Statistical determination of cost-effectiveness frontier based on net health benefits

• Published in: Health economics Vol. 11; no. 3; pp. 249 - 264
• Main Authors: Laska, Eugene M., Meisner, Morris, Siegel, Carole, Wanderling, Joseph
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.04.2002; Wiley Periodicals Inc
• Series: Health Economics
• Subjects: Algorithms; Antipsychotic Agents - economics; Antipsychotic Agents - therapeutic use; Confidence Intervals; Cost analysis; Cost benefit analysis; Cost effectiveness; Cost-Benefit Analysis - statistics & numerical data; cost-effectiveness analysis; Decision Making; familywise error rate; Health Care Costs - statistics & numerical data; Health care expenditures; Health economics; Health services; Humans; Models, Statistical; net health benefit; pointwise error rate; Program Evaluation - economics; Program Evaluation - methods; Randomized Controlled Trials as Topic - statistics & numerical data; Schizophrenia - drug therapy; Schizophrenia - economics; Statistical analysis; statistical frontier; Statistical methods; Statistics; Stochastic Processes; Treatment Outcome
• ISSN: 1057-9230; 1099-1050
• DOI: 10.1002/hec.659

Statistical methods are given for producing a cost‐effectiveness frontier for an arbitrary number of programs. In the deterministic case, the net health benefit (NHB) decision rule is optimal; the rule funds the program with the largest positive NHB at each λ, the amount a decision‐maker is willing to pay for an additional unit of effectiveness. For bivariate normally distributed cost and effectiveness variables and a specified λ, a statistical procedure is presented, based on the method of constrained multiple comparisons with the best (CMCB), for determining the program with the largest NHB. A one‐tailed t test is used to determine if the NHB is positive. To obtain a statistical frontier in the λ‐NHB plane, we develop a method to produce the region in which each program has the largest NHB, by pivoting a CMCB confidence interval. A one‐sided version of Fieller's theorem is used to determine the region where the NHB of each program is positive. At each λ, the pointwise error rate is bounded by a prespecified α. Upper bounds on the familywise error rate, the probability of an error at any value of λ, are given. The methods are applied to a hypothetical clinical trial of antipsychotic agents. Copyright © 2002 John Wiley & Sons, Ltd.