What is “Performance” in EMS? Part 1

It is that time of year for resolutions and reflection.  As I ponder this thought, the topic that sticks out to me is about what really constitutes a “High Performance EMS”.  As we look back over the past year of the High Performance EMS social network (including our Twitter and Facebook feeds as well as this blog) one of the recurring comments that disturbs me is that “response time doesn’t matter”.  This causes me concern in two ways – first, that the primary measure of performance is overwhelmingly always “response time” and the other is that this simple measure is deemed to not really be important.  So, for the next few posts, I will discuss various characteristics that I feel do matter in becoming a truly high performing EMS system.

Part 1: Response Time

This past February, Elsevier published an excellent newsletter (EMS Insider, Volume 39, Number 2) focused on EMS response times and included articles such as “The Great Ambulance Response Time Debate Continues” in which the author, Teresa McCallion, laid out many of the facts.  For instance, the article recites the “MedStar example” from Super Bowl XLV suggesting that “very few EMS calls” in that prospective two week study actually “required an immediate response.”  It is important to note that this statement did not go so far as to say that response time is meaningless in all cases – just that it is far less limited in most.  Then as counterpoint to dismissing response times altogether, the public conflict at EMSA in Oklahoma City was brought up where at least one politician complained of the number of excluded calls required in order to reach a 90% response time compliance rate.  This is only a single instance, but we all understand that it is certainly indicative of how the public measures the value we provide.  In the conclusion, Matt Zavadsky, MedStar EMS Associate Director for Operations, offered several good recommendations to improve patient outcomes and public understanding of the EMS system.  While I agree with nearly everything he said, I would really only argue with his statement that began, “There is no such thing as an inappropriate request for 9-1-1,” (which is a whole other topic) but then he added “there is such a thing as an inappropriate response to that request.”  I can only assume he was referring to the fact that accidents sometimes happen en route to calls.  While these incidents point out failures in judgement somewhere, it is not the “response” itself that is at fault.

Zavadsky also authored another article in that newsletter entitled “Response Time Realities: The Scientific Evidence.”  Interestingly, several of the studies he cites actually help to make the case for effectively reducing response times under 4 or 5 minutes in certain cases rather than eliminating the standards in general.  Furthermore, the quotes he uses from the 2008 “Gathering of Eagles” consortium position paper entitled “Prehospital Emergency Care” do not discount the time of a response, but instead point out the unsupportability of “over-emphasis on response-time interval metrics” compared to the “unintended, but harmful, consequences (e.g. emergency vehicle crashes) and an undeserved confidence in quality and performance.”  While I also cannot justify the 7:59 standard used in many urban areas, I also cannot condone apathy toward responding timely.  Maybe I am overly sensitive to the literal meaning of “response time doesn’t matter” when justified with the statement that the “golden hour” is just a myth.  For most of us, at least 10-20% of calls include a cardiac, respiratory, stroke or other event where time really is critical and we must be at the top of our game to prevent a death or minimize as much loss in quality of life as possible.

My concern in these arguments is an unstated bias that “response” means only the arrival of an ALS-experienced paramedic traveling with red lights and sirens from a fixed fire station.  Technically, “response” must be understood as simply the time between a call for emergency assistance and the initiation of appropriate necessary treatment.  For many calls, that care could be BLS-led in most circumstances assuming that the calls are appropriately triaged at dispatch.  Emergency Medical Dispatch itself even provides some level of immediate guidance in care with a response time of zero.  Additionally, the greater availability of defibrillators as well as more common knowledge of compression-only CPR means that initial emergency life-saving care can be initiated well before any ambulance arrives.  The existence of advanced telemedicine devices (such as the LifeBot-5) are also changing the rules by providing advanced medical consultation even more quickly in remote rural areas typically with far longer average ALS arrival times.

My point is not necessarily trying to get medical responsders moving faster, but to redefine response time not just as the metric for the ambulance arrival to justify budgets but as a factor that affects patient outcome.  There are many ways to achieve this goal and it begins as education within the system as well as with the public because technology is changing the dynamics.  Zavadsky’s points are valid.  Making defibrillators more available and teaching the public how to respond when a medical event is witnessed is critical.  Also while adding ambulances and staff to more locations would be another way to address reducing response time, it is not financially practical.  An effective alternative to achieve that same goal would be to position the responders closer to the call thereby minimizing distance and the associated need for risky driving.  Modern “dynamic system status management” practice has proven that response time can be shortened to most calls (at least 80-85%) without the need for excessive driving risk that places crews or the public in danger.  Improving performance means responding appropriately in less time – not necessarily just responding “faster.”  Technology can be evaluated as being “outcome-based” just the same as patient treatments.

Watch for future posts which will highlight other components of performance-based EMS beyond just measuring and improving response time.

 

6 Comments

  • Skip Kirkwood says:

    “…Modern “dynamic system status management” practice has proven that response time can be shortened to most calls (at least 80-85%)…”

    I seem to have missed that paper. Can you provide a citation?

    “Response time doesn’t matter” is simply a shorthand way of saying, “There is no evidence to suggest that shortening ambulance response time has a measurable effect on patient outcomes.” Which, as best I can tell, is true.

    The big farce is that “ambulance response time” has anything to do with “time until treatment.” In most cases, treatment can be delivered prior to ambulance arrival. The whole “8 minute” thing evolved when “defibrillator” meant “paramedics” which meant “ambulance” in Seattle in the late 1960s. No longer true. The cases where shortening the care continuum matters (STEMIs, trauma requiring immediate surgical intervention)are few, and could be perhaps addressed some other way than ambulance response time.

    • daleloberger says:

      Skip, thanks for requesting the clarification on my claim. The statement about affecting response time was based on research done at BCS comparing the “hotspot forecasts” using MARVLIS with actual call records. The ‘Demand Monitor’ module of MARVLIS forecasts a demand probability surface displayed in as a map in ‘Deployment Monitor’ with a range of colors signifying future demand. The “red zones”, typically covering about 6 to 10% of the geography of the overall service area, have been found to contain about 75% of the calls during the next hour. The next larger “yellow zone” is a ring that covers an additional 5-10% of future calls. This means that 80-85% of the calls that will be received can be accurately forecast within a small percentage of the service’s jurisdiction. Any system that can tell, with that level of certainty, where future calls will be initiated from will provide time to properly preposition vehicles to better respond safely and without undue haste that could increase risk to providers or the public. Actual customer findings over many years are primarily anecdotal, but widely confirm the actual test results using sample historic user data. Hopefully, a university can examine additional sample data to test these results more thoroughly and publish them independently.

      You are also absolutely correct in distinguishing “ambulance response time” from “time until treatment” because there is not always a direct correlation. As the public becomes better prepared to take action in time critical instances, such as cardiac arrest, before advanced emergency care arrives via an emergency dispatched vehicle, the odds of minimizing any loss in future quality of life for the patient is greatly increased. While great strides have been made in teaching the public, or even increasing the number of medically trained first responders, it is not yet prudent to assume such help will be generally available. The AHA “chain of survival” still includes a link for “early ALS intervention”. At this point, that level of medical training still comes most often in the ambulance.

  • MJMaz says:

    Isn’t a large (and important) part of this also public perception? The public, who by in large do not have a comprehensive understanding of prehospital medicine, likely find the “response time” metric the easiest way to measure EMS Performance. Right or wrong, thats the way it is.

    While, I agree that quick response times do not necessarily correlate to better patient outcome, or higher quality care; I do believe that quick response times provide an appropriate (and necessary) reassurance to our customers. Our customers expect competent medical care upon arrival of any ambulance, and with that in mind, the quicker we respond to their calls, the better we are doing (in their eyes).

    • daleloberger says:

      I completely agree here as well. Whether we like them or not – or even believe in them or not – response times are still being demanded and measured. And it is not just the public demanding quick response, service funding and grants are also sometimes tied to meeting these perceived “objective” measures.

      I have already mentioned that I cannot justify the common 7:59 response standard, but to say that the time any patient waits is not important is not realistic or service-oriented.

  • Darrell Roberts says:

    Interesting read, and while I agree with the article in principal, I think it fails to take into account a few facts that we deal with in the real world on the streets. I’ve been doing this for quite a number of years now and we all have seen run volumes increase while at the same time funding decreases and staffing levels in many cases also decrease. We are currently going through that now. EMS and the fire service as a whole is being squeezed to become more streamlined, do more with less to the point that it not only becomes a safety issue to the service, but to the customer as well. There’s only so many ambulances and fire trucks to go around and if everyone is tied up on less than critical calls someone is going to end up waiting perhaps during a true emergency. I think a better question might be how do we educate the public on the proper use of 911, which has now become the answer for everything from my toilet won’t flush to grandma isn’t breathing. Yes, there are true emergencies where fast response times and rapid intervention do make a difference, but there are too many times when 911 is used when there are other avenues of service more appropriate to the situation.

  • Rick Moyle says:

    As I read this article and the responses, I realize that I am a half step behind as usual. I have been thinking about this matter for a couple of years now. I agree with pretty much everything said. The general public has the perception that faster is better. When they call 911 it may not always be an emergency, but it is for them. I am seeing a much bigger increase in the number of people who really can not take care of themselves (all ages) and use the ER as a clinic, naturally the ambulance gets them seen faster, right? Much of that is lack of education. If we “educate” them and they do not call 911 where does that leave us in terms of liability when there is a negative outcome (maybe another topic)? Also, I agree that not all calls need lights and sirens. Priority dispatch can sort which calls get what type of response. Likewise trained EMD can give appropriate instructions over the phone while a response vehicle is en route. More available options today allow for a “non-emergent” ambulance response. Some of these are QRV and First Responder agencies (who are closer to most calls). The Fire Service normally abides by the NFPA recommendation of a Station for each 5 mile radius. Where I am the average Fire response is 4-5. EMS can be take anywhere from 2-15 minutes depending where they are coming from. My department runs ALS engines. We can manage any scene that comes up until an ambulance arrives. Another option to alleviate the burden on 911 ambulance is using private or third party services for the routine, non-emergent calls with a first responder standing by until they arrive.

3 Trackbacks

Leave a Reply

Your email address will not be published. Required fields are marked *

background image Blogger Img

High Performance EMS

High Performance Emergency Medical Services (HP-EMS) systems provide effective clinical care promoting positive patient outcomes and community wellness while maintaining a focus on improving economic efficiency of the system. This site is dedicated as a community seeking to increase agency performance by promoting useful information regarding the developing trends and improvements in the efficiency of delivering basic and advanced medical care in the field.

FireEMS Blogs eNewsletter

Sign-up to receive our free monthly eNewsletter

LATEST EMS NEWS

HOT FORUM DISCUSSIONS