Abstract:Video-based assessment (VBA) is gaining traction as an adjunctive, and potentially alternative, method to more traditional assessments of surgical technical skills during residency training and may someday play a role in professional certification. 1 In this issue of JAMA Network Open , Ketel et al 2 report the use of a previously developed, video-based, minimally invasive esophagectomy (MIE) competency assessment tool (CAT) 3 to examine the association between the intraoperative technical quality of how MIE is performed at 15 Dutch hospitals and postoperative complications. Similar to the initial seminal work reporting an association of technical quality with better patient outcomes, the authors used masked reviewers with expertise in MIE and found that hospitals with better overall technical quality in the performance of the operation had lower intraoperative and severe postoperative complication rates and better performance during the anastomosis phase was associated with lower leak rates. 2 This important work supports the concept that superior technical skill is associated with better patient outcomes. 4 ,5 It also builds on prior work by suggesting that expertise within the team (rather than just the individual) may be similarly critical. The association between better performance during the creation of anastomosis phase and lower postoperative leak rates is particularly compelling because it suggests that experts can accurately identify technical quality during the most critical steps of a procedure. However, as with anything, the devil is in the details. Therefore, we must consider the nuances of these data to fully understand their meaning and how they might eventually be operationalized. For example, hospitals that performed in the highest quartile of MIE-CAT score had a significantly lower rate of severe postoperative complications compared with those in the lowest quartile (18.7% in the highest quartile vs 39.2% in the lowest quartile). We should exercise caution, however, in interpreting this association because many postoperative complications may be affected by numerous other relevant patient-, process-, and hospital-level factors above and beyond the technical quality of the intraoperative care delivered. Another important nuance is the stratification of performance into quartiles for the purposes of this analysis. While there is certainly precedent for using this type of stratification and it is often necessary for presenting data in a cohesive and interpretable manner, our natural tendency is to equate lower-ranked performance with worse outcomes. However, this type of oversimplification can be problematic because technical ability spans a range (rather than a static rank), and increments within that range may not necessarily translate into clinically meaningful differences. Furthermore, performance may be associated differently with different outcomes, some of which may or may not be expected to be directly associated with better performance. For instance, in this study, R0 resection, lymph node yield, and 30-day mortality were not clearly associated with hospital MIE-CAT performance. While intraoperative technical quality may not necessarily directly affect postoperative mortality, it might have been reasonable to expect an association with R0 resection and lymph node yield. Because they weren't, this finding begs the question of why intraoperative and postoperative complications were associated while these other outcomes were not. It may simply be that these findings are a reminder that the goal is not, and should not be, numeric performance alone but, rather, clinically relevant competence. Because skill level represents a spectrum and a variety of skill levels may achieve the same desired result, as VBA becomes more common and takes on a larger role in evaluating surgical competence, it will be imperative to continue to anchor technical skill (as measured with these types of tools) to clinically relevant patient outcomes. A critical challenge highlighted by this study is the relatively labor-intensive nature of creating footage and completing assessments as part of VBA. That the authors cite an unfeasible workload as the main reason they could only use 2 representative videos to create a hospital-level assessment speaks directly to this concern. While the smaller team sizes at the Dutch hospitals included in this study may have facilitated a limited number of videos providing a generalizable representation of each hospital's performance, we cannot know what we haven't seen. With only 30 total videos assessed from the 15 hospitals that treated the 970 patients included in the study, potential explanations for these findings may be that (1) the authors have identified a tool that generalizes perfectly (with minimal data) to all the surgical team members and patients underg -Abstract Truncated-

Use of error management theory to quantify and characterize residents' error recovery strategies

Surgical experience and identification of errors in laparoscopic cholecystectomy

Using Artificial Intelligence to Gauge Competency on a Novel Laparoscopic Training System

Error identification and recovery by student nurses using human patient simulation: opportunity to improve patient safety

Observed rates of surgical instrument errors point to visualization tasks as being a critically vulnerable point in sterile processing and a significant cause of lost chargeable OR minutes

Video-based learning of coping strategies for common errors improves laparoscopy training—a randomized study

Medication Errors as a Marker of Resident Competency

Video-Based Assessment of Surgical Quality

The place of error in the teaching field of heart surgery

Medical Error: Using Storytelling and Reflection to Impact Error Response Factors in Family Medicine Residents

Evaluation and Implementation of a Mannequin-Based Surgical Simulator for Margin-Involving Eyelid Laceration Repair – a Pilot Study

Expanded Access to Video-Based Laparoscopic Skills Assessments: Ease, Reliability, and Accuracy

Validity and reliability evidence support task-specific metrics for laparoscopic fundoplication

Harnessing In Situ Simulation to Identify Human Errors and Latent Safety Threats in Adult Tracheostomy Care

Experiential Learning in Patient Safety: A Multi-Center Study Examining Emergency Medicine Residents' Situational Awareness of Hazards via Simulation

Judgment Errors in Surgical Care

Medical Error: Using Storytelling and Reflection to Impact Resident Error Response Factors

Improving Resident Comfort With Peripartum Cesarean Hysterectomy Through the Use of Low-Fidelity Simulation Models

Force-Sensing Enhanced Simulation Environment (forsense) for Laparoscopic Surgery Training and Assessment

Implementation of an Online Intraoperative Assessment of Technical Performance for Surgical Trainees

Quality and Safety of Minimally Invasive Surgery: Past, Present, and Future