OBJECTIVES: We aimed to characterize the demographics, diagnoses, and management of transferred patients who were directly discharged from the emergency department (ED) or admitted less than 24 hours.
METHODS: We conducted a retrospective, cross-sectional study of patients classified as interfacility ED transfers over a 12-month period in the Pediatric Health Information System database, an administrative database of 42 tertiary care pediatric US hospitals. The primary study outcomes were ED resource utilization at the receiving facility with a focus on children who were discharged directly from the ED or admitted less than 24 hours.
RESULTS: Overall, 24 905 interfacility transfers were identified, accounting for 1.3% of the ED volume of these academic pediatric centers. Of these, 24.7% were discharged directly from the ED and 17.0% were admitted for less than 24 hours. Among those directly discharged from the ED, the 3 most common complaints were orthopedic problems, nonsurgical abdominal pain, and viral gastroenteritis; 20.7% received no medical or procedural intervention. Among those admitted for less than 24 hours, the 3 most common complaints were orthopedic problems, traumatic head injury, and gastrointestinal conditions.
CONCLUSIONS: A significant proportion of interfacility transfers to academic pediatric EDs is discharged directly from the ED or is admitted for less than a day. These patients and their clinical outcomes provide insight into the educational needs and medical capabilities of referring hospitals and clinicians.
- ED —
- emergency department
- EM —
- emergency medicine
- ICD-9 —
- International Classification of Diseases, Ninth Revision
- PHIS —
- Pediatric Health Information System
What’s Known on This Subject:
Although many children are treated in general emergency departments, many such facilities have limited pediatric capabilities. Transfer to academic centers improves outcomes for critically ill patients, but transfers of noncritically ill children have not been well studied.
What This Study Adds:
Although more than half of these patients are seriously ill, many transferred patients are discharged directly from the emergency department or are admitted for less than 24 hours. Orthopedic problems, gastrointestinal conditions, and traumatic head injury are the most common complaints.
Over 27 million pediatric emergency department (ED) visits occur annually; of these, 89% occur in general EDs with 24% in nonurban areas.1 However, there is growing evidence that these departments may not be adequately prepared to meet the medical needs of pediatric patients. The Institute of Medicine revealed that only 6% of EDs are adequately equipped for pediatric emergencies.2 Additionally, Dharmer et al3 found decreased quality of care of pediatric patients by general emergency medicine (EM) physicians in rural, nonacademic centers, whereas Walls et al4 found that 22% of pretransport care from community hospitals is suboptimal with 69% of referring hospitals performing inappropriate care.
Concurrently, with rising concern for the quality of pediatric ED care outside of dedicated pediatric hospitals, there has been mounting evidence that transferring patients can improve their care. Previous studies have revealed that critically ill pediatric patients have improved outcomes when cared for in dedicated PICUs.5–9 Based on these studies, both the Society of Critical Care Medicine and the American Academy of Pediatrics have recommended regionalization of pediatric critical care services,10 which has served to encourage transfers to academic centers for care.
Beyond the evidence supporting the transfer of the critically ill children, little is known about the transfer of noncritically ill pediatric patients. In 1 study from a single ED, up to 31% of interfacility transfers were considered unjustified11; the authors of this study examined the overall ED population and did not limit the study to pediatric patients. A second study by Soundappan et al12 focused on pediatric trauma transfers to academic pediatric trauma centers and revealed that over 50% were considered unnecessary.
The question of what is a “justified” transfer is debatable. Community hospitals may have limitations that necessitate transfer such as a lack of a pediatric inpatient unit, lack of ED nurses with pediatric skill, or physician confidence with children. Rather than focusing on the minimization of “unnecessary” transfers, a better goal is to improve the care of pediatric patients outside of pediatric academic centers and continue to transfer patients requiring a higher level of care. The first step is to identify the limitations of community hospitals in providing pediatric care. One approach is to study interfacility transfers on a national level. By focusing on patients who ultimately had low medical needs, we can gain insight into targeted areas of pediatric care outside of academic facilities. Our goal was to evaluate pediatric interfacility transfers with a focus on the ED management and disposition of those who were directly discharged or admitted for less than 24 hours.
Data for this study were obtained from the Pediatric Health Information System (PHIS), an administrative database that contains inpatient, ED, ambulatory surgery, and observation data from 42 not-for-profit, tertiary care pediatric hospitals in the United States. These hospitals are affiliated with the Child Health Corporation of America (Shawnee Mission, KS), a business alliance of children’s hospitals. Data quality and reliability are ensured through a joint effort between the Child Health Corporation of America and participating hospitals. The data warehouse function for the PHIS database is managed by Thomson Reuters (Ann Arbor, MI). Participating hospitals provide discharge/encounter data including demographics, diagnoses, and procedures as well as resource utilization data (eg, pharmaceuticals, imaging, and laboratory) into PHIS. Data are deidentified at the time of data submission and are subjected to a number of reliability and validity checks before being included in the database.
We conducted a cross sectional study that was approved by the Institutional Review Board at Children’s Hospital Boston and by the administrators of PHIS. All ED patients under the age of 18 years old who arrived by interfacility transport and were reported to the PHIS database between October 1, 2009, and September 30, 2010, were eligible for inclusion in this study. The “source of admission” code classified the mode of arrival (eg, transfer, self-referral, etc) allowing for the identification of patients who were transferred. Any hospital whose source of admission code was missing for more than 50% of their ED patients was excluded from the analysis.
The primary study measures were diagnoses, ED management, and disposition. We used the International Classification of Diseases, Ninth Revision (ICD-9) codes,13 charges including procedure codes (Current Procedural Terminology codes), laboratory and radiology charges, and clinical billing. All ICD-9 codes in the data set were manually reviewed and categorized into broader diagnostic categories by one investigator and were reviewed by a second investigator. We compared diagnoses and the management of patients who were directly discharged with those admitted less than 24 hours and also with patients admitted more than 24 hours (because these strata likely reflect differences in medical need). We stratified the sample of transferred patients into 3 groups: (1) directly discharged; (2) those admitted less than 24 hours; and (3) those patients admitted more than 24 hours. We calculated standard descriptive statistics across our 3 groups for all outcomes. We used means for normally distributed variables, and medians and ranges for dimensional or continuous variables with nonnormal distributions, and frequencies and percents for binary variables. We performed the data analysis with Stata 12.0 (College Station, TX).
Overall Transport Rate and Disposition
Of 42 available hospitals, 13 hospitals were excluded due to invalid or missing source of admission code. For the remaining 29 hospitals, more than 95% of patients had a valid source of admission code. During the study period, 24 905 interfacility transfers were identified (1.3% of the overall ED volume of the 29 included hospitals, see Appendix 1). Of these transfers, 8.1% had either no disposition code or were missing the duration of admission and were excluded from the remainder of the analyses. Among the remainder, 24.7% were discharged directly from the ED, whereas 17.0% were admitted for less than 24 hours (Fig 1). The remaining 58.4% were admitted greater than 24 hours. Of those admitted less than 24 hours, 26.0% required a surgical intervention. Of those admitted greater than 24 hours, 16.0% required admission to the ICU and 26.1% required a surgical procedure.
Profile of Transferred Patients
Age, gender, and insurance type are outlined for the 3 disposition subgroups in Table 1. The patients in all 3 groups had an average age of 6 years. The most common gender and ethnicity in all 3 groups were boy and white, respectively. The most common time of arrival for all 3 groups was 4 pm to midnight. The most frequent diagnostic category among those either discharged directly from ED or admitted less than 24 hours was an orthopedic condition. In contrast, the most common diagnostic problem in those admitted longer than 24 hours was a general surgical condition. The most common ICD-9 codes for each diagnostic category are listed in Table 2.
Management and Care Received
Medications, diagnostic testing including laboratory and radiologic studies, and procedures performed were compared between the 3 groups (Table 3). Among those children directly discharged, 20.7% did not require any testing, procedure, or medications. An additional 33% of those directly discharged only received acetaminophen, ibuprofen, ondansetron, and/or a plain radiograph at the receiving hospital. In contrast, 97.3% of those admitted for <24 hours and nearly all (99.5%) of those admitted longer than 24 hours required at least a laboratory test, medication, or procedure. Among those directly discharged, the most common interventions were the administration of an intravenous medication or fluid (33.9%), plain radiograph (33.6%), or blood test (17.7%). Procedural sedation was the most common ED procedure (5.5%).
Disposition and Management of Transferred Patients Based on Diagnosis
Among patients diagnosed with an orthopedic condition and nonsurgical abdominal pain, 48.5% and 72.5%, respectively, were directly discharged from the ED. Additionally, 14.8% of those with nonsurgical abdominal pain did not have any procedures or laboratory tests performed or medications administered. Patients with pneumonia and bronchiolitis were frequently admitted for longer than 24 hours (80.6% and 82%, respectively) with the majority of patients receiving at least 1 intervention (only 3.2% and 5.7%, respectively, had no testing, medications, or procedures as illustrated in Table 4). (More information about the specific management of common diagnoses is available in Appendix 2).
Many community hospitals were found by the Institute of Medicine to be ill-equipped for pediatric patients.2 However, despite this finding, it is not ideal to transfer all children to pediatric academic centers. Regionalization is the most ideal route for critically ill pediatric patients, but this practice must be investigated for noncritically ill patients. Both the American Academy of Pediatrics and American College of Emergency Physicians recommend ongoing pediatric education of all ED providers.14,15 For this population in particular, educational interventions and outreach programs for community physicians and nurses to elevate the level of pediatric care in general EDs would be valuable. We examined pediatric interfacility transport patients because the services and diagnostic patterns of this population likely reflect the resource or education gaps that need to be strengthened. By describing the overall population of pediatric interfacility transfers with a specific focus on patients who are directly discharged from the accepting facility or are discharged in <24 hours, we attempted to gain further insight into the potential areas of improvement for both referring clinicians and facilities.
In this large sample size of pediatric interfacility transfers, ∼40% of patients were either discharged directly from the accepting hospital or hospitalized for <24 hours; of this group, over 20% received no medications, testing, or procedures. Of those directly discharged from the ED, gastroenteritis/dehydration, gastrointestinal complaints, and nonsurgical abdominal pain were the most common diagnostic categories. Our results differ from the only previous study of pediatric interfacility transfers to a single institution, in which they found respiratory distress, asthma, and seizures as the most common diagnoses among transfers but did not report the resulting disposition.4 In our sample, asthma and seizures were among the 10 most common diagnoses, and many of these required extended admissions. Many factors contribute to the decision to transfer a pediatric patient. In some cases, the transfer may result from a combination of education and capability deficits among community ED providers, as well as a lack of adequate pediatric support services at community hospitals. Such services might include skilled personnel for intravenous access or phlebotomy, pediatric specific radiologic services, or the ability to perform procedural sedation. Additionally, the transfer of children might be related to a perceived need for subspecialty consultation or potential admission that might be not available at the transferring hospital. Also, if the primary care provider is contacted by the community ED physician, the primary care provider may request a transfer to an academic pediatric hospital regardless of the care available or received at a community hospital. We also attempted to determine if there were certain ED procedures that might have necessitated transfer to a pediatric facility that could also be a point of education and intervention. These procedures included lumbar punctures, procedural sedation, laceration repair, and orthopedic procedures. However, only a small percentage of transferred patients had these procedures.
The purpose of our investigation was to identify areas in need of improvement in pediatric care in community hospitals. Through analysis and examination of the types of patients who are transferred and their clinical needs, resource and education needs will become more apparent. The authors of previous studies have attempted to identify areas of discomfort for general EM physicians.16–18 However, these studies were either limited in the scope of pediatric EM practice, focusing only on specific common pediatric problems16,17 or only asked about broad categories of pediatric EM.18 Additionally, these studies were surveys and did not analyze the specific areas identified by general EM physicians.
By determining problem areas based on practice patterns, we can take steps toward developing targeted interventions to enhance the care of these patients at their community institutions. This will also serve to potentially decrease duplication in testing and treatment, which ultimately wastes resources and can lead to low patient satisfaction. These interventions could be educational or administrative, with the goal of enhancing pediatric care and avoiding transfers of low acuity patients to other facilities. Based on our study, educational efforts should include abdominal pain, gastroenteritis and orthopedic problems, with additional consideration of asthma and seizures.
The American Academy of Pediatrics Committee on Pediatric Emergency Medicine recommends the development of telemedicine for promoting pediatric emergency care.19 The authors of preliminary studies have investigated the use of telemedicine for critical care/trauma consults20–24 as well as for subspecialty care in the office setting25–28 with promising results. In a study by Heath et al,20 looking at its use in pediatric critical care patients, both the referring and consulting physicians felt that the telemedicine consultation improved patient care. Although there are only a few studies of pediatric care in relation to the ED and most of these studies focused on critical care,20–23 telemedicine is a new technology that should be investigated as a helpful adjunctive service.
Future investigations should study transferring hospitals and their pediatric capability at all levels of care and across common pediatric medical and surgical conditions. Although we investigated a large number of pediatric hospitals, the next step would be to try to capture a broader understanding of pediatric transfers to all types of academic centers.
Our study has several limitations. First, the source of admission code, which identifies a patient as an interfacility transfer or not, is a new variable that was included in the PHIS database in October of 2009. As such, the reliability of this code needs to be formally measured even though the PHIS database has a rigorous data review and monitoring process. To further improve the quality of the data, we attempted to apply additional quality control measures by excluding those hospitals that had a significant portion of this variable missing.
Additionally, we only had information from the accepting hospital and do not have any information from the transferring hospital. We also were unable to derive the specific reason for transfer. Although we specifically studied transferred patients, we cannot estimate the proportion of pediatric patients requiring transfer; therefore, we can only infer the needs of referring hospitals by investigating those patients who are transferred. As such, we can only comment on the resulting treatment and resource utilization after transfer. Despite these limitations, this is the first study to analyze the outcome, management, and disposition of noncritically ill pediatric transfers on a national level.
Although most of pediatric emergency care is provided in general EDs, there has been an increasing effort to define quality metrics for pediatric emergency care. Pediatric interfacility transfers are a special group of patients who have not been well studied. Understanding the reasons for transfer in the context of the entire health-care system has the opportunity to improve pediatric care outside of academic centers. We noted that a large fraction of patients are directly discharged or admitted <24 hours with minimal intervention; this specific subpopulation might be the best target for educational intervention and development of pediatric capability outside of academic centers.
- Accepted March 2, 2012.
- Address correspondence to Joyce Li, Division of Emergency Medicine, Harvard Medical School, Children’s Hospital Boston, 300 Longwood Ave, Boston, MA 02115. E-mail:
Drs Li and Bachur conceived and designed the study; Dr Monuteaux was responsible for data acquisition and management; Drs Li, Monuteaux, and Bachur provided statistical advice on study design; and Dr Monuteaux performed analysis of data. Dr Li was responsible for article creation, and all authors contributed to the article review and revisions. Dr Li takes responsibility for the article as a whole.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
- Institute of Medicine Committee on the Future of Emergency Care in the United States Health System
- Bertazzoni G,
- Cristofani M,
- Ponzanetti A,
- et al
- ↵National Center for Health Statistics. ICD-9 coding manual. Available at: http://icd9cm.chrisendres.com/index.php?action=contents. Accessed March 1, 2011
- American Academy of Pediatrics Committee on Pediatric Emergency Medicine
- Yamamoto LG,
- American Academy of Pediatrics Committee on Pediatric Emergency Medicine
- Kofos D,
- Pitetti R,
- Orr R,
- Thompson A
- Marcin JP,
- Schepps DE,
- Page KA,
- Struve SN,
- Nagrampa E,
- Dimand RJ
- Sable CA,
- Cummings SD,
- Pearson GD,
- et al
- Marcin JP,
- Ellis J,
- Mawis R,
- Nagrampa E,
- Nesbitt TS,
- Dimand RJ
- Karp WB,
- Grigsby RK,
- McSwiggan-Hardin M,
- et al
- Pammer W,
- Haney M,
- Lmhc N,
- et al
- Copyright © 2012 by the American Academy of Pediatrics