Related Experiment Videos
Applying Bayesian ideas in drug development and clinical trials
D J Spiegelhalter1, L S Freedman, M K Parmar
1MRC Biostatistics Unit, Institute of Public Health, Cambridge, U.K.
Statistics in Medicine
|August 1, 1993
Summary
This study highlights using external evidence in clinical trial design and monitoring. It proposes an interpretation section in reports to bridge results and discussion with a Bayesian perspective.
Area of Science:
- Clinical trial methodology
- Biostatistics
- Evidence-based medicine
Background:
- Clinical studies are often isolated, neglecting valuable external evidence.
- Bayesian methods offer a formal framework to integrate prior knowledge into trial processes.
- Ethical and efficiency considerations in clinical trials necessitate robust evidence assessment.
Purpose of the Study:
- To present tools for assessing existing evidence in clinical trials.
- To illustrate the application of Bayesian approaches in trial design, monitoring, and reporting.
- To advocate for an 'interpretation' section in clinical reports.
Main Methods:
- Utilizing graphical displays to visualize evidence for treatment efficacy.
- Applying Bayesian statistical principles to trial data.
- Case study illustration with an ongoing osteosarcoma trial.
Main Results:
- Demonstrated methods for assessing cumulative evidence during a trial.
- Provided insights into ethical and efficiency trade-offs using graphical tools.
- Highlighted the utility of Bayesian analysis in informing trial decisions.
Conclusions:
- External evidence can be formally incorporated into clinical trial design, monitoring, and reporting.
- Graphical tools aid in understanding ethical and efficiency aspects of trial conduct.
- An 'interpretation' section in clinical reports, incorporating a Bayesian perspective, is recommended.