Michael Bronsert1, Abhinav B Singh2, William G Henderson3, Karl Hammermeister4, Robert A Meguid5, Kathryn L Colborn6. 1. University of Colorado Anschutz Medical Campus, Adult and Child Consortium for Health Outcomes Research and Delivery Science, Aurora, CO, USA; Surgical Outcomes and Applied Research Program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO, USA. Electronic address: Michael.Bronsert@CUAnschutz.edu. 2. Surgical Outcomes and Applied Research Program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO, USA. Electronic address: Abhinav.Singh@CUAnschutz.edu. 3. University of Colorado Anschutz Medical Campus, Adult and Child Consortium for Health Outcomes Research and Delivery Science, Aurora, CO, USA; Surgical Outcomes and Applied Research Program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO, USA; University of Colorado Anschutz Medical Campus, Colorado School of Public Health, Department of Biostatistics and Informatics, Aurora, CO, USA. Electronic address: William.Henderson@CUAnschutz.edu. 4. University of Colorado Anschutz Medical Campus, Adult and Child Consortium for Health Outcomes Research and Delivery Science, Aurora, CO, USA; Surgical Outcomes and Applied Research Program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO, USA; University of Colorado Anschutz Medical Campus, School of Medicine, Department of Cardiology, Aurora, CO, USA. Electronic address: Karl.Hammermeister@CUAnschutz.edu. 5. University of Colorado Anschutz Medical Campus, Adult and Child Consortium for Health Outcomes Research and Delivery Science, Aurora, CO, USA; Surgical Outcomes and Applied Research Program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO, USA. Electronic address: Robert.Meguid@CUAnschutz.edu. 6. University of Colorado Anschutz Medical Campus, Colorado School of Public Health, Department of Biostatistics and Informatics, Aurora, CO, USA. Electronic address: kathryn.colborn@CUAnschutz.edu.
Abstract
BACKGROUND: Using the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) complication status of patients who underwent an operation at the University of Colorado Hospital, we developed a machine learning algorithm for identifying patients with one or more complications using data from the electronic health record (EHR). METHODS: We used an elastic-net model to estimate regression coefficients and carry out variable selection. International classification of disease codes (ICD-9), common procedural terminology (CPT) codes, medications, and CPT-specific complication event rate were included as predictors. RESULTS: Of 6840 patients, 922 (13.5%) had at least one of the 18 complications tracked by NSQIP. The model achieved 88% specificity, 83% sensitivity, 97% negative predictive value, 52% positive predictive value, and an area under the curve of 0.93. CONCLUSIONS: Using machine learning on EHR postoperative data linked to NSQIP outcomes data, a model with 163 predictors from the EHR identified complications well at our institution.
BACKGROUND: Using the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) complication status of patients who underwent an operation at the University of Colorado Hospital, we developed a machine learning algorithm for identifying patients with one or more complications using data from the electronic health record (EHR). METHODS: We used an elastic-net model to estimate regression coefficients and carry out variable selection. International classification of disease codes (ICD-9), common procedural terminology (CPT) codes, medications, and CPT-specific complication event rate were included as predictors. RESULTS: Of 6840 patients, 922 (13.5%) had at least one of the 18 complications tracked by NSQIP. The model achieved 88% specificity, 83% sensitivity, 97% negative predictive value, 52% positive predictive value, and an area under the curve of 0.93. CONCLUSIONS: Using machine learning on EHR postoperative data linked to NSQIP outcomes data, a model with 163 predictors from the EHR identified complications well at our institution.
Authors: Adi V Gundlapalli; Guy Divita; Andrew Redd; Marjorie E Carter; Danette Ko; Michael Rubin; Matthew Samore; Judith Strymish; Sarah Krein; Kalpana Gupta; Anne Sales; Barbara W Trautner Journal: J Biomed Inform Date: 2016-07-09 Impact factor: 6.317
Authors: Grant B Weller; Jenna Lovely; David W Larson; Berton A Earnshaw; Marianne Huebner Journal: Stat Methods Med Res Date: 2017-03-01 Impact factor: 3.021
Authors: Julie A Choudhuri; Ronald F Pergamit; Jeannie D Chan; Astrid B Schreuder; Elizabeth McNamara; John B Lynch; Timothy H Dellit Journal: Infect Control Hosp Epidemiol Date: 2011-08 Impact factor: 3.254
Authors: Kathryn L Colborn; Michael Bronsert; Karl Hammermeister; William G Henderson; Abhinav B Singh; Robert A Meguid Journal: Am J Infect Control Date: 2018-12-04 Impact factor: 2.918
Authors: Mila H Ju; Clifford Y Ko; Bruce L Hall; Charles L Bosk; Karl Y Bilimoria; Elizabeth C Wick Journal: JAMA Surg Date: 2015-01 Impact factor: 14.766