Comparison of video-based observation and direct observation for assessing the operative performance of residents undergoing phacoemulsification training.

Purpose: To compare the video observation of procedural skills (VOPS) method with the direct observation of procedural skills (DOPS) method in the assessment of senior residents' performance utilizing the International Council of Ophthalmology's Ophthalmology Surgical Competency Assessment Rubric for phacoemulsification (ICO-OSCAR; phaco). Methods: This is a prospective comparative study conducted at a university-affiliated hospital. Six ophthalmology residents of postgraduate year 4 participated. Their performance in phacoemulsification was rated via DOPS and later in a masked manner through VOPS by a single faculty assessor. Results: Seventy-one surgeries were evaluated. There were no statistically significant differences between the scores of VOPS and DOPS regarding all ICO-OSCAR indices except "instrument insertion into the eye" in which DOPS had higher scores (P = 0.035). A significant correlation was observed in total scores of "task-specific" (r = 0.64, P < 0.001) and "global" (r = 0.38, P = 0.003) indices between VOPS and DOPS while some subscales did not show a correlation between the two methods of assessment. The Bland-Altman analysis demonstrated that nearly all data points of total "task-specific" and "global" scores fell within the 95% limits of agreement ([-5.84, 6.87] and [-4.78, 4.86], respectively). Conclusion: This study demonstrated that VOPS holds promise for a general rating of residents' performance.