Predicting Hospitalization and Outpatient Corticosteroid Use in Inflammatory Bowel Disease Patients Using Machine Learning.

Background Inflammatory bowel disease (IBD) is a chronic disease characterized by unpredictable episodes of flares and periods of remission. Tools that accurately predict disease course would substantially aid therapeutic decision-making. This study aims to construct a model that accurately predicts the combined end point of outpatient corticosteroid use and hospitalizations as a surrogate for IBD flare. Methods Predictors evaluated included age, sex, race, use of corticosteroid-sparing immunosuppressive medications (immunomodulators and/or anti-TNF), longitudinal laboratory data, and number of previous IBD-related hospitalizations and outpatient corticosteroid prescriptions. We constructed models using logistic regression and machine learning methods (random forest [RF]) to predict the combined end point of hospitalization and/or corticosteroid use for IBD within 6 months. Results We identified 20,368 Veterans Health Administration patients with the first (index) IBD diagnosis between 2002 and 2009. Area under the receiver operating characteristic curve (AuROC) for the baseline logistic regression model was 0.68 (95% confidence interval [CI], 0.67-0.68). AuROC for the RF longitudinal model was 0.85 (95% CI, 0.84-0.85). AuROC for the RF longitudinal model using previous hospitalization or steroid use was 0.87 (95% CI, 0.87-0.88). The 5 leading independent risk factors for future hospitalization or steroid use were age, mean serum albumin, immunosuppressive medication use, and mean and highest platelet counts. Previous hospitalization and corticosteroid use were highly predictive when included in specified models. Conclusions A novel machine learning model substantially improved our ability to predict IBD-related hospitalization and outpatient steroid use. This model could be used at point of care to distinguish patients at high and low risk for disease flare, allowing individualized therapeutic management.

[1]  A. Schoepfer,et al.  Accuracy of Four Fecal Assays in the Diagnosis of Colitis , 2007, Diseases of the colon and rectum.

[2]  Clement J. McDonald,et al.  Development of the Logical Observation Identifier Names and Codes (LOINC) vocabulary. , 1998, Journal of the American Medical Informatics Association : JAMIA.

[3]  A. Ananthakrishnan,et al.  Permanent Work Disability in Crohn's Disease , 2008, The American Journal of Gastroenterology.

[4]  William J. Tremaine,et al.  Update on the incidence and prevalence of Crohn's disease and ulcerative colitis in Olmsted County, Minnesota, 1940–2000 , 2007, Inflammatory bowel diseases.

[5]  S. Saini,et al.  Corticosteroid Use and Complications in a US Inflammatory Bowel Disease Cohort , 2016, PloS one.

[6]  Ken Kleinman,et al.  The prevalence and geographic distribution of Crohn's disease and ulcerative colitis in the United States. , 2007, Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association.

[7]  S. Saini,et al.  Cost utility of inflammation-targeted therapy for patients with ulcerative colitis. , 2012, Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association.

[8]  E. Loftus.  The burden of inflammatory bowel disease in the United States: a moving target? , 2007, Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association.

[9]  F. Casellas,et al.  Impairment of Health‐Related Quality of Life in Patients With Inflammatory Bowel Disease: a Spanish Multicenter Study , 2005, Inflammatory bowel diseases.

[10]  C. McDonald,et al.  LOINC, a universal standard for identifying laboratory observations: a 5-year update. , 2003, Clinical chemistry.

[11]  C. Porter,et al.  Direct health care costs of Crohn's disease and ulcerative colitis in US children and adults. , 2008, Gastroenterology.

[12]  R. Modigliani,et al.  A simple biological score for predicting low risk of short‐term relapse in Crohn's disease , 2006, Inflammatory bowel diseases.

[13]  S M Huff,et al.  A Characterization of Local LOINC Mapping for Laboratory Tests in Three Large Institutions , 2010, Methods of Information in Medicine.

[14]  Chuan-Fen Liu,et al.  Use of outpatient care in Veterans Health Administration and Medicare among veterans receiving primary care in community-based and hospital outpatient clinics. , 2010, Health Services Research.

[15]  Hyeon-Eui Kim,et al.  An approach to improve LOINC mapping through augmentation of local test names , 2012, J. Biomed. Informatics.

[16]  E. Loftus.  Clinical epidemiology of inflammatory bowel disease: Incidence, prevalence, and environmental influences. , 2004, Gastroenterology.

[17]  H. El‐Serag,et al.  Accuracy of Diagnostic Codes for Identifying Patients with Ulcerative Colitis and Crohn’s Disease in the Veterans Affairs Health Care System , 2014, Digestive Diseases and Sciences.

[18]  Ren Mao,et al.  Fecal calprotectin in predicting relapse of inflammatory bowel diseases: A meta‐analysis of prospective studies , 2012, Inflammatory bowel diseases.

[19]  Catherine Moore,et al.  Case Report: Standardizing Laboratory Data by Mapping to LOINC , 2006, J. Am. Medical Informatics Assoc..

[20]  Sijian Wang,et al.  Algorithms outperform metabolite tests in predicting response of patients with inflammatory bowel disease to thiopurines. , 2010, Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association.

[21]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[22]  P. Higgins,et al.  Machine Learning Algorithms for Objective Remission and Clinical Outcomes with Thiopurines , 2017, Journal of Crohn's & colitis.

[23]  C. McDonald,et al.  Logical observation identifier names and codes (LOINC) database: a public use set of codes and names for electronic reporting of clinical laboratory test results. , 1996, Clinical chemistry.

[24]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[25]  J. Mate,et al.  Fecal calprotectin and lactoferrin for the prediction of inflammatory bowel disease relapse , 2009, Inflammatory bowel diseases.

[26]  A. Franco,et al.  Diagnostic value of fecal leukocytes in chronic bowel diseases. , 1994, Sao Paulo medical journal = Revista paulista de medicina.

[27]  W. Sandborn,et al.  Predicting relapse in patients with inflammatory bowel disease: what is the role of biomarkers? , 2005, Gut.

[28]  E. Seidman,et al.  Clinical Utility of Fecal Biomarkers for the Diagnosis and Management of Inflammatory Bowel Disease , 2014, Inflammatory bowel diseases.