Observational Skill-based Clinical Assessment tool for Resuscitation (OSCAR): development and validation

The aim of the study reported here was to address the need to assess and train teamwork and non-technical skills in the context of Resuscitation. Specifically, we sought to develop a tool that is feasible to use and psychometrically sound to assess team behaviours during cardiac arrest resuscitation attempts.

To ensure validity, reliability, and feasibility, the Observational Skill based Clinical Assessment tool for Resuscitation (OSCAR) was developed in 3 phases. A review of the literature leading to initial tool development was followed by an assessment of face and content validity, and finally a thorough reliability assessment, using Cronbach's α to assess internal consistency and intraclass correlation to assess inter-rater reliability.

OSCAR was developed methodically, and tested for face and content validity. Cronbach's α results ranged from 0.736 to 0.965 demonstrating high internal consistency, and intraclass correlation results ranged from 0.652 to 0.911, all of which are strongly significant and indicate good inter-rater reliability.

On the basis of our results, we conclude that OSCAR is psychometrically robust, scientifically sound, and clinically relevant. We have developed the Observational Skill-based Clinical Assessment tool for Resuscitation (OSCAR) for the assessment of non-technical skills in Resuscitation teams. We propose the use of this tool in simulation and real Cardiac Arrest Resuscitation attempts to assess, guide and train non-technical skills to team members, to improve patient safety and maximise the chances of successful resuscitation.