Procedure-specific assessment tool for flexible pharyngo-laryngoscopy: gathering validity evidence and setting pass–fail standards

Abstract

Objective

The attainment of specific identifiable competencies is the primary measure of progress in the modern medical education system. The system, therefore, requires a method for accurately assessing competence to be feasible. Evidence of validity needs to be gathered before an assessment tool can be implemented in the training and assessment of physicians. This evidence of validity must according to the contemporary theory on validity be gathered from specific sources in a structured and rigorous manner. The flexible pharyngo-laryngoscopy (FPL) is central to the otorhinolaryngologist. We aim to evaluate the flexible pharyngo-laryngoscopy assessment tool (FLEXPAT) created in a previous study and to establish a pass–fail level for proficiency.

Methods

Eighteen physicians with different levels of experience (novices, intermediates, and experienced) were recruited to the study. Each performed an FPL on two patients. These procedures were video recorded, blinded, and assessed by two specialists. The score was expressed as the percentage of a possible max score. Cronbach's α was used to analyze internal consistency of the data, and a generalizability analysis was performed. The scores of the three different groups were explored, and a pass–fail level was determined using the contrasting groups' standard setting method.

Results

Internal consistency was strong with a Cronbach's α of 0.86. We found a generalizability coefficient of 0.72 sufficient for moderate stakes assessment. We found a significant difference between the novice and experienced groups (p < 0.001) and strong correlation between experience and score (Pearson's r = 0.75). The pass/fail level was established at 72% of the maximum score. Applying this pass–fail level in the test population resulted in half of the intermediary group receiving a failing score.

Discussion

We gathered validity evidence for the FLEXPAT according to the contemporary framework as described by Messick. Our results support a claim of validity and are comparable to other studies exploring clinical assessment tools. The high rate of physicians underperforming in the intermediary group demonstrates the need for continued educational intervention.

Conclusion

Based on our work, we recommend the use of the FLEXPAT in clinical assessment of FPL and the application of a pass–fail level of 72% for proficiency.

https://ift.tt/2rIEO6n