Abstract
Improving the efficiency of clinical trials is seen as a priority by funding bodies such as the United Kingdom’s National Institute for Health Research (NIHR), and adaptive clinical trials are seen as one way to achieve this. However, recent research by Flight et al. (2019) found that it is not common to incorporate the cost-effectiveness of the research process into their design. This talk considers whether recent innovations in Bayesian decision-theoretic methods for sequential experimentation (Pertile et al. (2014), Chick et al. (2017)) can improve the efficiency of sequential clinical trials. The methods provide a rule for stopping a two-armed clinical trial at interim analyses, informed by balancing the costs and benefits of continuing the trial with those of stopping. To attempt to answer the question in the talk's title, I present applications of the model to two retrospective case studies from the United Kingdom.
E. Chick, M. Forster, P. Pertile (2017). A Bayesian decision-theoretic model of sequential experimentation with delayed response. JRSSB, 2017; 79(5):1439–1462.
Flight, F. Arshad, R. Barnsley, K. Patel, S. Julious, A. Brennan, S. Todd (2019). A review of clinical trials with an adaptive design and health economic analysis. Value in Health. 2019;22(4):391—398.
Pertile, M. Forster, D. La Torre (2014). Optimal Bayesian sequential sampling rules for the economic evaluation of health technologies. JRSSA, 2014;177(2):419–438.
Organizzatore
Christian Hennig