Pediatric Transport Triage: Development and Assessment of an Objective Tool to Guide Transport Planning

Abstract

Objectives:

We developed a Pediatric Transport Triage Tool (PT3) to objectively guide selection of team composition and transport mode, thereby standardizing transport planning. Previously, modified Pediatric Early Warning Scores for transport have been used to assess illness severity, but not to guide transport decision-making.

Methods:

PT3 was created for pediatric transport by combining objective evaluations of Neurologic, Cardiovascular, and Respiratory (NCR) systems with a systems-based medical condition list to identify diagnoses requiring expedited transport and/or advanced team composition not captured by NCR systems alone. A scoring algorithm was developed to guide transport planning. Transport data (mode, team composition, time to dispatch, patient disposition and complications) were collected before and after PT3 implementation at a single tertiary center over an 18-month period.

Results:

We reviewed 2,237 inbound pediatric transports. Transport mode, patient disposition, and dispatch time were unchanged over the study period. Fewer calls using a transport nurse were noted after PT3 implementation (33.9% vs. 30%, p=0.05), with a trend toward fewer rotor-wing transports and transports requiring physicians. The majority of users, regardless of experience level, reported improved transport standardization with the tool. Need to upgrade team composition or mode during transport was not different during the study period. No adverse patient safety events occurred with PT3 use.

Conclusions:

PT3 represents an objective triage tool to reduce variability in transport planning. PT3 decreased resource utilization and was not associated with adverse outcomes. Teams with dynamic staffing models, various experience levels, and multiple transport modes may benefit from this standardized assessment tool.

INTRODUCTION

Interhospital transport plays an important role in facilitating access to healthcare services, particularly in pediatrics where acute specialized care is highly centralized. Risks associated with interhospital transport (e.g., lack of physiologic stability prior to or during transport, logistical limitations related to providing care in a moving vehicle) increase the potential for patient deterioration,^1–3 which has been estimated to occur in approximately 6–12% of transports.^3,4 Safe and efficient transport is facilitated through selection of an appropriate team composition and transport mode. However, planning for transport is a complex process which relies on accurate assessment of a child’s clinical condition, clear communication between referring providers and the transport team, resource availability, and environmental conditions.

Objective transport assessment tools have been used successfully to predict 7-day mortality risk in neonates^5,6 and physiologic stability in adults,^7,8 and to guide medical decision-making prior to adult interhospital transport.^8–11 Pediatric transport tools have been developed to predict severity of illness, need for pediatric intensive care unit (PICU) admission, and in-hospital mortality.^12–15 Pediatric Risk of Mortality (PRISM) scores underestimated both need for major interventions and occurrence of unplanned events during pediatric transport,^16,17 whereas simple physiologic pre-transport parameters correlated with such events.¹⁸ Pediatric Early Warning Scores (PEWS), first used in inpatient hospital populations to identify patients at risk for clinical deterioration,^19–21 have been modified and successfully used to determine illness severity during transport, but not to guide transport planning.¹⁵ Recently, Singh et al reported that a higher transport PEWS score was only modestly associated with in-transport critical events, and was inferior to the presence of pre-transport cardiovascular instability and need for mechanical ventilation.²²

Pediatric transport intake and dispatch at our institution is completed by team members with different levels of experience, leading to variable over-the-phone patient assessment and decision-making about transport mode and team composition. As efforts to reduce variation and standardize patient care have been shown to improve quality of care, efficiency, and resource utilization within the intensive care unit (ICU),²³ we hypothesized they would lead to similar outcomes in the transport setting. We sought to develop a triage tool that would limit variability during intake, standardize transport planning (regardless of provider background or experience), reduce risk of adverse events, and optimize resource utilization.

Our goals were to: 1) modify the PEWS to objectively guide pediatric transport planning, rather than specifically predict adverse events or clinical deterioration; and 2) evaluate the tool’s impact on transport practice and outcomes at our institution.

MATERIALS AND METHODS

Development and Implementation of the Pediatric Transport Triage Tool (PT3)

Pediatric Transport Service

Our pediatric transport team receives approximately 3,000 requests annually to transport both critically ill and non-critically ill patients one day to 22 years old. The majority of referring facilities are urban community hospitals located within a 25-mile radius from our institution; therefore most transports are performed by ground ambulance. Less than 10% of transports are performed by air ambulance (rotor- or, much less frequently, fixed-wing aircraft).

Our team’s composition varies, depending mainly on the patient’s severity of illness and age, but also on the mode of transport. The Maryland Institute for Emergency Medical Services Systems (MIEMSS) mandates that all pediatric and adult interhospital transports performed by ground ambulance be staffed by a team that includes a paramedic and an emergency medical technician (EMT). Therefore, our lower-acuity pediatric ground transports are generally performed by a team composed of a pediatric-trained paramedic and EMT only, whereas our critical care pediatric ground transports are performed by an advanced team, which also includes a pediatric transport nurse, with or without a pediatric-trained respiratory therapist (RT) and/or a pediatric critical care physician (almost always a fellow). While paramedics are required by MIEMSS to maintain certification for endotracheal intubation for on-scene events, physicians generally perform all advanced airway procedures during interhospital pediatric transport. As such, physicians are dispatched for all transports anticipated to require advanced airway skills or complex minute-to-minute medical management due to significant physiologic instability. Table 1 details team staffing models based on patient age and anticipated medical needs during ground transport. Our pediatric air transports are generally performed by a team composed of a flight paramedic and a flight nurse (employed by the air vendor company); however, air transports of younger or higher-acuity patients are usually joint missions performed by an advanced team, which also includes our institution’s pediatric transport nurse, with or without a physician.

Table 1.

Team compositions for interhospital transport by ground of patients one day to 22 years old

Team Composition	Transport Examples
EMT Paramedic	• Patients >28 days old unlikely to require critical care interventions during transport, such as:     ○ Certain surgical emergencies (e.g., ovarian/testicular torsion, digit amputation)     ○ Minor burns (not requiring fluid resuscitation, or airway management)     ○ Minor trauma (e.g., fractures, lacerations)
EMT Paramedic Pediatric transport nurse	• Patients likely to require titration of medical therapies and non-invasive respiratory support per internal pediatric transport protocols, such as:     ○ Bronchiolitis requiring HFNC support     ○ Burns requiring fluid resuscitation     ○ DKA requiring fluid and insulin titration     ○ Febrile neutropenia requiring antibiotics     ○ Status asthmaticus requiring continuous nebulized bronchodilators
EMT Paramedic Pediatric transport nurse Pediatric critical care physiciana	• Patients likely to require procedural interventions during transport, such as:     ○ Chest tube placement     ○ Intubation     ○ Pericardiocentesis • Patients likely to require extensive minute-to-minute medical management, and/or critical care interventions during transport, such as:     ○ Multisystem trauma with unstable hemodynamics     ○ Refractory shock requiring titration of vasopressors     ○ Severe traumatic brain injury requiring management of ICH
EMT Paramedic Pediatric transport nurse Transport respiratory therapist Pediatric critical care physiciana	• Patients with known or anticipated difficult airway and potential for airway compromise • Patients with PHTN requiring iNO titration +/− invasive/non-invasive respiratory support • Patients with upper airway obstruction requiring helium-oxygen mixtures (Heliox) • Special transports, such as:     ○ Patients with ARDS requiring HFV     ○ Patients in cardiac arrest requiring CPR     ○ Patients on ECMO or VAD support     ○ Patients requiring palliative critical care transports home for terminal extubation

ARDS = acute respiratory distress syndrome, CPR = cardiopulmonary resuscitation, DKA = diabetic ketoacidosis, ECMO = extracorporeal membrane oxygenation, EMT = emergency medical technician, HFNC = high flow nasal cannula, HFV = high frequency ventilation, ICH = intracranial hypertension, iNO = inhaled nitric oxide, PHTN = pulmonary hypertension, VAD = ventricular assist device.

^aPediatric critical care physicians are usually fellows. Pediatric critical care attendings staff only the ECMO transports.

Relatively few (17%) incoming pediatric transports are completed by private vehicles or outside transport services; most of these are low-acuity patients (usually older children or adolescents) deemed safe to transport one-way, without specialized pediatric or critical care transport expertise.

Transport triage is performed by transport nurses or pediatric critical care fellows who obtain demographic and clinical information, determine appropriate patient disposition, team composition, and transport mode; and then dispatch the team to retrieve the patient. Pediatric critical care physicians (attendings or fellows) provide medical control during transport. While our team members are generally experienced pediatric critical care providers, varied degrees of transport experience exist at all levels due to staffing turn-over, particularly among our physician trainees, transport nurses, and RTs.

Development of Initial PT3

The PT3 was created by modifying the physiologic parameters of the PEWS^19–21 for the Neurologic, Cardiovascular, and Respiratory (NCR) systems to include factors that would impact transport decision-making (e.g., need for sedatives and paralytics, vasoactive infusions, non-invasive or invasive respiratory support). Based on the NCR systems, a scoring algorithm (range: 0–9) was developed to guide selection of team composition and transport mode (including the option of one-way transport by private vehicle or an outside transport service for stable, low-acuity patients). Additionally, a list of medical conditions requiring either expedited transport or pediatric critical care transport expertise was created by the authors based on their transport experience and after consultation with specialists from a variety of pediatric services (e.g., critical care, cardiology, emergency medicine, gastroenterology, general surgery, hematology-oncology, neurology, neurosurgery, ophthalmology). This systems-based list incorporated conditions not captured by the NCR systems alone, and called for either “fastest mode of transportation” (A), or “fastest mode of transportation and advanced team composition” (B). The PT3 was developed with a patient data form to facilitate clinical data collection for completion of PT3 scoring (see Table, Supplemental Digital Content 1, which illustrates the initial Pediatric Transport Triage Tool and patient data collection form).

Pediatric Transport Team Training

All transport team nurses and pediatric critical care fellows received individual hands-on training prior to initial PT3 implementation. Team members were advised to follow the transport logistics (team composition and transport mode) recommended by the PT3 scoring, unless clinical judgment indicated that a faster mode or more advanced team were necessary. If in doubt, transport team members were instructed to choose a conservative approach with a faster mode of transportation, a more advanced team composition (including addition of a transport nurse, physician, or RT), or both a faster mode of transport and a more advanced team composition.

Initial Use and Modifications to the PT3

During a 7-month beta testing phase (June 1, 2014 – December 31, 2014) all transports were reviewed and feedback was collected from transport staff to identify safety risks, over- or under-triage, and uncertainty about scoring. As part of ongoing quality improvement, the tool underwent several minor modifications. The modified tool allowed providers to account for tachycardia due to fever, pain, or beta-agonist use (which consistently elevated PT3 scores, leading to over-triage); downgraded use of high-flow nasal cannula from the highest score of “3” to a “2” in the respiratory category (as transport resources were over-utilized for children on this mode of support); and included acidosis, hyperkalemia, anemia, and thrombocytopenia with bleeding on the list of significant diagnoses. Table 2 contains the final modified version of the PT3, including the scoring algorithm.

Table 2.

Pediatric Transport Triage Tool: Neurologic, Cardiovascular and Respiratory Systems, Significant Diagnoses List, and Scoring Algorithm

Pediatric Transport Triage Tool (PT3)
Neurologic, Cardiovascular and Respiratory (NCR) systems
System	Score
	0	1	2	3
Neurologic Baseline GCS	• Alert, playful, interactive • GCS 15 • At baseline	• Sleepy, but responds to interventions • Crying, but consolable • GCS 13–15	• Irritable, inconsolable • Cannot maintain tone • New focal deficit • GCS ≤ 12, or > 2 less than baseline	• Lethargic, confused • Loss of cough or gag • GCS ≤ 10, or > 3 less than baseline • Sedated or paralyzed
Cardiovascular Baseline HR Baseline BP	• Pink • Capillary refill 1–2 seconds • HR and/or BP < 10% above a or below baseline	• Pale • Capillary refill 3 seconds • HR and/or BP 10% above a or below baseline	• Mottled • Capillary refill 3–4 seconds • HR and/or BP 20% above a or below baseline • Unrepaired, or shunt dependent CHD	• Gray • Capillary refill <1, or >4 seconds • HR and/or BP 30% above a or below baseline • Requiring vasoactive medications • Unstable arrhythmias
Respiratory Baseline RR	• Normal RR • No WOB	• RR > 10/minute above normal • Accessory muscle use • Nasal flaring • Stridor with agitation • < 3L O2	• RR > 20/minute above normal • Stridor at rest • ≥ 3L O2 • HFNC • Continuous nebs • Tracheostomy/ventilator dependent, or respiratory muscle weakness with any change from baseline	• RR > 5/minute below normal • Hypoventilation with AMS • Grunting • NIPPV with BiPAP • MV
Significant Diagnoses List
Airway	Neurologic	Gastrointestinal	Surgical, Trauma and Burns	Other
• Pneumothorax (A) • Upper airway obstruction (B) • History of difficult airway with respiratory compromise (B) • Unstable mandibular fracture with potential for airway compromise (B)	• VPS malfunction (A) • New intracranial mass (A) • Intracranial hemorrhage (A) • Stroke (A) • Acute spinal cord injury, or disease (A) • Uncontrolled seizures (B)	• Esophageal foreign body (A) • Volvulus/malrotation (A) • Perforated appendix (A) • Incarcerated hernia (A)	• Burns > 5% BSA to face/neck, > 10% total BSA, or circumferential (A) • Orbital fracture with abnormal EOM (A) • Severe splenic/liver laceration (A) • Compartment syndrome (A) • Neurovascular compromise (A) • Testicular/ovarian torsion < 6 hours (A) • Amputation: digit (A); limb (B) • Inhalational injury/burn (B) • Multisystem trauma (B) • Penetrating injuries to head/chest/abdomen (B) • Pelvic fracture (B)	• Immunocompromised with potential infection (A) • Acute-onset visual loss (A) • Retrobulbar hematoma (A) • DKA with AMS (B) • Age ≤ 28 days (B) • Potassium ≥ 6.5 mmol/L (non-hemolyzed sample) (B) • Ph < 7.2 (B) • Hemoglobin ≤ 5 g/dL (B) • Platelets ≤ 20,000/μL with bleeding (B)
PT3 Scoring Algorithm b
Highest Single Category Score		Total Score		Significant Diagnoses List
3 = nurse±fellow call + fastest mode 2 = paramedic±nurse call + fastest mode 1 = paramedic call 0 = paramedic call, or one-way		≥ 5 = fellow call + fastest mode 4 = nurse±fellow call + fastest mode 3 = nurse call + fastest mode 2 = paramedic±nurse call, consider fastest mode 1 = paramedic call 0 = paramedic call, or one-way		A = fastest mode B = nurse±fellow + fastest mode

AMS indicates altered mental status; BiPAP, bilevel positive airway pressure; BP, blood pressure; BSA, body surface area; CHD, congenital heart disease; DKA, diabetic ketoacidosis; EOM, extraocular movements; GCS, Glasgow Coma Scale; HFNC, high-flow nasal cannula; HR, heart rate; MV, mechanical ventilation; NIPPV, non-invasive positive pressure ventilation; O2, oxygen; RR, respiratory rate; VPS, ventriculoperitoneal shunt; WOB. work of breathing.

^aIn the absence of fever, pain, or β-agonist use.

^bFor use: Providers score patient based on the NCR systems. Highest score in any single category (range: 0–3) and total score (range: 0–9) are calculated. The significant diagnoses list is reviewed and, if a diagnosis applies to the patient, an A or B score is assigned. Transport decision-making is determined by selecting the most advanced team composition and fastest transport mode as determined by the highest score in any of the three scoring areas (highest single category score, total score, significant diagnoses list)

Retrospective Data Collection

Retrospective data collection at this tertiary care center was performed for a 5-month period (January 1, 2014 – May 31, 2014) prior to PT3 use and a 6-month period (January 1, 2015 – June 30, 2015) after PT3 underwent modifications from the beta testing phase of the project. All patients one day to 22 years old transported to our institution were identified from our Pediatric Transport database. This dataset was cross-referenced with our Pediatric Emergency Department (PED) and PICU electronic medical record databases. In our transport system, a separate neonatal team transports premature and full-term newly borns from referring facilities to a group of level III and IV neonatal intensive care units (NICUs). For this study, the PT3 was used by the pediatric transport team only, therefore neonates transported to our NICU were excluded. We collected individual patient data regarding the transport process (team composition, transport mode, time to dispatch, complications during transport) and disposition after transport (PED, or direct admission to either pediatric ward or PICU). We also reviewed all denials from lack of transport resources or appropriate inpatient bed availability (sometimes due to inadequate staffing), as well as all “unanticipated PICU admissions” (patients triaged to the PED with anticipated subsequent admission to the ward, who required unplanned PICU admission within 24 hours due to unexpected acuity or progression of illness).

Transport Staff Survey

All transport nurses and PICU fellows were invited to complete an electronic survey at the conclusion of data collection (see Document, Supplemental Digital Content 2, which includes the survey instrument). Participants were asked to assess PT3 ease of use and impact on transport planning and resource utilization, and rate their confidence in decision-making and overall satisfaction with the tool.

Statistical Analysis

Descriptive statistics were expressed as mean ± standard deviation and number of patients (percentage in parenthesis). Poisson mean testing was used to compare the number of transports in the pre- and post-intervention periods. Fisher’s exact test and the Mann-Whitney U test were used to assess comparisons of categorical data and interval variables, respectively. All data analysis was completed using Stata V12.1 (StataCorp LP, College Station, TX). A two-sided p value of <0.05 was considered to be statistically significant. This study was acknowledged as a quality improvement project by The Johns Hopkins Hospital Institutional Review Board.

RESULTS

We reviewed 2,237 inbound pediatric transports: 958 from the pre-intervention period (pre-PT3) and 1,279 after modifications to the PT3 were made (post-PT3). Characteristics of the inbound transports during the study period are listed in Table 3. The mean number of monthly transports increased significantly over the study period (192 calls/month pre-PT3, 213 calls/month post-PT3; p=0.01).

Table 3.

Characteristics of Pediatric Transports during the Study Period

Pediatric Transport Data			Pre-PT3	Post-PT3	p Value
Total transfer requests a (monthly mean±SD)			1,137 (227.4±45)	1,478 (246.3±17)	0.04
Total inbound transports (monthly mean±SD)			958 (191.6±27.1)	1,279 (213.2±22.3)	0.01
	Patient triage disposition	PED, n (%)	788 (82.2)	1,027 (80.3)	0.28
		PICU, n (%)	166 (17.3)	214 (16.7)	0.73
		Other (Ward, OR, AED, NICU), n (%)	5 (0.5)	38 (3)	<0.001
	PICU admissions from transport	Total patients admitted to PICU b, n (%)	221 (23.1)	299 (23.4)	0.88
		“Unanticipated PICU admissions”, n (%)	28 (2.9)	41 (3.2)	0.81
	Transport team composition	JHH paramedic c, n (%)	729 (76.1)	978 (76.5)	0.84
		JHH pediatric transport nurse, n (%)	325 (33.9)	383 (30)	0.05
		JHH pediatric critical care physician, n (%)	82 (8.6)	92 (7.2)	0.23
	Transport mode	Ground, n (%)	874 (91.2)	1,186 (92.7)	0.21
		Air, n (%)	84 (8.8)	93 (7.3)	0.21
	Complications and adverse outcomes	Need to upgrade team composition, or transport mode, n (%)	2 (0.2)	0 (0)	0.18
		Patient safety report, n (%)	0 (0)	0 (0)
		Death in transport, n (%)	0 (0)	0 (0)
	Dispatch time, minutes, mean±SD		29.8±33.1	35.8±73.9	0.26
Denials, n (%)			52 (4.6)	40 (2.7)	0.007

AED indicates adult emergency department; JHH, Johns Hopkins Hospital; NICU, neonatal intensive care unit; OR, operating room; PED, pediatric emergency department; PICU, pediatric intensive care unit; PT3, Pediatric Transport Triage Tool.

^aIncludes transports to our institution (inbound transports), reverse transports to other facilities, consults only, and denials.

^bIncludes patients directly triaged to the PICU; patients with planned PICU admission, yet brought to the PED for trauma evaluation or due to lack of immediately available PICU bed; and unanticipated PICU admissions.

^cOur paramedics staff all ground transports unless patients are brought by private vehicle, one-way transport, or a different ambulance vendor.

Transport Mode and Team Composition

There were no statistical differences in the proportion of air transports (8.8% pre-PT3 vs. 7.3% post-PT3; p=0.21) and transports that included a paramedic (76.1% pre-PT3 vs. 76.5% post-PT3; p=0.84) or a physician (8.6% pre-PT3 vs. 7.2% post-PT3; p=0.23) over the study period. There were, however, fewer transports that included a nurse in the post-PT3 period (33.9% pre-PT3 vs. 30% post-PT3; p=0.05).

Patient Triage Location, Team Dispatch Time, and Patient Disposition

The proportions of patients triaged to the PICU (17.3% pre-PT3 vs. 16.7% post-PT3; p=0.73) and the PED (82.2% pre-PT3 vs. 80.3% post-PT3; p=0.28) were unchanged; however, the proportion of patients triaged to the pediatric ward increased (0.5% pre-PT3 vs. 3% post-PT3; p<0.001). There was no difference in the proportion of patients eventually admitted to the PICU (23.1% pre-PT3 vs. 23.4% post-PT3; p=0.88), or the percentage of “unanticipated PICU admissions” (2.9% pre-PT3 vs. 3.2% post-PT3; p=0.81). Mean dispatch time did not differ over the study period (29.8 ± 33.1 minutes pre-PT3 vs. 35.8 ± 73.9 minutes post-PT3; p=0.26).

Complications during Transport

There were no unplanned intubations, events requiring cardiopulmonary resuscitation, or deaths in transport during either period. There were also no adverse patient safety reports filed in the hospital voluntary event reporting system. Need to upgrade mode of transportation or team composition during transport was uncommon, and not significantly different (2 events pre-PT3 vs. 0 events post-PT3; p=0.18).

Transport Denials

Transport denials were more common prior to institution of the PT3 (pre-PT3 52 transports, 4.6% vs. post-PT3 40 transports, 2.7%; p=0.007). While lack of hospital or ICU beds prevented transport in the bulk of cases (90.4% pre-PT3 vs. 97.5% post-PT3), the percentage of cases denied due to lack of transport team availability decreased (pre-PT3 1.9% vs. post-PT3 0%), despite the increased transport volume.

Transport Staff Survey

Ninety-one percent (31/34) of transport staff (16/17 physicians, 15/17 nurses) completed the survey. Most agreed or strongly agreed that the PT3 tool was straightforward to use (26/31, 84%); not time-consuming (22/31, 71%); and helpful in making decisions about team composition (25/31, 81%) and transport mode (22/31, 71%). Moreover, many nurses agreed or strongly agreed that the tool reduced variability in transport planning (10/15, 66%) and disagreements about choice of team composition and transport mode (9/15, 60%), and therefore most felt that the tool increased their confidence in transport planning (11/15, 73%). The perceived disadvantages were that the PT3 led to unnecessary use of personnel (13/16 physicians, 81%; 6/15 nurses, 40%) and made planning for transport slower (7/31, 23%); however, data suggest reduced utilization of personnel and no significant difference in dispatch time.

DISCUSSION

Tools previously developed for pediatric transport have focused on patient assessment and predicting severity of illness. To our knowledge, no group has reported using such a tool to guide decision-making regarding team composition and transport mode. The complex logistics of pediatric interhospital transport depend upon a myriad of factors pertaining to the patient (e.g., age, weight, medical/surgical history); disease process (e.g., severity of illness, disease progression, complications); referring and receiving institutions (e.g., available resources, interfacility distance); environment (e.g., traffic, weather); and transport team’s resources (e.g., personnel, equipment/medications, transport vehicles) and capabilities (e.g., clinical and procedural skills, team dynamics, communication). Transport planning is further complicated by the need to accurately assess the patient from afar and anticipate complications during a potentially lengthy transport. The PT3 allows teams to approach transport planning objectively, and thus reduce variability. Standardized clinical pathways implemented in critical care and transport medicine have been shown to improve quality of care and efficiency, and reduce both adverse events and resource utilization,^23–28 hence the impetus to develop a tool targeting the pediatric interhospital transport triage and planning process.

In this study, we modified the PEWS for the transport environment. Other modifications have been shown to reliably predict severity of illness;¹⁵ however, they were found to be less helpful in anticipating adverse events in transport.²² Our tool included modifications of the NCR scoring system to account for factors impacting transport decision-making and added a list of conditions not captured by NCR scoring that would benefit from either expedited transport or an advanced transport team. Additionally, this tool was accompanied by a transport intake form to facilitate consistent clinical data collection to complete the PT3.

We noted that the number of patients transported increased over the study period. While our institution did not consistently use severity of illness scoring to monitor patient acuity at the time of this study, the proportion of patients admitted to the PICU from transport was unchanged, suggesting lack of dramatic differences in severity of illness over the study period. The percentages of patients triaged to the PED and PICU were unchanged, whereas the increased percentage of patients triaged to the pediatric ward reflects a hospital-wide effort to begin directly admitting stable patients. Prior to this effort, patients arriving from transport could only undergo evaluation in the PED or be directly admitted to the PICU.

We utilized “unanticipated PICU admissions” as a surrogate for potential inappropriate transport triage, as triage to the PED with need for subsequent unplanned PICU admission may indicate poor discrimination of the PT3 to identify critically ill patients. The percentage of these events was unchanged in the pre- and post-intervention periods. There was no increase in the frequency of reported adverse outcomes (need to upgrade team composition or transport mode, and/or unplanned intubation, cardiopulmonary resuscitation, or death in transport) with the use of the PT3; however, these events were rare prior to PT3 implementation. Unfortunately, it was not possible to capture other adverse events occurring during transport (e.g., physiologic deterioration, alteration in mental status, need to escalate respiratory support), which might better evaluate the tool’s impact on transport decision-making and safety. Such events, however, are also relatively rare, and thus may not be fully informative given their overall infrequency. Additionally, while the lack of adverse patient safety reports filed in the hospital voluntary event reporting system is encouraging, this may reflect incomplete institutional reporting of all safety events (e.g., medication errors, equipment malfunctions).

Our results indicate that resource utilization incrementally decreased with use of the PT3 tool. The percentage of nurses dispatched on transports was lower in the post-intervention period, albeit of borderline significance (p=0.05). Transports via rotor-wing vehicle were reduced, and use of physicians on transport was also less frequent, although not statistically significant. Given our large sample size and lack of changes impacting our transport program and catchment population over the study period, we have reason to believe that the PT3 may lead to improved use of resources, thus generating cost savings. The 1.5% fewer air transports (translating to 40.5 fewer flights per year) could yield yearly cost savings of ~$567,000, whereas the 4% fewer nurse calls (totaling 102 fewer nurse calls per year) could generate additional savings of ~$41,000 annually in our program. Moreover, judicious use of transport nurses may increase their availability for sicker transports, potentially leading to fewer denials for lack of team availability. Additionally, there were 1.4% fewer fellow calls in the post-intervention period. Whereas trainees have no billing privileges (hence no financial impact), dispatching fewer fellows on transport may improve PICU staffing and in-hospital throughput.

Standardization of care has been shown to enhance quality of care and resource utilization in the ICU²³; as such, quality improvement efforts often focus on minimizing variation to reduce risk to patients and ensure consistency in the approach to care. The most significant benefit of the PT3 may be the objective and systematic approach to patient triage and transport planning by standardizing the process regardless of provider background and experience. We found this to be quite valuable in our practice given the growing number of junior transport nurses on our team and the expected turn-over of physician trainees. Our survey data demonstrated that variation in transport planning was reduced. The tool was quick and easy to use, and was helpful in making decisions about team composition and transport mode. Particularly for nurses, confidence in making decisions increased. Moreover, we found the standardized PT3 intake form helpful in delivering complete and consistent handoff information (one of the reportable pediatric transport quality metrics³⁰), potentially minimizing communication errors and significant events.³¹ While some providers (mostly fellows) subjectively felt that the tool resulted in unnecessary use of personnel, our data indicates that resource utilization was, in fact, reduced. Free-text comments suggested that these perceptions were related to a transient increase in the proportion of physicians dispatched on transport during the beta-testing phase, which coincided with the outbreak of Enterovirus D68 and the start of respiratory virus season.

Important limitations of this study include the retrospective nature of the data collection (impacting our ability to detect all potential adverse outcomes) and the single center design. Testing the PT3 tool at other centers is essential, as we anticipate that minor scoring modifications may be necessary to accommodate differences between transport systems. Our dedicated pediatric transport program serves children of all ages with a wide range of illnesses and acuities from a busy urban community within a relatively discrete geographic area; hence our non-standardized staffing model and propensity towards ground transports. However, team composition and transport mode vary significantly between transport programs.³² Generally, programs transporting uniform populations of patients (neonates or adults) are more likely to choose a standard team composition, whereas those transporting a significant volume of pediatric patients often vary their compositions depending on the specifics of an individual transport (e.g., age, acute illness, comorbidities, location). A national survey of 229 unit-based and 106 dedicated neonatal transport programs in the United States identified 26 different team compositions used during neonatal transport.³³ Most unit-based programs surveyed (82%) were exclusively neonatal, whereas most dedicated programs (63%) were combined as neonatal-pediatric, neonatal-pediatric-adult, neonatal-maternal, or neonatal-adult. Many programs that transport older patients used non-physician team members but maintained the option to include a physician on select transports. While guidelines for determining team composition for pediatric transports have not been standardized nationwide, the PT3 may provide assistance by establishing objective criteria to include a physician on transport, utilize a smaller team, or choose a one-way transport.

We recognize that this tool was designed for transport systems with access to a variety of travel modes. Nevertheless, the PT3 targets the relative time it takes to complete the transport, rather than recommending a specific vehicle, therefore making it applicable to teams operating within various geographic areas with variable resources.

CONCLUSIONS AND FUTURE DIRECTIONS

The PT3 represents an objective tool to standardize and guide pediatric interhospital transport planning. The most notable benefit of the tool is its systematic approach, which limits variability in decision-making and reduces disagreements about team composition and transport mode. This tool may be particularly helpful for transport programs where providers have varied levels of experience. In addition, introducing the PT3 at referring hospitals should help standardize communication between sending and receiving providers and facilitate efficient patient assessment. Opportunities to utilize the PT3 via telemedicine or to modify the tool for the critically ill neonatal population also exist. Future studies should focus on prospectively tracking adverse outcomes, formally evaluating resource utilization, and analyzing costs. The PT3 should be used by different transport systems with varied staffing models and transport modes to determine if the tool can be adapted, with minor modifications, for their individual needs. We anticipate that such adjustments will likely not compromise the benefits of reduced variability in transport planning and potential improvements in resource utilization.

Supplementary Material

Supplemental Digital Content_1

Table that illustrates the initial Pediatric Transport Triage Tool and patient data collection form used during the transport referral intake process. Pdf

Supplemental Digital Content_2

Document which includes the Pediatric Transport Triage Tool survey instrument. pdf