Werner Vach1. 1. Clinical Epidemiology, Institute of Medical Biometry and Medical Informatics, Freiburg University Medical Center, Freiburg, Germany. Electronic address: wv@imbi.uni-freiburg.de.
Abstract
OBJECTIVES: Calibration is often thought to assess the bias of a clinical prediction rule. In particular, if the rule is based on a linear logistic model, it is often assumed that an overestimation of all coefficients results in a calibration slope less than 1 and an underestimation in a slope larger than 1. STUDY DESIGN AND SETTING: We investigate the relation of the bias and the residual variation of clinical prediction rules with the typical behavior of calibration plots and calibration slopes, using some artificial examples. RESULTS: Calibration is not only sensitive to the bias of the clinical prediction rule but also to the residual variation. In some circumstances, the effects may cancel out, resulting in a misleading perfect calibration. CONCLUSION: Poor calibration is a clear indication of limited usefulness of a clinical prediction rule. However, a perfect calibration should be interpreted with care as this may happen even for a biased prediction rule.
OBJECTIVES: Calibration is often thought to assess the bias of a clinical prediction rule. In particular, if the rule is based on a linear logistic model, it is often assumed that an overestimation of all coefficients results in a calibration slope less than 1 and an underestimation in a slope larger than 1. STUDY DESIGN AND SETTING: We investigate the relation of the bias and the residual variation of clinical prediction rules with the typical behavior of calibration plots and calibration slopes, using some artificial examples. RESULTS: Calibration is not only sensitive to the bias of the clinical prediction rule but also to the residual variation. In some circumstances, the effects may cancel out, resulting in a misleading perfect calibration. CONCLUSION: Poor calibration is a clear indication of limited usefulness of a clinical prediction rule. However, a perfect calibration should be interpreted with care as this may happen even for a biased prediction rule.
Authors: Kym I E Snell; Harry Hua; Thomas P A Debray; Joie Ensor; Maxime P Look; Karel G M Moons; Richard D Riley Journal: J Clin Epidemiol Date: 2015-05-16 Impact factor: 6.437
Authors: Ben Van Calster; Kirsten Van Hoorde; Yvonne Vergouwe; Shabnam Bobdiwala; George Condous; Emma Kirk; Tom Bourne; Ewout W Steyerberg Journal: Diagn Progn Res Date: 2017-02-08
Authors: M J McDonnell; S Aliberti; P C Goeminne; K Dimakou; S C Zucchetti; J Davidson; C Ward; J G Laffey; S Finch; A Pesci; L J Dupont; T C Fardon; D Skrbic; D Obradovic; S Cowman; M R Loebinger; R M Rutherford; A De Soyza; J D Chalmers Journal: Thorax Date: 2016-08-11 Impact factor: 9.139