Timothy D Imler1, Justin Morea2, Thomas F Imperiale3. 1. Division of Gastroenterology and Hepatology, Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana; Department of Medical Informatics, Regenstrief Institute, Indianapolis, Indiana; Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana. Electronic address: timler@iu.edu. 2. Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana. 3. Division of Gastroenterology and Hepatology, Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana; Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana; Center of Excellence for Implementation of Evidence-Based Practice, Richard L. Roudebush VA Medical Center, Indiana University, Indianapolis, Indiana.
Abstract
BACKGROUND & AIMS: With an increased emphasis on improving quality and decreasing costs, new tools are needed to improve adherence to evidence-based practices and guidelines in endoscopy. We investigated the ability of an automated system that uses natural language processing (NLP) and clinical decision support (CDS) to facilitate determination of colonoscopy surveillance intervals. METHODS: We performed a retrospective study at a single Veterans Administration medical center of patients age 40 years and older who had an index outpatient colonoscopy from 2002 through 2009 for any indication except surveillance of a previous colorectal neoplasia. We analyzed data from 10,798 reports, with 6379 linked to pathology results and 300 randomly selected reports. NLP-based CDS surveillance intervals were compared with those determined by paired, blinded, manual review. The primary outcome was adjusted agreement between manual review and the fully automated system. RESULTS: κ statistical analysis produced a value of 0.74 (P < .001) for agreement between the full text annotation and the NLP-based CDS system. Fifty-five reports (18.3%; 95% confidence interval, 14.1%-23.2%) differed between manual review and CDS recommendations. Of these, NLP error accounted for 30 (54.5%), incomplete resection of adenomatous tissue accounted for 14 (25.5%), and masses observed without biopsy findings of cancer accounted for 4 (7.2%). NLP-based CDS surveillance intervals had higher levels of agreement with the standard (81.7%) than the level agreement between experts (72% agreement between paired reviewers). CONCLUSIONS: A fully automated system that uses NLP and a guideline-based CSD system can accurately facilitate guideline-recommended adherence surveillance for colonoscopy.
BACKGROUND & AIMS: With an increased emphasis on improving quality and decreasing costs, new tools are needed to improve adherence to evidence-based practices and guidelines in endoscopy. We investigated the ability of an automated system that uses natural language processing (NLP) and clinical decision support (CDS) to facilitate determination of colonoscopy surveillance intervals. METHODS: We performed a retrospective study at a single Veterans Administration medical center of patients age 40 years and older who had an index outpatient colonoscopy from 2002 through 2009 for any indication except surveillance of a previous colorectal neoplasia. We analyzed data from 10,798 reports, with 6379 linked to pathology results and 300 randomly selected reports. NLP-based CDS surveillance intervals were compared with those determined by paired, blinded, manual review. The primary outcome was adjusted agreement between manual review and the fully automated system. RESULTS: κ statistical analysis produced a value of 0.74 (P < .001) for agreement between the full text annotation and the NLP-based CDS system. Fifty-five reports (18.3%; 95% confidence interval, 14.1%-23.2%) differed between manual review and CDS recommendations. Of these, NLP error accounted for 30 (54.5%), incomplete resection of adenomatous tissue accounted for 14 (25.5%), and masses observed without biopsy findings of cancer accounted for 4 (7.2%). NLP-based CDS surveillance intervals had higher levels of agreement with the standard (81.7%) than the level agreement between experts (72% agreement between paired reviewers). CONCLUSIONS: A fully automated system that uses NLP and a guideline-based CSD system can accurately facilitate guideline-recommended adherence surveillance for colonoscopy.
Authors: J Randall Curtis; Seelwan Sathitratanacheewin; Helene Starks; Robert Y Lee; Erin K Kross; Lois Downey; James Sibley; William Lober; Elizabeth T Loggers; James A Fausto; Charlotta Lindvall; Ruth A Engelberg Journal: J Palliat Med Date: 2017-11-28 Impact factor: 2.947
Authors: Timothy D Imler; Justin Morea; Charles Kahi; Eric A Sherer; Jon Cardwell; Cynthia S Johnson; Huiping Xu; Dennis Ahnen; Fadi Antaki; Christopher Ashley; Gyorgy Baffy; Ilseung Cho; Jason Dominitz; Jason Hou; Mark Korsten; Anil Nagar; Kittichai Promrat; Douglas Robertson; Sameer Saini; Amandeep Shergill; Walter Smalley; Thomas F Imperiale Journal: Am J Gastroenterol Date: 2015-03-10 Impact factor: 10.864
Authors: Melissa Magrath; Edward Yang; Chul Ahn; Christian A Mayorga; Purva Gopal; Caitlin C Murphy; Samir Gupta; Deepak Agrawal; Ethan A Halm; Eric K Borton; Celette Sugg Skinner; Amit G Singal Journal: J Natl Compr Canc Netw Date: 2018-11 Impact factor: 11.908
Authors: Roxanne Wadia; Mark Shifman; Forrest L Levin; Luis Marenco; Cynthia A Brandt; Kei-Hoi Cheung; Tamar Taddei; Michael Krauthammer Journal: AMIA Jt Summits Transl Sci Proc Date: 2017-07-26
Authors: Amit G Singal; Samir Gupta; Jeffrey Lee; Ethan A Halm; Carolyn M Rutter; Douglas Corley; John Inadomi Journal: Clin Gastroenterol Hepatol Date: 2014-12 Impact factor: 11.382