Table 1 Summary of main features of the traditional text-book Bayesian approach and the predictive Bayesian approach

Theoretical perspective

Application oriented perspective

Theoretical perspective

Application oriented perspective

Focus on fictional parameters

Average patient performance

Focus on observable quantities

Performance of the individual or group considered

Prior and posterior distributions

The experts' uncertainties about the unknown parameters

Prior and posterior distributions

The experts' uncertainties about the observable quantities

There exists underlying true probabilities

Probabilities and risks are unknown and need to be estimated

Probability is a subjective measure of uncertainty, conditional on the assessor's background information

Probability is assigned by the assessor

Model uncertainty exists

Model uncertainty needs to be covered by the analysis

Model uncertainty does not exist.

The accuracy of the models needs to be addressed