Design of Bayesian Clinical Trials with Clustered Data and Multiple Endpoints

2025-01-22

AI Summary

Conventional Monte Carlo simulation for evaluating operating characteristics (e.g., decision accuracy) and determining sample size in Bayesian clinical trials with clustered data and multiple endpoints is computationally expensive and inefficient. Method: We derive, for the first time, an analytical functional relationship between posterior probability and sample size within a Bayesian hierarchical framework that accommodates clustering and multiple endpoints. This enables full operating characteristic curve extrapolation from only two Monte Carlo simulations. We further quantify how simulation variability affects recommended sample sizes. Contribution/Results: By integrating Bayesian hierarchical modeling, cluster-aware inference, and theoretical derivation, our approach drastically reduces computational burden. It is validated on real-world cluster-randomized, adaptive, multi-endpoint Bayesian trials, demonstrating robustness and enabling rapid, reliable sample size determination without sacrificing statistical rigor.

Technology Category

• Machine Learning: Bayesian Learning
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Probabilistic Programming

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

In the design of clinical trials, it is essential to assess the design operating characteristics (i.e., the probabilities of making correct decisions). Common practice for the evaluation of operating characteristics in Bayesian clinical trials relies on estimating the sampling distribution of posterior summaries via Monte Carlo simulation. It is computationally intensive to repeat this estimation process for each design configuration considered, particularly for clustered data that are analyzed using complex, high-dimensional models. In this paper, we propose an efficient method to assess operating characteristics and determine sample sizes for Bayesian trials with clustered data and multiple endpoints. We prove theoretical results that enable posterior probabilities to be modeled as a function of the sample size. Using these functions, we assess operating characteristics at a range of sample sizes given simulations conducted at only two sample sizes. These theoretical results are also leveraged to quantify the impact of simulation variability on our sample size recommendations. The applicability of our methodology is illustrated using a current cluster-randomized Bayesian adaptive clinical trial with multiple endpoints.

Problem

• Efficiently assess Bayesian trial operating characteristics for clustered data
• Determine optimal sample sizes with multiple endpoints in clinical trials
• Reduce computational burden in evaluating complex high-dimensional models

Innovation

• Efficient Bayesian trial design for clustered data
• Model posterior probabilities as sample size functions
• Reduce simulations via two-sample-size approximation