The influence of real-time feedback on the quality of resuscitation: a prospective study comparing bystanders, paramedic course participants, and emergency physician trainee

The aim of this study was to analyze the potential benefits of real-time feedback in resuscitation training for participants in the prehospital emergency chain and to compare differences in the quality of chest compressions (CC) with and without feedback. The primary endpoint was to analyze the proportion of CC achieving the recommended depth (5-6cm) and frequency (100-120 / min) during two minutes of CC. This prospective cohort study compares bystanders (N = 75), paramedic trainees (N = 75), and emergency physician trainees (N = 75) with and without the feedback system of the Zoll X-Series®. Without feedback, paramedics (P) achieved the target compression frequency in 82.7 %, bystanders (B) in 49.8 % and emergency physician trainees (EP) in 75% (P vs. B, p < 0.001; EP vs. P, p = 0.759; EP vs. B, p = 0,217). There were no significant differences in target compression depth without feedback. With feedback, P achieved the compression frequency in 90.7%, B in 72.8% and EP in 91.4% (P vs. B, p < 0.001; EP vs. P, p = 0.425; EP vs. B, p < 0.001. With feedback, P achieved the compression depth in 56.9%, B in 47.3 % and EP in 75.1 % (P vs. B, p = 0.759; EP vs. P, p = 0.217; EP vs. B, p = 0.002). The results underscore the importance of real-time feedback in emergency medical training, especially for B. All cohorts showed significant improvement, indicating that feedback enhances CC and promotes skill development. Given the importance of high-quality CC, their early optimization in the training is essential. This highlights the need for standardized training concepts, including timing recommendations for feedback systems. Future studies should consider real-life pre-hospital conditions and investigate chest compression to validate transferability to real-life scenarios.

Methods
*Study design

*The prospective cohort study was approved by the ethics committee of “Medizinische Hochschule Brandenburg” (E-01-20200403).

*Participants/study population

*The recruitment and data collection were conducted at the state-recognized emergency medical school in Berlin. A total of 225 participants from first aid classes (B), paramedic classes (P), and emergency classes for physicians (EP) voluntarily participated in the study outside of class hours. Before the start of the study, the B participants acquired theoretical knowledge on performing chest compressions (CC). At the time of data collection, the P cohort had completed four weeks of theoretical and practical instructions and was about to begin hospital and ambulance internships. The EP cohort consisted of licensed physicians participating in an emergency physician class. 

*Study setting and procedure

*The resuscitation simulator AmbuMan® was equipped with the feedback system of the Zoll X-Series®. Initially participants performed a two-minute chest compression (CC) without the visibility of real-time feedback. After a break of two minutes, another CC for two minutes was performed, this time with real-time feedback.

*Data collection 

*Data collection took place from October 2020 to January 2021. The focus of the resuscitation analysis was on evaluating the compression depth within the target range and the compression rate within the target. The reference values are specified by the resuscitation guidelines [5]. The compression depth should be between 5–6 cm. The compression rate should be between 100–120 compressions per minute [5]. "Above the target range" referred to compressions with a depth >  6 cm or a compression rate > 120 compressions per minute. "Below the target range" was defined as a compression depth of less than 5 cm or a compression rate < 100 compressions per minute.

*Analysis

*The evaluation of the results was conducted using the "RescueNet Code Review" program, which compares compression depth and rate with the specified guideline parameters. Statistical analysis of the collected data was performed using the statistical software SPSS Version 28 (IBM). To describe the central tendency in cases of uneven data distribution or ordinal data, the median with the 25th–75th percentile was used. The comparison of the three independent groups was performed using the Mann-Whitney U test, as the data were not parametrically distributed. Categorical variables were compared using the Chi-square test. The results are presented in tables and graphs as box plots or as absolute numbers with percentage shares.