Reconciling Fractured COMMUNICATIONS Data

A frantic bystander activates the emergency response system by dialing 9-1-1. At the city's communication center, operated by the police department, personnel answer the call and begin entering information into a computer-aided dispatch (CAD) system. The bystander relates that the emergency involves a loved one in cardiac arrest, and the police phone the county 9-1-1 system, which handles the dispatching of first responders and EMS.

     A county call-taker gives prearrival instructions, reenters the information into her CAD system and triages the call's priority through a standardized emergency medical dispatch (EMD) process. To get units moving, she phones the city's fire department, which again reenters the information into its own CAD system to dispatch a first-responding engine. Rather than phoning the EMS agency directly, the county 9-1-1 system sends information electronically to the EMS CAD system. However, only a portion is sent, leaving EMS with incomplete information. The EMS CAD system dispatches EMS, and both first responders and EMS arrive to treat the patient. They find the bystander performing CPR. Unfortunately, the patient is no longer in a shockable rhythm, but EMS providers continue to deliver care and transport the patient to the closest hospital.

History of Interoperability
     Since September 11, 1996--that's right, 1996--professional organizations and federal agencies have emphasized the need for public-safety agencies and first responders to communicate with one another.^1-3 That means EMS has been challenging the issue of interoperability for more than 10 years. That same year, the Public Safety Wireless Advisory Committee (PSWAC) warned of poor interoperability in public-safety communications.⁴ This warning came five years before the September 11, 2001, terrorist attack on the World Trade Center, at which time interoperability was further recognized as a significant issue in public safety.

     Following the 2001 terrorist attack, the National Task Force on Interoperability (NTFI) was formed to examine the inability of public-safety agencies to communicate with one another in the field. The NTFI report, released in February 2003, explained the importance of, and hurdles associated with, achieving the vital communication of voice and/or data in real time.⁵ On any given EMS incident or patient encounter, communication may be needed with several agencies, even more so than for law enforcement and fire services. Therefore, the importance of resolving the communications problem in EMS is great. The Public Safety Wireless Network (PSWN) concurs that EMS is more greatly affected by the issue than other types of public-safety agencies.⁶ This is likely due to the position of EMS at the intersection of public-safety, public-health, law enforcement and emergency medical activities. Communication needs have a significant impact on EMS.

     The Department of Homeland Security has been charged with identifying and resolving these problems.⁷ It seems that much of its focus has been on disaster response and field communications. While it is important to note that disasters are high-impact events resulting in devastating effects, they are at low probability of occurring frequently.⁸ Although field communication is important in disaster preparedness, there are other examples of poor communications causing problems more frequently. The example of cardiac arrest is certainly one that occurs more frequently and potentially involves the participation of many agencies.

The CAD Interface
     Much discussion has focused on simple communication by radio and telephone among personnel in the field. But field communications alone, while vital to success in disaster responses, don't paint an accurate picture of how we can begin to repair the fractured EMS communications system. A more advanced, yet equally important means of communicating exists in the electronic transfer of data through computers.

     CAD systems are intended to reduce delays and enhance response capabilities by minimizing repetition of data entry.⁹ This is vital in disaster situations as well as for the day-to-day recording and tracking of information. It can allow for enhanced quality-assurance (QA) and research capabilities by creating a means to measure response time intervals and, barring patient outcomes, put together the pieces of a medical emergency into one unified record. So, not only are EMS communications important, but data contained in the CAD record is part of the important information shared between agencies.

     The difficulty of achieving interoperability lies in the daily utilization of multiple CAD systems to record information on a single incident. As in the example of cardiac arrest, city, county, fire and EMS CAD systems experience multiple points of call forwarding, data entry and call prioritization. If these systems are able to send information electronically, they establish a chain of communication, where each system becomes a link in that chain. When these systems are unlinked, each system in the network creates delays through telephone connectivity, dispatchers waiting on hold, and manual reentry of data. In addition, the information in unlinked networks is rarely fed backward to avoid repeating the delays for each event, and this prevents the existence of complete CAD records. Complete intervals are essential for understanding how the emergency care system response affects patient outcomes. In addition to completing the record of response intervals, the times obtained from different sources should be confirmed for accuracy and/or synchronization to each other.

     Beyond interfacing CAD systems with one another, there are issues with time discrepancies in the record. CAD systems may lack synchronization whether they are linked or unlinked. Lack of interoperability and inconsistencies in time synchronization prevent administrators and researchers from reconstructing the timelines of events that occur in the field. Variations in milliseconds or nanoseconds may be indiscernible; however, systems that are not synchronized by common sources may have drifts of greater than 1-2 minutes. The magnitude of time drifts is significant in measurement of system performance in time-sensitive events such as prehospital cardiac arrests. Furthermore, drifts can occur at irregular intervals, both positive and negative. Resolving the lack of electronic data transfer between communication centers may resolve synchronization issues, but measures should be taken to ensure that synchronization occurs not only to the same clock, but also utilizing the same method. Methods such as Internet programs or radio frequencies may be interrupted due to weather, connectivity, distance and other factors. A more advanced and reliable method is using satellites to ensure ongoing synchronization.

Resolving the CAD challenge
     So how do we resolve CAD interfacing and time synchronization? From a nonjurisdictional standpoint, it simply makes sense to have a unified CAD system. Many counties and municipalities believe they can ensure higher standards of communication if they manage their CADs themselves. While this may be true internally, it creates problems when responding agencies cross jurisdictional boundaries in answering calls. In these cases, an EMS provider that covers an entire county may receive dispatch information from a county as well as several municipal CAD systems. While this scenario may get first responders on the scene quicker, EMS is delayed, and the ability to put the event back together for system performance measurement is hindered.

     While a unified CAD may be ideal, fractured systems can still put the pieces back together by creating interoperability between CAD systems. CAD interfacing can be done regardless of hardware/software compatibility among manufacturers. Often, a standard data set is collected on all 9-1-1 calls and can be placed into a universally accepted format for data transfer, such as XML. Since each CAD system designs the order and specifications of its data fields uniquely, architects of one system can write an interface to convert its data into a format that can be received by the second system. The data from the first system can then be autopopulated into the second, which improves efficiency by reducing time and errors. The interface is complete when the transfer of information is bidirectional--i.e., the second CAD system converts its data into a format that can similarly be received by the first. The data is then automatically sent to the first CAD system, and both now have complete records of the event.

     "Once this interface is complete," says Tim Lewis, a communications consultant with Atlanta Communications who's involved with several municipal CAD systems operating within the same county, "the solution has now been created between the two CAD vendors anywhere in the world." This adaptive programming solution does not require the purchase of additional hardware. It does, however, require an interfacing effort to be made by the vendor on behalf of their public-safety customer.

     In a case of cardiac arrest, specific event information is needed to measure system performance and implement improvement strategies. In our scenario, data has been collected by four different CAD systems, but only one is sending electronic data in a forward direction; the data sent is not complete; the times recorded by the CAD systems are not in synch with one another; and processing delays occur through the use of multiple call-takers.

     When CAD systems are operated by police departments, the implications and uses of response intervals in measuring EMS system performance may not be fully appreciated. This can make it difficult to collect CAD data in an ongoing fashion. Sensitivity of data and fear of legal repercussions may drive a defensive posture. In any case, the scenario may only allow EMS to review the event without noting when the 9-1-1 system was activated by the bystander. Because of independently functioning CAD systems and the lack of interfaces among them, individuals within each CAD system can commit to providing their individual parts of the data set to reconstruct the timeline of the prehospital response.

Playing well with others
     The NTFI report mentioned earlier listed five reasons why public-safety agencies cannot communicate with one another: incompatible and aging equipment, limited funding, limited planning, inadequate radio spectrum, and lack of coordination and cooperation. There are many options for CAD systems to collect and communicate data in real time,¹⁰ but in making the data system work, emphasis should remain on this last element: coordination and cooperation. Whether in a disaster scenario or routine events, considerations must be made to include neighboring jurisdictions and their agencies. If agencies would consider these neighbors as they develop communications systems, information could flow readily and efficiently. How we "play with others" will determine our success in crisis situations and the progress of research that relies on such communication. Authors of the National EMS Research Agenda in 2001 called for public and private agencies to support EMS research, and recommended that evidence from research be required by EMS before implementation of new procedures.¹¹ This certainly includes allocation of resources to reduce response times in order to improve patient outcomes.

     It seems manufacturers might not have an interest in promoting such compatibility. Each CAD software program utilizes its own unique identifiers and incident numbering system. However, the public-safety customer should remain steadfast regarding the priority of sharing information. Setting up essential data elements for electronic transfer and designing the XML specifications can be relatively easy, but requires time by the CAD vendor. Using the National Emergency Medical Services Information System (NEMSIS) XML specification as a template can allow agency information technology gurus to assist in this process, which will catalyze vendor participation. As customers, communication centers must demand interoperability through the electronic forwarding and receiving of information that includes other CAD software incident numbers, call priority and location information, and synchronized processing, dispatch and response times.

EMS and the IOM
     Last June, the Institute of Medicine (IOM) released a wide-ranging set of reports on emergency care in America. The reports described both how poor public-safety communications will hamper responses to major incidents and how this deficiency contributes to the overall inability of the emergency system to prevent unnecessary deaths. One of the three reports was dedicated solely to EMS issues. There was significant cause for concern: The IOM described several defects in EMS communications and data systems and the barriers to achieving adequate communications and relay of data, and recommended that agencies involved in emergency care (e.g., hospitals, trauma centers, EMS, fire, police, emergency management and public health) integrate interoperable networking systems.⁸ The report specifically pointed out that there are purposes beyond disaster preparedness for addressing the issue of information sharing. The communications problem inhibits the ability to conduct effective prehospital research and develop QA measures to survey system performance.

     The IOM report suggested that attention be directed toward interoperability between emergency medical services, 9-1-1 centers, first responders and police.¹² The IOM identified the Cardiac Arrest Registry to Enhance Survival (CARES)--a collaborative effort between the CDC and Association of American Medical Colleges--as a model to identify, track and record prehospital cardiac arrest events, which ultimately completes the record by unifying CAD, first responder and EMS data along with hospital outcomes.^8,11,13

     CARES is a model of measuring system performance by allowing three data sets to come together and producing feedback reports. These reports can further assist EMS agencies in making improvements by allowing them to benchmark internally, within their agency, and externally, with similar agencies across the nation. Using Web-based technologies in partnership with Sansio,¹⁴ CARES securely stores event information and determines when hospital outcome information and CAD information are needed to complete the record. The important aspect of CARES is that in addition to EMS and hospital information, systems can take fractured communications data and bring it together in one record. Even if the system has only one participating CAD system and the communications data is already intact, CARES can help agencies by providing various reports based on that CAD information. Participating agencies can access the registry through accounts that are specific for their type of user (i.e., EMS, hospital or CAD), which will allow users to directly enter data into the CARES database. Each agency can also establish automatic XML data transfers to the database to streamline the process.

     Through CARES, EMS agencies can correlate response intervals with patient outcomes for cardiac arrest events. By unifying outcome data and communications data, intervals such as first responder and EMS response times can be shown to impact survival. CARES has been working with several agencies to improve interfacing and synchronization of CAD systems. As a neutral party, it has helped neighboring agencies come together and discuss issues of sharing and synchronizing CAD information. For more information, see https://mycares.net.

Design Challenges and Recommendations
     These recommendations are simple to lay out, but may be more difficult to achieve. Success will lead to resolving an entire network of operational challenges, and this is why efforts should be made to repair fractured communications data.

     Coordination--The lack of coordination and cooperation mentioned by the NTFI points out a reluctance to relinquish management and control. It is important to note that the emphasis is not for each entity to dissolve into a centralized system, but rather that the points at which they communicate should come from the same source. One could argue that many systemwide communication problems are the result of multiple agencies that disagree on who should manage communications. Regardless of which entity manages the communications system(s), the goal should remain that systems communicate with each other in real time using advanced data-transfer technology to maintain unified records of each event. By staying focused on this objective, communication centers could moderate what goes on in the field, and not regulate and restrict access to helpful information. Recommendations to correct issues of coordination are listed below:

• Establish a collaborative task force. Invite managers from CAD systems, EMS and first responders that interact with one another. This is especially important when dealing with jurisdictions that have multiple CAD systems.
• Determine each other's needs. Your agency's needs may be different than those of agencies with which you communicate.

     Interfacing--Some vendors have sought exclusivity with their customers through developing software programs that are limited in their use among other software and hardware systems. The lack of interface in electronic transfer of emergency communications data should be seen as unacceptable. The recommendations for interfacing issues are:

• Prioritize: Each CAD, EMS and first responder agency has someone familiar with information technology. Put them on the job to create interfaces using a familiar program such as XML.
• Make your demands. Address interfacing issues with your hardware/software vendor. As a customer, your needs are paramount. If a vendor can't provide what you need, someone else can and will. They will see that investing the time to meet your needs improves their marketability. It may just be that in their next upgrade, they include interfacing capabilities. In the interim, vendors should help you meet your needs of measuring system performance.
• Get in the game. Many systems want to remain self-contained when it comes to sending data electronically. But interfacing is no different than radio interoperability. We can't expect to communicate well if we don't want to turn on the radio.

Conclusion
     Will EMS, more than 10 years after the 1996 PSWAC report, remain crippled in its ability to relay information in an efficient manner? Overcoming these barriers will help achieve the ultimate goal of sharing information vital to successful field operations and evaluation of system performance. Putting the pieces together can be difficult. With adequate cooperation, interfacing and synchronization, improvements can be made by establishing and reviewing a complete record of every event.

References

1. National Public Safety Telecommunications Council, www.npstc.org.
2. Werner C, McEwen HR, Ward M, Nash G, Overby S. Public Safety Communications: What You Need to Know. Fire Rescue International presentation to the International Association of Fire Chiefs, Sept. 15, 2006.
3. Department of Homeland Security. Interoperable Communications Grants Fact Sheet, Sept. 25, 2003.
4. Public Safety Wireless Advisory Committee. Final Report, 1996.
5. National Taskforce on Interoperability. Why Can't We Talk? Working Together to Bridge the Communications Gap to Save Lives: A Guide for Public Safety Officials, 2003.
6. Murphy R, Siegle D. Speaking the same language: PSWN works to improve communications in EMS. Emerg Med Serv 30(8):39-41, 2001.
7. USA PATRIOT Act, HR3162 RDS, Oct. 24, 2001.
8. Institute of Medicine of the National Academies. Emergency Medical Services: At the Crossroads. Washington, DC: The National Academies Press, 2006.
9. Ludwig GG. Understanding computer-aided dispatching and interfaces. In Prehospital Care Administration, JEMS Communications, Carlsbad, CA, 1995.
10. McNally B, Kellermann A, Stokes A, Click L. CARES Reference Manual, https://mycares.net.
11. National Highway Traffic Safety Administration, Maternal and Child Health Bureau. National EMS Research Agenda, 2001.
12. Ott WE. The homeland security communications mistake. JEMS 28(10):90-1, 2003.
13. McNally B, Kellermann A. Boosting the Odds of Surviving Cardiac Arrest. Medscape General Medicine 8:3, 2006, www.medscape.com/viewarticle/542698.
14. Sansio. Strategic Partnerships, www.sansio.com/Partnerships.htm.

Allen Stokes, BSc, NREMT-P, is senior research project coordinator for prehospital and disaster medicine in the Department of Emergency Medicine at Emory University in Atlanta.