. Author manuscript; available in PMC: 2017 Oct 15.

Published in final edited form as: Stat Med. 2016 Jun 1;35(23):4136–4152. doi: 10.1002/sim.6997

Table 1. Use of mbc, c-mbc and commonly used concordance measures.

For proportional hazards regression models and logistic regression models, it is specified: how to measure concordance in an apparent validation setting (in patients whose data was used for model development); how to assess the influence of case-mix heterogeneity on concordance (Concordance assuming correct regression coefficients) in an external validation setting (in new patients); and how to measure concordance in an apparent validation setting.

	Apparent validation	External validation
	Concordance	Concordance assuming correct regression coefficients	Concordance
Proportional hazards regression models	c-index	case-mix corrected c-index	c-index
	Uno		Uno
	Gönen-Heller	Gönen-Heller
	mbc	mbc	c-mbc
Logistic regression models	c-index	case-mix corrected c-index	c-index
Logistic regression models	mbc	mbc	c-mbc