Quality of cardiopulmonary resuscitation: Degree of agreement between instructor and a feedback device during a simulation exercise

Background: High-quality cardiopulmonary resuscitation (CPR) saves lives after a cardiorespiratory arrest. Currently used feedback devices (FD) provide accurate information on CPR quality during training. However, there is no consensus in previous studies that have compared FD to the feedback provided by an instructor and most show methodological limitations. This study aimed to analyse the degree of agreement between an FD and an instructor in the assessment of high-quality CPR. Method: 60 undergraduate nursing students participated in a descriptive observational study. Variables related to the quality of chest compressions (CC) and ventilation were recorded. Students were evaluated during 2 minute compression/ventilation cycles by an expert instructor and using the CPR training torso, Little AnneTM QCPR (Laerdal Medical) with its associated QCPR Instructor App software for iOS. Results: The degree of agreement between instructor and FD assessments was moderate-good: CC rate per minute (Intracass correlation coeficiente [ICC] = 0.791), complete chest recoil (ICC = 0.437); CC depth (k = 0.804); CC with correct depth (ICC = 0.557); correct ventilations (k = 0.510); ventilations per cycle (ICC = 0.635); CC per cycle (ICC = 0.215); overall quality of CPR (ICC = 0.602). However, the degree of agreement should be considered poor since the limits were broad. Conclusions: Although there were discrepancies between the FD and the instructor, it would be advisable to follow a combined approach in CPR training, whereby the quantitative feedback supplied by the FD is complemented by the qualitative assessment of an instructor.