Application of Bayesian analyses to doubly randomized delayed start, matched control designs to demonstrate disease...

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Neuroscience

Background:

• Disease modification is a key goal for chronic progressive diseases, aiming to slow or reverse progression.
• Current analytical tools struggle to conclusively demonstrate disease modification.
• Traditional methods like slope comparisons have limitations in proving treatment efficacy.

Purpose of the Study:

• To propose a novel adaptation of the delayed start study design for establishing disease modification.
• To incorporate hierarchical Bayesian inference and posterior probabilities into study analysis.
• To compare the effectiveness of various statistical models for analyzing disease modification data.

Main Methods:

• A novel adaptation of the delayed start study design was developed.
• Hierarchical Bayesian inference and posterior probabilities were utilized.
• Simulations compared general linear models, repeated measures models, spline models, and model averaging.
• The approach was applied to the Disease Recovery Evaluation and Modification (DREaM) study for schizophrenia.

Main Results:

• Model averaging demonstrated superiority in accurately characterizing complex, real-world data.
• The proposed Bayesian methodology provides a robust framework for assessing disease modification.
• The approach was successfully applied to a real-world clinical trial design.

Conclusions:

• The novel Bayesian delayed start design offers a powerful tool for demonstrating disease modification.
• Model averaging is recommended for analyzing complex data in disease modification studies.
• Implementation in real-world trials like the DREaM study highlights practical considerations and the approach's value.