Skip to main content

The Trip Report: The Effect of CPR on Bystanders

Reviewed This Month

“Please. Don’t. Die.”: A Grounded Theory Study of Bystander Cardiopulmonary Resuscitation.

Authors: Mausz J, Snobelen P, Tavares W. 

Published in: Circ Cardiovasc Qual Outcomes, 2018 Feb.

A recent Trip Report discussed bystander CPR. This month we review another manuscript on the topic. We don’t normally review two manuscripts on the same topic so close together, but we all know how important bystander CPR is to improving a patient’s chances of survival from out-of-hospital cardiac arrest. 

In July’s Trip Report we discussed a manuscript that evaluated the quality of bystander CPR. That study used quantitative methods to determine that bystanders perform high-quality CPR. Authors evaluated how often bystanders met AHA guidelines for rate and depth. They calculated percentages and p-values with data obtained from databases. There was no interaction between the study team and the study subjects. 

This study, conversely, uses qualitative methods to evaluate the experience of bystanders who have performed CPR. There are no calculations and no large databases to query. Rather, the authors spoke face-to-face with bystanders and sought to understand how this experience impacted them. 


Lead author Justin Mausz and his coauthors performed this study in Ontario, Canada. The EMS system used for this study typically sees about 112,000 emergency calls per year, including approximately 80 cardiac arrests per year where a victim collapses in a public place. About half of these have bystander CPR performed. The authors reported that about 10–20 cases had a bystander delivering a shock from an AED. For purposes of this study, a bystander was defined as “an individual who discovered a victim of cardiac arrest and witnessed their collapse and attempted to intervene by calling 9-1-1, performing CPR, and using an AED.”

This system had recently implemented a program to track AED use by bystanders, with the goal of providing public education and safety planning, and providing emotional support and referral services to bystanders. Paramedics flagged incidents for follow-up by a community safety specialist. Study authors then contacted those bystanders who agreed to participate to obtain written consent and conduct recorded interviews and focus groups. The study took place from November 1, 2015 to November 1, 2016. 

The authors had an interview guide that asked questions regarding the emotional, physical and cognitive challenges of the experience, and what bystanders would have changed regarding their CPR training. However, the authors didn’t stick exclusively to this guide. Rather, they described their interviews and focus groups as semistructured. Authors recognized these events may have been distressing and prioritized emotional support over specific questions, allowing the participants to describe the event “in their own words, at their own pace, and in accordance with their own comfort levels.”

Analysis involved listening to the recordings and reading the transcripts multiple times to develop codes for categories and themes until a point of theoretical sufficiency was reached, or “the point at which progressively fewer new ideas were identified during data collection, the research team was satisfied that all relevant lines of inquiry had been pursued, and we had achieved a rich, deep description of the phenomenon.”


The authors indicated they achieved theoretical sufficiency after reviewing six out-of-hospital cardiac arrests that included a total of 15 involved bystanders (average age 45.6 years). Every bystander who was asked to participate agreed. They conducted a single one-on-one interview and five focus groups ranging from 2–7 participants.

All cardiac arrests occurred in the bystanders’ place of work, and all victims except one were coworkers. All bystanders had previous CPR training either through their job or a traditional class, but none had previously performed bystander CPR. Of the six cardiac arrests included in the study, four victims had ROSC on scene, and three survived. 

The three commonalities that arose from the interview and focus groups were: being called to act, taking action, and making sense of the experience. With regard to the first, the authors noted the participants indicated that “seeing a person collapse is distressing.” Things that were particularly unnerving to participants were “cyanosis, convulsions, incontinence, loss of muscle tone, snoring, and gasping for respirations.” They also noted that the victims’ face, eyes, and lifelessness were distressing. The authors noted the participants experienced senses of panic and urgency and were prompted by the awareness that someone had to act to try to save the victim’s life. 

While all participants overcame perceived barriers to taking action, a consistent barrier was fear of liability or discipline. Participants weren’t sure whether they would get in trouble for performing CPR with an expired certification or if their training wasn’t recent. Workplace policies and procedures were also a barrier, in one case resulting in an off-duty worker being called to report to work to assist with the response.

The authors also indicated that participants often mistook cyanosis, emesis, and gasping for obstructed airways. In one instance a worker delayed chest compressions and stopped CPR intermittently to perform back slaps. Fear of hurting the patient was a common theme, particularly when bystanders did not remember exact anatomical landmarks. Further, the chaos of the scenes was distressing, sometimes limiting the bystanders’ ability to hear instructions provided by their AED. One of the scarier statements indicated that two bystanders “attempted to cut an elongated S-shaped defibrillator pad into two separate pads because the defibrillator they had used during training used two smaller pads.”

Making sense of the experience was trying for some bystanders. Some were unable to drive home following the event because they were shaking so much. Some feared they weren’t allowed to speak about the incident to others. For those cases in which the patient died, participants wondered why their efforts “didn’t work.” It is also evident that everyone reacts differently to these events: Some indicated after the patient died that they were comfortable because they’d done their best to help, while others indicated they were “messed up,” unable to sleep, and experiencing symptoms of depression (even when the patient lived). 

CPR Can Be Stressful

This study is an important addition to the literature. CPR instructors may want to include more information about how cardiac arrest victims may present and fully address misconceptions regarding perceived barriers to act and the chaos that might be present. Further, while other studies have suggested long-term psychological struggles are rare among lay rescuers, instructors may want to explain that those who intervene may have trouble dealing with the event afterward. 

The authors did a commendable job explaining their study limitations, which included possible recall bias (difficulty accurately recollecting the event) and selection bias (only bystanders referred by paramedics on scene received follow-up). Also, because of the small number of participants, the results may not be generalizable. Finally, they noted that because all but one of the victims was a colleague of the bystander, the degree of stress may have been increased. 

July’s quantitative study suggested bystanders do perform quality CPR, and this qualitative study indicates we may need to do a better job providing information when training the public in CPR. For bystanders, providing CPR may be an extremely stressful, once-in-a-lifetime event. It might be a good idea to say “thank you” or “good job” to let them know what they just did was important to help the victim’s chances of survival. A few words might go a long way.  

Antonio R. Fernandez, PhD, NRP, FAHA, is research director at the EMS Performance Improvement Center and an assistant professor in the Department of Emergency Medicine at the University of North Carolina–Chapel Hill.


Back to Top