical significance in the univariable analyses. PICU LOS in d We also estimated the ratios of observed Mean (SD) 3.01 (7.5) 3.03 (6.5) .65 deaths to the expected deaths, known as the Median (IQR) 1.21 (1.9) 1.33 (2.0) ,.001 standardized mortality ratio, for the IQR, interquartile range; LOS, length of stay.

HOSPITAL PEDIATRICS Volume 8, Issue 7, July 2018 3

Downloaded from www.aappublications.org/news by guest on September 25, 2021 children admitted from children’s hospital ventilation during their first hour of transfers).20,21 In the current study, results EDs. Children from referring hospital EDs admission. Although the children admitted from a large multicenter national cohort were also more likely to be mechanically from referring hospital EDs had higher validate previous observations regarding ventilated within their first hour in the PICU observed mortality, their odds of mortality the differences in severity of illness between (28.4% vs 23.4%, P , .001) but less likely to were not significantly different than those of interhospital and intrahospital transfers; be admitted after a trauma activation (7.5% children admitted from children’s hospital however, there was no difference in risk- vs 14.2%, P , .001). Patients from referring EDs after adjusting for severity of illness. adjusted mortality between the 2 groups. hospital EDs were less likely to be admitted Taken together, these data demonstrate that Our findings could be attributed to the with a primary diagnosis of respiratory children who are admitted to PICUs after differences in patient management and illness (28.4% vs 34.1%, P , .001) and more being transferred from referring hospital processes of care at the referring and likely to be admitted for neurologic EDs have different clinical characteristics children’s hospital EDs16,19 and may include disorders (18.2% vs 14.0%, P , .001) as including diagnoses and severity of illness, such factors as differential adherence to compared with children transferred from but similar odds of mortality once adjusted published guidelines,2,12 availability of children’s hospital EDs (Table 2). for these differences in illness severity and pediatric subspecialists,13 availability of other confounders. Once adjusted for severity of illness as pediatric-specific resources,2 pediatric- measured by the PIM2-predicted ROM, age, We and others have reported differences specific expertise of staff,2 and timeliness of the need for mechanical ventilation during between children admitted to PICUs from triage and transfer. The higher severity of the first hour of PICU admission, and referring and children’s hospital EDs.14–16,19 illness of the children from referring EDs clustering by hospital site, the odds of PICU In a previous study, we reported that compared with children’s hospital EDs on mortality between children admitted from children admitted from a referring hospital presentation to the PICU could result from referring hospital EDs and children’s ED had higher rates of mechanical differences in stabilization of children before hospital EDs were not significantly different ventilation (33.5% vs 23.6%, P , .01) and transfer. Efforts have been made to optimize (Table 3) (odds ratio [OR]: 0.90; 95% vasoactive infusions (7.3% vs 5.2%, P , .01) outcomes in referring hospitals by fi con dence interval [CI]: 0.79 to 1.02; on the first day of PICU admission compared standardizing the care of specific disease 5 P .09). The standardized mortality ratios with children admitted from an ED within entities.10,22,23 A compelling case of outreach 14 of patient cohorts from referring hospital the same hospital as the PICU. In a efforts altering practice patterns and EDs (OR: 1.04; 95% CI: 0.98 to 1.11) and retrospective analysis of outcomes of improving outcomes is in the treatment of ’ childrens hospital EDs (OR: 0.97; 95% CI: 8897 patients admitted to a PICU at a pediatric septic shock.10,22,23 Early recognition fi 0.91 to 1.02) were not signi cantly different tertiary care center, patients transferred and goal-directed resuscitation of pediatric 5 (P .09). from a pediatric ward or ED of a referring and neonatal septic shock according to ’ hospital to the study hospitals PICU were Pediatric Advanced Life Support guidelines DISCUSSION sicker and had unadjusted mortality rates have been associated with significant This is the largest and most contemporary that were nearly twofold greater than those reduction in morbidity and mortality rates.22,23 comparison of pediatric patients admitted of patients admitted from the ED within the to US PICUs from referring and children’s study hospital.15 Other studies have Better stabilization of children at referring hospital EDs. Children admitted to PICUs reported poorer outcomes of adult patients EDs that lack a PICU may also be achieved from referring hospital EDs had higher transferred from outside hospitals by leveraging the expertise of pediatric illness severity as measured by a validated (interhospital transfers) compared with hospitalists and general pediatricians in the pediatric mortality prediction score and patients within the hospital needing a pretransfer treatment of children needing were more likely to undergo mechanical higher level of care (intrahospital higher levels of care in a PICU. Studies have revealed that hospitalists play an important role in improving patient flow, including TABLE 2 Comparison of Primary Diagnoses Between Children Admitted From Referring and patient handoffs, by appropriately triaging Children’s Hospital EDs patients and coordinating processes of Diagnosis Type Referring Hospital EDs, n (%) Children’s Hospital EDs, n (%) P care.24,25 Another possible solution to improve the pretransfer quality of care at Primary diagnoses ,.001 referring EDs is the use of telemedicine.26 Respiratory illness 8182 (28.7) 17 597 (34.1) Consultations provided remotely by Injury, poisoning, or ADRE 5536 (19.4) 9570 (18.6) pediatric intensivists at children’s hospitals Neurologic disorder 5173 (18.2) 7229 (14.0) via telemedicine lower the on-arrival illness Endocrine derangement 3316 (11.6) 4622 (9.0) severity of children transferred to the PICU Cardiovascular process 1550 (5.4) 2317 (4.5) and reduce unnecessary PICU transfers of Infectious disease 1298 (4.6) 2874 (5.6) children who can be appropriately treated at ADRE, adverse drug-related event. their local EDs.27,28 A combination of critical

4 EVANS et al

Downloaded from www.aappublications.org/news by guest on September 25, 2021 TABLE 3 Association Between PICU Mortality and Source of Admission, Adjusted for Covariates 4. Remick K, Kaji AH, Olson L, et al. Pediatric and Clustering by Hospital Type readiness and facility verification. Ann Patient Characteristics OR SE 95% CI P Emerg Med. 2016;67(3):320–328.e1 Admitted from a children’s hospital ED 0.90 0.06 0.79 to 1.02 .09 5. Tilford JM, Simpson PM, Green JW, PIM2 ROM, % 1.08 0.002 1.07 to 1.08 ,.001 Lensing S, Fiser DH. Volume-outcome Age, y 1.002 0.01 0.99 to 1.01 .62 relationships in pediatric intensive care Mechanical ventilation 6.73 0.52 5.78 to 7.84 ,.001 units. Pediatrics. 2000;106(2 pt 1):289–294 6. Dharmar M, Marcin JP, Romano PS, et al. Quality of care of children in the care telemedicine and a pediatric hospitalist EDs and/or during transport before emergency department: association with program at a community hospital improved admission. Our results may also be subject hospital setting and physician training. triage of critically ill children by reducing to lead-time bias, which may confound J Pediatr. 2008;153(6):783–789 unnecessary PICU tranfers.29 mortality risk predictions when quality and 7. Phillips MR, Machta RM, McLean SE, duration of care differ in different The results of the current study may also be Charles AG. Pediatric appendiceal settings.32,33 In addition, the current data set subject to differences in patient-level perforation rates are associated with a does not allow for identification of a characteristics such as diagnosis, insurance shortage of general surgeons and an relatively small cohort of children who are status, income, and education of parent(s) or increased rate of transfer to surgical transported from a referring hospital ED to guardian(s). Another explanation for our care in North Carolina. J Am Coll Surg. a children’s hospital ED without first results is that once a patient leaves the 2014;219(3):S77 referring facility, deterioration of physiologic determining the ultimate destination of the stability of a patient may occur during patient (pediatric ward versus PICU). 8. Nielsen KR, Migita R, Batra M, Gennaro transportation.30 Furthermore, the illness JL, Roberts JS, Weiss NS. Identifying CONCLUSIONS severity of children from referring hospital high-risk children in the emergency EDs may be impacted by the type of transport Using a large, national cohort of children department [published online ahead of team used (ie, dedicated pediatric critical admitted to a PICU over a 2-year period, we print February 10, 2015]. J Intensive care transport teams31 versus critical care found that children admitted to PICUs from Care Med. doi:10.1177/088506661557189 transport teams that transport both adults referring EDs had higher severity of illness 9. Gold DL, Mihalov LK, Cohen DM. and children). The effects of transport and on admission and different clinical Evaluating the pediatric early warning the type of transport team used are beyond diagnoses but similar risk adjusted score (PEWS) system for admitted the scope of this study and will need to be mortality. Additional study is warranted to patients in the pediatric emergency explored in future studies. determine if better adherence to existing department. Acad Emerg Med. 2014; guidelines, leveraging local resources, and 21(11):1249–1256 Our study is not without limitations, similar new technologies could lead to increased to other studies in which retrospectively stabilization of pediatric patients before 10. Cruz AT, Perry AM, Williams EA, Graf JM, collected data were used. Although the VPS transfer to a higher level of care. Wuestner ER, Patel B. Implementation of database attempts to maximize data goal-directed therapy for children with integrity through training modules, periodic REFERENCES suspected sepsis in the emergency audits, site visits, and automated algorithms 1. Bloom B, Jones LI, Freeman G. Summary department. Pediatrics. 2011;127(3). designed to capture aberrancies, health statistics for U.S. children: Available at: www.pediatrics.org/cgi/ information is gathered from multiple National Health Interview Survey, 2012. content/full/127/3/e758 institutions and by an even greater number – Vital Health Stat 10. 2013;(258):1 81 11. Brierley J, Carcillo JA, Choong K, et al. of personnel and is therefore prone to data 2. American Academy of Pediatrics; Clinical practice parameters for entry errors. Another limitation with our Committee on Pediatric Emergency hemodynamic support of pediatric and study is the possibility of a selection bias; Medicine; American College of neonatal septic shock: 2007 update from the VPS database tends to include larger the American College of Critical Care PICUs affiliated with academic medical Emergency Physicians; Pediatric Committee; Emergency Nurses Medicine [published correction appears centers that are more likely to be able to Association Pediatric Committee. Joint in Crit Care Med. 2009;37(4):1536]. Crit provide the most novel and regionalized of policy statement—guidelines for care of Care Med. 2009;37(2):666–688 services. This would lead to referring children in the emergency department. hospitals sending the most severely ill 12. Zee-Cheng JE, Webber EC, Abu-Sultaneh Pediatrics. 2009;124(4):1233–1243 children to the PICUs included in the S. Adherence to pediatric diabetic database. In addition, in the current study, 3. Institute of Medicine. Emergency Care ketoacidosis guidelines by community we do not control for variations in the for Children: Growing Pains. Washington, emergency departments’ providers. specific types of therapy administered in DC: National Academic Press; 2006 Int J Emerg Med. 2017;10(1):11

HOSPITAL PEDIATRICS Volume 8, Issue 7, July 2018 5

Downloaded from www.aappublications.org/news by guest on September 25, 2021 13. Stoddard JJ, Cull WL, Jewett EA, affects treatment. Pediatr Crit Care Med. 27. Dayal P, Hojman NM, Kissee JL, et al. Brotherton SE, Mulvey HJ, Alden ER; AAP 2007;8(2):91–95 Impact of telemedicine on severity of Committee on Pediatric Workforce 20. Hernandez-Boussard T, Davies S, illness and outcomes among children Subcommittee on Subspecialty McDonald K, Wang NE. Interhospital transferred from referring emergency Workforce. Providing pediatric ’ facility transfers in the United States: a departments to a childrens hospital subspecialty care: a workforce nationwide outcomes study. J Patient PICU. Pediatr Crit Care Med. 2016;17(6): analysis. Pediatrics. 2000;106(6): – Saf. 2017;13(4):187–191 516 521 1325–1333 21. Sokol-Hessner L, White AA, Davis KF, 28. Yang NH, Dharmar M, Kuppermann N, 14. Gregory CJ, Nasrollahzadeh F, Dharmar Herzig SJ, Hohmann SF. Interhospital et al. Appropriateness of disposition M, Parsapour K, Marcin JP. Comparison of transfer patients discharged by following telemedicine consultations critically ill and injured children academic hospitalists and general in rural emergency departments. transferred from referring hospitals internists: characteristics and Pediatr Crit Care Med. 2015;16(3): versus in-house admissions. Pediatrics. – outcomes. J Hosp Med. 2016;11(4): e59 e64 2008;121(4). Available at: www.pediatrics. 245–250 org/cgi/content/full/121/4/e906 29. Labarbera JM, Ellenby MS, Bouressa P, 22. Carcillo JA, Kuch BA, Han YY, et al. Burrell J, Flori HR, Marcin JP. The impact 15. Odetola FO, Rosenberg AL, Davis MM, of telemedicine intensivist support and a Clark SJ, Dechert RE, Shanley TP. Do Mortality and functional morbidity after use of PALS/APLS by community pediatric hospitalist program on a outcomes vary according to the source community hospital. Telemed J E Health. of admission to the pediatric intensive physicians. Pediatrics. 2009;124(2): – 2013;19(10):760–766 care unit? Pediatr Crit Care Med. 2008; 500 508 9(1):20–25 23. Han YY, Carcillo JA, Dragotta MA, et al. 30. Kanter RK, Tompkins JM. Adverse events during interhospital transport: 16. Shibata S, Khemani RG, Markovitz B. Early reversal of pediatric-neonatal physiologic deterioration associated Patient origin is associated with septic shock by community physicians is with pretransport severity of illness. duration of endotracheal intubation and associated with improved outcome. – Pediatrics. 1989;84(1):43–48 PICU length of stay for children with Pediatrics. 2003;112(4):793 799 status asthmaticus. J Intensive Care 24. Chadaga SR, Shockley L, Keniston A, et al. 31. Stroud MH, Sanders RC Jr, Moss MM, Med. 2014;29(3):154–159 Hospitalist-led medicine emergency et al. Goal-directed resuscitative 17. Bennett TD, Spaeder MC, Matos RI, et al; department team: associations with interventions during pediatric Pediatric Acute Lung Injury and Sepsis throughput, timeliness of patient care, interfacility transport. Crit Care Med. – Investigators (PALISI). Existing data and satisfaction. J Hosp Med. 2012;7(7): 2015;43(8):1692 1698 – analysis in pediatric critical care 562 566 32. Goldhill DR, Sumner A. Outcome of research. Front Pediatr. 2014;2:79 25. Howell E, Bessman E, Marshall R, Wright intensive care patients in a group 18. Slater A, Shann F, Pearson G; Paediatric S. Hospitalist bed management effecting of British intensive care units. Index of Mortality (PIM) Study Group. throughput from the emergency Crit Care Med. 1998;26(8):1337– PIM2: a revised version of the paediatric department to the intensive care unit. 1345 – index of mortality. Intensive Care Med. J Crit Care. 2010;25(2):184 189 33. Tunnell RD, Millar BW, Smith GB. The – 2003;29(2):278 285 26. Dharmar M, Romano PS, Kuppermann N, effect of lead time bias on severity of 19. Carroll CL, Smith SR, Collins MS, et al. Impact of critical care telemedicine illness scoring, mortality prediction and Bhandari A, Schramm CM, Zucker AR. consultations on children in rural standardised mortality ratio in intensive Endotracheal intubation and pediatric emergency departments. Crit Care Med. care—a pilot study. Anaesthesia. 1998; status asthmaticus: site of original care 2013;41(10):2388–2395 53(11):1045–1053

6 EVANS et al

Downloaded from www.aappublications.org/news by guest on September 25, 2021 Illness Severity of Children Admitted to the PICU From Referring Emergency Departments Jacqueline M. Evans, Parul Dayal, Douglas L. Hallam, JoAnne E. Natale, Pranav Kodali, Hadley S. Sauers-Ford and James P. Marcin Hospital Pediatrics originally published online June 1, 2018;

Updated Information & including high resolution figures, can be found at: Services http://hosppeds.aappublications.org/content/early/2018/05/30/hpeds. 2017-0201 Supplementary Material Supplementary material can be found at: http://hosppeds.aappublications.org/content/suppl/2018/05/30/hpeds. 2017-0201.DCSupplemental References This article cites 28 articles, 6 of which you can access for free at: http://hosppeds.aappublications.org/content/early/2018/05/30/hpeds. 2017-0201#BIBL Subspecialty Collections This article, along with others on similar topics, appears in the following collection(s): Critical Care http://www.hosppeds.aappublications.org/cgi/collection/critical_care _sub Emergency Medicine http://www.hosppeds.aappublications.org/cgi/collection/emergency_ medicine_sub Permissions & Licensing Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.hosppeds.aappublications.org/site/misc/Permissions.xht ml Reprints Information about ordering reprints can be found online: http://www.hosppeds.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news by guest on September 25, 2021 Illness Severity of Children Admitted to the PICU From Referring Emergency Departments Jacqueline M. Evans, Parul Dayal, Douglas L. Hallam, JoAnne E. Natale, Pranav Kodali, Hadley S. Sauers-Ford and James P. Marcin Hospital Pediatrics originally published online June 1, 2018;

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://hosppeds.aappublications.org/content/early/2018/05/30/hpeds.2017-0201

Data Supplement at: http://hosppeds.aappublications.org/content/suppl/2018/05/30/hpeds.2017-0201.DCSupplemental

Hospital Pediatrics is an official journal of the American Academy of Pediatrics. Hospital Pediatrics is owned, published, and trademarked by the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2018 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.

Downloaded from www.aappublications.org/news by guest on September 25, 2021