American Journal of Emergency Medicine 32 (2014) 199–202
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
How well do General EMS 911 dispatch protocols predict ED resource utilization for pediatric patients?☆ Stephanie J. Fessler, MD a,⁎, Harold K. Simon, MD, MBA a, b, d, Arthur H. Yancey II, MD MPH c, d, Michael Colman, MPA c, Daniel A. Hirsh, MD a, b a
Pediatric Emergency Medicine, Children's Healthcare of Atlanta, Atlanta, GA, USA Dept of Pediatrics, Emory University, Atlanta, GA, USA Grady Health Systems, Grady Emergency Medical Services, Atlanta, GA, USA d Department of Emergency Medicine, Emory University b c
a r t i c l e
i n f o
Article history: Received 13 June 2013 Received in revised form 22 September 2013 Accepted 23 September 2013
a b s t r a c t Introduction: The use of Emergency Medical Services (EMS) for low-acuity pediatric problems is well documented. Attempts have been made to curb potentially unnecessary transports, including using EMS dispatch protocols, shown to predict acuity and needs of adults. However, there are limited data about this in children. The primary objective of this study is to determine the pediatric emergency department (PED) resource utilization (surrogate of acuity level) for pediatric patients categorized as “low-acuity” by initial EMS protocols. Methods: Records of all pediatric patients classified as “low acuity” and transported to a PED in winter and summer of 2010 were reviewed. Details of the PED visit were recorded. Patients were categorized and compared based on chief complaint group. Resource utilization was defined as requiring any prescription medications, labs, procedures, consults, admission or transfer. “Under-triage” was defined as a “lowacuity” EMS transport subsequently requiring emergent interventions. Results: Of the 876 eligible cases, 801 were included; 392/801 had no resource utilization while 409 of 801 had resource utilization. Most (737/801) were discharged to home; however, 64/801 were admitted, including 1 of 801 requiring emergent intervention (under-triage rate 0.12%). Gastroenterology and trauma groups had a significant increase in resource utilization, while infectious disease and ear-nosethroat groups had decreased resource utilization. Discussion: While this EMS system did not well predict overall resource utilization, it safely identified most low-acuity patients, with a low under-triage rate. This study identifies subgroups of patients that could be managed without emergent transport and can be used to further refine current protocols or establish secondary triage systems. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The use of Emergency Medical Services (EMS) for low-acuity pediatric problems has been well documented [1-18]. The literature reports from 16% to 60% of all pediatric transports via EMS were deemed to be unnecessary based on medical caregiver opinion, critical outcomes from Pediatric Emergency Department (PED) visit, or PED visit billing codes [1-4]. These medical transports for low-level emergencies contribute to excessive health care costs by consuming
☆ Prior presentations: Information has not been presented as an article in the past. However, the information was presented in part as a platform presentation for the Pediatric Academic Society conference in April 2012 and a poster presentation for the Society of Academic Emergency Medicine conference in May 2012. ⁎ Corresponding author. Pediatric Emergency Medicine Fellow 170 Boulevard SE D104 Atlanta, GA 30312, USA. Tel.: +1 352 262 9698. E-mail address: [email protected] (S.J. Fessler). 0735-6757/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2013.09.018
a relatively scarce resource and causing delays in transport for those who have truly time-sensitive emergencies [5,6]. Attempts have been made to curtail the EMS transports of pediatric patients with stable conditions not requiring emergent intervention. These have included employing the on-scene emergency medical technician (EMT) to determine the necessity of transport for patients to an emergency care facility [3,7-10]. Unfortunately, the majority of these studies found that this strategy resulted in undesirable outcomes, such as discordance between initial EMT assigned triage level and ultimate clinical course, under-referral rates, or over-referral rates [7-11]. While there has not been much success with EMT-driven independent on-scene triage of pediatric patients, programmatic 91-1 dispatch protocols have been shown to effectively stratify adult callers [6,12-15,17]. In these studies, as well as those in the pediatric population, hospital-based resource utilization has been used as a reasonable surrogate measure of patient acuity [16]. However,
200
S.J. Fessler et al. / American Journal of Emergency Medicine 32 (2014) 199–202
insufficient data have been gathered to confirm the efficacy of EMS dispatch protocols in guiding appropriate use of EMS for pediatric patients. The primary objective of this study is to determine the PED resource utilization of pediatric patients who were categorized as “low-acuity” by local response codes that are based on a specific 9-1-1 dispatch protocol program and then subsequently transported by EMS to a PED. A secondary objective is to determine the “under-triage rate” for this group of patients—or the rate of those children initially categorized as “low-acuity” by the lowest of the 3 response codes based on initial 9-1-1 dispatch protocols but subsequently requiring emergent resources. 2. Methods In the City of Atlanta jurisdiction, the 9-1-1 public EMS system is urban-hospital based, and utilizes the National Academy of Emergency Dispatch: Medical Priority Dispatch System v12 to process and prioritize all of the calls, then dispatch ambulances to the scene for further patient assessment and subsequent transport. It is designed to prioritize the calls of both the adult and pediatric populations. This specific EMS service designates all patients as either “emergent, lifethreatening”, “emergent, not life-threatening”, or “non-emergent, not life-threatening.” The pediatric portion of the “non-emergent, not lifethreatening” designated group was chosen as our low-acuity cohort in this study. This study had IRB approval from both the institution which oversees the EMS transport system, as well as the institution for the receiving children’s hospitals. All pediatric patients (ages less than 19 years) in this low-acuity cohort who had EMS transport from the scene to one of 3 pediatric emergency departments (PED) of a large tertiary care pediatric healthcare system from January 2010 through October 2010 were identified. Consecutive cases were reviewed in three winter and three summer months (Starting Jan 1, 2010, and July 1, 2010, respectively) to account for seasonal variability in pediatric related illness and injuries. The initial EMS information included incident address, demographic information, and any upgrade in transport acuity through the course of their EMS care. The patients’ EMS records were matched to their subsequent PED records using patient name, date of birth and date of encounter. The EMS record and PED charts were manually reviewed and recorded by a single reviewer (author SF). The patient demographics and PED course were documented. These included patient age, race and payor status; triage nursing assigned chief complaint (CC) and PED nursing assigned triage level, initial vital signs, PED resource utilization, final diagnosis and disposition. The CCs were then grouped by systems (CC Index). For the purposes of this study, PED resource utilization was used as a surrogate measure of acuity level. The study definition of resource utilization
CC Index Chief complaints classified into groups CC group Triage nurse documented CC Fever, cold, flu symptoms Constipation, vomiting, diarrhea, abdominal pain, blood in stool, swallowed foreign body, gastrostomy tube complaint TRAUMA Any extremity injury, swelling or pain; bite, burn or laceration ENT Eye drainage, tooth pain, mouth or throat pain, ear pain, foreign body in ear/nose, neck or facial swelling NEURO Crying or fussy infant; lethargy, weakness; seizures, dizziness or fainting, headache RESP Cough, choking, fast or difficulty breathing, wheezing, grunting, apnea DERM Rash, bites, abscess GU Genital abrasions, swelling or pain; urinary, vaginal or testicular symptoms. PSYCH Suicidal or homicidal ideations, “mental breakdown” HEME Sickle cell pain OTHER Broken CVL, hyperglycemia, tachycardia, medical clearance for law, alleged sexual assault, call back for prescription ID GI
included: the patient was given any medication besides acetaminophen, ibuprofen, or diphenhydramine; the patient had any laboratory study performed other than a rapid strep; for a patient for whom any procedure was performed, a consult was obtained; or the patient was admitted or transferred. Emergent intervention was defined as any immediate resuscitation, intravenous fluids, intubation or CPR performed. Those patients who were transferred from another healthcare facility, left without being seen, or with incomplete records were excluded. CC groups were compared based on resource utilization. Under-triage rate was calculated, as defined above. Categorical variables were compared via χ 2 testing and continuous variables with the Student t test. Significance was established at a P b .05. All analysis was performed on IBM SPSS Statistics for Windows, Version 20.0. (IBM Corp, Armonk, NY). 3. Results There were 876 eligible cases of pediatric patients classified as “non-emergent, non-life threatening” by initial EMS dispatch response code and subsequently transported to the pediatric emergency department. Seventy-five cases were excluded, of which 65 did not have a matching PED chart, 5 patients left without being seen, and 5 were transferred from another healthcare provider. This left 801 cases as our study cohort. Of these 801 cases, 392 (49%) of 801 did not utilize PED based resources and were all discharged to home. The remaining 409 (51%) of 801 of cases had some degree of significant resource utilization as defined a priori for this investigation. While the majority of all patients (737/801 [92%]) were discharged to home, 64 of the total 801 (8%) patients were admitted (Fig.). There was one emergent procedure (intubation) in 801 cases for an emergent undertriage rate of 0.12%. No statistically significant difference was found between the resource utilization group and the no resource utilization group when compared on race, payor status, or gender. The majority of patients were African American with public or no insurance and had a slight male predominance. However, a statistically significant difference in median age was discovered in the resource utilization group vs no resource utilization group (66 vs 35 months respectively, P b .001) (Table 1). The most common CC categories that comprised 593 (74%) of 801 of the study population were infectious disease (ID), gastroenterology (GI), trauma, and ENT. The ID complaints consisted primarily of “fever”, GI complaints mostly of vomiting, and ENT complaints of “ear pain” (CC Index). Medications, imaging, labs, and procedures were the most commonly used resources. There were 293 (37%) of 801 patients that required medications other than ibuprofen, acetaminophen or diphenhydramine. The most common medications received were an antiemetic (92/293 (31%)) and opioid analgesic pain (76/293 (26%)), with the majority (264/293 (90%)) of these patients discharged to home. There were 155 (19%) of 801 patients who had radiographic imaging, of which 132 (91%) of 155 were plain films, and 120 (83%) of 155 of these patients were discharged to home. Laboratory studies were performed in the PED on 140 (17%) of 801 patients. Procedures were done for 118 (15%) of 801 patients, mostly orthopedic (44/118 [37%]) or for laceration/wound repair (31/114 [27%]). The remaining procedures were split between foreign body removal, incision and drainage, and others (mainly gastrostomy tube replacement) (Table 2). Significant differences were found between the no resource utilization group and the resource utilization group based on CC category. For those with a GI or trauma complaint, there were higher odds of requiring resource utilization, with OR 3.0 (95% CI 2.08-4.36) for the GI group related to the common use of ondansetron for vomiting; and OR 8.5 (5.00-14.51) for the trauma group—related to
S.J. Fessler et al. / American Journal of Emergency Medicine 32 (2014) 199–202
201
Fig. Distribution of Cases.
empirical antibiotics and intubation prior to transport to a pediatric intensive care unit with a multilobar pneumonia, empyema, and Streptococcus pneumonia bacteremia.
wound or orthopedic injuries. However, those with ID or ENT complaints had significant decreased resource utilization, OR 0.28 (0.20-0.41) and OR 0.3 (0.21-0.50), respectively (Table 3). There were 64 (8%) of 801 patients admitted from the emergency room. There were a variety of ages, CCs, and ultimate diagnoses. The overwhelming majority of these patients (63/64) were admitted to the floor for non-emergent treatments. Many required admission for intravenous antibiotics due to neonatal fever, skin or soft tissue infection, or urinary tract infection. A large portion of the admitted patients had pre-existing medical conditions, such as known psychiatric disorder, sickle cell disease, ventriculoperitoneal shunt, or short gut syndrome. Also included in this group was a patient with a retained nasal foreign body, a patient with a tibia fracture requiring surgical repair, one patient with ingestion of cocaine requiring observation, or prolonged post-ictal state after seizure. Of note, 32 (50%) of 64 were 24 months of age or less. There was one patient who required emergent intervention—a 2year-old girl with respiratory failure and sepsis. The CC of this 2-yearold female with sepsis and respiratory failure was 1 week of upper respiratory symptoms and progressive cough which resulted in her categorization by EMS dispatch as “non-emergent, not life-threatening.” She was not only under-triaged according to our study definition; her degree of illness was not readily recognized by EMS personnel. The EMTs involved in her evaluation and transport did NOT upgrade the child’s acuity status to a more urgent level after arrival to scene. No interventions were done by EMS en route. In the PED, the child had vital signs and physical exam findings consistent with decompensated shock and respiratory distress. She received an IO for access, rapid intravenous fluid resuscitation, nebulized medications,
In this investigation of this specific 9-1-1 EMS dispatch triage system, the dispatch protocols applied to pediatric patients resulting in classification as “non-emergent, not life-threatening” did not accurately predict PED resource utilization, based on study definitions, with 409/801 (51%) requiring some degree of PED based resource utilization. However, this system did accurately identify in most cases the lowestrisk patients requesting evaluation and transport to a PED for evaluation, evidenced by a low under-triage rate. While the emergent under-triage rate was only 0.12%, unrecognized significant illness can occur to a minimal degree, based on present protocols. As seen here, such occasional cases can also be undetected by EMTs. It is unclear as to why the degree of the one critical patient’s illness was not recognized by the EMS personnel involved in her care. Perhaps this is due to lack of pediatric experience among this team of medics. In addition, staff may be de-conditioned by the larger majority of patients they transport in the area that are truly not emergent. This study is limited due to being a retrospective chart review; however, it did not influence physician practice or triage protocols. The charts were reviewed in respective summer and winter months, a decision made a priori to chart review, in order to account for the extremes in seasonal variability. However, this may have resulted in
Table 1 Demographics
Table 2 Resource utilization
N = 801 Age in months median (IQR) Male African American Public/self-insured Hour of arrival median (IQR)
4. Discussion
No resources (n = 392)
Resources (n = 409)
P
Resource
n (of 801)
% of 801
Dispo: home
Dispo: admit
35 (9–51) 55% 95% 98% 14 (9–20)
66 (19–113) 56% 92% 95% 14 (10–20)
P b .001 P = .889 P = .097 P = .897 P = .812
Procedures Medications Imaging Labs Other Admit/transfer
118 293 155 140 93 64
15 37 19 17 11 8
113 264 114 89 54 0
5 29 41 51 39 64
202
S.J. Fessler et al. / American Journal of Emergency Medicine 32 (2014) 199–202
Acknowledgments
Table 3 Risk of resource utilization for most common CC groups CC group
# of 801 (%)
# with resource utilization (% of 801)
Odds ratio resource utilization, OR [95%CI]
ID GI Trauma ENT
182 169 131 111
51 121 114 31
0.3 3.0 8.5 0.3
(23) (21) (16) (14)
(6) (15) (14) (4)
[0.20-0.41] [2.08-4.36] [5.0-14.5] [0.21-0.5]
a cohort that may not fully represent all low acuity EMS patients during the remaining periods. In addition, while resource utilization was used as a surrogate for a measure of acuity, it is not a direct measure. However, it has been used as a valid surrogate in other studies [6,13,15,16]. It does have its limitations, however. For instance, the study definition of resource utilization included any prescription medication that was given, which commonly included odansetron for vomiting and oral hydrocodone for pain. The decision to administer those medications in these specific children’s hospitals is largely determined by nursing triage protocols, which, in some cases, may have been distributed to patients without an urgent/emergent need. For example, occasionally patients with a mild musculoskeletal injury received an opiate medication, or a child with post-tussive emesis was given odansetron. In these types of cases, these resources utilized may be falsely inflating the rate of resource utilization. The study population was predominantly composed of patients in an urban area, African-American, with public or no insurance. It would be difficult to generalize these results to all populations. These patients likely have complex socioeconomic factors leading to use of EMS transport for low-acuity problems. This may include lack of other modes of transportation to medical care, or lack of education and resources from their primary care offices “sick care” services. Further analysis investigating why families chose EMS transport as it relates to their medical needs and perceived available resources may provide insight. While the 9-1-1 EMS dispatch triage system did not well predict overall PED resource utilization, this study identifies subgroups of pediatric complaints, such as low acuity ENT concerns, that could potentially be managed without EMS transport to a PED. These groups present an opportunity in which access to care could be designed to more efficiently use the EMS system and potentially decrease the burden on PEDs and the EMS transport system for low-acuity pediatric patients. This information can help stakeholders as they further refine pediatric based 9-1-1 EMS dispatch protocols as well as design systems and alternative pathways for managing low-acuity pediatric illness and injury. Perhaps those patients less than 24months of age, patients with respiratory complaints, or patients with pre-existing medical conditions should be triaged at a higher priority. This would require triage acuity adjustment, or using pre-established hospital nursing triage call lines, such as in other cities across the world, and increasingly across the nation. [19-24].
Special acknowledgements to Grady Paramedics Tina Wright and Kimberly Dawn Brand, who compiled the initial EMS data. References [1] Camasso-Richardson, Wilde JA, Petrack EM. Medically unnecessary pediatric ambulance transports: a medical taxi service? Acad Emerg Med 1997;4:1137–41. [2] Kost S, Arruda J. Appropriateness of ambulance transportation to a suburban pediatric emergency department. Prehosp Emerg Care 1999;3:187–90. [3] Patterson PD, Baxley EG, Probst JC, et al. Medically unnecessary Emergency Medical Services (EMS) transports among children ages 0 to 17 years. Matern Child Health J 2006;10:527–36. [4] Richard J, Osmond MH. Management outcomes of pediatric patients transported by emergency medical services in a Canadian prehospital system. Can J Emerg Med 2006;8(1):6–12. [5] Reforming Emergency Care. Department of Public Health Publications. 2001 Oct. London, England. http://www.dh.gov.u k/prod_consum_dh/gro up s/ dh_digitalassets/@dh/@en/documents/digitalasset/dh_4058836.pdf. [6] Feldman MJ, Verbeek PR, Lyons DG, et al. Comparison of the medical priority dispatch system to an out-of-hospital patient acuity score. Acad Emerg Med 2008;13(9):954–60. [7] Haines CJ. Paramedic initiated non-transport of pediatric patients. Prehosp Emerg Care 2006;10(2):213–9. [8] Hauswald M. Can paramedics safely decide which patients do not need ambulance transport or emergency department care? Prehosp Emerg Care 2002;6:387–90. [9] Schaefer RA, Rea TD, Plorde M, et al. An Emergency Medical Services program of alternate destination of patient care. Prehosp Emerg Care 2002;6(3):309–14. [10] Schmidt TA, Atcheson R, Federiuk C, et al. Evaluation of protocols allowing emergency medical technicians to determine need for treatment and transport. Acad Emerg Med 2000;7(6):663–9. [11] Pringle RP, Carden DL, Xiao F, et al. Outcomes of patients not transported after calling 911. J Emerg Med 2005;28(4):449–54. [12] Foltin GF. Critical Issues in Urban Emergency Medical Services for Children. Pediatrics Supp:174–179. [13] Hinchey P, Meyers B, Zalkin J, et al. Low acuity EMS dispatch criteria can reliably identify patients without high acuity illness or injury. Prehosp Emerg Care 2007;11(1):42–8. [14] Palumbo L, Kubincanek J, Emerman C, et al. Performance of a system to determine EMS dispatch priorities. Am J Emerg Med 1996;14:388–90. [15] Shah MN, Bishop P, Lerner EB, et al. Validation of using EMS dispatch codes to identify low acuity patients. Prehosp Emerg Care 2005;9(1):24–31. [16] Simon HK, Hirsh DA, Rogers AJ, et al. Pediatric emergency department overcrowding: electronic medical record for identification of frequent, lower acuity visitors. can we effectively identify patients for enhanced resource utilization? J Emerg Med 2009;36(3):311–6. [17] Schmidt TA, Atcheson R, Federiuk C, et al. Hospital follow up of patients categorized as not needing an ambulance using a set of emergency medical technician protocols. Prehosp Emerg Care 2001;5:366–70. [18] Yarris LM, Moreno R, Schmidt TA, et al. Reasons why patients choose an ambulance and willingness to consider alternatives. Acad Emerg Med 2006;13(4): 401–5. [19] http://www.usatoday.com/news/health/2010-06-01-emsredirect01_ST_N.htm? csp=34news. [20] Hirsh DA, Simon HK, Massey R, et al. The host hospital 24-hour underreferral rate: an automated measure of call-center safety. Pediatrics 2007;119(6):1139–44. [21] Snooks H, Williams S, Crouch R, et al. NHS emergency response to 999 calls: alternatives for cases that are neither threatening nor serious. BMJ 2002;325(7359):330–3. [22] Smith WR, Culley I, Plorde M, et al. Emergency medical services telephone referral program: an alternative approach to non-urgent 911 calls. Prehosp Emerg Care 2001;5:174–80. [23] Keatinge D, Rawlings K. Outcomes of nurse-led telephone triage service in Australia. Int J Nurs Pract 2005;11(1):5–12. [24] Turner VF, Bentley PJ, Hodgson SA, et al. Telephone triage in Western Australia. MJA 2002;176(4):100–3.
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
How well do General EMS 911 dispatch protocols predict ED resource utilization for pediatric patients?☆ Stephanie J. Fessler, MD a,⁎, Harold K. Simon, MD, MBA a, b, d, Arthur H. Yancey II, MD MPH c, d, Michael Colman, MPA c, Daniel A. Hirsh, MD a, b a
Pediatric Emergency Medicine, Children's Healthcare of Atlanta, Atlanta, GA, USA Dept of Pediatrics, Emory University, Atlanta, GA, USA Grady Health Systems, Grady Emergency Medical Services, Atlanta, GA, USA d Department of Emergency Medicine, Emory University b c
a r t i c l e
i n f o
Article history: Received 13 June 2013 Received in revised form 22 September 2013 Accepted 23 September 2013
a b s t r a c t Introduction: The use of Emergency Medical Services (EMS) for low-acuity pediatric problems is well documented. Attempts have been made to curb potentially unnecessary transports, including using EMS dispatch protocols, shown to predict acuity and needs of adults. However, there are limited data about this in children. The primary objective of this study is to determine the pediatric emergency department (PED) resource utilization (surrogate of acuity level) for pediatric patients categorized as “low-acuity” by initial EMS protocols. Methods: Records of all pediatric patients classified as “low acuity” and transported to a PED in winter and summer of 2010 were reviewed. Details of the PED visit were recorded. Patients were categorized and compared based on chief complaint group. Resource utilization was defined as requiring any prescription medications, labs, procedures, consults, admission or transfer. “Under-triage” was defined as a “lowacuity” EMS transport subsequently requiring emergent interventions. Results: Of the 876 eligible cases, 801 were included; 392/801 had no resource utilization while 409 of 801 had resource utilization. Most (737/801) were discharged to home; however, 64/801 were admitted, including 1 of 801 requiring emergent intervention (under-triage rate 0.12%). Gastroenterology and trauma groups had a significant increase in resource utilization, while infectious disease and ear-nosethroat groups had decreased resource utilization. Discussion: While this EMS system did not well predict overall resource utilization, it safely identified most low-acuity patients, with a low under-triage rate. This study identifies subgroups of patients that could be managed without emergent transport and can be used to further refine current protocols or establish secondary triage systems. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The use of Emergency Medical Services (EMS) for low-acuity pediatric problems has been well documented [1-18]. The literature reports from 16% to 60% of all pediatric transports via EMS were deemed to be unnecessary based on medical caregiver opinion, critical outcomes from Pediatric Emergency Department (PED) visit, or PED visit billing codes [1-4]. These medical transports for low-level emergencies contribute to excessive health care costs by consuming
☆ Prior presentations: Information has not been presented as an article in the past. However, the information was presented in part as a platform presentation for the Pediatric Academic Society conference in April 2012 and a poster presentation for the Society of Academic Emergency Medicine conference in May 2012. ⁎ Corresponding author. Pediatric Emergency Medicine Fellow 170 Boulevard SE D104 Atlanta, GA 30312, USA. Tel.: +1 352 262 9698. E-mail address: [email protected] (S.J. Fessler). 0735-6757/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2013.09.018
a relatively scarce resource and causing delays in transport for those who have truly time-sensitive emergencies [5,6]. Attempts have been made to curtail the EMS transports of pediatric patients with stable conditions not requiring emergent intervention. These have included employing the on-scene emergency medical technician (EMT) to determine the necessity of transport for patients to an emergency care facility [3,7-10]. Unfortunately, the majority of these studies found that this strategy resulted in undesirable outcomes, such as discordance between initial EMT assigned triage level and ultimate clinical course, under-referral rates, or over-referral rates [7-11]. While there has not been much success with EMT-driven independent on-scene triage of pediatric patients, programmatic 91-1 dispatch protocols have been shown to effectively stratify adult callers [6,12-15,17]. In these studies, as well as those in the pediatric population, hospital-based resource utilization has been used as a reasonable surrogate measure of patient acuity [16]. However,
200
S.J. Fessler et al. / American Journal of Emergency Medicine 32 (2014) 199–202
insufficient data have been gathered to confirm the efficacy of EMS dispatch protocols in guiding appropriate use of EMS for pediatric patients. The primary objective of this study is to determine the PED resource utilization of pediatric patients who were categorized as “low-acuity” by local response codes that are based on a specific 9-1-1 dispatch protocol program and then subsequently transported by EMS to a PED. A secondary objective is to determine the “under-triage rate” for this group of patients—or the rate of those children initially categorized as “low-acuity” by the lowest of the 3 response codes based on initial 9-1-1 dispatch protocols but subsequently requiring emergent resources. 2. Methods In the City of Atlanta jurisdiction, the 9-1-1 public EMS system is urban-hospital based, and utilizes the National Academy of Emergency Dispatch: Medical Priority Dispatch System v12 to process and prioritize all of the calls, then dispatch ambulances to the scene for further patient assessment and subsequent transport. It is designed to prioritize the calls of both the adult and pediatric populations. This specific EMS service designates all patients as either “emergent, lifethreatening”, “emergent, not life-threatening”, or “non-emergent, not life-threatening.” The pediatric portion of the “non-emergent, not lifethreatening” designated group was chosen as our low-acuity cohort in this study. This study had IRB approval from both the institution which oversees the EMS transport system, as well as the institution for the receiving children’s hospitals. All pediatric patients (ages less than 19 years) in this low-acuity cohort who had EMS transport from the scene to one of 3 pediatric emergency departments (PED) of a large tertiary care pediatric healthcare system from January 2010 through October 2010 were identified. Consecutive cases were reviewed in three winter and three summer months (Starting Jan 1, 2010, and July 1, 2010, respectively) to account for seasonal variability in pediatric related illness and injuries. The initial EMS information included incident address, demographic information, and any upgrade in transport acuity through the course of their EMS care. The patients’ EMS records were matched to their subsequent PED records using patient name, date of birth and date of encounter. The EMS record and PED charts were manually reviewed and recorded by a single reviewer (author SF). The patient demographics and PED course were documented. These included patient age, race and payor status; triage nursing assigned chief complaint (CC) and PED nursing assigned triage level, initial vital signs, PED resource utilization, final diagnosis and disposition. The CCs were then grouped by systems (CC Index). For the purposes of this study, PED resource utilization was used as a surrogate measure of acuity level. The study definition of resource utilization
CC Index Chief complaints classified into groups CC group Triage nurse documented CC Fever, cold, flu symptoms Constipation, vomiting, diarrhea, abdominal pain, blood in stool, swallowed foreign body, gastrostomy tube complaint TRAUMA Any extremity injury, swelling or pain; bite, burn or laceration ENT Eye drainage, tooth pain, mouth or throat pain, ear pain, foreign body in ear/nose, neck or facial swelling NEURO Crying or fussy infant; lethargy, weakness; seizures, dizziness or fainting, headache RESP Cough, choking, fast or difficulty breathing, wheezing, grunting, apnea DERM Rash, bites, abscess GU Genital abrasions, swelling or pain; urinary, vaginal or testicular symptoms. PSYCH Suicidal or homicidal ideations, “mental breakdown” HEME Sickle cell pain OTHER Broken CVL, hyperglycemia, tachycardia, medical clearance for law, alleged sexual assault, call back for prescription ID GI
included: the patient was given any medication besides acetaminophen, ibuprofen, or diphenhydramine; the patient had any laboratory study performed other than a rapid strep; for a patient for whom any procedure was performed, a consult was obtained; or the patient was admitted or transferred. Emergent intervention was defined as any immediate resuscitation, intravenous fluids, intubation or CPR performed. Those patients who were transferred from another healthcare facility, left without being seen, or with incomplete records were excluded. CC groups were compared based on resource utilization. Under-triage rate was calculated, as defined above. Categorical variables were compared via χ 2 testing and continuous variables with the Student t test. Significance was established at a P b .05. All analysis was performed on IBM SPSS Statistics for Windows, Version 20.0. (IBM Corp, Armonk, NY). 3. Results There were 876 eligible cases of pediatric patients classified as “non-emergent, non-life threatening” by initial EMS dispatch response code and subsequently transported to the pediatric emergency department. Seventy-five cases were excluded, of which 65 did not have a matching PED chart, 5 patients left without being seen, and 5 were transferred from another healthcare provider. This left 801 cases as our study cohort. Of these 801 cases, 392 (49%) of 801 did not utilize PED based resources and were all discharged to home. The remaining 409 (51%) of 801 of cases had some degree of significant resource utilization as defined a priori for this investigation. While the majority of all patients (737/801 [92%]) were discharged to home, 64 of the total 801 (8%) patients were admitted (Fig.). There was one emergent procedure (intubation) in 801 cases for an emergent undertriage rate of 0.12%. No statistically significant difference was found between the resource utilization group and the no resource utilization group when compared on race, payor status, or gender. The majority of patients were African American with public or no insurance and had a slight male predominance. However, a statistically significant difference in median age was discovered in the resource utilization group vs no resource utilization group (66 vs 35 months respectively, P b .001) (Table 1). The most common CC categories that comprised 593 (74%) of 801 of the study population were infectious disease (ID), gastroenterology (GI), trauma, and ENT. The ID complaints consisted primarily of “fever”, GI complaints mostly of vomiting, and ENT complaints of “ear pain” (CC Index). Medications, imaging, labs, and procedures were the most commonly used resources. There were 293 (37%) of 801 patients that required medications other than ibuprofen, acetaminophen or diphenhydramine. The most common medications received were an antiemetic (92/293 (31%)) and opioid analgesic pain (76/293 (26%)), with the majority (264/293 (90%)) of these patients discharged to home. There were 155 (19%) of 801 patients who had radiographic imaging, of which 132 (91%) of 155 were plain films, and 120 (83%) of 155 of these patients were discharged to home. Laboratory studies were performed in the PED on 140 (17%) of 801 patients. Procedures were done for 118 (15%) of 801 patients, mostly orthopedic (44/118 [37%]) or for laceration/wound repair (31/114 [27%]). The remaining procedures were split between foreign body removal, incision and drainage, and others (mainly gastrostomy tube replacement) (Table 2). Significant differences were found between the no resource utilization group and the resource utilization group based on CC category. For those with a GI or trauma complaint, there were higher odds of requiring resource utilization, with OR 3.0 (95% CI 2.08-4.36) for the GI group related to the common use of ondansetron for vomiting; and OR 8.5 (5.00-14.51) for the trauma group—related to
S.J. Fessler et al. / American Journal of Emergency Medicine 32 (2014) 199–202
201
Fig. Distribution of Cases.
empirical antibiotics and intubation prior to transport to a pediatric intensive care unit with a multilobar pneumonia, empyema, and Streptococcus pneumonia bacteremia.
wound or orthopedic injuries. However, those with ID or ENT complaints had significant decreased resource utilization, OR 0.28 (0.20-0.41) and OR 0.3 (0.21-0.50), respectively (Table 3). There were 64 (8%) of 801 patients admitted from the emergency room. There were a variety of ages, CCs, and ultimate diagnoses. The overwhelming majority of these patients (63/64) were admitted to the floor for non-emergent treatments. Many required admission for intravenous antibiotics due to neonatal fever, skin or soft tissue infection, or urinary tract infection. A large portion of the admitted patients had pre-existing medical conditions, such as known psychiatric disorder, sickle cell disease, ventriculoperitoneal shunt, or short gut syndrome. Also included in this group was a patient with a retained nasal foreign body, a patient with a tibia fracture requiring surgical repair, one patient with ingestion of cocaine requiring observation, or prolonged post-ictal state after seizure. Of note, 32 (50%) of 64 were 24 months of age or less. There was one patient who required emergent intervention—a 2year-old girl with respiratory failure and sepsis. The CC of this 2-yearold female with sepsis and respiratory failure was 1 week of upper respiratory symptoms and progressive cough which resulted in her categorization by EMS dispatch as “non-emergent, not life-threatening.” She was not only under-triaged according to our study definition; her degree of illness was not readily recognized by EMS personnel. The EMTs involved in her evaluation and transport did NOT upgrade the child’s acuity status to a more urgent level after arrival to scene. No interventions were done by EMS en route. In the PED, the child had vital signs and physical exam findings consistent with decompensated shock and respiratory distress. She received an IO for access, rapid intravenous fluid resuscitation, nebulized medications,
In this investigation of this specific 9-1-1 EMS dispatch triage system, the dispatch protocols applied to pediatric patients resulting in classification as “non-emergent, not life-threatening” did not accurately predict PED resource utilization, based on study definitions, with 409/801 (51%) requiring some degree of PED based resource utilization. However, this system did accurately identify in most cases the lowestrisk patients requesting evaluation and transport to a PED for evaluation, evidenced by a low under-triage rate. While the emergent under-triage rate was only 0.12%, unrecognized significant illness can occur to a minimal degree, based on present protocols. As seen here, such occasional cases can also be undetected by EMTs. It is unclear as to why the degree of the one critical patient’s illness was not recognized by the EMS personnel involved in her care. Perhaps this is due to lack of pediatric experience among this team of medics. In addition, staff may be de-conditioned by the larger majority of patients they transport in the area that are truly not emergent. This study is limited due to being a retrospective chart review; however, it did not influence physician practice or triage protocols. The charts were reviewed in respective summer and winter months, a decision made a priori to chart review, in order to account for the extremes in seasonal variability. However, this may have resulted in
Table 1 Demographics
Table 2 Resource utilization
N = 801 Age in months median (IQR) Male African American Public/self-insured Hour of arrival median (IQR)
4. Discussion
No resources (n = 392)
Resources (n = 409)
P
Resource
n (of 801)
% of 801
Dispo: home
Dispo: admit
35 (9–51) 55% 95% 98% 14 (9–20)
66 (19–113) 56% 92% 95% 14 (10–20)
P b .001 P = .889 P = .097 P = .897 P = .812
Procedures Medications Imaging Labs Other Admit/transfer
118 293 155 140 93 64
15 37 19 17 11 8
113 264 114 89 54 0
5 29 41 51 39 64
202
S.J. Fessler et al. / American Journal of Emergency Medicine 32 (2014) 199–202
Acknowledgments
Table 3 Risk of resource utilization for most common CC groups CC group
# of 801 (%)
# with resource utilization (% of 801)
Odds ratio resource utilization, OR [95%CI]
ID GI Trauma ENT
182 169 131 111
51 121 114 31
0.3 3.0 8.5 0.3
(23) (21) (16) (14)
(6) (15) (14) (4)
[0.20-0.41] [2.08-4.36] [5.0-14.5] [0.21-0.5]
a cohort that may not fully represent all low acuity EMS patients during the remaining periods. In addition, while resource utilization was used as a surrogate for a measure of acuity, it is not a direct measure. However, it has been used as a valid surrogate in other studies [6,13,15,16]. It does have its limitations, however. For instance, the study definition of resource utilization included any prescription medication that was given, which commonly included odansetron for vomiting and oral hydrocodone for pain. The decision to administer those medications in these specific children’s hospitals is largely determined by nursing triage protocols, which, in some cases, may have been distributed to patients without an urgent/emergent need. For example, occasionally patients with a mild musculoskeletal injury received an opiate medication, or a child with post-tussive emesis was given odansetron. In these types of cases, these resources utilized may be falsely inflating the rate of resource utilization. The study population was predominantly composed of patients in an urban area, African-American, with public or no insurance. It would be difficult to generalize these results to all populations. These patients likely have complex socioeconomic factors leading to use of EMS transport for low-acuity problems. This may include lack of other modes of transportation to medical care, or lack of education and resources from their primary care offices “sick care” services. Further analysis investigating why families chose EMS transport as it relates to their medical needs and perceived available resources may provide insight. While the 9-1-1 EMS dispatch triage system did not well predict overall PED resource utilization, this study identifies subgroups of pediatric complaints, such as low acuity ENT concerns, that could potentially be managed without EMS transport to a PED. These groups present an opportunity in which access to care could be designed to more efficiently use the EMS system and potentially decrease the burden on PEDs and the EMS transport system for low-acuity pediatric patients. This information can help stakeholders as they further refine pediatric based 9-1-1 EMS dispatch protocols as well as design systems and alternative pathways for managing low-acuity pediatric illness and injury. Perhaps those patients less than 24months of age, patients with respiratory complaints, or patients with pre-existing medical conditions should be triaged at a higher priority. This would require triage acuity adjustment, or using pre-established hospital nursing triage call lines, such as in other cities across the world, and increasingly across the nation. [19-24].
Special acknowledgements to Grady Paramedics Tina Wright and Kimberly Dawn Brand, who compiled the initial EMS data. References [1] Camasso-Richardson, Wilde JA, Petrack EM. Medically unnecessary pediatric ambulance transports: a medical taxi service? Acad Emerg Med 1997;4:1137–41. [2] Kost S, Arruda J. Appropriateness of ambulance transportation to a suburban pediatric emergency department. Prehosp Emerg Care 1999;3:187–90. [3] Patterson PD, Baxley EG, Probst JC, et al. Medically unnecessary Emergency Medical Services (EMS) transports among children ages 0 to 17 years. Matern Child Health J 2006;10:527–36. [4] Richard J, Osmond MH. Management outcomes of pediatric patients transported by emergency medical services in a Canadian prehospital system. Can J Emerg Med 2006;8(1):6–12. [5] Reforming Emergency Care. Department of Public Health Publications. 2001 Oct. London, England. http://www.dh.gov.u k/prod_consum_dh/gro up s/ dh_digitalassets/@dh/@en/documents/digitalasset/dh_4058836.pdf. [6] Feldman MJ, Verbeek PR, Lyons DG, et al. Comparison of the medical priority dispatch system to an out-of-hospital patient acuity score. Acad Emerg Med 2008;13(9):954–60. [7] Haines CJ. Paramedic initiated non-transport of pediatric patients. Prehosp Emerg Care 2006;10(2):213–9. [8] Hauswald M. Can paramedics safely decide which patients do not need ambulance transport or emergency department care? Prehosp Emerg Care 2002;6:387–90. [9] Schaefer RA, Rea TD, Plorde M, et al. An Emergency Medical Services program of alternate destination of patient care. Prehosp Emerg Care 2002;6(3):309–14. [10] Schmidt TA, Atcheson R, Federiuk C, et al. Evaluation of protocols allowing emergency medical technicians to determine need for treatment and transport. Acad Emerg Med 2000;7(6):663–9. [11] Pringle RP, Carden DL, Xiao F, et al. Outcomes of patients not transported after calling 911. J Emerg Med 2005;28(4):449–54. [12] Foltin GF. Critical Issues in Urban Emergency Medical Services for Children. Pediatrics Supp:174–179. [13] Hinchey P, Meyers B, Zalkin J, et al. Low acuity EMS dispatch criteria can reliably identify patients without high acuity illness or injury. Prehosp Emerg Care 2007;11(1):42–8. [14] Palumbo L, Kubincanek J, Emerman C, et al. Performance of a system to determine EMS dispatch priorities. Am J Emerg Med 1996;14:388–90. [15] Shah MN, Bishop P, Lerner EB, et al. Validation of using EMS dispatch codes to identify low acuity patients. Prehosp Emerg Care 2005;9(1):24–31. [16] Simon HK, Hirsh DA, Rogers AJ, et al. Pediatric emergency department overcrowding: electronic medical record for identification of frequent, lower acuity visitors. can we effectively identify patients for enhanced resource utilization? J Emerg Med 2009;36(3):311–6. [17] Schmidt TA, Atcheson R, Federiuk C, et al. Hospital follow up of patients categorized as not needing an ambulance using a set of emergency medical technician protocols. Prehosp Emerg Care 2001;5:366–70. [18] Yarris LM, Moreno R, Schmidt TA, et al. Reasons why patients choose an ambulance and willingness to consider alternatives. Acad Emerg Med 2006;13(4): 401–5. [19] http://www.usatoday.com/news/health/2010-06-01-emsredirect01_ST_N.htm? csp=34news. [20] Hirsh DA, Simon HK, Massey R, et al. The host hospital 24-hour underreferral rate: an automated measure of call-center safety. Pediatrics 2007;119(6):1139–44. [21] Snooks H, Williams S, Crouch R, et al. NHS emergency response to 999 calls: alternatives for cases that are neither threatening nor serious. BMJ 2002;325(7359):330–3. [22] Smith WR, Culley I, Plorde M, et al. Emergency medical services telephone referral program: an alternative approach to non-urgent 911 calls. Prehosp Emerg Care 2001;5:174–80. [23] Keatinge D, Rawlings K. Outcomes of nurse-led telephone triage service in Australia. Int J Nurs Pract 2005;11(1):5–12. [24] Turner VF, Bentley PJ, Hodgson SA, et al. Telephone triage in Western Australia. MJA 2002;176(4):100–3.
