GI Trainees Research Articles

Impact of a deep learning assistant on the histopathologic classification of liver cancer

Artificial intelligence (AI) algorithms continue to rival human performance on a variety of clinical tasks, while their actual impact on human diagnosticians, when incorporated into clinical workflows, remains relatively unexplored. In this study, we developed a deep learning-based assistant to help pathologists differentiate between two subtypes of primary liver cancer, hepatocellular carcinoma and cholangiocarcinoma, on hematoxylin and eosin-stained whole-slide images (WSI), and evaluated its effect on the diagnostic performance of 11 pathologists with varying levels of expertise. Our model achieved accuracies of 0.885 on a validation set of 26 WSI, and 0.842 on an independent test set of 80 WSI. Although use of the assistant did not change the mean accuracy of the 11 pathologists (p = 0.184, OR = 1.281), it significantly improved the accuracy (p = 0.045, OR = 1.499) of a subset of nine pathologists who fell within well-defined experience levels (GI subspecialists, non-GI subspecialists, and trainees). In the assisted state, model accuracy significantly impacted the diagnostic decisions of all 11 pathologists. As expected, when the model’s prediction was correct, assistance significantly improved accuracy (p = 0.000, OR = 4.289), whereas when the model’s prediction was incorrect, assistance significantly decreased accuracy (p = 0.000, OR = 0.253), with both effects holding across all pathologist experience levels and case difficulty levels. Our results highlight the challenges of translating AI models into the clinical setting, and emphasize the importance of taking into account potential unintended negative consequences of model assistance when designing and testing medical AI-assistance tools.

P013 IBD 101: AN INTRODUCTORY COURSE FOR FIRST YEAR GASTROENTEROLOGY FELLOWS

Abstract Introduction Inflammatory bowel disease (IBD) management is changing at a rapid rate, with the emergence of multiple new therapies and evolving care goals. In addition, the quality and quantity of IBD fellowship education is highly variable, based on patient populations and institutional expertise. Therefore, it is crucial to develop improved ways to educate our trainees. Here, we describe an educational initiative designed to familiarize first year gastroenterology (GI) fellows to key concepts in the management of IBD. Methods IBD 101 is a one-day course designed to introduce first-year GI fellows to various clinical topics relevant to the management of IBD. The program was held on September 14, 2019. Fellows from U.S. allopathic GI training programs were selected by their training directors for participation. The course included small group didactic sessions led by expert faculty members and group observed structured clinical examinations (OSCEs), in which fellows observed encounters between expert faculty and standardized patients followed immediately by debrief sessions to reinforce key concepts. The topics included need for surgery in IBD, pregnancy and IBD, escalation of care for acute severe UC, initiating biologic therapy, treatment of mild-moderate IBD, treatment of moderate-severe IBD and managing loss of response to therapy. A review of supplemental opportunities for education in IBD was presented at the conclusion of the course. Pre- and post-course surveys using Likert scoring (1=“strongly disagree” through 4=“strongly agree”) were administered to assess baseline knowledge and educational impact of the course on each addressed topic. Results 55 fellows from 32 programs participated. 49/55 (89%) completed pre- and post-course surveys to assess the educational impact of the program on the didactic sessions and on the group OSCE format. 100% of fellows felt that the course content was appropriate for their scope of clinical practice. Substantial improvement in comfort with all addressed clinical topics was noted (figure 1). In the post-course survey, all fellows reported an improved ability to manage and treat patients with IBD. Comparing career interest in IBD, more participants expressed interest in pursuing a career in IBD after participating (pre: 63% vs post: 75%). 96% of attendees stated that they would strongly recommend this course to future GI fellows. Conclusions This single day course for first-year GI trainees was effective and well-received, and offers a novel intervention to address the challenges of IBD education and training. Follow-up of this cohort of trainees and expansion for next year is planned.

458 Implementation and Analysis of an Online Esophageal Motility Curriculum for Gastroenterology Fellows

INTRODUCTION: Compared with endoscopic training, motility education of gastroenterology (GI) fellows is limited at many training centers. To better prepare fellows for treating patients with esophageal motility (EM) disorders, an understanding of EM is essential. Novel educational technology may facilitate dissemination of standardized curricula to train GI fellows at programs lacking strong motility presence. METHODS: A pretest survey assessing comfort level and baseline interpretation of EM studies was sent to UCSF GI fellows in July 2017. An online application called Qstream was then implemented which sent 1-2 questions per week out of a specially designed question bank. Questions, with detailed explanations, were repeated or retired depending on response accuracy. A post-curricular test was sent in June 2018 to gather quantitative and qualitative data. The Qstream software was also mined for data on correct responses across multiple attempts at answering questions. Analysis was performed using qualitative assessment of feedback from participants and comparative statistics using a paired- t test with unequal variance and Chi-square test. RESULTS: Nine trainees completed a pretest survey and 6 trainees completed the posttest. Initially, only 33.3% of fellows were comfortable in interpreting EM studies, whereas following the curriculum 83.3% felt comfortable in their abilities (P = 0.097). Trainees' accuracy in interpreting and answering clinical questions increased from 77.8% to 80% on the posttest (P = 0.43). However, when using data obtained directly from the Qstream software, there was relative improvement (26.9%) from fellows' first attempt at answering questions to their final attempt. Trainees found the following aspects of the curriculum to be valuable: the ease of usability of the online questions, the flexibility in pacing, and the real time explanations of questions. CONCLUSION: A technology based curriculum was effective and efficient in educating GI fellows on core concepts of EM. The online nature of the curriculum was critical in accommodating the busy clinical schedules of our trainees. Fellows reported increased comfort in interpreting EM studies after completing the curriculum; lack of improvement on the post-test likely reflects their high performance on the pre-test, which was not very complex. We plan to continue offering this curriculum with increasingly challenging questions to allow for further stratification of data and overall improved competency in EM among GI trainees.

A multi-centre audit of excess steroid use in 1176 patients with inflammatory bowel disease.

Corticosteroids are central to inducing remission in inflammatory bowel disease (IBD) but are ineffective maintenance agents. To benchmark steroid usage in British outpatients and assess factors associated with excess exposure. We recorded steroid use in unselected IBD outpatients. Cases meeting criteria for steroid dependency or excess were blind peer reviewed to determine whether steroid prescriptions were avoidable. Associations between steroid use and patient/institutional factors were analysed. Of 1176 patients, 30% received steroids in the prior 12months. 14.9% had steroid dependency or excess, which was more common in moderate/severe ulcerative colitis (UC) than Crohn's disease (CD) (42.6% vs 28.1%; P=.027). Steroid dependency or excess was deemed avoidable in 49.1%. The annual incidence of inappropriate steroid excess was 7.1%. Mixed-effects logistic regression analysis revealed independent predictors of inappropriate steroid excess. The odds ratio (OR, 95%CI) for moderate/severe compared to mild/quiescent disease activity was 4.59 (1.53-20.64) for UC and 4.60 (2.21-12.00) for CD. In CD, lower rates of inappropriate steroid excess were found in centres with an IBD multi-disciplinary team (OR 0.62 [0.46-0.91]), whilst dedicated IBD clinics protected against inappropriate steroid excess in UC (OR 0.64, 95% CI 0.21-0.94). The total number of GI trainees was associated with rates of inappropriate steroid excess. Steroid dependency or excess occurred in 14.9% of British IBD patients (in 7.1% potentially avoidable). We demonstrated positive effects of service configurations (IBD multi-disciplinary team, dedicated IBD clinics). Routine recording of steroid dependency or excess is feasible and should be considered a quality metric.

Tu1006 A Comparison of Screening Colonoscopy Quality Indicators Among Gastroenterologists, Physician Assistants, and GI Trainee Endoscopists

Tu1006 A Comparison of Screening Colonoscopy Quality Indicators Among Gastroenterologists, Physician Assistants, and GI Trainee Endoscopists

Sa1128 - Musculoskeletal Injuries are Frequently Reported among GI Trainees: Results of a National Survey of Fellows and Program Directors

Sa1128 - Musculoskeletal Injuries are Frequently Reported among GI Trainees: Results of a National Survey of Fellows and Program Directors

Colonoscopy Procedural Volume Increases Adenoma and Polyp Detection Rates in GI Trainees: A Longitudinal Analysis: 2016 ACG Fellows-In-Training Award (Endoscopy Category) 2016 ACG Presidential Poster Award

Colonoscopy Procedural Volume Increases Adenoma and Polyp Detection Rates in GI Trainees: A Longitudinal Analysis: 2016 ACG Fellows-In-Training Award (Endoscopy Category) 2016 ACG Presidential Poster Award

Mo1842 Excess Steroid Use in IBD: Too Much, How Much and Why? Results From a UK Nationwide Audit

Introduction Corticosteroids are the cornerstone of inducing remission in IBD but are limited in their ability to maintain remission, and associated with significant side effects. This is the first nationwide outpatient study of steroid use in IBD and factors affecting their use. Methods We audited consecutive IBD patients attending clinics at 11 centres over 3 months using a web-based assessment tool. Cases meeting criteria for steroid excess (SE) as defined by ECCO guidelines were blind peer reviewed and classified as non-IBD use, unavoidable and inappropriate steroid excess (ISE). Associations between steroid use and patient and institutional factors were analysed. Results Of 1177 patients [48% CD, 49% UC, 3% IBD-U] 79% were in remission/mild disease, 18.5% had moderate and 2.5% severe disease. In the previous 12 months, 30% had received steroids, 13.8% had SE. Peer review revealed that SE was inappropriate in 51.2% of these (8% non-IBD use; 40.7% unavoidable). Excess steroid exposure was more common in patients with active UC compared to active CD (41.6% vs 26.6%; p = 0.02). In multivariate analysis, disease activity was a significant predictor of SE/ISE. In addition, being established on anti-TNF agents protected against SE and ISE in CD. Exposure to thiopurine (SE + ISE) and starting anti-TNF therapy (ISE) were associated with excess steroid use in UC. CD patients from centres with an IBD MDT were less likely to have SE, similarly CD patients in centres with combined surgical clinics were less likely to experience SE and ISE. Care in centres with dedicated IBD clinics was associated with less SE and ISE in patients with UC. Centres with large numbers of GI trainees showed higher rates of SE in UC and SE and ISE in CD. (All of above independent predictors on multi-variate with significance p Conclusion We identified inappropriate excess steroid use in 7% of UK IBD patients. Risk factors for steroid exposure differed between UC and CD, likely reflecting inter alia , differences in access to biologic drugs. Our study is the first to demonstrate positive effects of service configurations (IBD MDT, dedicated IBD clinics, combined surgical clinics) on treatment outcomes even after correction for differences in disease severity. There was an association between ISE in Crohn’s and the number of GI trainees per centre suggesting possible gaps in training. Routine recording of excess steroid exposure is feasible and should be considered as a quality marker for outcomes of IBD services. Disclosure of Interest None Declared

Status of Competency-Based Medical Education in Endoscopy Training: A Nationwide Survey of US ACGME-Accredited Gastroenterology Training Programs.

The Accreditation Council for Graduate Medical Education (ACGME) emphasizes the importance of medical trainees meeting specific performance benchmarks and demonstrating readiness for unsupervised practice. The aim of this study was to examine the readiness of Gastroenterology (GI) fellowship programs for competency-based evaluation in endoscopic procedural training. ACGME-accredited GI program directors (PDs) and GI trainees nationwide completed an online survey of domains relevant to endoscopy training and competency assessment. Participants were queried about current methods and perceived quality of endoscopy training and assessment of competence. Participants were also queried about factors deemed important in endoscopy competence assessment. Five-point Likert items were analyzed as continuous variables by an independent t-test and χ(2)-test was used for comparison of proportions. Survey response rate was 64% (94/148) for PDs and 47% (546/1,167) for trainees. Twenty-three percent of surveyed PDs reported that they do not have a formal endoscopy curriculum. PDs placed less importance (1—very important to 5—very unimportant) on endoscopy volume (1.57 vs. 1.18, P<0.001), adenoma detection rate (2.00 vs. 1.53, P<0.001), and withdrawal times (1.96 vs. 1.68, P=0.009) in determining endoscopy competence compared with trainees. A majority of PDs report that competence is assessed by procedure volume (85%) and teaching attending evaluations (96%). Only a minority of programs use skills assessment tools (30%) or specific quality metrics (28%). Specific competencies are mostly assessed by individual teaching attending feedback as opposed to official documentation or feedback from a PD. PDs rate the overall quality of their endoscopy training and assessment of competence as better than overall ratings by trainees. Although the majority of PDs and trainees nationwide believe that measuring specific metrics is important in determining endoscopy competence, most programs still rely on procedure volume and subjective attending evaluations to determine overall competence. As medical training transitions from an apprenticeship model to competency-based education, there is a need for improved endoscopy curricula which are better suited to demonstrate readiness for unsupervised practice.

Post-polypectomy Guideline Adherence: Importance of Belief in Guidelines, Not Guideline Knowledge or Fear of Missed Cancer.

Adherence to post-polypectomy surveillance guideline recommendations is suboptimal. Surveillance is frequently over- and under-recommend, resulting in strained colonoscopy capacity, potential risks without expected benefits, and missed opportunities for colorectal cancer risk reduction. To identify factors associated with adherence to post-polypectomy surveillance guidelines. We conducted a three-phase study with a retrospective review of usual care post-polypectomy surveillance recommendations through medical chart abstraction (Phase I), prospective online physician survey (Phase II), and analysis of survey-based and other physician-based predictors of usual care surveillance recommendations (Phase III). Subjects included patients who underwent usual care colonoscopy 2011-2012 (Phases I and III) and gastroenterology (GI) attendings and fellows (Phases II and III). We identified rates of recommendations consistent with guideline adherence, surveillance overuse, and surveillance underuse based on usual care medical chart documentation and physician survey, as well as predictors of physician adherence to guidelines. We reviewed 640 patient charts for 28 survey respondents. Rates of usual practice recommendations consistent with guideline adherence, surveillance overutilization, and underutilization were 84, 13, and 3%, respectively. At survey, 82% of physicians were concerned about missed cancer. Eleven percentage believed that guidelines were not aggressive enough. GI trainees were 2.5 times more likely to issue guideline-adherent recommendations [OR 2.5, 95% CI (1.5-4.2)]. Disagreement with guideline aggressiveness was independently associated with 40% lower likelihood of adherence [OR 0.6, 95% CI (0.4-0.8)]. Belief in the appropriate aggressiveness of guidelines and trainee position, but not fear of missed cancer or guideline knowledge, was associated with adherence to post-polypectomy surveillance guidelines.

U.S. Gastroenterology Fellowship Programs Are Not Practicing Competency Based Medical Education in Endoscopy Training: National Survey of GI Program Directors and GI Trainees

Introduction: ACGME emphasizes importance of medical trainees meeting performance benchmarks and demonstrating readiness for unsupervised practice. The aims were to describe current methods of endoscopy training and assessment of competence and compare GI trainee (GIT) and GI program director (GPD) perceptions of current endoscopy curriculum quality and what is valued in competence assessment.Table 1Methods: ACGME accredited GPDs and GITs completed an online survey of domains relevant to endoscopy training and competency assessment. Content validity was verified by expert panel of 13 academic GIs. IRB approved. Results: Sixty-four percent (94/148) of GPDs and 47% of GITs (546/1167) responded. Only 77% of GPDs reported having a formal endoscopy curriculum and only 39% of GITs are aware of a formal curriculum. According to GPDs, the vast majority of programs assess competence by procedure volume (85%) and attending evaluations (96%). Minority use independent procedure completion rates (38%), adverse event rates (32%), skills assessment tools (30%) or validated quality metrics (28%). Ten percent of GITs are unaware how endoscopy competence is assessed. Only 64% of GPDs reported requiring fellows to meet specific benchmarks by year of training and only 65% of GPDs and GITs reported that current competency assessment by year of training is good or very good. GPDs rated their programs as having better quality of endoscopy training compared to GITs (tab1A) and GPDs rate quality of feedback better than GITs. Both GITs and GPDs value the importance of quality metrics in assessment of competence, GITs more so than GPDs (tab1B). Conclusion: Although both GITs and GPDs rate quality of current endoscopy training highly (GPD>GIT) and both value quality metrics (GIT>GPD), minority of programs assess competence by year of training or metrics that demonstrate readiness for independent practice. There is a need for improved endoscopy curricula to include competency-based measures which are better suited to demonstrate readiness for unsupervised practice.

Su1091 State of Trainee Inflammatory Bowel Disease Education in the United States: A National Survey

Introduction: The incidence of Inflammatory Bowel Disease (IBD) and the complexity of its management are increasing. Identifying shortcomings and defining the core content of IBD education in gastroenterology (GI) fellowship training in the United States (US) are critical to preparing trainees for this challenge. The aim of this study was to perform a needs assessment survey of US GI fellowship program directors (PD) and trainees to understand the current state of IBD training in the US. Methods: A 15-question PD survey and 19question trainee survey was performed using Survey Monkey. Survey content was developed and reviewed together with nationally recognized IBD experts and optimized in a pilot trial with recently graduated GI trainees. Final questionaires were sent to all US PDs with the request to forward the trainee survey to their respective program trainees. Data were analyzed using excel and SAS v9.3. Results: Surveys were completed by 43/161 (27%) PDs and 116 trainees. The majority of PDs (39/43) worked at university or university-affiliated hospitals. IBD specific inpatient and/or outpatient rotations were offered by 63% of PD institutions. All trainee years were equally represented. While most trainees were or with their inpatient and outpatient IBD training, a significant subgroup was unsure or felt their inpatient (33%) or outpatient (43%) training was inadequate. Only 28% of trainees were satisfied with their current level of IBD exposure during training. Monthly IBD didactics was the only variable independently associated with satisfaction with current level of training (OR 4.6, 95% CI 1.8-11.8). Most trainees (62%) were not interested in pursuing an IBD career, but 10% were considering applying for an advanced IBD fellowship. Strikingly, 57% would consider an IBD focus, if offered, in the context of a 3 year GI fellowship. We examined trainee comfort by different competencies in IBD management (Figure 1). Trainees reported the most confidence or comfort in the management of inpatient CD or UC. In contrast, fewer than half the trainees reported comfort in the management of pouch or stoma issues or the pregnant IBD patient. The proportion of PDs viewing a competency as essential training strongly correlated with trainee comfort in that area (correlation co-efficient 0.75, p < 0.001; Figure 2). Conclusions: Over one-third of participating US GI trainees did not feel confident or mostly comfortable with their level of IBD training. More than half would consider choosing an IBD focus during GI fellowship, if offered. There is a wide spectrum of comfort regarding different competencies in IBD management suggesting that specific areas may require additional focus during training. These findings can provide the basis for the development of an IBD core competency curriculum for US GI fellowship programs.

Mo1501 Attending Endoscopists and Gastroenterology (GI) Trainees Have Similar Performance When Characterizing Diminutive Colorectal Polyps With Narrow Band Imaging (NBI)

Mo1501 Attending Endoscopists and Gastroenterology (GI) Trainees Have Similar Performance When Characterizing Diminutive Colorectal Polyps With Narrow Band Imaging (NBI)

Limited Formal Training in Optically-enhanced Endoscopy: A Single Center Survey Analysis

Purpose: Colonoscopy is the gold standard screening modality for colorectal cancer but miss rates up to 25% have been reported. Novel technologies assist real-time polyp histology evaluation such as confocal endomicroscopy and endocytoscopy along with the contemporary NBI, I-Scan and FICE functions. The aim of this survey was to investigate trainee fellow exposure and demand of novel optically-enhanced endoscopy (OEE) techniques. Methods: A 10-question web-based survey was distributed to 34 gastroenterologists (17 fellows and 17 attendings) at a university-based gastroenterology program. Each of the survey participants have experience in performing screening colonoscopy and polyp evaluation but variable experiences with OEE. Results: We received a 74% overall response rate which included 15 fellows and 9 attending physicians. The GI trainee response rate was 94% and the attending response rate was 53%. The trainees comprised six physicians in their first year, four in their second year, and six in their third year. Overall, 68% of the survey participants reported a cecal intubation rate >90%, and in 88% an estimated cecal withdrawal times greater than 6 minutes. Forty percent of the survey reported high confidence in classifying polyps as hyperplastic or adenomatous. A total of 68% reported limited formal training with OEE either during GI training or as part of continued medical education. An overall 80% of respondents felt formal training in these modalities would be useful; since only 40% currently utilize these techniques in polyp evaluation. Predictably, attendings characterized polyp histology with a higher confidence than fellows (78% vs. 19%). Only 31% of the trainee fellows reported significant formal exposure to OEE but 63% use these technologies to assist polyp evaluation. These values were similar for attending physicians with one-third reporting formal training in OEE and nearly two-thirds utilizing these technologies in polyp evaluation. Overall, 75% of fellows reported formal training in these technologies would be useful in their training and 89% of the experienced attending physicians stated that formal training in these technologies would be useful. Conclusion: A majority of GI trainees and GI attendings surveyed reported limited formal training in OEE techniques and believe that formal training in these technologies would be useful during GI fellowship. The limitations of this study include its single institution, survey-based format, and lack of correlation between self-reported polyp characterization confidence and technology utilization rates with actual clinical encounters. Further studies are needed to explore the learning curve for novel endoscopic technologies and the role of these technologies in GI fellowship training.Table: Table. Survey demographics and response rate

PTH-139 A Unique Combined Gastroenterology/Rheumatology Clinic: the first Year

IntroductionArticular problems affect many patients with inflammatory bowel disease (IBD) and joint symptoms are often difficult to control despite the therapeutic strategies aimed at controlling gut inflammation (1). Patients with...

PTU-042 Endoscocopic Training in upper Gastrointestinal Bleeding (Ugib): a BSG Regional and National Audit

IntroductionUGIB is a common emergency frequently requiring endoscopic intervention. Training in therapeutic endoscopy for UGIB is not mandatory. Furthermore UGIB endoscopic experience may be diminished by the European Working Time...

PTU-015 Trainee Endoscopic Procedures by Deanery and Year of Training in the United Kingdom: August 2011 to July 2012. are we doing Enough?

IntroductionThe burden of acute medicine on specialty trainee registrars (StR) is rising. Anecdotal evidence suggests that fewer endoscopic procedures are being performed by gastroenterology (GI) StR’s and that trainees although...

Tu1440 Advanced Depressed Colorectal Cancer Shows a High Rate of Lymphatic Invasion, Venous Invasion and Nodal Metastasis Compared With Protruded-Type

were asked questions regarding polyp characteristics, suspicion for malignancy, and recommendations for resection. Results: The survey response rate was 154/ 330 (46.7%). Seventy-eight percent of respondents were attending physicians (91 GI and 29 surgery) and 22% were GI trainees. Among the GI attendings, 24 (26.3%) self-identified as SCPs. Endoscopic removal was recommended by 86.3% of respondents for the 5 colorectal adenomas compared to 63.5% for the polyp with adenocarcinoma (P 0.001). SCP were least likely to recommend surgical removal of the four colon adenomas (3.1%) while surgeons were most likely (17.2%, P .008). Surgeons were also more likely to recommend surgical resection of the rectal polyp than GIs (44.8% vs. 14.3%, P 0.004). Among all respondents, surgical resection was recommended more often for nonpolypoid (Paris O-IIA) compared to polypoid (Paris Is or Ips) lesions (20% vs. 9.6%, P 0.003). No differences in recommendations for endoscopic versus surgical resection were observed based on years in practice, polyp location (right versus left colon), or patient ASA class. Non-SCP GIs recommended referral to an SCP for complex polypectomy in 33% of polyps. Referral by a non-SCP GI to an SCP was more likely to be recommended for non-polypoid lesions (47.3%) than polypoid lesions (33.0%, P 0.02); however, polyp location was not significantly associated with referral to an SCP. Conclusions: In this large survey of GIs and surgeons, physician specialty was a strong determinant for the decision of endoscopic versus surgical resection of complex polyps. Surgeons were most likely to recommend surgical resection of complex non-malignant colorectal polyps, whereas SCPs recommended surgery the least frequently. Among polyp characteristics, non-polypoid morphology was the sole predictor of a recommendation for surgical resection or referral to an SCP by a non-SCP GI for endoscopic resection.

Mo1106 GI Trainee Status Guideline Knowledge and Confidence in Guidelines Are Associated With Evidence-Based Recommendations for Post-Polypectomy Colonoscopy Surveillance

Mo1106 GI Trainee Status Guideline Knowledge and Confidence in Guidelines Are Associated With Evidence-Based Recommendations for Post-Polypectomy Colonoscopy Surveillance

Employment Prospects and Trends for Gastroenterology Trainees in Canada: A Nationwide Survey

Many gastroenterology (GI) trainees face a variety of barriers to stable employment and are finding it increasingly difficult to secure employment in their chosen field. To elucidate factors that contribute to the burden of unemployment and underemployment, and to examine solutions that may remedy this growing problem in the field of GI. A nationwide survey of current, incoming and recently graduated individuals of GI training programs in Canada was conducted. Trainees in pediatric GI programs and those enrolled in subspecialty programs within GI were also included. The response rate was 62%, with 93% of respondents enrolled in an adult GI training program. Many (73%) respondents planned to pursue further subspecialty training and the majority (53%) reported concerns regarding job security after graduation as contributory factors. Only 35% of respondents were confident that they would secure employment within six months of completing their training. Regarding barriers to employment, the most cited perceived reasons were lack of funding (both from hospitals and provincial governments) and senior physicians who continue to practice beyond retirement years. Sixty-nine per cent perceived a greater need for career guidance and 49% believed there were too many GI trainees relative to the current job market in their area. Most residents had a contingency plan if they remained unemployed >18 months, which often included moving to another province or to the United States. GI trainees throughout Canada reported substantial concerns about securing employment, citing national retirement trends and lack of funding as primary barriers to employment. Although these issues are not easily modifiable, certain problems should be targeted including optimizing training quotas, tailoring career guidance to the needs of the population, and emphasizing credentialing and quality control in endoscopy.