Alexander Crispin1, Carsten Klinger2, Anna Rieger3, Brigitte Strahwald3,4, Kai Lehmann5, Heinz-Johannes Buhr2, Ulrich Mansmann3. 1. IBE - Institute of Medical Information Processing, Biometry, and Epidemiology, Ludwig-Maximilians-Universität Munich, Marchioninistrasse 15, 81377, Munich, Germany. cri@ibe.med.uni-muenchen.de. 2. Deutsche Gesellschaft für Allgemein- und Viszeralchirurgie, Haus der Bundespressekonferenz, Schiffbauerdamm 40, 10117, Berlin, Germany. 3. IBE - Institute of Medical Information Processing, Biometry, and Epidemiology, Ludwig-Maximilians-Universität Munich, Marchioninistrasse 15, 81377, Munich, Germany. 4. Cognomedic GmbH, Medical Valley Center, Henkestrasse 91, 91052, Erlangen, Germany. 5. Charité-Universitätsmedizin Berlin, Hindenburgdamm 30, 12203, Berlin, Germany.
Abstract
PURPOSE: The purpose of this study is to provide a web-based calculator predicting complication probabilities of patients undergoing colorectal cancer (CRC) surgery in Germany. METHODS: Analyses were based on records of first-time CRC surgery between 2010 and February 2017, documented in the database of the Study, Documentation, and Quality Center (StuDoQ) of the Deutsche Gesellschaft für Allgemein- und Viszeralchirurgie (DGAV), a registry of CRC surgery in hospitals throughout Germany, covering demography, medical history, tumor features, comorbidity, behavioral risk factors, surgical procedures, and outcomes. Using logistic ridge regression, separate models were developed in learning samples of 6729 colon and 4381 rectum cancer patients and evaluated in validation samples of sizes 2407 and 1287. Discrimination was assessed using c statistics. Calibration was examined graphically by plotting observed versus predicted complication probabilities and numerically using Brier scores. RESULTS: We report validation results regarding 15 outcomes such as any major complication, surgical site infection, anastomotic leakage, bladder voiding disturbance after rectal surgery, abdominal wall dehiscence, various internistic complications, 30-day readmission, 30-day reoperation rate, and 30-day mortality. When applied to the validation samples, c statistics ranged between 0.60 for anastomosis leakage and 0.85 for mortality after rectum cancer surgery. Brier scores ranged from 0.003 to 0.127. CONCLUSIONS: While most models showed satisfactory discrimination and calibration, this does not preclude overly optimistic or pessimistic individual predictions. To avoid misinterpretation, one has to understand the basic principles of risk calculation and risk communication. An e-learning tool outlining the appropriate use of the risk calculator is provided.
PURPOSE: The purpose of this study is to provide a web-based calculator predicting complication probabilities of patients undergoing colorectal cancer (CRC) surgery in Germany. METHODS: Analyses were based on records of first-time CRC surgery between 2010 and February 2017, documented in the database of the Study, Documentation, and Quality Center (StuDoQ) of the Deutsche Gesellschaft für Allgemein- und Viszeralchirurgie (DGAV), a registry of CRC surgery in hospitals throughout Germany, covering demography, medical history, tumor features, comorbidity, behavioral risk factors, surgical procedures, and outcomes. Using logistic ridge regression, separate models were developed in learning samples of 6729 colon and 4381 rectum cancerpatients and evaluated in validation samples of sizes 2407 and 1287. Discrimination was assessed using c statistics. Calibration was examined graphically by plotting observed versus predicted complication probabilities and numerically using Brier scores. RESULTS: We report validation results regarding 15 outcomes such as any major complication, surgical site infection, anastomotic leakage, bladder voiding disturbance after rectal surgery, abdominal wall dehiscence, various internistic complications, 30-day readmission, 30-day reoperation rate, and 30-day mortality. When applied to the validation samples, c statistics ranged between 0.60 for anastomosis leakage and 0.85 for mortality after rectum cancer surgery. Brier scores ranged from 0.003 to 0.127. CONCLUSIONS: While most models showed satisfactory discrimination and calibration, this does not preclude overly optimistic or pessimistic individual predictions. To avoid misinterpretation, one has to understand the basic principles of risk calculation and risk communication. An e-learning tool outlining the appropriate use of the risk calculator is provided.
Entities:
Keywords:
Colorectal cancer; Model validation; Risk calculator; Statistical prediction models
Authors: Dawn Stacey; France Légaré; Krystina Lewis; Michael J Barry; Carol L Bennett; Karen B Eden; Margaret Holmes-Rovner; Hilary Llewellyn-Thomas; Anne Lyddiatt; Richard Thomson; Lyndal Trevena Journal: Cochrane Database Syst Rev Date: 2017-04-12
Authors: Elise H Lawson; Xue Wang; Mark E Cohen; Bruce Lee Hall; Howard Tanzman; Clifford Y Ko Journal: J Am Coll Surg Date: 2011-04-05 Impact factor: 6.113
Authors: Takayuki Anazawa; Jennifer L Paruch; Hiroaki Miyata; Mitsukazu Gotoh; Clifford Y Ko; Mark E Cohen; Norimichi Hirahara; Lynn Zhou; Hiroyuki Konno; Go Wakabayashi; Kenichi Sugihara; Masaki Mori Journal: Medicine (Baltimore) Date: 2015-12 Impact factor: 1.817
Authors: Delphina Gomes; Lien Le; Ulrich Mansmann; Regina Ensenauer; Sarah Perschbacher; Nikolaus A Haas; Heinrich Netz; Uwe Hasbargen; Maria Delius; Kristin Lange; Uta Nennstiel; Adelbert A Roscher Journal: BMC Med Date: 2022-04-14 Impact factor: 8.775