Related Experiment Videos
On the probability of cost-effectiveness using data from randomized clinical trials
1Department of Clinical Epidemiology and Biostatistics, HSC-2C, McMaster University, 1200 Main Street West, Hamilton, ON, L8N 3Z5, Canada. willana@mcmaster.ca
BMC Medical Research Methodology
|November 1, 2001
Summary
A new method quantifies the probability of a treatment being cost-effective, independent of sample size. This approach offers a superior alternative to acceptability curves for health economic evaluations.
Area of Science:
- Health Economics
- Clinical Trial Analysis
- Statistical Modeling
Background:
- Acceptability curves are used to estimate the probability of a treatment's cost-effectiveness.
- Existing methods often depend on the amount of evidence, specifically sample size.
- These methods express certainty in a treatment's cost-effectiveness based on available data.
Purpose of the Study:
- To propose an alternative method for quantifying treatment cost-effectiveness probability.
- To develop a measure that is independent of sample size.
- To provide a more interpretable metric for cost-effectiveness.
Main Methods:
- Introduction of a novel parameter for cost-effectiveness probability.
- Utilization of non-parametric statistical methods for estimation.
- Development of point and interval estimation techniques.
Main Results:
- The proposed parameter is independent of sample size.
- Non-parametric methods yield estimators and confidence intervals for the incremental cost-effectiveness ratio.
- An illustrative example is presented.
Conclusions:
- The new parameter is superior to acceptability curves due to its sample size independence.
- The parameter can be interpreted as the proportion of patients benefiting from the new therapy.
- Non-parametric methods facilitate estimation, variance calculation, confidence intervals, and hypothesis testing.