. Author manuscript; available in PMC: 2017 Nov 3.

Published in final edited form as: Cancer. 2010 Jul 29;116(23):5420–5431. doi: 10.1002/cncr.25500

Table 5.

Fitted Bayesian logistic regression model for chronic GVHD (N=43; 2 patients were inevaluable due to early deaths).

Variable	Mean	SD	Posterior 95% Credible Interval		Probability of a Beneficial Effect
			2.50%	97.50%

Intercept	0.582	0.779	−0.887	2.111	-
Azacitidine dose	−0.0145	0.036	−0.083	0.057	0.658
Number of cycles	−0.439	0.311	−1.073	0.159	0.928