The goal of any medical service is to provide the patient with the right resources at the right time. For many in our society any wait for anything is too long, hence the concept of a “Starbucks on every corner.”¹ Although it is arguable that there is a greater need for emergency medical services (EMS) than for coffee, very few EMS agencies could afford to place an ambulance on every corner. Because of this EMS have used various deployment strategies to meet their patient care goals.

Deployment, according to one definition from the Free Dictionary, means, “To distribute (persons or forces) systematically or strategically.”² Historically, ambulances were dispatched from fixed bases, most commonly from fire stations or funeral homes, depending on the local model. These stations were usually cited based on political subdivisions that have little or nothing to do with the needs of the community. In many areas, this practice continues today. In rural areas, and areas served by volunteer agencies, basing the ambulance at a station is the only system that makes sense. In urban areas, however, it is better to site equipment based on patient demand, rather than provider convenience. In these areas, as Overton and Gunderson have noted, the pattern of EMS usage resembles that of a police, rather than a fire department.³ Because of this, many high-volume urban organizations have adopted plans where ambulances are dispersed to predetermined “posts” around the service area. These posts may be a specific corner or perhaps a mall parking lot; in some cases the ambulance crew is given a general geographic area in which they are to be located. EMS agencies using this model of dynamic deployment are often described as “high performance” although this term may be a misnomer.

• 
  

  Define deployment and posting.

  
  
• 
  

  Describe the phases of ambulance response.

  
  
• 
  

  Describe how red-lights-and-siren responses affect the ambulance response.

  
  
• 
  

  Define system status management.

  
  
• 
  

  Describe the basic research concepts behind SSM.

  
  
• 
  

  Describe how technologies have changed SSM.

EMS systems have historically been designed to address two major ­problems: cardiac arrest and motor vehicle trauma. Spaite et al note that cardiac arrest accounts for a very small number of ambulance responses, and that there is no good data on the effects of EMS system ^­components on trauma. He contends, “… Most prehospital trauma research has emphasized the wrong issues, asked the wrong questions, and used the wrong methods.”⁴ An attempt to rectify the lack of information on trauma patients demonstrated that there was no evidence supporting time sensitivity for trauma mortality.⁵ The problem, according to Spaite and his colleagues, is that EMS research has historically emphasized individual diseases and interventions, rather than system issues.

Many EMS agencies have a response standard written into their contract. Commonly such a clause says that the service must respond to a call within a specified number of minutes. Frequently the clause says that such a response is expected 90% to 95% of the time. There may be monetary penalties associated with failure to meet these expectations. Eight minutes, 90% of the time, is frequently used as a standard. These standards are usually based on data for cardiac arrest survival.⁶ Other services may have expectations that there will be a BLS response within 4 minutes and an ALS response within 8 minutes, 90% of the time. These standards are based on the National Fire Protection Association’s (NFPA) 2010 response standards (NFPA 1710).⁷ These specify that the service shall provide an AED response within 240 seconds of travel, following 60 seconds of “turnout” time. Aside from Spaite’s contention that the basis for this standard is a very small part of the EMS realm,⁴ it also may not take into account all of the pieces that need to fall into place to access the patient, care for the patient, and transfer them to the hospital. A recent publication by the United Kingdom’s National Audit Office reflects this concern, stating, “Performance over the last decade has been driven by response time targets and not outcomes.… Its ­existence in isolation from more direct measures of patient ­outcomes has, however, created a narrow view of what constitutes ‘good’ ­performance, and skewed the ambulance services’ approach to performance measurement and management.”⁸

The University of Arizona group tried to bring some of this science into the medical realm in a 1993 article in which they defined and validated a model for the time intervals in an EMS response.⁹ This article defined 10 intervals between the event and the return to service. This granularity allows EMS to more appropriately address the problems that may occur as EMS agencies try to provide care to patients in a timely fashion. For example, many providers report that they are “on scene” when they park their vehicle at the patient’s residence. This would be a useful basis for comparison if everyone were in a single floor, single family dwelling, and there was a standard way to label dwellings. Unfortunately, most areas have a mix of single family and multiple family dwellings. The time to access a patient in a single-family ranch is very different than the time to locate a patient in a 20-story urban high rise. Understanding the difference between the response interval and the patient access interval is important for EMS managers looking to improve their systems. It is also important to clinicians and politicians who may be trying to explain why their survival statistics are not as good as other agencies’ numbers.

Most systems do not use this level of detail. Instead they employ several strategies to minimize the response and transport intervals, paying little attention to the other eight intervals. The most common method to minimize the response and transport intervals is through the use of fast driving, facilitated by the use of red lights and a siren. Although this is how the public sees EMS,^10,11 it is unclear whether such use produces significant time savings. Ho and his group noted the use of red lights and sirens in both urban¹² and rural¹³ environments saved time. A group from East Carolina University attempted to match the route taken by an ambulance and found very little difference between emergency response and routine traffic.¹⁴ O’Brien et al in Louisville employed a similar technique. They found that there was a time saving associated with the use of red lights and siren (RLS), but that the time savings did not translate into any clinically important interventions upon arrival at the ­hospital.¹⁵ Others have reported a similar experience.¹⁶ A UK study suggests that EMS providers feel that the current performance standards place them at jeopardy.¹⁷ Blackwell and Kaufman noted that in his system in Mecklenburg County, North Carolina, unless patients could be reached within 5 minutes there was no evidence that there would be a survival improvement.¹¹ A study of the use of RLS in pediatric patients showed that over 1/3 of the pediatric patients were unnecessarily subjected to an emergency response.¹⁸ In addition to the questionable efficacy from the use of RLS transport, there are concerns that care of the patient within the vehicle may suffer. Although ambulances are being designed to keep the caregiver restrained while performing patient care, it is still difficult to perform even basic procedures with the rapid stops and quick starts that often accompany RLS transports. The risk involved with these emergency responses is substantial. A group from Milwaukee found that over an 11-year period most of the crashes and fatalities in ambulances occurred during an emergency response.¹⁹ “Wake effect” crashes, which occur due to the distraction produced by the emergency vehicle’s passage, further compound the problems caused by this response.²⁰ Current and future work on reduction of response times will likely focus on ­prediction algorithms and the concepts of dynamic posting and system status management.^21,22

The concept of the high-performance EMS system using system status management (SSM) is commonly attributed to Jack Stout, an economist. He looked at the EMS system using tools usually applied to industry and developed these concepts along with the public utility model for EMS.²³ Stout reveals his background when he says that the purpose of SSM is to match supply and demand.²⁴ He notes that SSM does not require that ambulances be moved around, describing roving ambulances as “a waste [of] human energy, fuel, and money.” The primary difference between the systems that Stout describes and the historical fixed base model is that Stout advocates placing bases or posts based on patient need, not political subdivisions. This typically involves basing positioning of ambulances and staffing on historical geographical call data with variance in time, day of the week, and seasons. This strategy may also take into consideration types of calls, key locations, highway access, traffic patterns, and weather issues.

Bledsoe has criticized SSM, saying, “I was surprised that there was no scientific evidence to support the practice,” and claiming that it “increases work stress” and “causes an increase in vehicle maintenance and miles traveled.”²⁵ As his references Dr Bledsoe refers to a single study of one EMS system, and concludes by saying that, “I can’t document the following statement with science just with experience and emotion. I believe that employee satisfaction, morale and pay are generally lower in systems that use SSM, while employee turnover, stress and physical ailments are higher.” Fortunately, there is a solid basis for SSM.

Most of the information regarding SSM is derived from the operations management literature. The classic use of SSM has been in the area of industrial transportation and in manufacturing plant design. ReVelle and his colleagues explained this problem as having the, “objective of maximizing some measure of utility to the owners while at the same time satisfying constraints on demands and other conditions.”²⁶ Clearly this fits the description of an EMS system as well as it fits a manufacturing operation.

One of the things that makes EMS deployment complex is the nature of the service. EMS can, however, be looked at in a manner similar to other service industries. The consumer can control the demand for some goods. The provider may schedule other services. Table 14-1 provides examples of this.

The basic problem for EMS is therefore the unpredictability of customer demand. Other issues, such as traffic and weather, as well as the variable nature and severity of the patient’s complaint further complicate this model. Although the geographic distribution of calls and their nature seem completely random, there are ways to predict these incidents and use this information to appropriately distribute EMS resources (Figure 14-1).