Validation for a scoring system of the ALS cardiac arrest simulation test (CASTest).

AIM The cardiac arrest simulation test (CASTest) assesses resuscitation knowledge and skills during a simulated cardiac arrest. The aim of this study is to validate an alternative scoring system for measuring individual candidate performance during research involving the CASTest. METHODS The performance of 537 participants was measured using the new scoring system. Evidence of internal structure was sought by comparing the score with global rating of performance and pass/fail decision; identification of participants with instructor potential, skill tests and MCQ scores. Relationships between CASTest score, profession and seniority were also examined. RESULTS Global assessment of performance identified 413 passes (76.9%) and 124 fails (23.1%). CASTest score was significantly higher in those that passed than in those that failed (median 77 vs 62.5, P<0.0001). There were no differences between professions. Senior staff performed slightly better than junior staff (median 74 and 72 respectively, P=0.01). Excellent participants (identified as having instructor potential) scored significantly higher than the other participants (median 94 and 72 respectively, P<0.0001). A strong correlation was demonstrated between domains in the CASTest (rho 0.72-0.82, P<0.01). Other assessment outcomes for the ALS course correlated poorly with CASTest scores (rho 0.27-0.37, P<0.01). CONCLUSION This new simple scoring system can be used to better characterise performance on the ALS course CASTest than the current binary pass-fail outcome.