Measuring Quality of Colonoscopy Care Using Natural Language Processing

Colonoscopy is the predominant method for screening for colorectal cancer in the United States. Yet, its effectiveness is limited by its variation in performance. For example, the rate at which physicians detect cancer precursors (called adenomas) during a colonoscopy has been shown to vary three-fold from one physician to another. For example, a patient whose colonoscopy is performed by a physician with a low adenoma detection rate has a higher risk of subsequent colorectal cancer.

Our work has centered on measuring colonoscopy quality and understanding what drives the variation in performance. The major innovation of this work is to use natural language processing (NLP) to measure quality. NLP is a field of computer science in which a computer is trained to “read” text to identify relevant data. We developed and validated one of the first NLP-based computer software applications that analyzes colonoscopy and associated pathology reports. We have applied this NLP program to all the colonoscopy and pathology reports from a two year period at four clinical sites around the U.S.

In this research, we addressed a number of topics:

  • Is public reporting of colonoscopy quality associated with quality improvement?
  • How should one measure colonoscopy quality?
  • What are the challenges in applying NLP programs to several clinical sites?
  • What physician characteristics are associated with higher quality?
  • How much does the pathologist drive variation in colonoscopy quality?
  • Ertem FU, Ladabaum U, Mehrotra A, Tehranian S, Shi Z, Saul M, Morris M, Crockett SD, Schoen RE. Incidence of interval colorectal cancer attributable to an endoscopist in clinical practice. Gastrointest Endosc. 2018;88(4):705-711.e1. PMCID: PMC6139042.
  • Ezaz G, Leffler DA, Beach S, Schoen RE, Crockett SD, Gourevitch RA, Rose S, Morris M, Carrell DS, Greer JB, Mehrotra A. Association between endoscopist personality and rate of adenoma detection. Clin Gastroenterol Hepatol. 2018; 2018 Oct 13. pii: S1542-3565(18)31140-6.
  • Gourevitch RA, Rose S, Crockett SD, Morris M, Carrell DS, Greer JB, Pai RK, Schoen RE, Mehrotra A. Variation in Pathologist Classification of Colorectal Adenomas and Serrated Polyps. Am J Gastroenterol. 2018;113(3):431-439. PMCID: PMC6049074.
  • Marcondes FO, Gourevitch RA, Schoen RE, Crockett SD, Morris M, Mehrotra A. Adenoma Detection Rate Falls at the End of the Day in a Large Multi-site Sample. Dig Dis Sci. 2018;63(4):856-859.
  • Crockett SD, Gourevitch RA, Morris M, Carrell DS, Rose S, Shi Z, Greer JB, Schoen RE, Mehrotra A. Endoscopist factors that influence serrated polyp detection: a multicenter study. Endoscopy. 2018;50(10):984-992. PMCID: PMC6160341.
  • Mehrotra A, Morris M, Gourevitch RA, Carrell DS, Leffler DA, Rose S, Greer JB, Crockett SD, Baer A, Schoen RE. Physician characteristics associated with higher adenoma detection rate. Gastrointest Endosc. 2018;87(3):778-786.e5. PMCID: PMC5817032.
  • Abdul-Baki H, Schoen R, Dean K, Rose S, Leffler A, Kuganeswaran E, Morris M, Carrell D, Mehrotra A. Public Reporting of Colonoscopy Quality Is Associated with an Increase in Endoscopist Adenoma Detection Rate, Gastrointestinal Endoscopy. 2015 Oct;82(4):676-82. doi: 10.1016/j.gie.2014.12.058. PubMed PMID: 26385276; PubMed Central PMCID: PMC4575767.
  • Marcondes F, Dean KM, Schoen R, Leffler D, Rose S, Morris M, Mehrotra A. The Impact of Exclusion Criteria on a Physician’s Adenoma Detection Rate, Gastrointestinal Endoscopy. 2015 Oct;82(4):668-75. doi: 10.1016/j.gie.2014.12.056. PubMed PMID: 26385275; PubMed Central PMCID: PMC4575765.
  • Mehrotra A, Dellon ES, Schoen RE, Saul M, Bisheshari F, Farmer C, Harkema H. Applying a natural language processing tool to electronic health records to assess performance on colonoscopy quality measures. Gastrointestinal Endoscopy, 2012;76(6):1233-1239. PMC Journal – In Process
  • Harkema, H, Chapman WW, Saul M, Dellon ES, Schoen RE, Mehrotra A. Developing a natural language processing application for measuring the quality of colonoscopy procedures. Journal of the American Medical Informatics Association: JAIMA. 2011;18 Suppl 1:i150-i156.