Analyzing Clinical Trials with Competing Risks
Author Information
Author(s): Tai Bee-Choo, Wee Joseph, Machin David
Primary Institution: National University of Singapore
Hypothesis
How can we effectively design and analyze clinical trials that involve competing risks endpoints?
Conclusion
The cause-specific hazard analysis is more efficient for analyzing competing risks outcomes when treatment does not affect the competing event's hazard.
Supporting Evidence
- The treatment effect on distant metastasis was significant with a hazard ratio of 0.43.
- The cause-specific hazard analysis required fewer subjects than the subdistribution hazard analysis.
- Adjusting for nodal status and tumor size did not materially alter the results.
Takeaway
This study helps doctors understand how to analyze clinical trials where patients might experience different types of failures, like cancer returning or spreading.
Methodology
The study used statistical models to analyze data from a clinical trial of patients with nasopharyngeal cancer, focusing on competing risks outcomes.
Potential Biases
Potential biases may arise from the assumptions made in the statistical models used.
Limitations
The study may not generalize to all types of cancer or clinical trials.
Participant Demographics
Patients with advanced (non-metastatic) nasopharyngeal cancer.
Statistical Information
P-Value
0.002
Confidence Interval
95% CI 0.25 - 0.72
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website