We know that victims of cardiac arrest have a much higher chance of survival with bystander CPR. In a study of more than 17,000 bystander-witnessed cardiac arrests, the Swedish Cardiac Arrest Registry found CPR done by lay rescuers doubled the survival rate (6.2%) and CPR administered by health-care professionals tripled the survival rate (10.8%) at one month when compared to no bystander CPR.(1) If we look more closely at the benefit of high-quality bystander CPR, it has been shown that “good” CPR—defined as providing a palpable pulse and chest rise with ventilation—led to a 23% survival to hospital discharge, compared to 1–6% for patients with no or poor CPR.(2)

The definitions of “good” and “poor” CPR used in the above-mentioned study are empiric but reflect a field study using humans showing differences in outcomes. The 2005 American Heart Association (AHA) Guidelines for CPR and emergency cardiac care more objectively define our current standard for “high-quality” CPR based mostly on animal data. They include four independent, interrelated CPR benchmarks.(3)

  • Minimal “hands-off” time, i.e., minimal interruptions and maximized efficiencies of ventilation, rhythm-check and pulse check pauses;
  • Proper compression rate;
  • Proper compression depth and allowance of full chest recoil; and
  • Proper ventilation, which is being redefined but must include avoidance of hyperventilation.(4)

We should focus on simultaneously improving all four variables that affect cardiac arrest outcomes, because improvement of one or two aspects with suboptimal execution of others will likely negate a positive effect.(4-8) It’s the responsibility of the quality improvement (QI) effort to provide a framework of education, training and measurement that ensures we meet these four parameters to achieve better outcomes in out-of-hospital cardiac arrest. In addition, we should consider how technology may aid this effort.

Defining the QI Process
The first step in any QI initiative is to define the goal and plan for achieving it through a combination of training, education or improved technology. After implementation, we must study any change to confirm improvement is achieved. Finally, the process change that leads to an improvement must be institutionalized, and the changes that didn’t should be altered or abandoned.

As we try to improve quality we must measure our performance against our benchmark and give feedback to providers. The process should start with goal definition and provider education and training. It’s important to note that education involves providing evidence of new knowledge as a rationale for change while training involves the hands-on integration of new protocols and building the muscle memory involved in providing CPR at the proper rate and depth, as well as the feel of proper ventilation volume and rate. The provider has to have a “trained intuition” for what proper depth and rate feels like. Modern training manikins provide feedback more effectively than simple trainer coaching. Technology seems to be on our side during training for compressions and ventilation.(9)

After the initial rollout, the method for measuring the effect of the change must be in place. Inability to monitor the effects of system changes reduces their potential benefit by blinding the system to their true nature. Success and failure of these initiatives can be inaccurately assigned in equal measure by our inability to measure what we’ve done. These techniques aren’t mutually exclusive; rather they’re conceivably complementary. It’s important to also consider the “shelf life” for this training, and to know when and how to retest and retrain providers.

We can’t assume the devices that have shown benefit for manikin training will translate to improved CPR on our patients. Education and training alone may not lead to good CPR in the field. Extensive evidence confirms that CPR performance in the real world has been suboptimal.(9,10) Therefore, oversight, feedback and, when possible, measurement of provided CPR must be part of our QI loop. We must determine the available methods and whether they’re effective.

Delivering Feedback
Probably the most commonly used QI tool for the critique of CPR on real patients is direct observation with post-event debriefing. This technique is a good means to reinforce obvious positives and to use less-positive aspects as opportunities for improvement on the next event. Historically, this is a subjective exercise most often led by a supervisor who shares their observations as the “fly on the wall.”

An enhanced version of this technique uses recording tools that generate a report of the entire event from the time the monitor/defibrillator was attached to the patient. Studies provide good evidence supporting this technique to improve CPR performance. The use of a debriefing intervention with actual CPR quality data improved CPR guidelines compliance and led to an improvement in ROSC in one in-hospital study.(10)

A variety of technologies can improve compliance with our goal of high-quality CPR. They range from simple metronomes and crew coaching to post-event review of downloaded data, solutions based on accelerometers and impedance changes detected by sophisticated cardiac monitors.

Monitor/defibrillator vendors offer software that records these data. These products show the underlying rhythm, ventilations, compressions, delivered shocks and other biometric data, such as end-tidal CO2 and blood pressure. This can be shared with the team during the post-event review. This approach requires no special additional devices in its basic form and runs silently in the background during the event. A continuous record is generated, which can be annotated by the reviewer to indicate such things as ROSC and ventilation.

Of those devices using compression sensors on the chest, the depth of compression and allowance of full chest recoil by the rescuer are additional recorded features. The software typically allows the generation of a report for the event. For example, a report with the percentage of hands-off time, average compression rate and ventilation frequency helps evaluate areas for further improvement. This type of product can help a system study its CPR delivery, providing a recorded baseline performance level that can be compared against subsequent performance after changes in training or protocols. But can technology help us with real-time delivery of CPR on our patients, and is there good prehospital scientific evidence of safety and efficacy?

Metronomes, which have been integrated into some AEDs and defibrillators, are an effective method of improving delivery of CPR. Small studies have shown improvement in the consistency of chest compressions when a simple metronome was used to pace CPR.(11,12) Therefore, the evidence shows that a simple, low-tech device can be of some help to improve the real-time performance of CPR on patients. Evidence also shows manikins that use a metronome or equivalent for rate and feedback for depth and full chest recoil of CPR lead to better compliance with established optimal performance.(6) This is true for ventilation as well, although this type of manikin training can be enhanced by concurrent instructor moderation.

More sophisticated means for giving detailed feedback during the performance of CPR are available as well. The resistance to the flow of electrical current through the chest, or impedance, changes predictably during ventilation. Monitoring thoracic electronic impedance measured between the defibrillator pads can record ventilation frequency.

Direct CPR feedback in the field is provided by monitor/defibrillator accessories that measure acceleration when a force is applied. Most use a device called an accelerometer, although at least one device relies on direct mechanical compression. Manufacturers have calibrated the amount of desired sternal depression for adults so the end-user gets feedback on the depth and rate of their compressions. This equipment can consist of a device placed between the heel of the CPR provider’s hand and the victim’s sternum, or a one-piece combination of the monitor/defibrillator pads with a built-in accelerometer, which goes over the sternum. Real-time coaching is then built into the device with lights, graphs, tones or voice prompts, or a combination of the four. CPR performance may also be recorded in a data management program for post-event review. Electronic coaching is now available to advise and provide warnings on compression depth, rate, hands-off time and ventilation frequency.

The New Standard?
With this type of technology available, should we consider real-time feedback the new standard for CPR delivery? Are there solid prehospital data to support our adoption of it? Are there any known or potential downsides?

Some studies show accelerometer technology overestimates compression depth when CPR is performed on a mattress, regardless of whether a backboard is placed.(13-15) Perkins showed that when CPR was performed on manikins on a bed with a foam or inflatable mattress, the accelerometer significantly overestimated the sternal compression, with compression of the underlying mattress accounting for 35–40% of the recorded sternal deflection. Placing a backboard improved the accuracy by a small margin, causing an additional 1.9–2.6 mm of deflection depending on whether a narrow or wide backboard was used. The authors relate the loss of compression depth seen in this study to animal studies in which a similar reduction led to a 65% relative reduction in cardiac output.

Had we adopted such technology unaware of this caveat, our data might show excellent CPR performance but our survival rates would presumably suffer. Given the infrequency with which most field resuscitations are performed on a mattress, this specific limitation is less likely to affect field providers than hospital providers. But could there be other unknown issues for what seems like otherwise promising technology?

Are we at risk for information overload with the compression-to-compression review of CPR performance? Does performance improve knowing that a poor last compression occurred? Does the extra piece of equipment add to the complexity of the scene? Is there any loss of compression energy imparted to activate the sensors? If so, does this have any adverse effect on rescuer stamina? There’s very limited data so far in the prehospital literature to provide rational guidance.

A study of three ambulance services in Norway found the introduction of CPR performance evaluation with automated corrective feedback had no effect on CPR quality. Depth of compression, however, significantly improved at one site.(16) The authors concluded that, without implementation strategies to change prevailing attitudes and practice, feedback alone was not sufficient to improve CPR.

Answers to at least some of these questions might come from the Resuscitation Outcomes Consortium trial. Part of this study tests whether a specific feedback device affects cardiac arrest outcome as well as CPR performance. The primary endpoint is the rate of ROSC compared to a baseline established during the first phase of the trial. Secondary endpoints include spontaneous circulation on hospital arrival, survival to discharge, complete release of compression, rate of ventilation, and adherence to guidelines for CPR fraction, rate and depth of compression.(17)

Our systems try to optimize response to cardiac emergencies in a time frame that will make a difference. We’re expected to arrive with the equipment, techniques and experience necessary to achieve success. We expend a significant amount of our resources attempting to improve the chance of survival from out-of-hospital cardiac arrest. Every system wants to excel in the treatment of these patients; some consider their outcomes an indicator of the overall quality of their systems.

EMS QI programs should prioritize their often-limited resources to problems that are simultaneously urgent, important and amenable to study through a set of accepted standards. QI efforts centered on how we deliver CPR to our patients seem to be a natural area for us to reinvigorate our programs and ensure that we’re providing the best quality CPR. We have better knowledge than ever of how to define this quality benchmark.

We’re being offered a variety of technologies that can help us achieve our ultimate goal of saving lives. It’s up to EMS leaders and providers to determine which of these approaches is the best fit for our own agencies and make demonstrable high-quality CPR a top priority.


Disclosure: The author has reported no conflicts of interest with the sponsor of this supplement.


  1. Herlitz J, Svensson L, Holmberg S, et al: “Efficacy of bystander CPR: Intervention by lay people and by health care professionals.” Resuscitation. 66(3):291–295, 2005.
  2. Wik L, Steen PA, Bircher NG: “Quality of bystander cardiopulmonary resuscitation influences outcome after prehospital cardiac arrest.” Resuscitation. 28(3):195–203, 1994.
  3. ECC Committee, Subcommittees and Task Forces of the AHA: “2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.” Circulation. 112(24 Suppl):IV1–203, 2005.
  4. Aufderheide TP, Sigurdsson G, Pirrallo RG, et al: “Hyperventilation-induced hypotension during cardiopulmonary resuscitation.” Circulation. 109(16):1960–1965, 2004.
  5. Abella BS, Alvarado JP, Myklebust H, et al: “Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest.” JAMA. 293(3):305–310, 2005.
  6. Wik L, Kramer-Johansen J, Myklebust H, et al: “Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest.” JAMA. 293(3):299–304, 2005.
  7. Eftestøl T, Sunde K, Steen PA. “Effects of interrupting precordial compressions on the calculated probability of defibrillation success during out-of-hospital cardiac arrest.” Circulation. 105(19):2270–2273, 2002.
  8. Yu T, Weil MH, Tang W, et al: “Adverse outcomes of interrupted precordial compression during automated defibrillation.” Circulation. 106(3):368–372, 2002.
  9. Yeung J, Meeks R, Edelson D, et al: “The use of CPR feedback/prompt devices during training and CPR performance: A systematic review.” Resuscitation. 80(7):743–751, 2009.
  10. Edelson DP, Litzinger B, Arora V, et al. “Improving in-hospital cardiac arrest process and outcomes with performance debriefing.” Archives of Internal Medicine. 168(10):1063–1069, 2008. 
  11. Jost D, Banville I, Degrange H, et al: “Metronome use to improve CPR by firefighters during out-of-hospital cardiac arrest.” Academic Emergency Medicine. 15(s1): S21–S22, (Abstract), 2008.
  12. Berg RA, Sander AB, Milander M, et al: “Efficacy of audio-prompted rate guidance in improving resuscitator performance of cardiopulmonary resuscitation on children.” Academic Emergency Medicine. 1(1):35–40, 1994.
  13. Perkins,GD, Kocierz L, Smith SC, et al: “Compression feedback devices overestimate chest compression depth when performed on a bed.” Resuscitation. 80(1):79–82, 2009.
  14. Nishiki AN, Nysaether J, Sutton R, et al: “Effect of mattress deflection on CPR quality assessment of of older children and adolescents.” Resuscitation. 80(5):540–545, 2009.
  15. Monsieurs KG: “Chest compressions on mattresses: Time to achieve sufficient depth.” (Editorial). Resuscitation. 80(5):503–504, 2009.
  16. Olasveengen TM, Tomlinson AE, Wik L, et al: “A failed attempt to improve quality of out-of-hospital CPR through performance evaluation.” Prehospital Emergency Care. 11(4):427–433, 2007. 
  17. Clinicaltrials.gov: “Secondary Prevention of Small Subcortical Strokes Trial (SPS3).”  http://clinicaltrials.gov/ct/show/NCT00059306