Spaite 1995 Annals Of Emergency Medicine

CONCEPTS

Uniform Prehospital Data Elements and Definitions: A Report From the Uniform Prehospital Emergency Medical Services Data Conference From the Arizona Emergency Medicine Research Center, University of Arizona, Tucson*; EM5 Data Systems, Inc, Phoenix, Arizona~; Department of Emergency Medicine, Stanford UniversitS, EMS Bureau, Montana Department of Health and Environmental 5dences, State of Montana II, EMS Division, National Highway Traffic Safety Administration~; Division of Trauma and EMS, Health Resources and Services Administration#; and Uniformed Services University* *; National Center for Injury Prevention and Control, Centersfor Disease Control and Prevention"~; and PACE Evaluation, SepuIveda Veterans' Administration Medical Center, Sepulveda, California.~

Daniel Spaite, MD, FACEP* Ronald Benoit, BS* Douglas Brown, CEP* Richard Cales, MD, FACEP§ Drew Dawsonll Chuck Glass ~ Christoph Kaufmann, MD, MPH, FACS# ** Daniel Pollock, MD # Susan Ryan11 Elizabeth M Yano, PhD§§

Receivedfor publication November 9, 1994. Accepted for publication November 14, 1994. Fundingfor this project was supplied by NHTSA, contract no. DTNH22-92-C-05314, Uniform Prehospital EMS Data Conference. Copyright © by the American College of Emergency Physicians.

One of the distinct and universal aspects of emergency medical service (EMS)is the belief that before its implementation many people were dying or being killed by ill-equipped, poorly trained "hearse drivers" and that this tragic state of affairs has been rectified by the advances in the prehospital phase of care. Except for cases of nontraumatic, out-of-hospital cardiac arrest there is almost no convincing scientific evidence to prove that prehospital care has had an impact on morbidity or mortality. At the very foundation of this problem is the lack of a set of broadbased, well-conceived, accurate, reliable, uniform EMS data. Many attempts have been made to develop a uniform EMS data set, but without a national consensusthese have not achieved wide distribution. In 1992, with the assistance of the National Highway Traffic Safety Administration, the national consensus process began with a series of meetings involving many EMS agencies and organizations. This culminated in August 1994 with the development of an 81-item uniform EMS data set. We detail the prior attempts at data set development and outline the process leading to the this uniform, national EMS data set. [Spaite D, Benoit R, Brown D, Cales R, Dawson D, Glass C, Kaufmann C, Pollock D, Ryan S, Yano EM: Uniform prehospital data elements and definitions: A report from the uniform prehospital emergency medical services data conference. Ann Emerg Med April 1995;25:525-534.] INTRODUCTION The development of modern emergency medical services (EMS) systems has had a profound impact on the expectations of American citizens of immediate access to and continuous availability of emergency medical care outside the confines of traditional health care facilities. The development of sophisticated prehospital care systems has also created a unique culture among those who provide this care. One of the distinct and nearly universal aspects of this culture is the intense belief that in the "old days,"

APRIL 1995

25:4

ANNALS OF EMERGENCY MEDICINE

52 5

EMS DATA Spaite et aI

people died because of or were killed by ill-equipped, poorly trained "hearse drivers," a state of affairs, most assume, that has been rectified by the advent of modern EMS systems. Advances in prehospital treatment have markedly decreased the morbidity and mortality due to serious illness and injury Furthermore, many believe that providing care in the field is more advantageous to the patient than subsequent inhospital care, even if it is optimal. This sense of a lifesaving mission has captured public interest and commanded societal resources that can only be measured in the tens of billions of dollars over the past two decades. Amazingly, except for the case of nontraumatic, outof-hospital cardiac arrest in the urban setting, there is almost no convincing scientific evidence to prove that prehospital medical care has had an impact on the morbidity or mortality of illness or i n j u ~ At the very foundation of this problem lies the lack of a broad-based, wellconceived, accurate, reliable collection of uniform EMS data. One of the most fundamental of all current needs in EMS is access to prehospital patient and system information from the wide variety of systems that exist in this country. Without uniform EMS data, there is no hope that the many pressing questions surrounding outcome, impact, and cost-effectiveness will be answered. Akhough [he mere existence of a national uniform data set and definitions alone is not sufficient to answer such questions, the absence of one makes it impossible. THE HISTORY OF EMS DATA COLLECTION To fully appreciate the problems related to [he lack of EMS data collection, it is important to recognize that this lack is not a new issue. The landmark paper "Accidental Death and Disability: The Neglected Disease Of Modern Society," published in 1966 by the National Academy of Sciences (NAS) and the National Research Council (NRC), firs[ noted the lack of adequate prehospital data collection: Data are lacking on which to determine the number of individuals whose lives are lost or injuries are compounded by misguided attempts at rescue or first aid, absence of physiciansat the scene of injury, unsuitable ambulanceswith inadequate equipment and untrained attendants, lack of trafficcontrol, or the lack of voice communicationsfacilities.~ The paper recommended the development of hospitalbased trauma registries [o collect information from both the prehospital and hospital settings "as a mechanism for

526

the continuing description of the natural history of the various forms of injuries." The first federal legislation to address EMS, the Highway Safety Act of 1966 (much of which was based on the NAS/NRC paper) was essentially devoid of language directed at the formation of prehospital data collection systems. However, EMS system evaluation guidelines, based on demonstration projects, were published in 1974 by the National Highway Traffic Safety Administration (NHTSA) in the Highway Safety Program Manual 11 and were recommended to the Governors' Highway Safety Representatives for use in their EMS programs. On the basis of these guidelines and demonstration projects in New York, NHTSA published specific ambulance and emergency department report forms in Appendix S in the Highway Safety Program Manual 11. But, as Boyd points out, "These funds, except in few i n s t a n c e s . . , did not stimulate the development of medically accountable EMS systems of care. ''2 The first federal legislation to specifically address data collection in EMS was the Emergency Medical Services System Act of 1973 (EMSSA). This legislation stimulated comprehensive EMS systems development throughout the nation and stipulated that every EMS system comprise 15 mandatory system components• Component 11 was "Standard Record Keeping," which was to: provide for a standardized patient record keeping system meeting appropriate standards as established by the Secretary, which records shall cover the treatment of the patient from initial entry into the system through his discharge from it, and shall be consistent with ensuing patient records used in follow-up care and rehabilitation of the patient• •

.

.

NHTSA and the Department of Health and Human Services (DHH) proceeded to encourage and urge their respective EMS constituencies to use prescribed standard data elements for EMS reporting and evaluation. Each regional system was to provide for a coordinated record keeping system including linked prehospital, hospital, and critical care records. To facilitate patient care evaluation, the data elements were also to be consistent with those in patient records used in follow-up care and rehabilitation. The minimal patient record was to comprise dispatcher records, ambulance records (both basic life support and advanced life support [ALS]), and all hospital records. Key elements of a coordinated EMS record were specified. However, the EMS regions that applied for or received federal funds experienced major problems in implementing this component. The myriad problems earned this component the reputation of being the most difficult of all the EMS systems components to

ANNALS OF EMERGENCY MEDICINE

25:4

APRIL 1995

EMS

DATA

5paite et al

implement) Ironically, despite the proposed complexity and significance of "standard record keeping," none of the seven federal EMS Demonstration Projects, which preceded EMSSA, evaluated the feasibility of carrying out this component. While developing the Standard Record Keeping component, the federal government contracted with Macro Systems, Inc, to develop the original EMS minimum data set (MDS). The Health Services Administration (HSA), which was responsible for implementing the EMSSA, determined it was not feasible to impose standardized recordkeeping forms and procedures on grant recipients. <5 Consequently, HSA recommended that 20 data elements serve as a minimum data set for all EMS systems. 5 The data set was never widely distributed or officially incorporated into EMS data collection systems to any significant degree. Furthermore, it failed to resolve problems associated with data collection or analysis. However, a systems evaluation workbook was published in 1976 to promote efficient data collection techniques and appropriate evaluation methods# In the 1976 amendments to EMSSA, Congress mandated a study to identify the category of patients to be included in a uniform reporting system that would evaluate the effectiveness of EMS in reducing death and disability Congress expressed the hope that "the development of a uniform reporting system would discourage the development of a multiplicity of incompatible reporting systems that might prove inadequate for evaluating the effectiveness of EMS systems and result in unnecessary duplication of effort." In 1977, the HSA responded to the congressional mandate by initiating "program abstracts" to evaluate EMS systems. The mandatory abstracts specified groups of patients that received care in the seven critical care areas of EMS systems (major trauma, burns, spinal cord injuries, myocardial infarction, poisonings, high-risk obstetrics/neonates, psychiatric) would be used as "tracers" for outcome determination. The outcomes of these patients would supposedly help the evaluation of the effectiveness of an entire EMS system. In a 1978 study performed to validate the data abstraction method, it was determined that statistics describing the incidence of patients failing within the critical care categories could be reliably collected from EMS systems, thereby meeting HSA reporting requirements. However, it was also concluded that HSA had failed to clearly state the data definitions and instructions necessary to collect appropriate data to evaluate outcomes. 7

APRIL 1995

25:4

ANNALS OF EMERGENCY MEDICINE

In 1981, supported by a grant from the NHTSA, the Health Operations Research Group at the University of Pittsburgh, through the Pennsylvania EMS Division, examined the feasibility and components of an ALS MDS for the Commonwealth of Pennsylvania. s The research group used a standardized consensus format to develop an extensive and comprehensive list of 86 data points for inclusion on the patient care record (PCR). Despite its comprehensive nature, this MDS was never significantly used. A National EMS Management Information Workshop was convened in 1983 to evaluate issues in EMS management information systems (MIS). Two of the four paper sessions focused on data needs and data sources. Most of the papers presented described the development of statewide uniform EMS PCRs9-1 t and noted that linkage with hospital discharge data and police records would facilitate system evaluation.9,zl, 12

In the late 1980s, the American Society for Testing and Materials (ASTM) Committee F-30.03.03 on EMS/MIS attempted to develop an MDS. Although several years of committee work yielded many drafts, the ASTM consensus process did not lead to a final, approved product. In 1990, Hedges and Joyce proposed an MDS based on much of the aforementioned historical work. ~3 Despite these attempts to develop an MDS, a forum was never provided to assure broad input from the enormous cadre of stakeholders. Thus none of the data sets received widespread attention at the local, state or, regional level. Because it had become clear that failure to provide an effective process at the national level would doom MDS development to an endless cycle of failure, NHTSA solicited support from other federal agencies to aid in developing a consensus forum (Table 1). THE PURPOSES OF PREHOSPITAL DATA COLLECTION

The main purposes of prehospital data collection include the following. (1) Data collected in the field represent the legal documentation of the patient encounter. Independent of all other issues, this information serves the primary purpose of being the prehospital patient medical record. (2) In many systems, prehospital information is necessary for billing of services provided to the patient. Thus the financial resources to maintain the system rely directly on the collection of this information. (3) The PCR serves as the foundation for most system quality-improvement programs. (4) Accurate prehospital data collection is fundamental to all system evaluations

52 7

EMS DATA Spaite et al

and alterations. (5) Research necessary to answer important questions depends on prehospital data. This is especially true in attempts to identify cost-effectiveness and impact on patient outcome. (6) The evaluation of broad public health issues is growing increasingly dependent on prehospital data. (7) The ability to properly allocate societal resources must be tied to accurate prehospital information analyzed from local, regional, state, and national perspectives. OBSTACLES TO UNIFORM PREHOSPITAL DATA COLLECTION AND USE

Despite a widespread sense among EMS professionals that uniform data collection and reporting are important to the future of EMS, little progress has been made in this endeavor. The main reason is that EMS funding, administration, expansion, system maintenance, protocol development, and medical direction often remain under local control. Few would argue that this should change, because of the diversity of geography, demographics, resources, and patient populations found among EMS systems. Consequently, few local system administrators or medical directors have a perspective or concern beyond the problems and issues that they face in their own small subsystems. Thus nearly no one has been concerned with helping answer questions that are not of obvious local importance. In essence, each system °'owns" its own data (what is collected, how it is collected, how it is used), and its global importance is an issue of little concern. Unfortunately, many essential questions can onIy be answered by the compilation and sharing of meaningful information from many EMS microcosms. 14-~6 Limited resources pose a second barrier to uniform data collection and reporting. This is true on several fronts. First, large-scale collection, reporting, and aggreTable 1.

Sponsoringfederal agencies. National Highway Traffic Safety Administration, US Department of Transportation The following organizations of the US Department of Health and Human Services: Division of Trauma and Emergency Medical Systems, Health Resources and Services Administration National Center for Injury Prevention and Control, Oenters for Disease Control NationaF Heart, Lung, and Blood Institute, National Institutes of Health Maternal and Child Health Bureau, Health Resources and Services Administration Office of Rural Health Policy, Health Resources and Services Administration Office of Coverage and Eligibility Policy, Health Care Finance Administration Office of Science and Data Development, Administration for Health Care Policy and Research US Fire Administration, Federal Emergency Management Agency

528

gation of data would require substantial financial resources. Because such resources have not been forthcoming from federal or state governments, funding would have had to come from local and regional entities. Clearly, this has not occurred; nor would it have been predicted, for reasons detailed earlier. Second, even if adequate financial resources were available, few local EMS system personnel have the information-management experience or expertise in development and implementation to produce a system that is appropriate and responsive to local and national needs. Third, even in systems where appropriate expertise exists, few are associated with research institutions where experience in system evaluation, medical informatics, statistical analysis, and cost-effectiveness analysis exists.lZ It is not surprising that local EMS agencies have seen little need to develop sophisticated and expensive data collection processes that would remain virtually unused. The final barrier to the development of uniform data collection and reporting has been the lack of a lead federal agency to direct a national consensus process. To remove this barrier and to help assure widespread acceptance, NHTSA built a coalition of federal agencies and nonfederal organizations. With a national, consensus-based uniform data set established, it is hoped that substantial efforts will be made to deal with the other barriers at the federal, state, regional, and local levels. DEVELOPMENT OF THE DATA SET STRAWMAN

The Uniform Prehospital EMS Data Conference was convened in an effort to establish a consensus-based national EMS data set. During the development process, attendees evaluated current prehospital data elements, building and refining definitions as appropriate. The conference established, as a goal, the development of consensus statements Table 2.

No@derd stakeholder groups represented ~n planning. American Academy of Pediatrics American Ambulance Association American College of Emergency Physicians American College of Surgeons American Hospital Association American Society for Testing Materials EMS Data Systems Incorporated National Association of EMS Physicians National Association of EMTs National Association of State EMS Directors National Council of State EMS Training Coordinators International Association of Fire Chiefs

ANNALS OF EMERGENCY MEDICINE 25:4

APRIL 1995

EMS D A T A

Spaite et aI

that would lead to the implementation of a widely accepted and utilized data set to be modified and refined through research. The national consensus process began in 1992 with the convening of a series of meetings including many stakeholder organizations (Tables 1 and 2). During this process, a subgroup of the planning committee was given the task of developing a "strawman" document to include potential "core" and "supplemental" data elements and definitions. A conceptual change led to the terminology of uniform data set (UDS) instead of minimum data set. This change reflected several issues: First, the term "minimum" was a misnomer, given the broad scope of the project and the predicted final product. With the inclusion of both core and supplemental elements, the data set could no longer be considered minimum. Second, the term "minimum" might imply a relatively fixed, if not stagnant, group of elements. It was clearly understood from the outset that the consensus group product would only represent a beginning, with the full intent for future evaluation, revision, and improvement. Third, dealing only with the minimum data set would continue to leave the supplemental elements in the same quagmire of local, regional, and state variation that had typified the previous data set. Finally, compatibility with the monumental work done by the Utstein Consensus Conference zs for prehospital cardiac arrest data was important, even though some of these data elements could not be considered core for many local E M S systems. CRITERIA FOR INCLUSION OF PREHOSPITAL DATA ELEMENTS

During UDS strawman development, meetings of the data set development task group provided for extensive discussion of each potential data element. The rationale for inclusion of a given data element are listed in Table 3. For a data element to be regarded as core, it was mandatory that it be considered an essential part of the medical record in any local EMS system. Many of the core elements had other rationales for inclusion, but appropriateness for the medical record was a prerequisite. Although one of the intended uses of the uniform data is evaluation of system-related questions, usefulness for evaluative or epidemiologic purposes was not sufficient to qualify a data element as core. The rationale was simple: If data elements were considered core despite the fact that their primary usefulness was for research or evaluation, the likelihood of their widespread acceptance would be compromised.

APRIL 1995

25:4

ANNALS OF EMERGENCY MEDICINE

The inclusion of supplemental data elements in the consensus process was important. Although some might argue that only core elements should have been included, exclusion of the supplemental elements would have been detrimental in several ways. First, many of the supplemental elements are extremely important for EMS system evaluation and quality improvement at the local, regional, and state levels. However, many of the supplemental elements could not realistically be considered core because it would be unfeasible to collect them in all systems. Second, many riMS systems alrea@ collect many of the supplemental elements. Failure to include these as a part of the uniform data set might discourage systems from continuing to collect this important information. Third, inclusion of the supplemental data elements may serve to stimulate future improvements in EMS data collection. We hope that many EMS systems will consider collecting some or all of the supplemental elements in an effort to provide more detailed and useful information for quality improvement and research. Finally, inclusion of the supplemental elements may provide valuable insight into future core elements. THE UNIFORM PREHOSPITAL EMS DATA ELEMENT CONFERENCE

In August 1993, the consensus conference convened in Arlington, Virginia. A diverse group of participants represented essentially every conceivable organization with an interest in EMS. The data set strawman was made available to all conference participants before their arrival. A National Institutes of Health consensus process was used to develop the final product. 19 Aided by the Data Set Development Task Group, a panel of 12 expert referees (Table 4) heard testimony during open forums on each of the data elements and definitions. Having heard the extensive testimony, the expert panel convened to develop the final data set and definitions (Appendix). It should be noted that the consensus panel changed the term "core" to ~essential" in the final document. According to the consensus panel, an essential data element is one that is cruTable 3. Rationale for inclusion in data set. Medical record System evaluation Quality improvement Billing Medicolegal Research

529

EMS DATA 5paite et aI

cial for the basic operations of an EMS service and that can serve a purpose at the regional or national level. The term "supplemental," which had been used by the data set development task group, was changed to "desirable." The panel noted that desirable elements may well be critical to local operations but may not be considered critical in all jurisdictions or situations. One significant advantage to this change is the fact that the terms are both descriptive and essentially self-defining. It was the consensus of the panel that these recommendations should be implemented at the state, regional, or local level as appropriate. The purpose of a uniform data set and definitions is to provide common terminology and definitions to be used in the evaluation of EMS. Although the data set is useful in describing the prehospital aspects of care, it lacks outcome measures that would strengthen the evaluation process. It is to be considered a first step in the process of EMS system evaluation. LINKAGE OF PREHOSPITAL DATA WITH OTHER INFORMATION SOURCES

Even if prehospital data collection and management were being performed perfectly on a national scale, it would not be adequate, because even well-conceived prehospital databases carried out in a vacuum are of limited use. Simply stated, prehospital information that remains unlinked with hospital and autopsy outcome data has limited meaning. In fact, the failure of this linkage is one of the main reasons that so little evidence exists with regard to the impact of prehospital care on patient outcome. 2° Although few people involved in the prehospital, hospital, and rehabilitative aspects of patient care would argue against the importance of linkage, the obstacles to information sharing are formidable. In addition to the Table 4.

Composition of conference consensus panel. State EMS director State EMS training coordinator Ambulance service manager--private EMS manager--public, fire-based Regional/county EMS manager EMS medical director Emergency physician Emergency nurse Trauma systems director Epidemiologist EMS researcher Panel chairperson

530

barriers of patient confidentiality and anonymity of agencies and personnel, most EMS agencies do not employ personnel with the technical expertise to properly link separate databases. The complexities and logistics of emergency medical system informatics are certainly not trivial. It is hoped that the parallel work toward a trauma care system uniform data set under the auspices of the Division of Trauma and EMS at the Health Resources and Services Administration will provide the opportunity to develop a model for linkage of prehospital and hospital information. Optimally, future cooperation between federal and nonfederal agencies will provide the support necessary to successfully deal with this enormous task. THE FUTURE: IMPLEMENTATION OF THE UNIFORM PREHOSPITAL DATA SET

As stated previously, the product of the consensus conference should only be considered a beginning. Although it is hoped that the substantial progress made by the conference has yielded a product that is both useful and valuable, research and evaluation of the UDS must begin immediately Pilot testing of the data set in frontier, rural, suburban, and urban settings will be necessary to determine the feasibility of widespread use. In addition, the process of aggregating, analyzing, and using information at the regional, state, and federal levels will be extremely important. If quality information is gathered but not shared, the global impact of this data set will be only minimal. It is important to note that the national conference did not develop consensus on the data item dictionary. For instance, the data element "Signs and Symptoms Present" was discussed and considered during the consensus process. It did receive a definition and a priority assignment of essential. However, the items on the data item content list (ie, abdominal pain, back pain, childbirth) were not formal subjects of the consensus conference. Therefore this article does not include the extensive data dictionary issues, only cursory comments and examples. The Division of EMS at NHTSA continues to work with the consensus panel and the data set development task group to resolve these remaining issues before publication of the final federal document. Copies of that document will be available through the EMS Division of NHTSA. Finally, the existence of a widely accepted uniform data set does not ensure that accurate and complete information will be collected in the field. 21-22 Much work remains to be done in evaluating the many barriers to the process of obtaining high-quality information collected in the prehospital setting. If this process results only in the collec-

ANNALS OF EMERGENCY MEDICINE 25:4 APRIL 1995

EMS DATA

Spaite et al

tion and reporting of a large amount of bad data, it will simply lead to bad conclusions that are believed with greater fervency. It is the responsibility of EMS professionals at every level to ensure that such a situation does not occur. Future conferences should be convened to review the experience of systems that implement the UDS. This experience, combined with research and evaluation of the data set, will form the basis for revision and improvement. It is hoped that this process, along with improved linkage of information, will yield answers to the fundamental questions about the cost-effectiveness and outcome impact of EMS systems.

10. Houston J: Consensusdesign and development of an effective, workable management information system for a rural EMS system: The Dartmouth system revisited. In Dethlefs WC, Ham KS (eds}, Proceedingsof the NationalEMSManagementInformationSystemsWorkshop. Harrisburg, Pennsylvania,October 1983. 11. Ham KS: Consensusdevelopment for identification of data elements and implementation among diverse regional EMS systems, in Dethlefs WC, Ham KS (ads), Proceedingsof the NationalEMSManagementInformationSystemsWorkshop.Harrisburg, Pennsylvania,October 1983. 12. Larson D: Statewide collection of EMS-related information: The Utah vehicular trauma study, in Dethlefs WC, Ham KS (eds), Proceedingsof the NationalEMSManagementInformation SystemsWorkshop.Harrisburg, Pennsylvania,October 1983. 13. HedgesJR, Joyce SM: Minimum data set for EMS report form: Historical development and future implications, PrehospDisasterMad 1990;5:383-388. 14. Cummins DO: Section Ih Meving toward uniform reporting and terminology. Ann EmergMed 1993;22:33-36. 15. HedgesJR: Beyond Utstein: Implementation of a multiseurce uniform data base for prehospita[ cardiac arrest research.Ann EmergMed1993;22:41-46.

REFERENCES 1. National Academy of Sciencesand the National ResearchCouncil: Accidental Death and Disability: The Neglected Disease of MOdernSociety. Washington, DC. September 1966. 2. Boyd DR (ed):A symposium on the Illinois trauma program: A systems approachto the care of the critically injured. J Trauma1973;13:275,320. 3. Boyd DR: The history of emergencymedical systems (EMS) in the United States of America, in Beyd DR, Edlich RF, Sylvia M (eds), SystemsApproachto EmergencyMedical Care.Norwalk, Connecticut, Appleton-Century-Crofts, 1983. 4. Steele R: Developmentof a Minimal Data Set for EmergencyServices Patient RecordKeeping. Macro Systems, Inc. Preparedfor the Health Services Administration, National Technical Information Services (NTIS),July 1974. 8. Birch H: Guidelines for Patient Record Keeping Systems for EmergencyMedical Services.Vol Ih Model System Guidelines for Patient RecordKeeping and Management Reporting. Macro Systems, Inc. Preparedfor the Health ServicesAdministration, NTIS, September 1974,

16. Cummins RO: The Utstein style for uniform reporting of data from out-of- hospital cardiac arrest. Ann EmergMad 1993;22:37-40. 17. Spaite DW, ValenzuelaTD, Meislin HW: Barriers to EMS system evaluation: Problemsassociated with field data collection. PrehospDisasterMad 1993;8(suppl):S35-S40. 18, Cummins RO, ChamberlainDA, Abramsen NS, et at: Recommendedguidelines for uniform reporting of data from out-of-hospital cardiac arrest: The Utstein style. Ann EmergMed 1991;29:861-874. 19. Guidelines for the selection and management of consensusdevelopment conferences: Office of Medical Applications of Research,National Institutes of Health. 20. Maid R: The utilization of public health research models in the evaluation of EMS systems. San Francisco,California, 23rd Annual Meeting of the Society for Academic Emergency Medicine, May 1993. 21. MossessoVN: The most neglected tool in EMS: The clock. Ann EmergMed1993;22:13111312.

6. Evaluationworkbook for EMS. US Department of Health, Educationand Welfare, Public Health Service, Health Services Administration, Division of EmergencyMedical Services. DHEW Publications {HSA)76-2021, August 1976.

22. Spaite DW, Han[onT, Cfiss EA, et ah Prehospita[data entry compliance by paramedics after institution of a comprehensiveEMS data collection tool. Ann EmergMeg1990;19:1270-1273.

7. Reliability of accuracy,completeness, and comparability of the emergencymedical services systems data needed to meet reporting requirements of Public Laws 93-154 and 94-573. Arthur Young and Co. Final report {draft) submitted to the Health Services Administration January 1980.

The authors thank the members of the Data Conference Panel: Chairman, J Michael Dean, MD; Jim Dowser; Rick Buell; Herbert Garrison, MD; W Briggs Hopson, MD; Leonard Inch; Chief Jack Krakeel; Ronald MaiD, DO; Peter Pens, MD, FACEP; Wade Spruill, Jr; Patricia Waller, PhD; and Marie Wilson, RN.

8. Wolfe H, Shuman LJ, Esposite G, et ah DevelopmentefALSMinimumDataSetandBLS Algorithms. University of Pittsburgh, August 1981. 9. Johnson SW: Developmentof a uniform ambulance run report and the use of nominal and delphi group process techniques to reach consensus,in DetNefs WC, Ham KS (eds), Proceedingsof the NationalEMSManagementInformationSystemsWorkshop.Harrisburg, Pennsylvania, October 1983.

Reprint no. 47/1/62558 No reprints available from the author.

Appendix.

Uniform EMS data elements and definitions. Data Element

Priority

Definition

Comment

Address (or best approximation) where patient was found or, if no patient, address to which the unit responded City or township (if applicable) where patient was found or to which unit responded (or best approximation} County or parish (if applicable} where patient was found or to which unit responded (or best approximation)

Free text

1.

Incident address

Essential

2.

Incident city

Essential

3.

Incident county

Essential

APRIL 1995

25:4

ANNALS OF EMERGENCY MEDICINE

Numeric entry Numeric entry

53 1

EMS DATA Spa~te et aZ

Appendix, continued Data Element

Priority

Definition

Comment

State, territory, or province (or District of Columbia) where patient was found or to which unit responded Type of location of incident Date of onset of symptoms or injury date

Alphanumeric entry

4.

Incident state

Essential

5. 6.

Location type Onset date

Essential Desirable

7. 8.

Onset time Date incident reported

Desirable Essential

9.

Time incident reported

Essential

10.

Time dispatch notified

Essential

Time of first connection with EMS dispatch

11. 12.

Date unit notified Time unit notified

Desirable Essential

13.

Time unit responds

Essential

Date on which response unit is notified by EMS dispatch Time at which response unit is notified by EMS dispatch in seconds and clock synchronization strongly encouraged. Time at which response unit begins physical motion

14.

Time of arrival at scene

Essential

15.

Time of arrival at patient

Desirable

16.

Time unit leaves scene

Essential

Time when response unit begins physical motion from scene

17.

Time of arrival at destination

Essential

Time when patient arrives at destination or transfer point

18.

Time back in service

Essential

19. 20. 21. 22.

Lights and sirens to scene Service type Incident number Responsenumber

Essential Essential Essential Essential

Time that response unit is back in service and available for response Use of lights and sirens en route to scene Type of service requested Unique number for each incident reported to dispatch Unique number for each individual response by a response unit to an incident

23. 24.

PCR number Agency/unit number

Essential Essential

25. 26.

Vehicle type First crew member number

Essential Essential

27.

Second crew member number

Essential

28.

Third crew member number

Desirable

29.

Crew member l type

Essential

Personnel certification/license level of crew member

30. 31. 32. 33. 34. 35. 36.

Crew member 2 type Crew member 3 type Patient name Patient street address City of residence County of residence State

Essential Desirable Essential Desirable Desirable Desirable Desirable

Personnel certification/license level of crew member Personnel certification/license level of crew member Patient name Patient's street address (if applicable) Patient's city or township of residence (if applicable) County or parish where patient resides (if applicable) State, territory, or province (or District of Columbia) where patient resides

5 32

Time of onset of symptoms or injury time Date the call was first received by a public safety answering point (PSAP) or other designated entity Time call was first received by PSAP or other designated entity

Time EMS unit stops physical motion at scene (last place that the unit or vehicle stops before assessing patient) Time at which response personnel establish direct contact with patient

Unique number for each PCR Number identifies the agency and unit responding to an incident Type of vehicle that responded to an incident Personnel certification/license number for first crew member Personnel certification/license number for second crew member Personnel certification/license number for third crew member

Examples: residence, public building, farm May differ from the date of EMS response. May be numerically coded.

Starting point of the EMS response. Measurement in seconds and clock synchronization strongly encouraged. Permits assessment of delays between time incident reported and the notification of EMS dispatch. Measurement in seconds and clock sychronization strongly encouraged. May be numerically coded. Permits measurement of response and possible delays. Measurement Permits measurement of interval he~een notification and actual mobilization of response unit. Measurement in seconds and clock synchronization strongly encouraged. Measurement in seconds and clock synchronization strongly encouraged. Identifies earliest time at which EMS care can actuafly begin. Measurement in seconds and clock synchronization strongly encouraged. Measurement in seconds and clock synchronization strongly encouraged. May reflect time of rendezvous with another EMS unit. Measurement in seconds and clock synchronization strongly encouraged. Permits measurement of total out-of-service interval. Yes, no, upgraded, downgraded Items such as scene, interfacility, standby Alphanumeric entry Each unit responding to a single incident would have the same incident number but a unique response number. Useful in constructing agency or unit-specific reports. Examples: ground, rotorcraft, fixed-wing, other. Identifies personnel involved in response, patient care, or both.

A given agency may desire to list three or mere personnel. The ability to list at least two is essential. Examples: first responder, emergency medical technician basic, paramedic, nurse

Free text Free text; patient's place of residence, if known Numeric entry Numeric entry Alphanumeric entry

ANNALS OF EMERGENCY MEDICINE 25:4

APRIL 1995

EMS DATA 5pa~te et aI

Appendix, continued Data Element

Priority

Definition

Comment

37.

Zip code of resident

Essential

Zip code of patient's residence

38. 39. 40. 41. 42. 43.

Telephone number Social Security number Date of birth Age Gender Race/ethnicity

Desirable Desirable Essential Desirable Essential Essential

Patient's main telephone number Patient's Social Security number Patient's date of birth Patient's age or best approximation Gender of patient Patient's ethnic origin

44.

Destination/transferred to

Essential

45.

Destination determination

Essential

Health care facility or prehospital unit/home that received patient from EMS responder providing record Reason a transport destination was selected

46. 47.

Lights, sirens used from scene Essential Incident/patient disposition Essential

Use of lights, sirens, or both from the scene Result of EMS response

48. 49.

Chiefcomplaint Causeof injury

Desirable Essential

Statement of problem by patient or other person External cause of injury

50.

Provider impression

Essential

Previder's clinical impression that led to the management given to the patient (treatments, medications, procedures)

51.

Preexisting condition

Essential

Preexisting medical conditions known to the provider

52.

Signsand symptoms

Essential

Signs and symptoms reported to or observed by provider

53.

Injury description

Essential

Olinical description of injury type and body site

54.

Injury intent

Desirable

Intent of individual inflicting injury

55.

Safety equipment

Essential

Safety equipment in use by patient at time of injury

56.

Factorsaffecting EMS delivery of care

Desirable

Special circumstances affecting EMS response or delivery of care

57. 58. 59.

Alcohol/drug use Time of first CPR Providerof first CPR

Essential Desirable Desirable

Suspected alcohol or drug use by patient Best estimate of time of first CPR Person who performed first CPR on patient

Numeric entry; county can be derived from zip code. Numeric entry Numeric entry, nine digits Numeric entry Numeric entry, three-digit field Male, female, unknown American Indian/Alaska Native; Asian/Pacific islander; black non-Hispanic, black Hispanic, white non-Hispanic, white Hispanic, unknown Examples: hospital (specify), medical office, morgue, airport, other EMS responder Examples: patient choice, closest facility, managed care, specialty resource center Yes, no, upgraded, downgraded Examples: canceled, patient refusal, no treatment required, treated and transferred to other EMS provider, treated and transported Free text Items should be consistent with E-codes in ICD-9. Although such detail may not be appropriate, the data items should be compatible with and collapsible to the E-code cause of injury list. Primary, single most pertinent clinical assessment. Examples: abdominal pain, airway obstruction, allergic reaction Examples: asthma, diabetes, chronic obstructive pulmonary disease Examples: back pain, bloody stools, headache. Should be compatible with ICD-9 codes. List of all injuries sustained by injury type (amputation, blunt, gunshot) and body site Intentional, unintentional, unknown, not applicable Examples: none used, shoulder belt only, child safety seat, helmet, eye protection Examples: adverse weather, vehicle problems, language barrier, hazardous environment, combative patient Yes, no, unknown, not applicable

60.

Time CPR discontinued

Desirable

61.

Time of witnessed cardiac arrest Witness of cardiac arrest

Desirable Desirable

Time at which medical control or responding EMS unit terminated resuscitation efforts in the field Best estimate of time of witnessed cardiac arrest (if known and applicable) Person who witnessed the cardiac arrest

Desirable

Time of first defibrillatory shock

Desirable

65.

Timeof first defibrillatory shock Returnofspontaneous circulation Pulserate

66.

Initial heart rhythm

Desirable

67. 68.

Rhythmat destination Respiratory rate

Desirable Essential

Whether a palpable pulse or blood pressure was restored after cardiac arrest and resuscitation in the field. Patient's palpated or auscultated pulse rate, expressed in number per minute Initial monitored heart rhythm as interpreted by EMS personnel Monitored cardiac rhythm on arrival at destination Unassisted patient respiratory rate, expressed as number per minute

62. 63. 64.

Essential

APRIL 1995 25:4 ANNALS OF EMERGENCY MEDICINE

Bystander, EMS responder, not applicable, unknown

Bystander, EMS responder, not witnessed, not applicable, unknown Measurement in seconds and clock synchronization strongly encouraged Numeric, not obtained, unknown, not applicable Use current advanced cardiac life support (ACLS) terms and definitions. Use current ACLS terms and definitions.

533

EMS DATA Spaite et al

Appendix, continued Data Element

Priority

Definition

Comment

69.

Respiratory effort

Patient's respiratory efforc

Normal; increased, not labored; increased and labored or decreased and fatigued; absent; not assessed

70. 71. 72.

Systolic blood pressure Diastolic blood pressure Skinperfusion

73. 74. 75. 76. 77. 78. 79.

Glasgow eye-opening component Glasgow verbal component Glasgow motor component Glasgow Coma RevisedTrauma Score Procedureor treatment name Procedureattempts

Essential for children, desirablefor adults Essential Desirable Essential for children, desirable for adults Essential Essential Essential Desirable Desirable Essential Desirable

80. 81.

Medication name Treatment authorization

Essential Desirable

934

Patient's systolic blood pressure Patient's diastolic blood pressure Patient skin perfusion, expressed as normal or decreased Patient's eye-opening component of the Glasgow Coma Scale Patient's verbal component of the Glasgow Coma Scale Patient's motor component of the Glasgow Coma Scale Patient's total Glasgow Coma Scale score (total) Patient's Revised Trauma Score Identification of procedure attempted of performed on patient Total number of attempts for each procedure attempted, regardless or success Medication name Indicates the type, if any, of treatment authorization

Compatible with ICD-9 procedure classification (P codes) Examples: protocol (standing orders), online (radio/telephone), on-scene physician, written orders, unknown, not applicable

ANNALS OF EMERGENCY MEDICINE 25:4 APRIL 1995