• Megan R Silas, Philippe Grassia, and Alexander Langerman.
• Department of Surgery, Operative Performance Research Institute, University of Chicago, Chicago, Illinois; Division of Biological Sciences, Pritzker School of Medicine, University of Chicago, Chicago, Illinois.
• J. Surg. Res. 2015 Aug 1;197(2):272-6.

BackgroundVideo recording in the operating room (OR) has many potential applications for research, quality improvement, and education. Routine video recording is limited by patient and staff privacy concerns, but this may be overcome by creating videos that are deidentified but still retain sufficient video data for intended applications. It is unknown what effect video processing may have on staff perceptions of identifiability in video recordings; this study was designed to investigate this effect.MethodsWe presented OR staff members with short clips of the same staged surgical procedure, each representing different data formats or processing (unaltered video, blurred faces, infrared, and point clouds). Staff rated each video on a Likert scale (1 = anonymous, "no one could identify me"; 10 = not anonymous, "it would be easy to identify me) and provided qualitative comments.ResultsEighty-three staff members agreed to participate. The mean response for the unaltered, blurred faces, infrared, and point cloud videos were 7.05, 3.75, 5.77, and 1.41, respectively (all P < 0.001), demonstrating that postprocessing methods impact perceived anonymity. Staff roles (surgeons versus anesthesiologists versus nurses) were not significantly associated perceptions of identifiability (P ≥ 0.16).ConclusionsThis study demonstrates that surgical video postprocessing affects OR staff members' perceptions of anonymity and that it is possible to produce videos that retain details about surgical activity while still being perceived as anonymous. These findings are highly relevant to any study that uses video for quality improvement or health care research by providing the first normative data on "deidentification."Copyright © 2015 Elsevier Inc. All rights reserved.

23 keywords...

hide…