Validating peer-led assessments of CPR performance

This study compares CPR performance outcomes between objective assessment through manikin generated metrics on real-time performance to subjective Instructor-led assessments.

Background: A patient’s survival from cardiac arrest is improved if they receive good quality chest compressions as soon as possible. During cardiopulmonary resuscitation (CPR) training subjective assessments of chest compression quality is still common. Recently manikins allowing objective assessment have demonstrated a degree of variance with Instructor assessment. The aim of this study was to compare peer-led subjective assessment of chest compressions in three groups of participants with objective data from a manikin.

Method: This was a quantitative multi-center study using data from simulated CPR scenarios. Seventy-eight Instructors were recruited, from different backgrounds; lay persons, hospital staff and emergency services personnel. Each group consisted of 13 pairs and all performed 2 min of chest compressions contemporaneously by peers and manikin (Brayden PRO®). The primary hypothesis was subjective and objective assessment methods would produce different test outcomes.

Results: 13,227 chest compressions were assessed. The overall median score given by the manikin was 88.5% (interquartile range 71.75–95), versus 92% (interquartile range 86.75–98) by observers. There was poor correlation in scores between assessment methods (Kappa 0.051 – þ0.07). Individual assessment of components
within the manikin scores demonstrated good internal consistency (alpha ¼ 0.789) compared to observer scores (alpha ¼ 0.011).

Conclusion: Observers from all backgrounds were consistently more generous in their assessment when compared to the manikin. Chest compressions quality influences outcome following cardiac arrest, the findings of this study support increased use of objective assessment at the earliest opportunity, irrespective of background.