What’s the best way to structure our EMS system?

By Eric DuBois,

I will start with a question that came up in conversation with Soovin after our most recent lecture.  Is it better to base our emergency response system on the type of patient that is most common or the type of patient that is least likely to survive?

My research for a long time since arriving at UW-Madison has focused on ambulance response during severe weather.  One of the chief problems that needs to be considered in this context is the expected survival of different patient types.  If someone is going to have to wait an extra 20 minutes during a blizzard for an ambulance, it had better be someone who isn’t likely to suffer any long term ill-effects.

On a normal service day, by far the most common patient is a trauma victim.  Yet, here’s the rub:  The vast majority of studies show that prehospital response times are not a very good indicator of trauma patient survival (For more see [1-3] below).  Further, “most ambulance services report that less than 10% of their calls are true life and death situations.”  So clearly our system is over-serving the vast majority of patients.

On the other hand are those few patients in dire need of medical attention, for whom a timely response can be critically important.  With survival rates of less than 6%, surely cardiac arrest patients deserve our concern too?  It is these very patients that have driven much of the literature and the current expectation that service times are so critically important.

Even if we decide it’s better to find the efficient solution for patient survival, this does nothing to answer the overall question of equity.  Worse yet, just because we have the theoretical resources to best serve cardiac arrest patients, doesn’t mean that we will actually serve them any better.  Sanghavi, et al found that the more advanced ALS ambulances actually did worse than BLS [4].  So just having the equipment and people in place may not be enough.

Now there are plenty of non-causal reasons that this could occur. For instance, ALS usually serve worse patients than BLS, and it is only possible to control so much for the actual severity of the patient with the data usually available.  Additionally, it should be considered that if the paramedics and EMTs aren’t well practiced in using advanced life support, due to the infrequency of such calls, trying them on such time-sensitive patients could just make the situation worse.

Beyond the question of what we view our ideal to be is the practical realities of the situation.  The American public expects to receive speedy service whether they have a cardiac arrest or a broken leg. It’s something we expect and it’s something for which we hold our leaders accountable. If we are to be honest with ourselves, it’s hard to argue that we wouldn’t want the quickest response possible when we are the ones dialing 911, no matter what the objective research says on the matter.

As with most questions in public OR, there is no simple answer to this one.  With multiple stakeholders and unclear measures, it is up to us as a community to develop best approaches.  I look forward to diving into these complex issues this semester and hope you’ll come along.  So what do you think?  Is there a better way to handle EMS calls than we currently do?  More importantly, is such a way practically implementable?


[1] Blackwell, T. H., Kline, J. A., Willis, J. J., & Hicks, G. M. (2009). Lack of association between prehospital response times and patient outcomes.Prehospital Emergency Care13(4), 444-450.

[2] Newgard, C. D., Schmicker, R. H., Hedges, J. R., Trickett, J. P., Davis, D. P., Bulger, E. M., … & Brown, T. B. (2010). Emergency medical services intervals and survival in trauma: assessment of the “golden hour” in a North American prospective cohort. Annals of emergency medicine55(3), 235-246.

[3] Pons, P. T., Haukoos, J. S., Bludworth, W., Cribley, T., Pons, K. A., & Markovchick, V. J. (2005). Paramedic response time: does it affect patient survival?. Academic Emergency Medicine12(7), 594-600.

[4] Sanghavi, P., Jena, A. B., Newhouse, J. P., & Zaslavsky, A. M. (2015). Outcomes after out-of-hospital cardiac arrest treated by basic vs advanced life support. JAMA internal medicine, 175(2), 196-204.



4 thoughts on “What’s the best way to structure our EMS system?

  1. This is a nice article on how public perception and public stakeholders guide model choice – thanks for the links! This reminds me of something. When I give talks I am often asked if there are a lot more cardiac events when it is snowing due to people shoveling the snow. The answer is yes: there are more cardiac events but I’m not sure if it is due to shoveling (maybe it’s due to stressing over bad snow removal on the roads?). I’m looking forward to more posts on the role of public stakeholders in the modeling process.


  2. Thank you for talking about this topic as I was kind of looking forward to it. I think that is a very interesting and tricky question; which group of patients have representativeness in the analysis, either the common one, or the urgent one.
    Additionally, it is good to know but also quite dissapointing to get to know one more article that says ALS does not perform better than BLS, in treating cardiac arrest patients. I’ve also seen an article about the impact of ALS for trauma patients, and they also concluded that ALS did not decrease mortality or morbidity(Stiell et al, 2008. The OPALS Major Trauma Study: impact of advanced life-support on survival and morbidity).


  3. Great post, Eric! You raise some really important and really difficult questions. Before this class, it never occurred to me how difficult it could be to correctly allocate ambulance resources. If I were to break my leg, as much as I would like to say I’d give up my spot on an ambulance to someone more “in need” (e.g., someone in cardiac arrest), I suspect in the moment I’d be inclined to behave very selfishly. Optimizing systems involving human elements is such a difficult task. I think there probably IS a better way to handle emergency response, but would we ever be able to convince people to change to a new system? That’s a purely rhetorical question, of course, but I think human nature is a big reason resource allocation problems are so tricky.

    Speaking of, I saw an ad for a TV special on “Texting 911.” I didn’t get to watch the special, but it’s got me wondering what the implications of a texting system for emergencies would be. In your research, have you come across anything like that? Do you have any thoughts on whether it would be beneficial or harmful? It seems to me that it would only increase the stress on an already overloaded system, but maybe there are advantages I’m missing.


    1. I had not given much thought to the issue until you brought it up. It has never really come up as an issue in my research.

      Text-to-911 seems to be a reasonably good way to help combat domestic abuse or gunmen- instances where you don’t want to heard calling 911. It’s also certainly better than having to locate a TTY device for the deaf.

      However, with funding levels where they are, many systems have trouble managing voice calls as it is, nevermind having to deal with texts. Apparently Minneapolis has only managed to start upgrading to the 21st century within the last couple years. So I can certainly see this having some negative impacts. Even beyond the money required to upgrade the system, this could lead to less information about a patient, and when in doubt, EMS likes to over-serve. So that means higher ALS utilization and a bigger probability of something going wrong for those that do need help.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s