EMS and Medical Surveillance

Terrorism preparedness has become a reality. Many EMS systems have made changes to improve their responses to terrorism that have also improved their daily operations. These include increased use of personal protective equipment (particularly respiratory protection), scene awareness, and decontamination and care of the hazmat patient.

Part of EMS terrorism preparation is establishing relationships with public health. Public health, rather than EMS, is primarily responsible for bioterrorism events. Bioterrorism may be difficult to differentiate from common diseases, and victims may be far from the site of their exposure before becoming ill. With emergency departments overcrowded and understaffed, EMS systems running at peak capacity and the primary care system strained, an uncommon disease marking a bioterrorism event may not be detected early in its course.

Public health performs surveillance of community health and disease. This surveillance is also used as a tool in the early detection of and response to bioterrorism. Although some regions may have effective real-time syndromic surveillance systems, most do not. There is a lack of state and national coordination of surveillance systems that leads to gaps in the critical disease monitoring and prevention process.

EMS commonly uses sophisticated information technology resources, particularly in public safety access points and dispatch centers. These resources have allowed some EMS agencies to participate in syndromic surveillance. These organizations participate in regional, state, and national bioterrorism surveillance activities while using the data for their own specific projects.

Surveillance

Surveillance is the monitoring of a population for changes from a predetermined norm or standard, with notification of appropriate individuals if a change is noted.¹ Medical surveillance (or biosurveillance) systems look at the health and disease status of a population. They detect and analyze diseases or symptom clusters (syndromes) to see if they are occurring more often than would be normally expected.

For a surveillance system to work, you need to know how often a specific event actually occurs in your community (a baseline standard). For example, the baseline standard for smallpox is that there are no cases. A single case would be considered a terrorist event and a national public health emergency. Viral gastroenteritis, on the other hand, is so common that a large increase above the baseline would be necessary before it could be considered unusual.

Surveillance systems must look for "important" diseases or syndromes and be able to detect changes from the baseline. The way a case is defined (signs and symptoms, clinical findings and laboratory studies) greatly influences the surveillance system. During influenza season, for example, it could be almost impossible to tell inhalational anthrax from influenza if the anthrax case definition were "fever and respiratory symptoms." That is why the CDC case definition includes the chest x-ray findings of widened mediastinum-and that's why surveillance systems must monitor for that result as well. Otherwise, the anthrax case probably would not be recognized by the surveillance system. In addition, a new disease like SARS might not be identified as unusual if it weren't different from other similar diseases (e.g., influenza, pneumonia) in terms of severity, symptoms, mortality or season of presentation.

Surveillance systems must also notify the appropriate individuals quickly. Most health departments receive notable laboratory results as they become available but do not have around-the-clock staffing, so response delays may exist. These public-health surveillance systems are not real-time and are slow to process information, resulting in delays in detection. Although these delays may be fine for traditional diseases, they are unacceptable for highly communicable bioweapons in the era of rapid world travel. There are EMS systems using continuous call data monitors (described below) that achieve near real-time surveillance. Additionally, a few emergency department-based systems and "drop-in" surveillance systems at special events have almost-real-time capability.

Traditional surveillance works by looking for a disease with a "gold standard" laboratory test confirmation. A patient must have an exposure, develop signs and symptoms, see a clinician who suspects the disease and have the appropriate laboratory test ordered. This process can take days or even weeks, depending on the disease, the suspicion of the practitioner and the availability of the laboratory study. Since only true cases are identified by the laboratory study, this method does not produce many false alarms. However, extensive training of EMS providers on signs and symptoms of bioterrorism diseases would be required for field personnel to effectively participate in this type of surveillance.

Syndromic surveillance takes a broader view. Rather than looking for a specific disease, such as anthrax, syndromic surveillance attempts to identify cases in which a patient has the signs and symptoms of that disease-like cough, fever and upper respiratory tract infection. A blood culture or Gram stain is not necessary for the system to be alerted. This approach is much more sensitive (likely to pick up a disease) but also more likely to generate "false positive" alarms for incidents that aren't real. Other terms used for syndromic surveillance include early warning systems, prodromic surveillance, outbreak detection systems, nontraditional surveillance, prediagnostic surveillance, health indicator surveillance and information systems-based sentinel surveillance.^1-3

By looking at symptoms and clinical diagnoses within "syndrome groups," syndromic surveillance picks up many different diseases (e.g., every case of influenza as well as every case of anthrax). Table 1 lists the CDC-defined syndrome classes for possible bioterrorism events.

Syndromic surveillance depends in large part on having people or systems capable of separating signals ("interesting" diseases that must be identified) from "noise" (the background of "noninteresting" diseases that is always present). Epidemiological investigators and infectious-disease physicians play an important role in this process.

Data Sources

Traditional data sources for surveillance systems are hospital, community and public-health laboratories. These sources are not timely. Nontraditional methods of detecting diseases are being used more commonly and include both clinical and nonclinical markers. Although these methods are less specific (they may pick up diseases other than the ones they're looking for), they are much faster in detecting disease.

Nontraditional clinical data comes from a number of sources, including EMS, emergency departments, urgent care centers, primary care offices and veterinary clinics.^3-10 Table 2 lists major sources of information and the types of clinical information they provide. The quality of the information varies. The final diagnosis is not particularly important, since syndromic surveillance systems cluster similar ICD-9 diagnosis codes into single syndrome groups. Chief complaint and final diagnosis have been shown to correlate well with each other, at least within syndromic groups,^11,12 which makes emergency department and EMS call center chief-complaint logs useful.

If there are many baseline cases of "normal" diseases in a syndrome group (e.g., vomiting and diarrhea in the gastrointestinal syndrome group), it may take a large number of extra cases from a bioterrorism event for the surveillance system to detect them (see Figure 1 for a numeric example).

There is also nontraditional, nonclinical data that may signify outbreaks, although the usefulness of this type of data is highly variable. For example, sales of over-the-counter (OTC) anti-influenza medications have been shown to increase about four days before emergency department visits increase for influenza.¹³ Selecting the right marker is critical for using this type of data. Studies have evaluated school absenteeism,¹⁴ purchase of OTC medications,¹⁵ nurse hotline calls and prescription medication sales.³ The most effective surveillance systems combine traditional and nontraditional clinical and nonclinical data, with different levels of importance being placed on each.

Surveillance systems can be thought of as either permanent systems or temporary "drop-in" systems for special events or circumstances. Permanent systems exist at local, regional, state and federal levels.¹⁶ They are usually run by health departments, which collect and analyze the data and perform investigations if needed. Although data may be collected continuously, it may not be analyzed on an around-the-clock basis, with deficits particularly during nights and weekends. Most current systems rely on traditional and some nontraditional clinical data. The majority of programs looking at nontraditional clinical and nonclinical data are pilot programs.

Some EMS systems have also begun to use their own fixed systems, monitoring EMS calls (types and geographic distribution). These systems are usually integrated into their communities' permanent public-health surveillance systems. New York City has used an EMS dispatch-based syndromic surveillance system for "flulike illness" since 1998 that is part of a much larger surveillance system.¹⁷ This program is a joint project of the NYC Department of Health and the Fire Department of New York. This data has a high sensitivity for severe illness. Other uses of the system have included examining the epidemiology of drug overdoses, monitoring for suicidal ideation after the World Trade Center attacks, monitoring heat-related disease and several other projects. The Seattle Fire Department's 9-1-1 dispatch monitoring and analysis project has also been effective in quickly picking up outbreaks of influenza.¹⁸

An EMS-based technology is Stout Solutions' FirstWatch, a dispatch-based syndromic surveillance system that is integrated into the Medical Priority Dispatch ProQA System and interfaced with CAD, ePCR, hospital diversion and other EMS data sources, as well as hospital ED and poison control data sources. Deployed so far in 54 agencies in North America, FirstWatch has demonstrated its sensitivity by detecting a flu outbreak in Oklahoma City,¹⁹ a carbon monoxide poisoning epidemic in Kansas City and a Norwalk virus outbreak in a senior community in Nevada. Although the system was initially designed to detect symptoms associated with a bioterrorism event, it allows customization to survey for other events, such as pediatric drownings and potential arson outbreaks, as well as response times and other performance criteria. Such systems are limited only by the imaginations of their users. When set to alert for events of high concern (e.g., bioterrorism events) but also analyze all data collected, EMS-based surveillance systems become powerful tools to positively impact both community well-being and responder performance and safety.

In addition to fixed systems, special circumstances (large events, disasters, increased threat levels) may call for "drop-in" surveillance in which systems are added to the existing public-health surveillance network. These non-permanent systems are left in place for the duration of the event or occurrence, and then, after a suitable post-event period (the incubation period of the disease being watched for), the system is dropped out.¹ Such systems have been used during the 2000 Sydney Olympics,²⁰ the 2002 Salt Lake City Winter Olympics,²¹ the Seattle World Trade Organization meeting, the 2000 Democratic and Republican National Conventions, and in New York City after September 11, 2001,^22,23 as well as in other venues.²⁴ These systems, although oriented toward bioterrorism, have picked up other communicable diseases; this demonstrates that such systems work. Major disadvantages of drop-in systems are that they are temporary and cannot be easily expanded throughout the community.

Practical Issues

There are many limitations to establishing a surveillance system. The greatest is probably cost. Surveillance systems are expensive to set up, particularly if the data being collected must be manually entered or is not already captured for other reasons. In addition, the system must be maintained and routinely monitored.

With surveillance-system emphasis on bioterrorism identification and response, timeliness is of utmost importance. Traditional systems often have delays due to the type of data they collect (e.g., confirmatory lab results), how the data is collected (mail, electronic transfer, fax, etc.) and when the system is monitored (around the clock versus 9 a.m.-4:30 p.m. on non-holiday weekdays). EMS dispatch-based systems have the advantage of operating in an around-the-clock environment and so are essentially "real-time" systems.

The way data is collected is also a concern. From a cost and sustainability viewpoint, the more ways in which a piece of data is used, the better. ED chief-complaint logs are required by law, and therefore using them for surveillance adds no additional data-collection burden; most dispatch centers already routinely maintain 9-1-1 complaint logs as well. Electronic data capture, particularly from multi-use information forms that allow fields to be "plucked" and electronically compiled, is probably the easiest and most cost-effective way to collect data.

The question of when to sound an alert is also of importance (see Figure 1). The lower the requirements to trigger an alert, the more sensitive the system is, but the higher the cost (more false alarms). On the other hand, higher requirements may delay detection and result in higher morbidity, mortality and cost in the long run. This is a delicate balance, and setting the appropriate levels relies on the experience of system designers and managers.

Perhaps the most important practical issue is that there have been no actual surveillance systems in place during the few real bioweapon events that have occurred, so they may not work.²⁵ The use of surveillance in EMS and public health for bioterrorism preparedness is theoretical and based on surveillance system performance during naturally occurring outbreaks. Although studies of non-bioterrorism events suggest that real-time syndromic surveillance systems detect outbreaks earlier than traditional methods (or no methods at all), it is unclear if the same is true for bioterrorism events. This is a question that can only be answered if two identical bioterrorism events happen at the same time in two communities, one with and one without a surveillance system. Nonetheless, the non-bioterrorism (dual-use) applications of surveillance systems are sufficient to justify their use.

Getting There

For EMS agencies and emergency departments that wish to be involved in surveillance, the most important step is to form a partnership with the public-health community. Although many of the non-bioterrorism applications already described may be useful internally for an agency or department, surveillance itself, particularly in the realm of bioterrorism, is ultimately the responsibility of public-health agencies. Interoperability and nonduplication of data is critical. Therefore, a multiagency approach with local, regional or state departments of public health is required.

Conclusion

For both bioterrorism events and routine community health monitoring, surveillance offers many opportunities. Although there are a number of limitations, technology and research should overcome these. EMS systems can become valuable tools for community surveillance. Through partnerships with public health, EMS agencies are given the opportunity to broadly impact and improve the health and safety of the communities for which they are responsible.

In-Depth Supplements

FirstWatch Trigger Examples and Ideas

Case Study: Successful Real-World Test of High-Tech Tool for Detecting All-Hazards Events in Seattle's Puget Sounds Region

References

1. Henning KJ. "Syndromic Surveillance," in Microbial Threats to Health: Emergence, Detection, and Response, Smolinski MS, Hamburg MA, Lederberg J, eds. Washington, DC, The National Academies Press, 2003.
2. Lober WB, Karras BT, Wagner MM, Overhage JM, et al. Roundtable on bioterrorism detection: Information system-based surveillance. J Amer Med Informatics Assoc 9:105-15, 2002.
3. Lombardo J, Burkom H, Elbert E, Magruder S, et al. A systems overview of the electronic surveillance system for the early notification of community-based epidemics (ESSENCE II). J Urban Health 80:i32-i42, 2003.
4. Greenko J, Mostashari F, Fine A, Layton M. Clinical evaluation of the emergency medical services (EMS) ambulance dispatch-based syndromic surveillance system, New York City. J Urban Health 80(s):i50-i56, 2003.
5. Hirson JM. The rationale for developing public health surveillance systems based on emergency department data. Acad Emerg Med 7:1,428-32, 2000.
6. Hutwagner L, Thompson W, Seeman GM, Treadwell T. The bioterrorism and response early aberration reporting system (EARS). J Urban Health 80(s):189-96, 2003.
7. Irvin CB, Nouhan PP, Rice K. Syndromic analysis of computerized emergency department patients' chief complaints: An opportunity of bioterrorism and influenza surveillance. Ann Emerg Med 41:447-52, 2003.
8. Lazarus R, Kleinman KP, Dashevsky I, DeMaria A, Platt R. Using automated medical records for rapid identification of illness syndromes (syndromic surveillance): The example of lower respiratory infection. BMC Public Health 1:9, 2001.
9. Lober WB, Trigg LJ, Karras BT, Bliss D. Syndromic surveillance using automated collection of computerized discharge diagnoses. J Urban Health 80(s):i97-i106, 2003.
10. Suyama J, Sztajnkrycer M, Lindsell C, Otten EJ, et al. Surveillance of infectious disease occurrences in the community: An analysis of symptom presentation in the emergency department. Acad Emerg Med 10:753-63, 2003.
11. Begier EM, Sockwell D, Branch LM, Davies-Cole JO, et al. The national capital region's emergency department syndromic surveillance system: Do chief complaint and discharge diagnosis yield different results? Emerg Infect Dis 9:393-96, 2003.
12. Mocny M, Cochrane DG, Allegra JR, et al. A comparison of two methods of biosurveillance of respiratory disease in the emergency department: Chief complaint vs ICD9 diagnosis code. Acad Emerg Med 10:513 (abstract), 2003.
13. Welliver RC, Cherry JD, Boyer KM, Deseda-Tous JE, et al. Sales of nonprescription cold remedies: A unique method of influenza surveillance. Ped Res 13:1,015-17, 1979.
14. Lenaway DD, Ambler A. Evaluation of a school-based influenza surveillance system. Public Health Reports 110:333-37, 1995.
15. Goldenberg A, Shmueli G, Caruana RA, Fienberg SE. Early statistical detection of anthrax outbreaks by tracking over-the-counter medication sales. Proceedings of the National Academy of Sciences of the United States of America 99:5,237-40, 2002.
16. Lewis M, Pavlin J, Mansfield J, O'Brien S, et al. Disease outbreak detection system using syndromic data in the greater Washington DC area. Am J Preventive Med 23:180-86, 2002.
17. Mostashari F, Fine A, Das D, Adams J, Layton M. Use of ambulance dispatch data as an early warning system for communitywide influenza like illness, New York City. J Urban Health 80(s):i43-i49, 2003.
18. Dockrey MR, Trigg LJ, Lober WB. An information system for 911 dispatch monitoring and analysis. Proceedings, AMIA Symposium, 2002.
19. Associated Press. "High Tech Tool Warns of Flu Outbreaks," USA Today, Nov. 28, 2003.
20. Jorm LR, Thackway SV, Churches TR, Hills MW. Watching the games: Public health surveillance for the Sydney 2000 Olympic games. J Epidemiol Community Health 57:102-8, 2003.
21. Gesteland PH, Wagner MM, Chapman WW, Espino JU, et al. Rapid deployment of an electronic disease surveillance system in the state of Utah for the 2002 Olympic winter games. Proceedings, AMIA Symposium, 2002.
22. Centers for Disease Control and Prevention. Syndromic surveillance for bioterrorism following the attacks on the World Trade Center. MMWR 51(special issue):13-15, 2002.
23. Das D, Weiss D, Mostashari F, Treadwell T, et al. Enhanced drop-in syndromic surveillance in New York City following September 11, 2001. J Urban Health 80(s):i76-i88, 2003.
24. Osaka K, Takahashi H, Ohyama T. Testing a symptom-based surveillance system at high-profile gatherings as a preparatory measure for bioterrorism. Epidemiol Infect 129:429-34, 2002.
25. Reingold A. If syndromic surveillance is the answer, what is the question? Biosecurity and Bioterrorism 1:77-81, 2003.

Suggested Reading

• www.cdc.gov/epo/dphsi/syndromic/index.htm.
• Barthell EN, Cordell WH, Moorhead JC, Handler J, et al. The frontlines of medicine project: A proposal for the standardized communication of emergency department data for public health uses including syndromic surveillance for biological and chemical terrorism. Annals Emerg Med 39:422-9, 2002.
• Broome C, Horton H, Tress D, Lucido SJ, Koo D. Statutory basis for public health reporting beyond specific diseases. J Urban Health 80(s):i14-i22, 2003.
• Mandl KD, Overhage JM, Wagner MM, et al. Implementing syndromic surveillance: A practical guide informed by early experience. J Am Med Inform Assoc 11:141-50, 2004.
• Mostashari F, Hartman J. Syndromic surveillance: A local perspective. J Urban Health 80(s):i1-i7, 2003.
• Pavlin JA, Mostashari F, Kortepter MG, Hynes NA, et al. Innovative surveillance methods for rapid detection of disease outbreaks and bioterrorism: Results of an interagency workshop on health indicator surveillance. Am J Public Health 93:1,230-35, 2003.
• Teich JM, Wagner MM, Mackenzie CF, et al. The informatics response in disaster, terrorism, and war. J Am Med Inform Assoc 9:97-104, 2002.
• Wagner MM, Tsui FC, Espino JU, Dato VM, et al. The emerging science of very early detection of disease outbreaks. J Public Health Management Practice 7:51-59, 2001.