Census tracts with very low population density are geographically very large. Inclusion of those tracts makes some of the service area polygons in our analysis highly irregular in shape. This irregularity represents an artifact of variation in census tracts and not misclassification in our analysis. See interactive maps.
Matthew B. Klein, C. Bradley Kramer, Jason Nelson, Frederick P. Rivara, Nicole S. Gibran, Thomas Concannon. Geographic Access to Burn Center Hospitals. JAMA. 2009;302(16):1774–1781. doi:10.1001/jama.2009.1548
Author Affiliations: Burn Center (Drs Klein and Gibran and Mr Kramer), Division of Plastic Surgery (Dr Klein), Department of Pediatrics and Harborview Injury Research and Prevention Center (Dr Rivara), University of Washington, Seattle; and Institute for Clinical Research and Health Policy Studies, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts (Mr Nelson and Dr Concannon).
Context The delivery of burn care is a resource-intensive endeavor that requires specialized personnel and equipment. The optimal geographic distribution of burn centers has long been debated; however, the current distribution of centers relative to geographic area and population is unknown.
Objective To estimate the proportion of the US population living within 1 and 2 hours by rotary air transport (helicopter) or ground transport of a burn care facility.
Design and Setting A cross-sectional analysis of geographic access to US burn centers utilizing the 2000 US census, road and speed limit data, the Atlas and Database of Air Medical Services database, and the 2008 American Burn Association Directory.
Main Outcome Measure The proportion of state, regional, and national population living within 1 and 2 hours by air transport or ground transport of a burn care facility.
Results In 2008, there were 128 self-reported burn centers in the United States including 51 American Burn Association–verified centers. An estimated 25.1% and 46.3% of the US population live within 1 and 2 hours by ground transport, respectively, of a verified burn center. By air, 53.9% and 79.0% of the population live within 1 and 2 hours, respectively, of a verified center. There was significant regional variation in access to verified burn centers by both ground and rotary air transport. The greatest proportion of the population with access was highest in the northeast region and lowest in the southern United States.
Conclusion Nearly 80% of the US population lives within 2 hours by ground or rotary air transport of a verified burn center; however, there is both state and regional variation in geographic access to these centers.
According to the American Burn Association, more than 500 000 burn injuries occur in the United States each year, causing approximately 4000 burn-related deaths.1 Death most commonly results from residential fires with smaller numbers from motor vehicle collisions, electrical injuries, or other burn etiologies. More than 40 000 patients are admitted to hospitals each year for treatment of burn injury.1 The delivery of optimal burn care to these patients is a resource-intensive endeavor requiring specialized equipment and experienced personnel. These resources are typically available only at dedicated burn centers.
In 2008, there were 128 self-identified burn centers in the United States, of which only 51 were verified by the American Burn Association.2 A recent analysis reported that only 22% of patients admitted to a hospital for burn care are admitted to verified burn centers.3 The verification process is the only recognized mechanism by which the quality of burn care provided at a center can be assessed and confirmed. Verified centers are located in 25 states and the District of Columbia and proximity to a verified burn center varies greatly around the country. Therefore, long-range transport of an individual who sustains a burn injury may be necessary to obtain definitive specialized care.
The optimal distribution of burn centers has long been debated.4 Several aspects of early postinjury burn care including airway assessment, estimation of burn extent and depth, and initiation of fluid administration are important to initial survival. Errors by physicians in estimating burn size and depth occur commonly and may lead to complications associated with either overresuscitation or underresuscitation.5- 9 Therefore, timely access to a burn center may benefit burn patients. In addition, with an anticipated shortage of qualified burn surgeons and the increasing complexity of medical and surgical aspects of burn care,10 there is a need to determine how best to ensure that all US residents have access to quality definitive care.
As a first and necessary step in this process, we sought to evaluate state, regional, and national access to burn centers in the United States. We assessed access to both verified and nonverified centers by ground and rotary air transport using national census, air medical transport, and street network databases.
We performed a cross-sectional analysis of access to burn centers, using US census tract-level data from 2000. We used approaches from previous resource allocation models to conduct our analyses.11,12 We estimated the percentage of the population living within 1 or 2 hours by ground transport and 1 or 2 hours by ground or rotary air transport of a burn care facility. We also assessed whether extended transport of 4 hours by ground transport could reach significantly more of the population than in the rotary air analyses. We assumed all transport occurred from an individual's home to the nearest burn center, measured by travel time. Our analysis included all US residents (N=281 421 906), excluding those living in US territories.
We examined access to burn centers that were verified by the American Burn Association as of 2008.2 We separately examined access to self-reported burn centers (ie, all centers listed in the American Burn Association directory), whether verified or not. Hospitals' latitude and longitude coordinates were obtained from the 2008 American Hospital Association annual survey. Each center was entered into a geographic information system (GIS) (Environmental Systems Research Institute ArcMap 9.2, Redlands, California) for subsequent analyses.
We summarized population data at the census tract level and entered it into the GIS. The population of each census tract was assigned to the tract's centroid (the geographic location that represents the “mean center” of a polygon). Tract-level data were aggregated at the state, regional, and national levels in separate analyses. For regional analyses, we used the 4 census regions: Midwest, Northeast, South, and West. Tracts were divided into 3 groups: urban, suburban, and rural, corresponding to tertiles of population density measured in people per square mile. We used these groups to adjust prehospital elapsed time estimates by urban density.
We also examined the number of burn care beds in each state relative to population. Bed numbers are self-reported in the American Burn Association directory and may represent either total number of dedicated beds for burn patients or potentially available burn beds given an unexpected surge in patients.
To determine the shortest driving times between census tract centroids and burn centers, we analyzed road network and speed limit data from the ArcGIS StreetMap data set using the ArcEditor (Environmental Systems Research Institute ArcMap 9.2) network analyst extension. For each estimated trip, we added extra time to account for dispatch of the emergency medical service (EMS) vehicle (1.4 minutes for urban and suburban tracts and 2.9 minutes for rural tracts), time from EMS depot to scene (total time was multiplied by a constant of 1.6, 1.5, or 1.4 for urban, suburban, or rural tracts, respectively), and time spent on scene (13.5 minutes for urban and suburban tracts and 15.1 minutes for rural tracts). These constants were derived in a meta-analysis of empirically determined prehospital care times for trauma.13
We used Atlas and Database of Air Medical Services (ADAMS) 2007 address and coordinate data to identify helipad locations and to estimate straight-line distances between 3 points: (1) base helipad, (2) tract centroid, and (3) burn center. Air Medical Services typically serve multiple hospitals with different aircraft.14 The ADAMS database allowed us to match burn centers, air medical service providers, base helipads, and available helicopters. Helicopter speed was estimated by taking the average speed of the service provider's fleet as recorded by ADAMS. Burn centers that were not matched with specific air medical service providers in the ADAMS database were matched with all base helipads located within the 1- or 2-hour travel time limit. We added a set of constants to account for dispatch time (3.5 minutes) and time spent preparing a landing zone (21.6 minutes).13 We assumed that ground transport could be dispatched concurrently with rotary transport, allowing air and ground vehicles to meet at the scene.15 Maps of the ground and rotary air coverage areas for verified and nonverified centers were generated using ArcMap version 9.2 software.
In 2008, there were 51 verified and 128 self-reported burn centers in the United States. A total of 782 helipads and 804 rotary wing helicopters served these centers. Nationally, 25.1%, 46.3%, and 67.7% of the US population lived within 1, 2 and 4 hours by ground transport, respectively, of a verified center and 41.1%, 68.4%, and 90.9% lived within 4 hours of any center, whether verified or not (Table 1). By air transport, 53.9% and 79.0% of the population lived within 1 and 2 hours, respectively, of a verified center, and 75.3% lived within 1 hour and more than 96.4% lived within 2 hours of any self-reported center (Table 1).
Access by ground transport to verified and self-reported burn centers varied by region (Table 1). Coverage was highest in the Northeast, with 40.2%, 72.7%, and 94.1% of the population living within 1, 2, and 4 hours by ground transport to a verified center. Coverage was lowest in the South, with 10.7%, 23.5%, and 46.2% living within 1, 2, and 4 hours, respectively, of a verified center. Access by ground transport to any self-reported burn center was highest in the Northeast (52.7%, 83.1%, and 98.0%) and lowest in the South (27.4%, 54.8%, and 87.6%).
Access by air transport followed the same regional pattern (Table 1). Coverage was highest in the Northeast, with 82.8% and 99.2% of the population living with 1 and 2 hours of a verified center. In the South, 30.2% and 62.0% lived within 1 and 2 hours of a verified center. Access by air transport to any self-reported burn center was also highest in the Northeast (89.9% and 99.9%) and lowest in the South (63.5% and 96.2%).
Geographic access to burn centers also varied by state (Table1). None of the population in 18 states lived within 2 hours by ground transport of a verified burn center, whereas more than 80% of the population in 5 other states and the District of Columbia lived within 2 hours by ground transport of a verified center. More than 90% of the population of Connecticut, Massachusetts, New Jersey, Rhode Island, and the District of Columbia lived within 2 hours by ground transport of any self-reported center, whereas none of the population of Montana and North Dakota lived within 2 hours of any self-reported center by ground or air transport.
None of the population in 6 states had 2-hour access by air transport to a verified center while 80% or more of the population in 29 states and the District of Columbia had 2-hour access by air transport. The entire population of 20 states and the District of Columbia lived within 2 hours by air transport of any self-reported center. Less than 20% of the population in 3 states had 2-hour access by air transport to any self-reported center.
Overall, the number of burn care beds was 0.65 beds per 100 000 US population. The number of beds varied by region, from 0.54 in the South to 0.81 in the Midwest. There was also variability in the number of beds by state, from 1.64 beds/100 000 population in Nebraska to no beds in the states Delaware, Idaho, Mississippi, Montana, New Hampshire, North Dakota, and Wyoming (Table 2).
Geographic access to verified burn centers and to self-designated burn facilities varies greatly across states and regions in the United States. A minority of the US population lives within 2 hours by ground transport of an American Burn Association–verified burn center or within 1 hour by ground transport of both verified and nonverified burn centers. To reach a verified burn care center within 2 hours, one-third of the US population must be transported by air. Extended transport of up to 4 hours by ground offers poorer coverage than 2-hour air transport. Population coverage is highest in the Northeast and lowest in the South, and there is substantial variation within regions. When including nonverified centers in the analysis, the proportion of population living near a burn facility increased by both ground transport and air transport, with more than 68% of the population living within 2 hours by ground transport of any center and more than 96% living within 2 hours by rotary air transport.
The ideal geographic distribution of burn centers has long been debated. The Figure illustrates that some regions are well served by verified centers, whereas others are relatively underserved (see also interactive maps). In fact, Zonies et al recently reported that more than half of the US population with burn injuries in 2001 and 2004 received definitive care at either nonverified burn centers or other hospitals not listed in the American Burn Association directory.3 This finding is likely attributable in part to geographic access to centers.
While this study did not address quality of care, MacKenzie, et al16 have previously demonstrated that trauma care provided at nondesignated centers was associated with worse outcomes than care at level I trauma centers. To our knowledge, there are no studies comparing long-term outcomes of burn injury based on verification status of treatment facility. However, the verification process is currently the only recognized mechanism by which quality of burn care can be assessed and ensured. Verification criteria include a number of specific requirements, including minimum number of acute burn admissions (≥50 annually); dedicated staff of physical and occupational therapists, psychologists, dietitians and social workers; prehospital triage and transfer guidelines; processes for quality of care review; and a director who has sufficient training or experience in burn care. In addition, verification must be renewed every 3 years.17 Although any hospital can designate itself a burn care facility, quality of care according to American Burn Association standards can be ensured only at centers that have successfully undergone the verification process. During the past decade, the number of verified centers has increased from 34 in 2000 to 51 in 2008; however, there are still more self-designated facilities than verified centers.
To address the need for better access to verified centers in some regions of the country and the apparent excess of centers in other regions, Warden and Heimbach4 have advocated that a system of regionalized burn care be developed to ensure a rational distribution of verified centers relative to population density. Regionalization may be reasonable from a resource distribution standpoint including both personnel and equipment; however, defining the optimum number of centers per population or per geographic area will prove challenging.
The variation in baseline geographic access rates found in this study may be an influential predictor of optimal regionalization strategy. For states and regions with a relatively high baseline rate of access, the best strategy for improving access and reducing time to definitive care may involve optimization of air and ground EMS systems. For states and regions with a relatively low baseline rate of access, the best strategy may involve construction or verification of new regional burn care facilities. Although we did not find burn centers that had no rotary air transport coverage, this strategy would be one way to increase coverage in the future if that were the case. To understand the best strategies for specific regions, one approach would be to model the predicted effect of system change on patient access rates, quality of care, and outcomes.
In addition to issues related to geographic distribution of centers, the optimum period in which burn patients require access to burn centers also remains to be defined. Burn care differs from other types of trauma in that there is no “golden hour,” but several aspects of early care can be critical to good outcomes. Failure to secure an airway adequately or diagnose related trauma can lead to adverse outcomes, including death. Inadequate burn size estimates occur commonly by health care professionals who are relatively inexperienced in the care of burn patients. These incorrect estimates can lead to underresuscitation and overresuscitation. Underresuscitation can lead to hypovolemic shock and renal failure, whereas overresuscitation can lead to abdominal and extremity compartment syndromes, as well as pulmonary edema and acute respiratory distress syndrome.18- 21
Based on the findings of this study, several hours of transport may be required for a burn patient to reach a verified or nonverified burn center. In addition, several centers provide coverage for large geographic regions—particularly in the western part of the United States. General trauma patients with delays in receipt of definitive care at trauma centers experience longer intensive care unit or hospital lengths of stay, increased number of complications, hospital changes, and mortality.22- 26 In one study, pediatric burn patients with delayed transfer to a verified burn center had longer lengths of stay and higher rates of infectious outcomes.27 However, in previous studies, we reported that a regional transfer and out-of-hospital transport protocol could be implemented safely for burn patients requiring long transport times to a verified center so that the critical aspects of acute postinjury care could be guided by clinicians who are experienced in caring for patients with burn injuries.28,29 As a result of these well-established protocols, patients with burns who were transferred to our verified burn center several hours after initial assessment at preliminary care facilities had outcomes similar to patients who were admitted directly to our center.29
There are several limitations to this study. First, fixed-wing transport data were not available nationally. Fixed-wing aircraft are typically used for patients requiring transport over longer distances and also tend to travel at higher speeds. Therefore, a larger geographic area could be covered within 2 hours by fixed-wing aircraft rather than by rotary aircraft. Additional factors that need to be considered in calculating access by fixed-wing aircraft include warm-up time prior to takeoff and need for appropriate landing zones. Nonetheless, larger geographic areas could be covered in shorter time. Second, listings of verified centers in the American Burn Association directory change from year to year, and our results would change if a substantial number of nonverified centers were to become verified. Finally, the data on burn bed availability are self-reported and have not been verified. It is possible that these reported numbers may not accurately reflect available staff to care for patients if all beds were filled.
In conclusion, this study demonstrates that nearly 80% of the population lives within 2 hours by ground transport or rotary air transport of a verified burn center. However, there exists substantial state and regional variation in geographic access to these centers. While the optimal distribution of burn centers relative to population and area remains to be determined, these data provide important information about population access that may be used to guide resource allocation in burn care.
Corresponding Author: Matthew B. Klein, MD, MS, UW Burn Center, Harborview Medical Center, 325 Ninth Ave, PO Box 359796, Seattle, WA 98104 (firstname.lastname@example.org).
Author Contributions: Dr Klein had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Klein, Kramer, Rivara, Gibran, Concannon.
Acquisition of data: Klein, Kramer, Nelson, Concannon.
Analysis and interpretation of data: Klein, Kramer, Nelson, Concannon.
Drafting of the manuscript: Klein.
Critical revision of the manuscript for important intellectual content: Klein, Kramer, Nelson, Rivara, Gibran, Concannon.
Statistical analysis: Klein, Kramer, Nelson, Concannon.
Obtained funding: Klein.
Administrative, technical, or material support: Klein, Kramer, Nelson, Concannon.
Study supervision: Klein, Concannon.
Financial Disclosures: None reported.
Funding/Support: This work was supported by the National Center for Research Resources grant 1KL2RR025015-01, Harborview Injury Prevention and Research Center (HIPRC) grant R49/CE000197 from the Centers for Disease Control and Prevention, and by the David and Nancy Auth-Washington Research Foundation Endowment.
Role of the Sponsor: The study sponsors had no role in the design and conduct of the study; in the collection, analysis, management, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Disclaimer: Dr Rivara, an editorial board member for JAMA, was not involved in the editorial review of or decision to publish this article.