Hi Patricio, I don't have an answer to your question for you, and I haven't even read up on the topic very recently, but here's what I do have to say. There is more to the question than just the ACLS units. Mickey Eisenberg, M.D. has written a couple of editions of a book, called Resuscitate!, I think, that describes all the complex variables a community has to take into account to achieve high rates of survival for out-of-hospital cardiac arrest. My information may be out of date, so I am not going to give specifics, but in one US city with an all-paramedic system the rates are quite low. Other systems achieve higher rates. A tiered response system, where first responders, such as fire fighters and police have AEDs, who can arrive on the scene before ambulances, plus other elements, such as wide-spread and targeted CPR training (you could see a paper on my profile, Are We Training the Right People Yet? and more recent papers citing it, such as papers by Robert Swor) and dispatcher CPR to make the system work. Eisenberg, makes it clear that each community has its own solution, too. Seattle is extremely unusual, in part because of Mickey Eisenberg's efforts over decades. The ratio you quote is probably just a small piece of explaining that survival rate, and I can guarantee that just replicating the ratio will not reproduce the survival rate. Bob

Thank you. I share the conclusion that survival Eisenberg is correlated with the type of community develop life. how amazing are the different results within the USA (see graph of Medic One report 2013). By the same token if one seeks to stabilize each link in the chain of survival (community, AED, BLS, ACLS ...), we should reach some minimum to be a reference for countries like ours: with low Gross Domestic Product (GDP) or unequally distributed.

That's a tough one. Inputs for an EMS planner would include: steady-state population versus "business hours" / commuter population, age, local risk factors, acuity, types of presenting complaints, traffic patterns, presence / absence of public access defibrillation, access to trauma centers / regionalized trauma / stroke / STEMI / Sepsis / post-ROSC centers, resource availability, population tolerance of risk, revenue base. Geospatial Information systems (GIS) can be very helpful here, if available.

I wouldn't use the Seattle model as a bench mark unless your only goal is cardiac arrest survival rates. The king county model leaves many patients in need of ALS interventions in the hands of BLS providers (first hand experience). I've always felt that a good starting point is 1 ALS ambulance per 30,000 population. After that I would factor in economics, age demographics, etc...

In our experience (Genoa, Italy), where the ratio means ALS and inhabitants is 1: 130,000 the survival rate of patients in cardiac arrest (with any heart rhythm) on field is 13% at 30 days. For all the details you can refer to my thesis: DOI: 10.13140 / 2.1.4653.600 or a technical report DOI: 10.13140/2.1.4737.6163

NFPA 1710 provides a general framework for response times, which may help your question.  However, I do not think there is, or should be, a unit per population ratio.  Different demographic factors influence the amount of calls for service per capita.  There are a lot of a good, practical, statistic programs that can plot calls for service by number, frequency, time of day and area that are useful for determining the level and amount of service provided.