Panel: CPR performance is progressing, but more needs to be done
Dana Edelson, MD, and Vinay Nadkarni, MD
The results of the first-ever, multinational attitudinal survey concerning the performance of CPR by healthcare professionals showed that a mere 25 percent of healthcare professionals actually perform CPR in accordance with the American Heart Association (AHA) guidelines. Following the results' presentation, an expert panel discussed the ramifications of the survey, as well as potential solutions to improve the overall delivery of CPR.

The results were presented concurrently at the AHA 2009 Scientific Sessions. The panel of CPR experts included Robert O’Connor, MD, from the University of Virginia Medical Center; Dana Edelson, MD, from the University of Chicago Hospital; and Vinay Nadkarni, MD, from the University of Pennsylvania Hospital.

The survey of 1,023 respondents (454 U.S. healthcare professionals and 569 healthcare professionals from the U.K., France and Germany)—completed by the CPR Improvement Working Group—reported discrepancies between perceptions of CPR performance and statistics.

Following is part of the panelists' discussion, which included the current state of CPR, the discrepancies uncovered by the survey when it comes to perception and reality of how healthcare professionals perform CPR and ways to improve the quality of CPR performance through training, debriefings and feedback devices.

Q: Are we making progress when it comes to improving CPR performance among healthcare professionals?

Edelson: Over the past four years, we’ve implemented multiple programs in my institution to try to improve CPR quality. We’ve instituted instantaneous feedback, along with debriefing, and saw improvement in CPR quality. Now, CPR quality, in terms of compression depth and rate and ventilation rate, is significantly better. We get a pulse back in almost 60 percent of cases, which is a 50 percent improvement in pulse rate prior to implementing these programs.

Q: Can training result in chest compressions that are closer to the ideal rate?

Edelson: If you train people to deliver the proper rate of compressions, it can be done. Almost 100 percent of people deliver the proper depth post-training.

There were a couple of scientific papers that came out a few years ago that found EMS chest compressions were not being delivered about 50 percent of the time, even when they accounted for legitimate reasons to have the responder’s hands off of the chest. The assumption was that CPR performance is better in the hospital. However, in a mannequin simulation, when the providers tried to adhere to the guidelines, they spent more time off the chest than on the chest.

As important as CPR is to survival, defibrillation is also important because longer interruptions between CPR and defibrillation decrease survival rates. The good thing is that the survey showed that 93 percent of the respondents understood that the amount of hands-off time is something that should be minimized.

Q: Regarding the survey findings, is it the reality that a lot of centers are not using feedback? If so, what can you suggest for training?

O’Connor: I think you’re shedding light on the gap between perception and performance, which is a very common phenomenon. This is another example of where, when one has a tool to measure what one is doing, one becomes aware of the problem and now has the ability to change.

In my hospital, we have the same perception/reality gap phenomenon. When we first measured and used several measuring devices to track depth rate and interruptions of chest compressions, we thought we were doing a very good job.

However, we found that even people who did it well, had about a 77 percent compliance with guidelines. When they turned on the feedback and listened to the debriefings, the performance improved to 88 percent compliance. There is hope that we can improve, but it seems that we need some help to be able to figure out just how to comply with the guidelines.

Q: How do you handle debriefing and ongoing training with your team in the hospital?

Edelson: Resuscitation is surprisingly complicated. There are so many moving pieces, so many things that need to happen at the same time and quickly because somebody’s life depends on it.

We know for example with chest compressions, that if responders do CPR for a while and if you ask them at what point they got tired and the compression got shallow, they say about seven minutes into the simulation. However, we can tell – in terms of their actual compressions – that after one minute, there is decay already.

We can’t be expected to be reliable in a setting like that and to objectively evaluate our own performance.

Q: Have you found that your staff is willing to learn, that they seem to welcome the additional training?

Nadkarni: When we first set out to track and debrief, we were very concerned that the practitioners would feel like Big Brother was watching them and that they wouldn’t want it. They have been doing CPR for 15 or 20 years and they know how to do it. And they question why somebody now is trying to criticize what they are doing.

In some ways, it was a wake-up-call for us, for those who make the guidelines, to listen to the perceptions that are out there and to be careful not to overload [staff] with too many instructions. Once those principles are recognized and embraced, then you convince the staff.

In our hospital, my experience has been that physicians and EMS quickly realize that his topic is important, that we want to perfect the principles of resuscitation, and that we don’t just want to test them on the specific number of compressions.

Q: How do you employ feedback and debriefing and make the distinctions between the two?

Edelson: We make the distinction that feedback is given in real time and debriefing is post-event. In our institution, we don’t do it immediately after an event. We think about it as a “Monday morning quarterback session.” One of the nice things that these monitoring devices allow you to do is to actually extract the data later.

Q: Could you talk about stationary CPR versus moving CPR, for example, on stairs, to the ambulance, etc.?

O’Connor: We have started to run the entire resuscitation onsite until we get don't get the pulse back. It’s been well documented that you can’t do CPR in the back of an ambulance. You are also exposing the providers to potential harm. It’s an unstable environment, for instance, if you crash or take a turn too fast.

Maybe there is an opportunity in the future to use automated devices to somehow alert you when you are over-breathing or under-breathing/over-compressing or under-compressing.

Q: Can you talk about the survey and any potential biases?

Edelson: This survey offers the best case scenario for bias. People who are more apt to respond are the people who are more passionate about CPR. For example, the statistic that 15 percent of people use instantaneous feedback is likely to be higher than in reality.

Q: What about the country comparisons the survey makes?

Nadkarni: When looking at the results of the survey, it was tempting to draw comparisons, but in reality, there were more similarities globally than expected. This underscores the fact that it’s a global issue, at least in the countries that were surveyed.

Q: How would you evaluate the role of feedback on patient safety?

Nadkarni: Particularly in communities where survival rates are low and very few systems are in place, implementing a systematic approach and using the tools that are available to improve the quality of the process of care of CPR can dramatically increase survival. Where systems are already in place and working well, it is difficult to implement small changes and make a difference. Perhaps we have to be more artful at understanding the physiology of the patient and guiding the resuscitation to that.

Edelson: I think the future is bright, and I think we can use these tools to personalize our care, to improve it. It’s a global problem, but a local solution. In almost every case, I’m sure that the solutions that seem to have worked in your hospital are the same principles that are working in my hospital, but the individual interventions have been very different. I don’t think there’s a simple “one size fits all solution” that’s likely to solve the problem.

Q: How can we create an improvement plan for the local hospital?

Edelson: Every system is different and you need to understand how it’s different to create the best plan. First, you need to measure where you are, know that you have a problem, design a plan that fits your system, institute that plan and then continue to measure for feedback and tweak the plan.

We are fortunate now to be in a position where we have the tools to be able to measure performance and enact changes. The key to aiding practitioners who are implementing the guidelines is to use measurement/feedback tools and to keep it simple. The key to being able to aid CPR practitioners do their job better is learning exactly who they are and what their current perceptions of good CPR are.

The science of resuscitation itself is very crude and in the early stages of development; the science of implementation is even younger. It’s a long process from guidelines being issued and their full implementation. We can accelerate that process by encouraging the use of quality measurement/feedback tools to coach and train people to do a better job.

Simplicity is key as it’s too difficult to think about lots of complicated actions when in the midst of performing CPR.

The article is collated from the transcript of the panel discussion, with help from the CPR Improvement Working Group. More information about the survey results is available here.