Bayesian decision procedures for binary and continuous bivariate dose-escalation studies