. "GB - Spojen\u00E9 kr\u00E1lovstv\u00ED Velk\u00E9 Brit\u00E1nie a Severn\u00EDho Irska" . . . . . "11320" . . "Nowadays, most clinical trials are conducted in different centers and even in different countries. In most multi-center studies, the primary analysis assumes that the treatment effect is constant over centers. However, it is also recommended to perform an exploratory analysis to highlight possible center by treatment interaction, especially when several countries are involved. We propose in this paper an exploratory Bayesian approach to quantify this interaction in the context of survival data. To this end we used and generalized a random effects accelerated failure time model. The generalization consists in using a penalized Gaussian mixture as an error distribution on top of multivariate random effects that are assumed to follow a normal distribution. For computational convenience, the computations are based on Markov chain Monte Carlo techniques. The proposed method is illustrated on the disease-free survival times of early breast cancer patients collected in the EORTC trial 10854."@en . . "1"^^ . "Baseline; treatment; effect; heterogeneity; survival; times; between; centers; using; random; effects; accelerated; failure; model; flexible; error; distribution"@en . "Statistics in Medicine" . "Baseline and treatment effect heterogeneity for survival times between centers using a random effects accelerated failure time model with flexible error distribution" . . "16"^^ . "3"^^ . . . "Baseline and treatment effect heterogeneity for survival times between centers using a random effects accelerated failure time model with flexible error distribution" . . "RIV/00216208:11320/07:00005649!RIV08-MSM-11320___" . . . . "411393" . "Z(MSM0021620839)" . "RIV/00216208:11320/07:00005649" . . . "26" . "[E0B4BC4D3934]" . . "5457;5472" . . "Nowadays, most clinical trials are conducted in different centers and even in different countries. In most multi-center studies, the primary analysis assumes that the treatment effect is constant over centers. However, it is also recommended to perform an exploratory analysis to highlight possible center by treatment interaction, especially when several countries are involved. We propose in this paper an exploratory Bayesian approach to quantify this interaction in the context of survival data. To this end we used and generalized a random effects accelerated failure time model. The generalization consists in using a penalized Gaussian mixture as an error distribution on top of multivariate random effects that are assumed to follow a normal distribution. For computational convenience, the computations are based on Markov chain Monte Carlo techniques. The proposed method is illustrated on the disease-free survival times of early breast cancer patients collected in the EORTC trial 10854." . . . . "30" . "Kom\u00E1rek, Arno\u0161t" . . "Baseline and treatment effect heterogeneity for survival times between centers using a random effects accelerated failure time model with flexible error distribution"@en . . . "Heterogenita v baselinu a efektu o\u0161et\u0159en\u00ED mezi centry pro \u010Dasy p\u0159e\u017Eit\u00ED p\u0159i pou\u017Eit\u00ED AFT modelu s n\u00E1hodn\u00FDmi efekty a flexibiln\u00EDm chybov\u00FDm rozd\u011Blen\u00EDm"@cs . . "Heterogenita v baselinu a efektu o\u0161et\u0159en\u00ED mezi centry pro \u010Dasy p\u0159e\u017Eit\u00ED p\u0159i pou\u017Eit\u00ED AFT modelu s n\u00E1hodn\u00FDmi efekty a flexibiln\u00EDm chybov\u00FDm rozd\u011Blen\u00EDm."@cs . . "Baseline and treatment effect heterogeneity for survival times between centers using a random effects accelerated failure time model with flexible error distribution"@en . . "Heterogenita v baselinu a efektu o\u0161et\u0159en\u00ED mezi centry pro \u010Dasy p\u0159e\u017Eit\u00ED p\u0159i pou\u017Eit\u00ED AFT modelu s n\u00E1hodn\u00FDmi efekty a flexibiln\u00EDm chybov\u00FDm rozd\u011Blen\u00EDm"@cs . . . "0277-6715" .