Construct validity of the ovine model in endoscopic sinus surgery training.

ABSTRACT

OBJECTIVES/HYPOTHESIS: To demonstrate construct validity of the ovine model as a tool for training in endoscopic sinus surgery (ESS). STUDY DESIGN: Prospective, cross-sectional evaluation study. METHODS: Over 18 consecutive months, trainees and experts were evaluated in their ability to perform a range of tasks (based on previous face validation and descriptive studies conducted by the same group) relating to ESS on the sheep-head model. Anonymized randomized video recordings of the above were assessed by two independent and blinded assessors. A validated assessment tool utilizing a five-point Likert scale was employed. Construct validity was calculated by comparing scores across training levels and experts using mean and interquartile range of global and task-specific scores. Subgroup analysis of the intermediate group ascertained previous experience. Nonparametric descriptive statistics were used, and analysis was carried out using SPSS version 21 (IBM, Armonk, NY). RESULTS: Reliability of the assessment tool was confirmed. The model discriminated well between different levels of expertise in global and task-specific scores. A positive correlation was noted between year in training and both global and task-specific scores (P < .001). Experience of the intermediate group was variable, and the number of ESS procedures performed under supervision had the highest impact on performance. CONCLUSIONS: This study describes an alternative model for ESS training and assessment. It is also the first to demonstrate construct validity of the sheep-head model for ESS training.