Identifying Workflow Disruptions in Robotic-Assisted Bariatric Surgery: Elucidating Challenges Experienced by Surgical Teams

Between 1999 and 2018, the prevalence of obesity among U.S. adults rose substantially. Bariatric surgery, predominantly performed laparoscopically, is effective, but the adoption of robotic-assisted surgery (RAS) in bariatrics has been slower than in other specialties. Robotic bariatric surgery (RBS) offers technical advantages (e.g., articulated wrist movements) that may help overcome challenges in patients with larger livers, excess intra-abdominal fat, and thicker abdominal walls. However, RAS introduces costs and operational challenges, including team separation, communication barriers, constrained workspaces, and new training requirements. A human factors engineering (HFE) approach, particularly the observation and classification of surgical flow disruptions (FDs)—deviations from the natural progression of an operation potentially compromising safety or efficiency—can elucidate systemic challenges of RBS. Prior research links increased FDs with higher workload and stress in OR personnel. The present study investigates how RBS affects the surgical work system by identifying and characterizing FDs across procedural phases to inform interventions that improve workflow, safety, and efficiency.

Design: Direct observational study of robotic-assisted bariatric surgery (RBS) cases using a human factors framework. Setting and period: Three hospital sites in California (a 968-bed nonprofit teaching medical center; a 145-bed nonprofit community hospital; and a 229-bed acute care hospital) between October 2019 and March 2022. Only procedures using the da Vinci Xi robot were included. COVID-19 restrictions limited elective case volume. Ethics: Institutional Review Board approvals obtained (site 1/2: Pro00056245; site 3: study 271); study deemed exempt for informed consent as only de-identified data were collected and no patient interaction occurred. Observers and training: Observers received ~25 hours of training, including literature review, surgical video review, practice identifying/classifying FDs, and shadowing to use the data collection tool. Data collected: Real-time FD observations during each case; de-identified patient data included age, sex, BMI, and ASA physical status classification. Procedure phasing and taxonomy: Each procedure divided into five phases (per Kanji et al.). FDs were time-stamped, described, categorized into one of nine categories (coordination, communication, environment, equipment, external factors, other, patient factors, surgical task considerations, training), and assigned a severity score (0 = potential impact; 1 = direct impact on process; 2 = direct or significant risk of direct impact on patient safety). Tools: Macro-enabled Microsoft Excel tool used for FD recording and classification. Analysis: Computed totals, averages, SDs, and CIs for FDs. Calculated FD rates per hour in each surgical phase; excluded severity 0 events from rate analyses. Identified most frequent FD category and conducted sub-analysis using open card sorting. Two researchers developed sub-categories from 100 FD narratives, then categorized an additional 100 to assess inter-rater reliability (Cohen’s kappa = 0.75). Remaining narratives were categorized individually.

• Sample: 29 RBS procedures (sleeve gastrectomy, sleeve with hiatal hernia repair, gastric band removal, revision of sleeve, and Roux-en-Y gastric bypass). Mean case duration 148.87 min (SD ± 49.44). Most patients had BMI > 40 kg/m^2 (55.17%) and ASA > 2 (68.97%).
• FD totals and rates: 1733 FDs observed; mean 59.76 FDs per procedure (CI ± 7.29). Overall FD rate 25.05 per hour (CI ± 2.77). Overall, an FD occurred approximately every 2.4 minutes.
• Phase-specific FD rates: Highest in phase 5 (patient closure to wheels out): M = 30.00 FDs/h (CI ± 6.03); and phase 2 (start of insufflation to surgeon on console/robot docking): M = 29.37 FDs/h (CI ± 4.01).
• Category frequencies overall: Coordination M = 9.63 FDs/h (CI ± 1.72); Environment M = 5.23 (CI ± 1.02); Communication M = 3.56 (CI ± 0.74) were most frequent.
• Category by phase highlights:
  • Coordination FDs were highest across phases, especially phase 2 (robot docking): M = 14.28 FDs/h (CI ± 3.11), occurring about once every 4 minutes during docking.
  • Phase 3: top categories were Coordination M = 7.26 (CI ± 2.01), Communication M = 5.00 (CI ± 1.35), Training M = 3.17 (CI ± 1.46).
  • Phase 5: top categories were Environment M = 13.05 (CI ± 4.57), Coordination M = 12.17 (CI ± 3.60), Communication M = 1.69 (CI ± 1.16).
• Coordination sub-analysis (n = 670 coordination FDs): 14 sub-categories identified. Most frequent were Unavailable staff/instruments (n = 152, 22.59%), Readjustments of equipment (n = 96, 14.26%), Handling supplies (dropping/picking up/putting away items) (n = 70, 10.40%), Assistance/support required (n = 64, 9.51%), and Inadvertent/incorrect actions (n = 54, 8.02%).

RBS exhibited a higher frequency of flow disruptions than previously reported in gynecology, urology, and trauma surgery. The most disruption-prone phases were patient transfer/transport (phase 5) and robot docking (phase 2). In phase 5, constrained environments due to accumulated equipment and cables, combined with the need for additional personnel to transfer patients with obesity, increased both environmental and coordination FDs. In phase 2, preparing the OR for docking and maneuvering the robot within confined spaces, while contending with patient positioning (e.g., extended arm, liver retractor), led to frequent coordination and environmental disruptions. The predominance of coordination FDs reflects demands introduced by robotic technology and the need for new technical skills and planning strategies, such as ensuring appropriate instruments and accommodations (e.g., bed extenders, support belts, chest rolls) are available and that docking paths are clear. The coordination sub-analysis underscores that unavailability of staff or instruments and the need to readjust equipment are primary contributors to workflow delays. These disruptions may cascade, producing additional delays and risks. Systems-focused interventions are suggested, including team training (e.g., the “RAS Olympics” to build technical and non-technical skills), structured preoperative briefings to clarify instruments and layout, and TeamSTEPPS communication practices (closed-loop communication, callouts) to maintain shared situational awareness, particularly with the surgeon at the console.

Using a human factors approach, the study identified frequent flow disruptions in robotic-assisted bariatric surgery, concentrated in patient transfer/transport and robot docking phases. Coordination-related issues—especially unavailable staff/instruments and equipment readjustments—were major contributors. Implementing systems-level interventions, such as targeted team training, structured preoperative briefings, and robust communication practices, may mitigate these disruptions, reducing intraoperative delays, costs, and risks while improving quality of care.

• Limited sample size (29 cases) and inclusion of five RBS procedure types with differing workflows.
• Multi-site study with variability in OR size, layout, and staff configurations; COVID-19 constrained elective case numbers; only da Vinci Xi platform included.
• Observational design susceptible to Hawthorne effect, though direct observation was considered optimal for capturing real-world workflow.
• Staff experience was not measured, potentially influencing FD rates by phase.
• No laparoscopic bariatric surgery control group; findings focus on characterizing RBS workflow challenges rather than direct method comparison.
• To address variability in phase duration and procedure mix, FD rates (rather than totals) were used to enhance generalizability.