FOCUS ON MASS CASUALTY INCIDENTS Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

EPIDEMIOLOGY OF MASS CASUALTY INCIDENTS IN THE UNITED STATES Ellen Schenk, MPH, ASPPH/NHTSA, Public Health Fellow, Gamunu Wijetunge, MPM, NRP, N. Clay Mann, PhD, MS, E. Brooke Lerner, PhD, Anders Longthorne, Drew Dawson ABSTRACT

their perception of the event to produce incidence rates of MCIs per 100,000 population and MCIs per 1,000 9-1-1 calls requesting EMS service. We conducted descriptive analyses to characterize the MCIs by geographic location, incident type, and time of day as well as the MCI patients by demographic and health information. We used chi-squared tests to compare response delays and two-tailed t-tests to compare system response times between EMS responses documented as MCIs and those not. Results. Among the 9,776,094 EMS responses in the 2010 National EMS Database, 14,504 entries were documented as MCI. These entries represented an estimated 9,913 unique MCIs from the National EMS Database: 39.1% occurred in the South Atlantic region of the United States where only 19.1% of the population resides, 60.9% occurred in an urban setting, and 58.4% occurred on a street or highway. There were an estimated 13,677 MCI patients. The prehospital EMS personnel’s primary impressions of the patients ranged from electrocution (0.01%) to traumatic injury (40.7%). Of the patients with a primary impression of injury (N = 7,960), motor vehicle traffic crash was the cause of injury for 62.7%. Among the MCI EMS responses, 47.6% documented experiencing a response delay compared to only 12.3% of non-MCI EMS responses. Conclusions. This study demonstrates the range of health conditions and characteristics of EMS responses that EMS personnel perceive as MCIs, suggests that response delays are common during MCIs, and indicates there may be underreporting of all persons involved in an MCI. The National EMS Database is useful for describing MCIs and may help guide national leadership in strengthening EMS system preparedness for MCIs. Key words: emergency medical services; prehospital; paramedic; EMS personnel; preparedness

Objective. We sought to characterize and estimate the frequency of mass casualty incidents (MCIs) occurring in the United States during the year 2010, as reported by emergency medical services (EMS) personnel. Methods. Using the 2010 National EMS Database of the National Emergency Medical Services Information System (NEMSIS), containing data from 32 states and territories, we estimated and weighted the frequency of MCIs documented by EMS personnel based on

Received August 12, 2013 from the Office of Emergency Medical Services, National Highway Traffic Safety Administration, U.S. Department of Transportation, Washington, DC (ES, GW, DD), Intermountain Injury Control and Research Center, Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, Utah (NCM), Department of Emergency Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin (EBL), and National Center for Statistics and Analysis, National Highway Traffic Safety Administration, U.S. Department of Transportation, Washington, DC (AL). Revision received December 11, 2013; accepted for publication December 16, 2013. This publication was supported by Cooperative Agreement Number DTNH22-09-H-00262 from the U.S. Department of Transportation, National Highway Traffic Safety Administration (NHTSA), and the Association of Schools and Programs of Public Health. The findings and conclusions of this publication do not necessarily represent the official views of NHTSA or ASPPH. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. The authors would like to thank the staff of the NHTSA Office of EMS for review of this manuscript and support of this project during Ms. Schenk’s fellowship. We would also like to thank Barry Eisemann with the NHTSA National Center for Statistics and Analysis for his support in retrieving and managing the data files as well as Bernice Boursiquot, former ASPPH/NHTSA Public Health Fellow, for her assistance with the SAS coding and data analysis.

PREHOSPITAL EMERGENCY CARE 2014;18:408–416

INTRODUCTION Emergency medical services (EMS) systems plays a key role in the management of any mass casualty incident (MCI). The most important consideration when defining an MCI is not the absolute number of victims, but exceeding the medical system’s ability to meet the

Address correspondence to Ellen Schenk, MPH, ASPPH/NHTSA, Public Health Fellow, NTI-140, W44-232, 1200 New Jersey Ave, SE, Washington, DC 20590, USA. e-mail: [email protected] doi: 10.3109/10903127.2014.882999

408

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

E. Schenk et al.

409

MASS CASUALITY INCIDENTS

health-care needs of the MCI victims utilizing locally available medical resources.1 It is critical that EMS systems and EMS personnel be prepared for MCIs. Sixty-one percent of prehospital EMS personnel perceive that it is likely that their health-care facility would be involved in an MCI within the next three years.2 However, our current understanding of MCIs in the United States is often limited to individual incidents or local geographic regions, since most incidents are documented through published case reports of specific events or other forms of anecdotal information.3–6 There is a need for better understanding of MCIs across the United States to create an evidence base for EMS preparedness planning at the national level that can be effectively utilized at the local level.7 Plans to prepare the nation against health threats should be based on the best available evidence and should be evaluated for effectiveness. Health system preparedness decision-making should be based on research that analyzes and forecasts vulnerabilities across the nation and that develops performance metrics.8 The National Emergency Medical Services Information System (NEMSIS), with funding support from the National Highway Traffic Safety Administration (NHTSA), is an effort to standardize information collected by EMS personnel who respond to emergency calls and to develop an aggregate data set encompassing EMS data from every U.S. state and territory.9,10 In 2010, 32 states and territories submitted EMS data to the National EMS Database.11 This repository offers the opportunity to study MCIs in the United States across a broader geographic scale. In this study, we sought to use the National EMS Database to estimate, describe, and calculate the rate of MCIs to which EMS responded in the United States in 2010. We undertook this descriptive study to assess the current state of national EMS data related to MCIs and to provide information to help guide future research.

METHODS Study Design This retrospective cross-sectional study utilized the 2010 National EMS Database available from the NEMSIS Technical Assistance Center (TAC). Institutional review board exemption for this study was obtained through the University of Utah’s School of Medicine. This study also met the exemption criteria for the Federalwide Assurance (FWA) for the Protection of Human Subjects through Title 45, Part 46 of the Code of Federal Regulations.

Study Setting The NEMSIS provides standard definitions and formats for over 400 data elements (version 2.2.1), with

83 of these variables comprising the National EMS Database. States participating in the NEMSIS project coordinate with local prehospital EMS personnel to promote consistent patient care documentation using standardized computer software programs conforming to NEMSIS data element standards. EMS data collection begins at the local level. When prehospital EMS personnel interact with a patient, an electronic patient care report (ePCR) is generated that contains information on the patient’s demographics, health emergency, transport to a medical facility, and any treatment en route. When multiple vehicles respond to the same incident, each EMS vehicle may submit data to the state and thus be represented in the National EMS Database. Of the states participating in the NEMSIS project, the lead EMS office in each state aggregates local EMS data into a statewide data set, and subsequently exports the 83 national variables to the National EMS Dataset, the data of which are referred to as the National EMS Database. The NEMSIS defines an EMS response as any request for service, regardless of the outcome of the response. However, states may further refine this definition and the NEMSIS accepts these data (e.g., only service requests resulting in patient contact). The NEMSIS does not define case inclusion criteria for the National EMS Database; the project accepts all data meeting the states’ individual inclusion requirements. States may submit data with less than 100% of EMS agencies participating in the state registry. The NEMSIS is financially supported by the U.S. Department of Transportation’s NHTSA with prior funding support from the U.S. Department of Health and Human Services’ Health Resource and Services Administration as well as other state and federal partners.10 The NEMSIS TAC (University of Utah School of Medicine, Salt Lake City, UT) aggregates and maintains the National EMS Database, which is based on the list of variables, the National EMS Dataset, submitted by the states.

Study Population We studied all data from the 2010 National EMS Database containing information on 9,776,094 EMS responses submitted by 32 states and territories during the 2010 calendar year. We analyzed the subset of EMS responses recorded as MCIs for this study. Data Use Agreements between the NEMSIS Technical Assistance Center and each state precluded the identification of individual EMS personnel or agencies in published reports.

Variables The primary outcomes of this study were the estimated frequency of MCIs as well as the estimated number of patients involved in MCIs. The NEMSIS version 2.2.1 utilizes the following definition of an MCI: “A mass casualty incident is defined as an

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

410 event which generates more patients at one time than locally available resources can manage using routine procedures or resulting in a number of victims large enough to disrupt the normal course of emergency and health care services and would require additional nonroutine assistance.”12,13 It is unknown how individual EMS personnel are trained to use this definition, but for each response, EMS personnel must document whether or not the response is considered an MCI. The NEMSIS contains information on the age, sex, race, ethnicity, EMS personnel’s primary impression, primary symptom, and cause of injury for patients involved in MCIs. The NEMSIS also lists the time of day, incident location, time and date that the EMS unit was notified by dispatch, and the incident five-digit zip code for each EMS activation. The National EMS Database documents population setting (urbanicity) using United States Department of Agriculture (USDA) and Office of Management and Budget (OMB) definitions.14 The NEMSIS contains information on times of dispatch, turnaround, response, or scene delays during the EMS response, as documented by EMS and dispatch personnel. The National EMS Database derives the following elapsed time intervals in minutes associated with system response: dispatch center time interval, the difference between the unit notified by dispatch date/time and the Public Safety Answering Point (PSAP) call date/time; chute time interval, the difference between the unit notified by dispatch date/time and the unit en route date/time; system response time interval, the difference between the unit notified by dispatch date/time and the unit arrived on scene date/time; scene time interval, the difference between the unit arrived on scene date/time and the unit left scene date/time; scene to patient time interval, the difference between the unit arrived on scene date/time and the arrived at patient date/time; transport time interval, the difference between the unit left scene date/time and the patient arrived at destination date/time; and total call time interval, the difference between the unit back in service date/time and the unit notified by dispatch date/time.15 We used these variables to describe the circumstances and the population of patients who were identified by EMS responders as being involved in an MCI.

Primary Data Analysis We analyzed the data using descriptive statistics, expressing the results using frequencies, percentages, and exact 95% confidence intervals, as well as χ 2 and two-sample t-tests using Statistical Analysis Software (SAS) 9.3. Since an MCI could result in multiple EMS responses due to multiple calls to more than one EMS agency, the frequency of MCIs in the 2010 National EMS Database was estimated by identifying exact matches among

PREHOSPITAL EMERGENCY CARE

JULY/SEPTEMBER 2014

VOLUME 18 / NUMBER 3

EMS responses based on the five-digit zip code of the incident location as well as the date and time that the unit was notified by dispatch. The element in the NEMSIS that records the date and time that the unit was notified by dispatch was found to have clusters of entries within the same zip codes that matched exactly down to the second, and the nearest cluster was hours or days apart, so an exact match rather than an interval of time was used to match entries in estimating the number of MCIs. All matching EMS responses were assigned the same unique MCI ID number. Since more than one EMS agency could have contact with a single patient, resulting in multiple EMS responses for a specific patient, the number of MCI patients in the National EMS Database was estimated by identifying exact matches among entries based on the assigned MCI ID number as well as recorded sex, race, and age. Entries that matched exactly on these elements were assigned the same unique patient ID number, generated by the researchers for this study. To calculate an incidence rate of MCIs per 100,000 population, the estimated number of MCIs found from the seven states that submitted 100% of their EMS runs (Alabama, Arkansas, Hawaii, Maine, Minnesota, North Carolina, and Oklahoma) to the 2010 National EMS Database was divided by the total estimated population of 29,028,803 for these states in 2010.16 In terms of calculating an incidence rate of MCIs by EMS call volume, the estimated number of MCIs from the seven states that submitted 100% of EMS runs was divided by a median number of 310,000 9-1-1 calls by state requesting EMS service in 2010,17 multiplied by seven. To calculate an estimate of the number of MCIs in the nation for 2010, the observed incidence rate of MCIs per population in the seven states was multiplied by a total population of 281,421,906 persons in the United States in the year 2010.18 Exact binomial 95% confidence intervals were calculated for the incidence rates and the estimate of MCIs in the nation. In this study, we sought to characterize each of the MCIs in the National EMS Database, based on a number of variables, including geographic location and time of day. The analysis included comparing the information contained in entries assigned the same MCI ID. When event level information among entries differed, the response that was documented in the greatest number of run reports was used, unless other variables for the same entry clearly supported a specific response. When the documented frequency of an actual value was equal to the documented frequency of a null value (i.e., “not available,” “not applicable,” or “not recorded”) in one record but had a recorded value in another record, the recorded value was utilized. Frequencies and percentages of MCIs were classified by U.S. Census Divisions,14 urbanicity, incident location, and time of day, using standard NHTSA categories.19 The median and range for the

E. Schenk et al.

411

MASS CASUALITY INCIDENTS

TABLE 1. Estimated mass casualty incidents (N = 9,913) by U.S. Census Division from the 2010 National EMS Database Characteristics

Estimated MCIs in the 2010 National EMS Database

% of total estimated MCIs

Estimated U.S. population in 201015

% of Total estimated U.S. population

750 601 287 1,089 400 58 3,872 17 2,436 403

7.6 6.1 2.9 11.0 4.0 0.6 39.1 0.2 24.6 4.1

46,439,372 18,460,132 44,488,019 22,133,139 10,878,089 49,996,985 59,916,816 3,721,978 20,537,086 36,480,581

14.8 5.9 14.2 7.1 3.5 16.0 19.1 1.2 6.6 11.7

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

U.S. Census Division13 East North Central East South Central Mid Atlantic Mountain New England Pacific South Atlantic Territories West North Central West South Central

States within each U.S. Census Division that submitted data to the National Emergency Medical Services Dataset in 2010: • East North Central: Illinois, Michigan • East South Central: Alabama, Mississippi, Tennessee • Mid Atlantic: New Jersey, Pennsylvania • Mountain: Colorado, Idaho, New Mexico, Utah, Nevada • New England: Maine, New Hampshire • Pacific: Alaska, Hawaii, Washington • South Atlantic: Florida, North Carolina, South Carolina, West Virginia • Territories: Northern Mariana Islands, Virgin Islands • West North Central: Iowa, Nebraska, Kansas, North Dakota, South Dakota • West South Central: Arkansas, Oklahoma

number of EMS responses to a specific incident as well as the estimated number of patients per MCI was calculated. This study characterized the patients documented as being in the 2010 National EMS Database, based on a number of variables, including demographic and health information. The information contained in entries assigned the same patient ID was compared. When information among entries differed, the same process described above was used to determine which response to include in the analysis. Descriptive statistics included information on EMS personnel’s primary impression, primary symptom, and cause of injury. Cases with missing or invalid values for the response time variables were excluded from analysis. Bivariate analyses were conducted using χ 2 tests to study the system response delays and Wilcoxon rank sum tests to analyze the mean system response time intervals.

RESULTS In 2010, 32 states and 3,529 local EMS agencies submitted data to the National EMS Database, resulting in 9,776,094 EMS responses. Among these responses 14,504 (0.15%) were recorded as MCIs. Among the 14,504 documented EMS responses to MCIs, it was estimated that there were 9,913 unique incidents and 13,677 unique patients. The following seven states submitted 100% of EMS runs to the National EMS Database and accounted for 32.9% of the data: Alabama, Arkansas, Hawaii, Maine, Minnesota, North Carolina, and Oklahoma (Appendix A, available online). Among the 3,206,077 EMS responses in the

seven states with 100% reporting, 5,198 (0.2%) were recorded as MCIs. Among the 5,198 MCI EMS responses, there were 3,768 estimated unique MCIs and 4,978 estimated unique MCI patients. An observed rate of 13.0 MCIs per 100,000 population (95% CI: 12.60/100,000–13.4/100,000) was identified in these seven states. A rate of 1.7 MCIs per 1,000 9-1-1 calls requesting EMS service (95% CI: 1.7/1,000–1.8/1,000) TABLE 2. Estimated mass casualty incidents (N = 9,913) by urbanicity and incident location from the 2010 National EMS Database Characteristics

Urbanicity Rural Suburban Urban Wilderness Missing Incident location Farm Health-care facility (clinic, hospital, nursing home) Home/residence Industrial place and premises Lake/river/ocean Mine or quarry Other location Place of recreation or sport Public buildings (schools, government offices, etc.) Residential institution (nursing home, jail/prison, etc.) Street or highway Trade or service (business, bars, restaurants, etc.) Not applicable Not available Not known Not recorded Not reporting

No.

%

1,648 1,619 6,039 556 51

16.6 16.3 60.9 5.6 0.51

27 727

0.3 7.3

1,807 59 26 10 187 169 214

18.2 0.6 0.3 0.1 1.9 1.7 2.2

162

1.6

5,791 230 68 63 233 135 5

58.4 2.3 0.7 0.6 2.4 1.4 0.1

412

PREHOSPITAL EMERGENCY CARE

JULY/SEPTEMBER 2014

VOLUME 18 / NUMBER 3

25

n=2005

20

n=1773 n=1588

15

n=1332 n=1085

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

10

n=863 n=460

5

0

%

n=807

00:00:00AM 3:00:01AM 6:00:01AM 9:00:01AM 12:00:01PM 3:00:01PM 6:00:01PM 9:00:01PM to to to to to to to to 3:00:00AM 6:00:00AM 9:00:00AM 12:00:00PM 3:00:00PM 6:00:00PM 9:00:00PM 23:59:59PM

FIGURE 1. Time of day of estimated mass casualty incidents (n = 9,913) from the 2010 National EMS Database.

was calculated. The incidence rate of MCIs per population was applied to the total 2010 estimated population of the country to produce an estimate of 36,529 MCIs (95% CI: 35,431–37,626) in the United States for the year 2010. Among the 9,913 estimated MCIs in the 2010 National EMS Database, the majority occurred in the South Atlantic region of the United States (Table 1), according to the U.S. Census Division element reported to the National EMS Database. The proportion of MCIs in the South Atlantic region was much higher than demonstrated in other census regions of the United States. The majority of the incidents occurred in the evening hours (Figure 1) and in an urban setting (Table 2). Of note, the median number of EMS responses per incident and the median number of patients per incident was 1 (Table 3). Patients documented as being in an MCI had a wide range of ages, and 30.75% of

the patients were pediatric (defined as age 18 years or less). Over half (51.3%) of the patients were female, while 45.2% were male, and the majority of patients were white (Table 4). The EMS personnel’s primary impressions of the patients ranged from electrocution (0.01%) to traumatic injury (40.7%), while 29.8% of patients’ impressions were recorded as not applicable or not available (Table 5). Pain (39.8%) was the most commonly recorded patient primary symptom. The cause of injury for 62.7% of estimated patients with possible injuries (N = 7,960) was motor vehicle traffic crash (Table 6). All types of response system delays studied were found to be significantly associated with MCI documentation (Table 7). MCIs were 8.63 times as likely to have recorded a dispatch delay compared with non-MCIs. There were also statistically significant differences in most mean response time interval components between MCIs and non-MCIs (Table 8).

TABLE 3. Responses and estimated patients per mass casualty incident from the 2010 National EMS Database Frequency

Mean

Median

Mode

Std Dev

Min

Max

14,084

1.0

EMS responses per estimated mass casualty incident 1.0 1.0 0.2

1.0

51.0

13,677

1.4

Estimated patients per estimated mass casualty incident 1.0 1.0 0.2

1.0

6.0

35.6

Age (years) of estimated patients involved in a mass casualty incident 30.0 18.0 23.8

3.6a

110.0

13,181 a

Patient age is in days.

E. Schenk et al.

413

MASS CASUALITY INCIDENTS

TABLE 4. Demographic characteristics of estimated patients (N = 13,677) involved in mass casualty incidents from the 2010 National EMS Database

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

Characteristics

Sex Female Male Not applicable Not available Not known Not recorded Not reporting Race American Indian or Alaskan Native Asian African American Native Hawaiian or other Pacific Islander White Other race Not applicable Not available Not known Not recorded Not reporting

No.

%

7,013 6,184 175 65 48 116 76

51.3 45.2 1.3 0.5 0.44 0.8 0.6

220 90 2,038 21 7,650 779 593 223 985 393 685

1.6 0.7 14.9 0.2 55.9 5.7 4.3 1.6 7.2 2.9 5.0

DISCUSSION While several studies and case reports have described MCIs in the United States, documentation has been limited to single or small-scale geographic incidents.3−6, 20 This study used national data to estimate the annual incidence and characteristics of MCIs in the United States. The median number of patients per incident found in this study from the 2010 National EMS Database was much lower than expected and was consistently low when validated with the data from the seven states that submitted 100% of EMS responses. Further validation analyses were conducted by matching the field values for incident zip code and the date and time that the unit was notified by dispatch from the MCI subset to the larger 2010 National EMS Database. While some additional EMS responses were found, the median number of estimated patients per incident was consistent with that calculated from the MCI subset. There are several factors that could explain this finding. First, unique patient needs, resource or personnel gaps in the local EMS system, or special environmental conditions could be causing fewer patients to overwhelm the EMS system. This potential explanation is supported by a significantly higher frequency of EMS system delays documented for MCIs. Second, not all patients involved in a particular MCI may be documented (e.g., fatalities or minor injuries/illnesses). In addition, some state and local EMS systems have protocols that permit EMS personnel to reduce the time devoted to patient documentation and tracking during large or complex incidents.21 Such documentation exemptions could also explain the high percentage of

TABLE 5. Presentation of estimated patients (N = 13,677) involved in mass casualty incidents from the 2010 National EMS Database Characteristics

EMS personnel’s primary impression Abdominal pain/problems Airway obstructions Allergic reaction Altered level of consciousness Behavioral/psychiatric disorder Cardiac arrest Cardiac rhythm disturbance Chest pain Diabetic symptoms/hypoglycemia Electrocution Hyperthermia Hypothermia Hypovolemic shock Inhalation injury/toxic gas Obvious death Poisoning/drug ingestion Pregnancy/obstetrics delivery Respiratory arrest Respiratory distress Seizure Sexual assault/rape Smoke inhalation Stings/venomous bite Stroke/CVA Syncope/fainting Traumatic injury Vaginal hemorrhage Not applicable Not available Not known Not recorded Not reporting Primary symptom Bleeding Breathing problem Change in responsiveness Choking Death Device/equipment problem Diarrhea Drainage/discharge Fever Malaise Mass/lesion Mental/psych Nausea/vomiting Pain Palpitations Rash/itching Swelling Transport only Weakness Wound None Not applicable Not available Not known Not recorded Not reporting

No.

%

318 10 36 285 197 69 79 325 75 1 58 16 33 76 108 206 33 9 273 72 12 30 4 59 130 5,562 6 1,443 2,630 303 1,099 120

2.3 0.1 0.3 2.1 1.4 0.5 0.6 2.4 0.5 0.0 0.4 0.1 0.2 0.6 0.8 1.5 0.2 0.1 2.0 0.5 0.1 0.2 0.0 0.4 1.01 40.7 0.0 10.6 19.2 2.2 8.0 0.9

533 414 750 6 163 3 17 8 38 52 6 160 168 5,439 33 25 114 101 292 351 1,289 1,078 860 517 1,108 152

3.9 3.0 5.5 0.0 1.2 0.0 0.1 0.1 0.39 0.4 0.0 1.2 1.2 39.8 0.21 0.2 0.8 0.7 2.1 2.6 9.4 7.9 6.3 3.8 8.1 1.1

414

PREHOSPITAL EMERGENCY CARE

TABLE 6. Mechanism of injury for estimated patients with possible injuries (N = 7,960) involved in mass casualty incidents from the 2010 National EMS Database

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

Characteristics

Cause of injury Aircraft-related crash Bicycle crash Bites Chemical poisoning Drowning Drug poisoning Electrocution (nonlightening) Excessive cold Excessive heat Falls Fire and flames Firearm assault Firearm injury (unintentional) Firearm self-inflicted Machinery injury Mechanical suffocation Motor vehicle non-traffic crash Motor vehicle traffic crash Motorcycle crash Pedestrian traffic crash Stabbing/cutting accident Stabbing/cutting assault Struck by blunt/thrown object Venomous stings (plants/animals) Othera Not applicable Not available Not known Not recorded Not reporting a

No.

%

18 6 13 19 1 8 1 3 14 306 37 41 1 10 15 4 383 4,993 155 25 9 33 480 3 24 822 208 90 157 81

0.2 0.1 0.2 0.2 0.0 0.1 0.0 0.0 0.2 3.8 0.5 0.5 0.0 0.1 0.2 0.1 4.8 62.7 1.9 0.1 0.4 6.0 0.0 0.3 10.3 2.6 1.1 2.0 1.0 0.23

Other includes water transport crashes, smoke inhalation, etc.

missing data related to the EMS personnel’s primary impression found in this study and could indicate biases in the types of patients that are reported. For example, EMS personnel might first respond to and

JULY/SEPTEMBER 2014

VOLUME 18 / NUMBER 3

document high-priority patients, such as children or the severely injured, and be less likely to document other patients when responding to an overwhelming incident. However, at least one patient in each MCI is likely to be documented, suggesting that the estimate of MCIs may be more accurate than the estimate of the number of MCI patients per incident. The inconsistency with which patients may be documented in MCIs highlights the need for development of minimum documentation standards as well as mechanisms to support adequate documentation and exchange of data during incidents that overwhelm EMS systems. Data and information exchange can directly impact the course of treatment for patients in the most resource-efficient manner. Lack of documentation hinders the course of treatment for patients and the ability to come up with data-driven conclusions. Therefore, improving the quality of EMS data is critical, and ensuring the quality of the information in the National EMS Database is a first step toward enacting databased policy changes. Third, the statistically significant differences in response delays and system time intervals between EMS responses that are documented as MCI and those that are not highlight an important finding and suggest that factors causing system delays, rather than the absolute number of patients, may be influencing EMS personnel’s documentation that a response is part of an MCI. Future research should be conducted to understand the factors that affect system delays as well as other elements that contribute to an EMS personnel’s perception of and documentation that an EMS response is an MCI. These factors might include the number of patients, the types of health emergencies, scarcity of resources to care for specific patient(s), and other

TABLE 7. Chi-squared tests of system response delay variables with documentation of the response being a mass casualty incident (MCI) from the 2010 National EMS Database Delay type

Dispatch delay MCI = yes MCI = no Turn-around delay MCI = yes MCI = no Response delay MCI = yes MCI = no Scene delay MCI = yes MCI = no

Total Delay responses (yes and none)

Total (delay = yes)

%

Risk ratio

p-Value

12,500 3,302,305

3,771 114,414

30.2 3.5

8.63

< 0.0001

151,274 4,450,833

6,372 590,364

42.1 13.3

3.17

< 0.0001

159,894 4,849,664

5,891 702,915

36.8 14.5

2.54

< 0.0001

165,624 4,892,073

7,886 602,528

47.6 12.3

3.87

< 0.0001

Definitions of system delay components in the National Emergency Medical Services Information System14 : • Dispatch delay = delays associated with the dispatch of the EMS unit to the patient encounter, including a caller being uncooperative, high call volume, language barrier, inability to obtain location, no units available, scene safety not secure for EMS, technical failure (e.g., computer, phone), etc. • Turn-around delay = delays associated with the EMS unit associated with the patient encounter, including clean-up, decontamination, documentation, emergency department overcrowding, equipment failure, equipment replenishment, staff delay, vehicle failure, etc. • Response delay = response delays of the unit associated with the patient encounter, including crowd, directions, distance, diversion, hazmat, safety, staff delay, traffic, vehicle crash, vehicle failure, weather, etc. • Scene delay = scene delays of the unit associated with the patient encounter, including crowd, directions, distance, diversion, extrication > minutes, hazmat, language barrier, safety, staff delay, traffic, vehicle crash, vehicle failure, weather, etc.

E. Schenk et al.

415

MASS CASUALITY INCIDENTS

TABLE 8. Wilcoxon rank sum tests of documentation of the response being a mass casualty incident (MCI) with system component response time intervals from the 2010 National EMS Database EMS response documented as a MCI Response time component (min)

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

Dispatch center time interval Chute time interval System response time interval Scene time interval Scene to patient time interval Transport time interval Total call time interval

EMS response documented as not a MCI

No. responses

Time mean (SD)

No. responses

Time mean (SD)

p-Value

8,359 10,864 19,891 18,221 11,286 16,132 20,544

9.4 (9.2) 3.2 (5.0) 12.8 (34.6) 34.1 (63.0) 7.2 (30.9) 21.0 (43.8) 93.6 (92.7)

2,452,789 4,502,571 6,954,682 6,046,020 4,365,379 5,598,239 7,487,179

3.1 (6.4) 2.8 (4.5) 12.1 (37.9) 18.6 (37.6) 3.3 (10.5) 18.8 (31.0) 71.3 (88.8)

< 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001

Definitions of response components in the National Emergency Medical Services Information System14 : • Dispatch center time interval = unit notified by dispatch date/time – Public Safety Answering Point (PSAP) call date/time • Chute time interval = unit notified by dispatch date/time – unit en route date/time • System response time interval = unit notified by dispatch date/time – unit arrived on scene date/time • Scene time interval = unit arrived on scene date/time – unit left scene date/time • Scene to patient time interval = unit arrived on scene date/time – arrived at patient date/time • Transport time interval = unit left scene date/time – patient arrived at destination date/time • Total call time interval = unit back in service date/time – unit notified by dispatch date/time

attributes of the environment or situation. A better understanding of how EMS personnel perceive and document MCIs can contribute to improvements in preparedness for MCIs, including the education of personnel. Further research should also be conducted to compare the EMS system response to MCIs with EMS responses that are not recorded as MCIs in order to identify system gaps for strengthening EMS response to MCIs. Future studies based on the National EMS Database could further contribute to national preparedness planning for EMS. The diversity of EMS personnel’s primary impressions reported as MCIs in this study suggests that the causes for EMS systems to become overwhelmed are considerably variable and may be specific to each system. Furthermore, this study suggests that EMS personnel perceive that their systems are being overwhelmed on a day-to-day basis with relatively small numbers of patients compared to large-scale disasters. Lastly, since motor vehicle crashes were found to be a predominant cause of MCIs, there is the potential for highway safety organizations to collaborate with EMS stakeholders to strengthen EMS preparedness and response efforts specifically related to motor vehicle crashes. There are several limitations to this study. First, the National EMS Database is not currently a populationbased and not yet a nationally representative sample. The 2010 National EMS Database contains data from only 32 states and territories, not all of which reported 100% of EMS responses. The National EMS Database does not include data from agencies whose data collection efforts are not NEMSIS compliant or where agency data are not aggregated to a state dataset. To mitigate potential bias from this limitation, the estimate of MCIs was calculated from the states with 100% reporting from EMS agencies. Furthermore, the conclusions drawn from the descriptive analyses were based primarily on the percentages, rather than the absolute frequencies of MCIs or patients. To identify po-

tential bias from EMS runs not submitted to the National EMS Database, the results of the descriptive statistics were validated with the data from the states with 100% reporting. With the exception of urbanicity, similar trends in the percentages of the descriptive statistics and the average number of patients per incident were found in the states with 100% reporting. Lastly, the subjectivity of the NEMSIS definition of MCI presents challenges with understanding and measuring the variable’s validity and reliability. The degree of measurement error on the results is unknown.

CONCLUSION Understanding the types of MCIs in the United States to which prehospital EMS personnel respond is critical for developing national policies and programs for EMS preparedness. This study demonstrated the utility of the National EMS Database for preparedness planning. This study found small-scale MCIs to be relatively frequent, with thousands occurring in the United States each year and an observed incidence rate of 13.0 MCIs per 100,000 population from the states that submitted 100% of EMS runs to the National EMS Database in 2010. A great diversity was found in this study regarding the types of patients involved in MCIs, the health emergencies of which ranged from electrocution to traumatic injury, yet there was a significant association of response time intervals and system delays with EMS responses documented as MCI. Given that the majority of MCIs occurred in urban settings, in the evening hours, and on streets or highways, EMS offices should also collaborate with state highway safety offices and state highway transportation officials in preparedness planning. This study showed the need for development of minimum documentation criteria and mechanisms to support adequate documentation during an MCI. There is also a need for further investigation into EMS personnel’s perceptions

416 and documentation of MCIs, especially with regard to factors that affect system response times and delays.

Prehosp Emerg Care Downloaded from informahealthcare.com by Nyu Medical Center on 06/14/15 For personal use only.

References 1. Doyle CJ. Mass casualty incident: Integration with prehospital care. Emerg Med Clin North Am. 1990;8:163–75. 2. Chaput CJ, Deluhery MR, Stake CE, Martens KA, Cichon ME. Disaster training for prehospital providers. Prehosp Emerg Care. 2007;11:458–65. 3. Branas CC, Sing RF, Perron AD. A case series analysis of mass casualty incidents. Prehosp Emerg Care. 2000;4:299–304. 4. Eastman AL, Rinert KJ, Nemeth IR, Fowler RL, Minei JP. Alternate site surge capacity in times of public health disaster maintains trauma center and emergency department integrity: Hurricane Katrina. J Trauma. 2007;63:253–7. 5. Irvin CB, Atas JG. Management of evacuee surge from a disaster area: solutions to avoid non-emergent, emergency department visits. Prehosp Disaster Med. 2007;22:220–3. 6. Sloan HM. Responding to a multiple casualty incident: room for improvement. J Emerg Nursing. 2011;37:484–6. 7. U.S. Department of Health and Human Services. National Health Security Strategy. Available at: www.phe.gov/prepar edness/planning/authority/nhss/strategy/documents/nhssfinal.pdf. Accessed March 8, 2013. 8. American Medical Association/American Public Health Association. Improving Health System Preparedness for Terrorism and Mass Casualty Events. Available at: www.amaassn.org/resources/doc/cphpdr/final summit report.pdf. Accessed September 1, 2012. 9. Dawson DE. National Emergency Medical Services Information System. Prehosp Emerg Care. 2006;10:314–6. 10. National Highway Traffic Safety Administration Office of Emergency Medical Services. Available at: www.ems.gov. Accessed May 6, 2013. 11. National Emergency Medical Services Information System. Available at: www.nemsis.org. Accessed August 18, 2012. 12. National Association of State EMS Officials. Extended Definition Document NEMSIS/NHTSA 2.2.1 Data Dictionary. Available at: www.nemsis.org/referenceMaterials/ documents/Data Managers Council - Data Definitions Proj ect Final Ve. 000.pdf. Accessed October 10, 2012. 13. National Emergency Medical Services Information System. NEMSIS Data Dictionary V 2.2.1. Available at: www.nemsis.

PREHOSPITAL EMERGENCY CARE

14.

15.

16.

17.

18.

19.

20.

21.

22.

JULY/SEPTEMBER 2014

VOLUME 18 / NUMBER 3

org/v2/downloads/documents/NEMSIS Data Dictionary v2. 2.1 04092012.pdf. Accessed September 1, 2012. U.S. Census Bureau. Census Regions and Divisions of the United States. Available at: www.census.gov/geo/www/ us regdiv.pdf. Accessed March 8, 2013. National Emergency Medical Services Information System. NEMSIS Research Data Set v.2.2.1 2011 User Manual. Available at: www.nemsis.org/reportingTools/documents/ NEMSISRDS2212011UserManual.pdf Accessed March 8, 2013. U.S. Census Bureau. Population Estimates. Available at www.census.gov/popest/data/state/totals/2011/index.html. Accessed October 2, 2012. Federal Interagency Committee on Emergency Medical Services. 2011 National EMS Assessment. U.S. Department of Transportation, National Highway Traffic Safety Administration, DOT HS 811 723. Washington, DC, 2012. Available at www.ems.gov. Wang HE, Mann NC, Jacobson K, Dai M, Mears G, Smyrski K, Yealy DM. National characteristics of emergency medical services responses in the United States. Prehosp Emerg Care. 2013;17:1–7. National Highway Traffic Safety Administration Data Resource Website. Available at: www-nrd.nhtsa.dot.gov/CATS/ index.aspx. Accessed March 12, 2013. Kahn CA, Schultz CH, Miller KT, Anderson CL. Does START triage work? An outcomes assessment after a disaster. Ann Emerg Med. 2009;54:424–30. Snohomish County. Multiple Casualty Incident Plan. Available at: www.snocountychiefs.org/documents/50-0111 pre incident planning.pdf. Accessed September 12, 2012. National Emergency Medical Services Information Systems (NEMSIS): Current Composition of NEMSIS Data Warehouse. Available at www.utahdcc.org/Reports/Pages/Report. aspx?ItemPath=%2fNEMSIS%2fView+National+Reports%2f Data+Quality%2fCurrent+Composition+of+the+NEMSIS+ Data+Warehouse. Accessed August 15, 2012.

SUPPLEMENTAL MATERIAL AVAILABLE ONLINE Appendix A: Composition of the 2010 NEMSIS Data Warehouse. Supplementary material can be viewed and downloaded at http://informahealthcare.com/pec