Skip to main content English


Explained variation in shared frailty and PH mixed models

Explained variation measures the relative gain in predictive accuracy when prediction based on prognostic factors replaces unconditional prediction. The factors may be measured on different scales or may be of different types (dichotomous, qualitative, or continuous). Thus, explained variation permits to establish a ranking of the importance of factors, even if predictive accuracy is too low to be helpful in clinical practice.

The provided software implements the extension of the explained variation measure by  Schemper and Henderson (2000) to accommodate random factors, such as center effects in multicenter studies. This permits a direct comparison of the importance of centers and of other prognostic factors.

While the current SAS version 9.4 restricts random effects in Cox models to random intercepts (i.e. the shared frailty model), the implementation in R also allows specification of more general proportional hazards mixed models (PHMM).


Gleiss, A, Gnant, M, Schemper, M (2018):
"Explained variation in shared frailty models", Statistics in Medicine. 2018; 37:1482 - 1490.



    Download: & pev_frailty_1.4 R