Cost-Effective Clinical Trial Design: Application of a Bayesian Sequential Stopping Rule to the ProFHER Pragmatic Trial

We investigate value-based clinical trial design by applying a Bayesian decisiontheoretic model of a sequential experiment to data from the ProFHER pragmatic trial. In the first applied analysis of its kind to use research cost data, we show that the model’s stopping policy would have stopped the trial early, saving about 5% of the research budget (approximately £73,000). A bootstrap analysis based on generating resampled paths from the trial data suggests that the trial’s expected sample size could have been reduced by approximately 40%, saving an expected 15% of the budget, with 93% of resampled paths making a decision consistent with the result of the trial itself. Results show how substantial benefits to trial cost stewardship may be achieved by accounting for research costs in defining the trial’s stopping policy and active monitoring of trial data as it accumulates