Cardiac Arrest Survival Rates

The Eisenberg Model


A 1993 University of Washington study of 1,667 cardiac arrests linked survival of cardiac arrest to the time that elapsed before the initiation of three critical interventions: CPR, defibrillation and advanced cardiac life support. From this landmark study, researchers produced a model for predicting cardiac arrest survival rates, known as the Eisenberg Model. Because it clearly links response time to the probability of survival, the Eisenberg Model has become a standard method for measuring effectiveness in the delivery of pre-hospital emergency medical services. The Eisenberg model equation is summarized to the right.

 

As the table to the right indicates, a 9-minute response time means that CPR is not initiated until at least 10 minutes have elapsed from the time of cardiac arrest; 11 minutes have elapsed before defibrillation; and 13 minutes have elapsed before ACLS care is initiated, resulting in an expected patient survival rate of only 4.6 percent.  Conversely, a 4-minute fire department response– with CPR initiated in 5 minutes, defibrillation in 6 minutes, and ACLS in 7 minutes– results in patient survivability rates of over 34%.


Quick Navigation

GIS Web-Mapping

The following map represents the GIS analysis of the Eisenberg model, using the existing deployment model. This model represents a dynamic view of response time, as it accounts for unit capability and expected tasks to be performed within a minute of arrival. This model assumes that all Engine companies are Basic Life Support Units and all Rescue squads are Advanced Life Support Units.

 


Download the map in a printable PDF format.

Interpretting the Results

The preceding map examines the predicted survivability of persons suffering out of hospital cardiac arrest based on current and proposed deployment models. The results are summarized by ranges of predicted survival rate as shown in the table below:

 

Table 1: Population by Predicted Survival Rate


Predicted Survival Rate

Population Covered Under Current Deployment Model

0

16

1-10%

63

11-20%

919

21-25%

5,666

26-30%

11,610

31-35%

13,047

36-40%

11,178

41-45%

8,539

 

 

The following tables indicates predicted survivability rates for cardiac arrest patients, based upon the Eisenberg formula in 5, 6, and 7 minutes, respectively.


“Effect of Response Times on Cardiac Patient Survival Rates”[1]

Fire Dep’t.

Response Time

Initiation of CPR

Time to

Defibrillation

Time to Advanced Cardiac Life Support (ACLS)

Predicted

Survival Rate/

All Cardiac Arrest (percentages)

9 minutes

10 minutes

11 minutes

13 minutes

4.6%

4 minutes

F.D. EMT:

5 minutes

11 minutes

12 minutes

18.2%

4 minutes

F.D. EMT:

5 minutes

F.D. EMT-D:

6 minutes

11 minutes

25.8%

4 minutes

F.D. EMT:

5 minutes

F.D. EMT-D:

6 minutes

F.D. Paramedic:

7 minutes

34.3%

 

This scenario requires two firefighters to provide CPR, one to prepare the AED and analyze the results of an electrocardiogram (ECG) report, and one to prepare for and initiate advanced cardiac life support measures, such as advanced airway management, I.V. therapy, and pharmacological interventions.  This breakdown of the expected capabilities of a medical alarm assignment requires a minimum contingent of four EMS personnel to arrive at the scene of a cardiac arrest within 5 minutes of receiving an alarm.  Most experts agree that four responders (at least two trained in ACLS and two trained in BLS) are the minimum required to provide ACLS to cardiac arrest victims[2].

 

As the table indicates, a 9-minute response time means that CPR is not initiated until at least 10 minutes have elapsed from the time of cardiac arrest; 11 minutes have elapsed before defibrillation; and 13 minutes have elapsed before ACLS care is initiated, resulting in an expected patient survival rate of only 4.6 percent.  Conversely, a 4-minute fire department response– with CPR initiated in 5 minutes, defibrillation in 6 minutes, and ACLS in 7 minutes– results in patient survivability rates of over 34%.

 

Put another way, based upon Eisenberg’s maximum percentage survival rate of 67%, the following conclusions can be reached:

• A 9-minute initial arrival time prior to pre-hospital emergency medical intervention gives the patient only a 1 in 15 chance of survival.

• A 4-minute arrival by firefighters, with the initiation of CPR in 5 minutes, increases the probability of patient survivability to 1 in 4.

• Firefighters delivering defibrillation within 6 minutes increases the probability of patient survivability to 1 in 3.

• Firefighters trained as paramedics, and delivering cardiac medication within 7 minutes, increases the probability of patient survivability to 1 in 2.

The simple reduction of 4 to 5 minutes in the response time through the use of cross-trained firefighters has a substantial impact of increased patient survival, with improved patient outcomes for each increase in level of pre-hospital training that firefighters receive.  The Eisenberg Model supports the findings published in the Journal of the American Medical Association, which concluded that “two-tier systems in which the first responders are trained in early defibrillation are most effective in providing rapid Advanced Cardiac Life Support.”[3]

 

It is clear that the quick arrival of an appropriate number of adequately trained personnel deploying with lifesaving medical resources is critical to increasing survivability from cardiac arrest and traumatic injury.  For these reasons, this analysis recommends every engine company be staffed with four full-time firefighters, all of which are trained, at a minimum, to the level of EMT-B.  Inasmuch as an increase in survivability correlates with the degree to which firefighters are trained in emergency medicine, the fire service should pursue efforts to ensure that, of the four firefighters assigned to all engine companies, two firefighters should be certified as EMT-Paramedics (EMT-Ps). 



[1] M.P. Larsen, M.S. Eisenberg, et al., “Predicting Survival from Out-of-Hospital Cardiac Arrest: A Graphic Model,” Annals of Emergency Medicine 22, no. 11 (November 1993): 1652 – 8.

[2] The Journal of the American Medical Association (October 28, 1992): 2291.

[3] The Journal of the American Medical Association (October 28, 1992): 2290.