Confidence intervals for cost/effectiveness ratios

Abstract

The reduction of costs is becoming increasingly important in the medical field. The relevant topic of many clinical trials is not effectiveness per se, but rather cost‐effectiveness ratios. Surprisingly, no statistical tools for analyzing cost‐effectiveness ratios have been provided in the medical literature yet. This paper explains the gap in the literature, and provides a first technique for obtaining confidence intervals for cost‐effectiveness ratios. The technique does not use sophisticated tools to achieve maximal optimality, but seeks for tractability and ease of application while still satisfying all formal statistical requirements.