Bayesian design and analysis of two-arm cluster randomised trials using assurance: Extension to binary outcomes and comparison of Markov chain Monte Carlo and Integrated Nested Laplace Approximations.

Puntos clave

• The aim is to extend fully Bayesian designs for two-arm cluster randomised trials specifically to binary outcomes and compare different analytical methods.
• Development of a fully Bayesian design framework for cluster randomised trials.
• Utilization of Integrated Nested Laplace Approximations for improved efficiency in binary outcome assessment.
• Comparison of Markov Chain Monte Carlo methods with Integrated Nested Laplace Approximations for analysis.
• Integrated Nested Laplace Approximations yield efficient assurance assessments for binary outcomes.
• The Bayesian approach maintains accuracy in inference for large cluster trials.
• Sample size determination effectively incorporates uncertainty in model parameters.