I was a frustrated brand-new paramedic in the late 1970s. I felt that old-school fire commanders were mangling fire-based EMS. When WE became the bosses, it would all be better.
Ah, the idealism of youth. By the late 1990s, my cohort was in middle and upper management. Paramedic-experienced executive fire officers were still screwing up fire-based ambulance service.
Why? One of the topics not covered in chief fire officer training is ambulance deployment.
An ambulance is not a fire truck with a stretcher
One of the big-city fire department takeovers of a municipal ambulance service a dozen years ago included an attempt to manage ambulance deployment at the battalion level. It works for suppression companies; they remain within their first-alarm area, save for natural disasters and the occasional greater alarm fire.
It did not work for the ambulances. Stephen Dean described the situation in his analysis of a big city’s fire-based ambulance service:
“There are times when the system has excess capacity and also times when the system is overtaxed. When ambulances complete a call, they attempt to return to their stations. During busy periods, the units seldom make it back to the station before they receive another call.”1
The San Francisco Fire Department (SFFD) moved from fire station-based ambulances to dynamically deployed units to improve response times during budget reductions. A 2010 report showed that SFFD was meeting its goal of a 10-minute response time to non-life threatening emergencies in eight of 10 battalions.2
Dynamic deployment, while common in third-service, private and for-profit services, is rare in fire-based ambulance systems. The 2010 Kansas City takeover of Metropolitan Ambulance Services Trust (MAST) pulled ambulances from dynamic deployment and into fire-station based deployment.
One year after the takeover in Kansas City, the Kansas City Star noted that the fire department failed to meet ambulance response time requirements for both life-threatening and non life-threatening incidents in its first year of operations.3
Last month the acting Kansas City fire chief unveiled four pilot projects designed to improve EMS delivery.4 The city is not returning to dynamic ambulance deployment.
The impact of cardiac arrest outcomes on ambulance deployment
Most deployment research focuses on Out-Of-Hospital Cardiac Arrest (OOHCA) results. Policymakers suggest that a good OOHCA outcome is an indication of a good ambulance service5.
Former Kansas City Fire Chief Smoky Dyer used OOHCA research to support his decision to stop maintaining an ambulance response time of 8:59 minutes to priority 1 calls 90 percent of the time. Dyer noted that their implementation of fire company-delivered, compression-only resuscitation almost doubled the number of patients showing a return of spontaneous circulation (ROSC).
Some of the OOHCA research could benefit other patients. A study using geospatial-time analysis resulted in a relatively low-cost ambulance deployment that significantly reduced response times without changes to the number of ambulances or crews at the Singapore Civil Defense Force (SCDF).
SCDF provides firefighting, rescue and emergency ambulance services; mitigates hazardous materials incidents; and formulates, implements and enforces regulations on fire safety and civil defense shelter matters. For this study, the number of ambulance bases expanded from 17 to 32.
Fatal flaw: ambulance staffing as an incremental fire department service
In examining big-city fire departments, many suffer from the idea that ambulance service is an incremental activity. True, in the 1960’s, but that contention is obsolete due to three workload factors: non-fire responses, crumbling infrastructure and EMS workload.
1. Non-fire responses
The urban environment is jammed with monitored smoke, waterflow and carbon monoxide alarms. Responding to activated alarms is a significant activity in high-rise, business and industrial areas. The first winter after the Illinois Carbon Monoxide Alarm Detector Act went into effect had Chicago truck companies up all night responding to CO alarms.
Investigating “hazardous conditions,” from odors, white powder, oil-in-creek, water conditions, etc., represents a time-intensive fire company activity.
2. Crumbling infrastructure
Age, recession and deferred maintenance result in structural collapses, arson fires, failing water supply and inoperative fire protection systems. Only 15 percent of firefighters work full-time in urban centers. Their activity and working conditions are significantly different from the rest of the industry.
3. EMS workload
Urban ambulance workload started to climb in the late 1990s. While West Coast fire departments went on a paramedic hiring spree, their East Coast counterparts closed fire companies to staff additional ambulances. On both coasts, the EMS workload tended to grow faster than the staffing of additional units.
Chief fire executives can provide smart ambulance service, but it requires knowledge and skills beyond the traditional professional development pathway.
References
1. Dean, Stephen S. Why the Closest Ambulance Cannot be Dispatched in an Urban Emergency Medical Services System. Prehospital and Disaster Medicine, 2008, 23, pp 161-165
2. Begin, Brett (2010 January 31) Ambulance response leveling off. San Francisco Examiner.
3. Editorial (2011 April 23) Disturbing failure on KC ambulance response times. Kansas City Star.
4. Horsley, Lynn (2012 September 10) KC unveils ways to improve ambulance service. Kansas City Star.
5. Persse, David E., Craig B. Key, Richard N. Bradley, Charles C. Miller, Atul Dhingra. Cardiac arrest survival as a function of ambulance deployment strategy in a large urban emergency medical services system. Resuscitation, 2003, 59, pp 97/104
6. Ong, Marcus Eng Hock, Tut Fu Chiam, Faith Suan Peng Ng, Papia Sultana, Swee Han Lim, Benjamin Sieu-Hon Leong, Victor Yeok Kein Ong, Elaine Ching Ching Tan, Lai Peng Tham, Susan Yap, V. Anantharaman. Reducing Ambulance Response Times Using Geospatial–Time Analysis of Ambulance Deployment. Academic Emergency Medicine, 2010, 17, pp 951–957
