Assessing North Texas Regional Trauma Handoffs: A Multicenter Mixed-Methods Needs Assessment.

INTRODUCTION: Trauma video review of Emergency Medical Services (EMS) handoffs demonstrates frequent problems including interruptions and incomplete information transfer. This study aimed to perform a regional needs assessment of handoff perceptions and expectations to guide future standardization efforts. METHODS: A multidisciplinary team of trauma providers through consensus building created an anonymous survey which was then distributed through the North Central Texas Trauma Regional Advisory Council and four regional level-1 trauma institutions. Qualitative data underwent content analysis; quantitative data are presented with descriptive statistics. RESULTS: Survey responses (n = 249) were submitted by trauma nurses (38%), EMS (24%), emergency physicians (14%), and trauma physicians (13%). Median overall handoff quality was rated well (4, scale 1-5) despite some variability between hospitals (3, scale 1-5). The top five most important handoff details were the same for both stable and unstable patients: primary mechanism, blood pressure, heart rate, Glasgow Coma Scale, and location of injuries. While providers felt neutral about the data order, the vast majority supported immediate bed transfer and primary survey in unstable patients. The majority of receiving providers report interrupting handoff at least once (78%); and 66% of EMS clinicians found interruptions disruptive. Content analysis revealed top priority categories for improvement: environment, communication, information relayed, team dynamics, and flow of care. CONCLUSION: Although our data demonstrated satisfaction and concordance with respect to the EMS handoff, 84% of EMS clinicians reported some to high amounts of variability across institutions. Gaps in the development of standardized handoffs identified include exposure, education, and enforcement of these protocols.

The case for a new post-graduate hernia designation: a review of fellowship council case logs from the past twelve-years.

BACKGROUND: The Fellowship Council (FC) is a robust accreditation body with numerous fellowships; however, no specific criteria exist for hernia fellowships. This study analyzed the case log database to evaluate trends in fellowship exposure to hernia repairs. METHODS: FC hernia case log records (2007-2019) were coded as inguinal or ventral hernias and with or without mesh repair. Retrospective analysis examined total hernia repairs logged, type of repair, program designation, and robotic adoption. Robotic adoption was categorized by quartiles of program performance according to the final year of analysis (2018-2019); yearly performance was then graphed by quartiles. RESULTS: Over this twelve-year period, 93,334 hernia repairs, 5 program designations, 152 unique programs and 1,558 unique fellows were analyzed. The number of fellows grew from 106 (2007-2008) to > 130 (2018-2019). Total hernias repairs per fellow increased from an average of 41.2 in 2007-2008 to 75.7 in 2018-2019 (183.7%). Open and robotic hernia repairs increased by 241.9% and 266.3%, respectively; laparoscopic hernia repairs decreased by 14.8%. Inguinal and ventral hernia repairs comprised 48.1% and 51.9% of total cases, respectively. Advanced GI/MIS and Advanced GI/MIS/Bariatrics programs logged the majority of hernia repairs (86.0-90.2%). 2014 began an exponential rise in robotic adoption, with fellows averaging < 1 robotic repairs before and > 25 repairs in 2019. A significant difference was found between all groups when comparing quartiles of robotic adopters (median robotic repairs per fellow; IQR): first quartile (72.0; 47.9-108.8), second quartile (25.5; 21.0-30.6), third quartile (13.0; 12.0-14.3) and fourth quartile (3.5; 0.5-5.0) (p-value < 0.05). CONCLUSIONS: This twelve-year analysis shows a near doubling in the growth of total hernia repairs, with a decrease in laparoscopic repairs as robotic repairs increased. These data show the importance of hernia repairs in FC fellows' training and warrant further granular analysis to determine specific accreditation criteria for hernia fellowship designations.

Developing artificial intelligence models for medical student suturing and knot-tying video-based assessment and coaching.

BACKGROUND: Early introduction and distributed learning have been shown to improve student comfort with basic requisite suturing skills. The need for more frequent and directed feedback, however, remains an enduring concern for both remote and in-person training. A previous in-person curriculum for our second-year medical students transitioning to clerkships was adapted to an at-home video-based assessment model due to the social distancing implications of COVID-19. We aimed to develop an Artificial Intelligence (AI) model to perform video-based assessment. METHODS: Second-year medical students were asked to submit a video of a simple interrupted knot on a penrose drain with instrument tying technique after self-training to proficiency. Proficiency was defined as performing the task under two minutes with no critical errors. All the videos were first manually rated with a pass-fail rating and then subsequently underwent task segmentation. We developed and trained two AI models based on convolutional neural networks to identify errors (instrument holding and knot-tying) and provide automated ratings. RESULTS: A total of 229 medical student videos were reviewed (150 pass, 79 fail). Of those who failed, the critical error distribution was 15 knot-tying, 47 instrument-holding, and 17 multiple. A total of 216 videos were used to train the models after excluding the low-quality videos. A k-fold cross-validation (k = 10) was used. The accuracy of the instrument holding model was 89% with an F-1 score of 74%. For the knot-tying model, the accuracy was 91% with an F-1 score of 54%. CONCLUSIONS: Medical students require assessment and directed feedback to better acquire surgical skill, but this is often time-consuming and inadequately done. AI techniques can instead be employed to perform automated surgical video analysis. Future work will optimize the current model to identify discrete errors in order to supplement video-based rating with specific feedback.

Survey of the 2020 Fellowship Council application and match process and the impact of COVID-19.

BACKGROUND: The interview process represents a necessary but potentially resource intensive process from applicant and program perspectives. This study aimed to identify opinions of the 2020 Fellowship Council (FC) application and match process and in-cycle transition to virtual interviews due to the COVID-19 pandemic. METHODS: Surveys were developed to assess the interview process and were distributed by the FC to all applicants and fellowship programs. Completion was voluntary and data (median [IQR] reported) were anonymous. RESULTS: Applicant response was 53%. Applicants submitted 27.5 (13.25-40) applications, were offered 10 (4-17) interviews, and ranked 10 (5-15) programs. Due to COVID-19, 74% of interview plans changed. Applicants completed 30% of their planned in-person interviews. For decision-making, 90% felt that in-person and 81% virtual interviews were sufficiently informative. Expected cost was $4750 ($2000-$6000) vs. actual cost $1000 ($250-$2250), (p < 0.05). Expected missed work-days were 10 (5-16) versus actual 3 (0-6.25) (p < 0.05). For future interviews, 44% of applicants preferred in-person after virtual pre-interviews, 29% preferred virtual only, and 18% preferred in-person only. Program response was 38%. Programs received 60 (43-85.5) applications, offered 20 (15-26) interviews, completed 16 (12.5-21) interviews, and ranked 14 (10-18) candidates. For decision-making, 92% of programs felt in-person versus 71% virtual interviews were sufficiently informative. Person-hours were greater for in-person 48 (27.5-80) versus virtual 24 (9-40) interviews (p < 0.05). For future interviews, 38% of programs preferred in-person after virtual pre-interviews, 31% preferred in-person only, and 21% preferred virtual only. CONCLUSION: Despite pandemic changes, 81% of applicants and 71% of programs felt they gained sufficient information from virtual sessions to create rank lists. Virtual interviews had lower costs and fewer missed work-days for applicants and decreased resource usage for programs. The majority of both groups preferred either solely virtual or virtual pre-interview followed by in-person interview formats. Virtual interviews should be incorporated into future fellowship application cycles.

Thriving or surviving? A critical examination of funding models for fellowship council fellowships.

BACKGROUND: Since 1997, the Fellowship Council (FC) has evolved into a robust organization responsible for the advanced training of nearly half of the US residency graduates entering general surgery practice. While FC fellowships are competitive (55% match rate) and offer outstanding educational experiences, funding is arguably vulnerable. This study aimed to investigate the current funding models of FC fellowships. METHODS: Under an IRB-approved protocol, an electronic survey was administered to 167 FC programs with subsequent phone interviews to collect data on total cost and funding sources. De-identified data were also obtained via 2020-2021 Foundation for Surgical Fellowships (FSF) grant applications. Means and ranges are reported. RESULTS: Data were obtained from 59 programs (35% response rate) via the FC survey and 116 programs via FSF applications; the average cost to train one fellow per year was $107,957 and $110,816, respectively. Most programs utilized departmental and grants funds. Additionally, 36% (FC data) to 39% (FSF data) of programs indicated billing for their fellow, generating on average $74,824 ($15,000-200,000) and $33,281 ($11,500-66,259), respectively. FC data documented that 14% of programs generated net positive revenue, whereas FSF data documented that all programs were budget-neutral. CONCLUSION: Both data sets yielded similar overall results, supporting the accuracy of our findings. Expenses varied widely, which may, in part, be due to regional cost differences. Most programs relied on multiple funding sources. A minority were able to generate a positive revenue stream. Although fewer than half of programs billed for their fellow, this source accounted for substantial revenue. Institutional support and external grant funding have continued to be important sources for the majority of programs as well. Given the value of these fellowships and inherent vulnerabilities associated with graduate medical education funding, alternative grant funding models and standardization of annual financial reporting are encouraged.

At-home medical student simulation: achieving knot-tying proficiency using video-based assessment.

Purpose: Due to the pandemic, we restructured our medical student knot-tying simulation to a virtual format. This study evaluated curriculum feasibility and effectiveness. Methods: Over 4 weeks, second-year medical students (n = 229) viewed a video tutorial (task demonstration, errors, scoring) and self-practiced to proficiency (no critical errors, < 2 min) using at-home suture kits (simple interrupted suture, instrument tie, penrose drain model). Optional virtual tutoring sessions were offered. Students submitted video performance for proficiency verification. Two sets of 14 videos were viewed by two surgeons until inter-rater reliability (IRR) was established. Students scoring "needs remediation" attended virtual remediation sessions. Non-parametric statistics were performed using RStudio. Results: All 229 medical students completed the curriculum within 1-4 h; 1.3% attended an optional tutorial. All videos were assessed: 4.8% "exceeds expectations", 60.7% "meets expectations", and 34.5% "needs remediation." All 79 needing remediation due to critical errors achieved proficiency during 1-h group sessions. IRR Cohen's κ was 0.69 (initial) and 1.0 (ultimate). Task completion time was 56 (47-68) s (median [IQR]); p < 0.01 between all pairs. Students rated the overall curriculum (79.2%) and overall curriculum and video tutorial effectiveness (92.7%) as "agree" or "strongly agree". No definitive preference emerged regarding virtual versus in-person formats; however, 80.2% affirmed wanting other at-home skills curricula. Comments supported home practice as lower stress; remediation students valued direct formative feedback. Conclusions: A completely virtual 1-month knot-tying simulation is feasible and effective in achieving proficiency using video-based assessment and as-needed remediation strategies for a large student class.