Defining valid, reliable, defensible, and generalizable standards for the evaluation of learner performance is a key issue in assessing both baseline competence and mastery in medical education. However, prior to setting these standards of performance, the reliability of the scores yielding from a grading tool must be assessed. Accordingly, the purpose of this study was to assess the reliability of scores generated from a set of grading checklists used by non-expert raters during simulations of American Heart Association (AHA) MegaCodes.
The reliability of scores generated from a detailed set of checklists, when used by four non-expert raters, was tested by grading team leader performance in eight MegaCode scenarios. Videos of the scenarios were reviewed and rated by trained faculty facilitators and by a group of non-expert raters. The videos were reviewed “continuously” and “with pauses.” Two content experts served as the reference standard for grading, and four non-expert raters were used to test the reliability of the checklists.
Our results demonstrate that non-expert raters are able to produce reliable grades when using the checklists under consideration, demonstrating excellent intra-rater reliability and agreement with a reference standard. The results also demonstrate that non-expert raters can be trained in the proper use of the checklist in a short amount of time, with no discernible learning curve thereafter. Finally, our results show that a single trained rater can achieve reliable scores of team leader performance during AHA MegaCodes when using our checklist in continuous mode, as measures of agreement in total scoring were very strong (Lin’s Concordance Correlation Coefficient = 0.96; Intraclass Correlation Coefficient = 0.97).
We have shown that our checklists can yield reliable scores, are appropriate for use by non-expert raters, and are able to be employed during continuous assessment of team leader performance during the review of a simulated MegaCode. This checklist may be more appropriate for use by Advanced Cardiac Life Support (ACLS) instructors during MegaCode assessments than current tools provided by the AHA.