February 5, 1997

Reporting Cost-effectiveness Analyses With Confidence-Reply

Author Affiliations

Harvard University Boston, Mass
University of Wisconsin Madison
University of Minnesota Minneapolis
McMaster University Hamilton, Ontario
Rutgers University New Brunswick, NJ
Office of Disease Prevention and Health Promotion US Public Health Service Washington, DC for the Panel on Cost-Effectiveness in Health and Medicine

JAMA. 1997;277(5):375. doi:10.1001/jama.1997.03540290027021

In Reply.  —The Panel on Cost-Effectiveness in Health and Medicine has recommended that summary tables reporting incremental cost-effectiveness exclude ratios for dominated alternatives.1,2 Drs Sacristán and Obenchain correctly point out that, even if a medical intervention is judged on the basis of baseline estimates to be dominated by another intervention (eg, to have higher cost but produce less health benefit than the other), the conclusion of dominance may be premature owing to uncertainties in baseline parameters. Sacristán and Obenchain would recommend use of CIs for reported cost-effectiveness ratios on all alternatives.Our specific recommendation that cost-effectiveness ratios should not appear for dominated alternatives is intended to reduce confusion concerning the interpretation of cost-effectiveness results tables. The calculation of base-case results for the efficient options (those not dominated under the main assumptions in the analysis) is less transparent if ratios for dominated programs are included in cost-effectiveness tables.Sacristán and Obenchain's main