Yvonne Vergouwe1, Patrick Royston, Karel G M Moons, Douglas G Altman. 1. Julius Center for Health Sciences and Primary Care, University Medical Centre Utrecht, Str 6.131, P.O. Box 85500, 3508 GA, Utrecht, The Netherlands. y.vergouwe@umcutrecht.nl
Abstract
OBJECTIVE: To illustrate the sequence of steps needed to develop and validate a clinical prediction model, when missing predictor values have been multiply imputed. STUDY DESIGN AND SETTING: We used data from consecutive primary care patients suspected of deep venous thrombosis (DVT) to develop and validate a diagnostic model for the presence of DVT. Missing values were imputed 10 times with the MICE conditional imputation method. After the selection of predictors and transformations for continuous predictors according to three different methods, we estimated regression coefficients and performance measures. RESULTS: The three methods to select predictors and transformations of continuous predictors showed similar results. Rubin's rules could easily be applied to estimate regression coefficients and performance measures, once predictors and transformations were selected. CONCLUSION: We provide a practical approach for model development and validation with multiply imputed data. Copyright 2010 Elsevier Inc. All rights reserved.
OBJECTIVE: To illustrate the sequence of steps needed to develop and validate a clinical prediction model, when missing predictor values have been multiply imputed. STUDY DESIGN AND SETTING: We used data from consecutive primary care patients suspected of deep venous thrombosis (DVT) to develop and validate a diagnostic model for the presence of DVT. Missing values were imputed 10 times with the MICE conditional imputation method. After the selection of predictors and transformations for continuous predictors according to three different methods, we estimated regression coefficients and performance measures. RESULTS: The three methods to select predictors and transformations of continuous predictors showed similar results. Rubin's rules could easily be applied to estimate regression coefficients and performance measures, once predictors and transformations were selected. CONCLUSION: We provide a practical approach for model development and validation with multiply imputed data. Copyright 2010 Elsevier Inc. All rights reserved.
Authors: Paul K J Han; Minjung Lee; Bryce B Reeve; Angela B Mariotto; Zhuoqiao Wang; Ron D Hays; K Robin Yabroff; Marie Topor; Eric J Feuer Journal: J Pain Symptom Manage Date: 2011-11-08 Impact factor: 3.612
Authors: David J Casarett; Sue Farrington; Teresa Craig; Julie Slattery; Joan Harrold; Betty Oldanie; Jason Roy; Richard Biehl; Joan Teno Journal: J Palliat Med Date: 2012-05-14 Impact factor: 2.947
Authors: Jimme K Wiggers; Bas Groot Koerkamp; Robert J Coelen; Erik A Rauws; Mark A Schattner; C Yung Nio; Karen T Brown; Mithat Gonen; Susan van Dieren; Krijn P van Lienden; Peter J Allen; Marc G H Besselink; Olivier R C Busch; Michael I D'Angelica; Robert P DeMatteo; Dirk J Gouma; T Peter Kingham; William R Jarnagin; Thomas M van Gulik Journal: Endoscopy Date: 2015-09-18 Impact factor: 10.093
Authors: K Vuong; B K Armstrong; M Drummond; J L Hopper; J H Barrett; J R Davies; D T Bishop; J Newton-Bishop; J F Aitken; G G Giles; H Schmid; M A Jenkins; G J Mann; K McGeechan; A E Cust Journal: Br J Dermatol Date: 2019-09-22 Impact factor: 9.302
Authors: Ruud G Nijman; Yvonne Vergouwe; Henriëtte A Moll; Frank J Smit; Floor Weerkamp; Ewout W Steyerberg; Johan van der Lei; Yolanda B de Rijke; Rianne Oostenbrink Journal: Pediatr Res Date: 2017-11-08 Impact factor: 3.756