Constructing confidence intervals for cost-effectiveness ratios: an evaluation of parametric and non-parametric techniques using Monte Carlo simulation
✍ Scribed by Andrew H. Briggs; Christopher Z. Mooney; David E. Wonderling
- Publisher
- John Wiley and Sons
- Year
- 1999
- Tongue
- English
- Weight
- 176 KB
- Volume
- 18
- Category
- Article
- ISSN
- 0277-6715
No coin nor oath required. For personal study only.
✦ Synopsis
The statistic of interest in most health economic evaluations is the incremental cost-e!ectiveness ratio. Since the variance of a ratio estimator is intractable, the health economics literature has suggested a number of alternative approaches to estimating con"dence intervals for the cost-e!ectiveness ratio. In this paper, Monte Carlo simulation techniques are employed to address the question of which of the proposed methods is most appropriate. By repeatedly sampling from a known distribution and applying the di!erent methods of con"dence interval estimation, it is possible to calculate the coverage properties of each method to see if these correspond to the chosen con"dence level. As the results of a single Monte Carlo experiment would be valid only for that particular set of circumstances, a series of experiments was conducted in order to examine the performance of the di!erent methods under a variety of conditions relating to the sample size, the coe$cient of variation of the numerator and denominator of the ratio, and the covariance between costs and e!ects in the underlying data. Response surface analysis was used to analyse the results and substantial di!erences between the di!erent methods of con"dence interval estimation were identi"ed. The methods, both parametric and non-parametric, which assume a normal sampling distribution performed poorly, as did the approach based on simply combining the separate intervals on costs and e!ects. The choice of method for con"dence interval estimation can lead to large di!erences in the estimated con"dence limits for cost-e!ectiveness ratios. The importance of such di!erences is an empirical question and will depend to a large extent on the role of hypothesis testing in economic appraisal. However, where it is suspected that the sampling distribution is skewed, normal approximation methods produce particularly poor results and should be avoided.