EVALUATING AN ONLINE FREEWARE CALCULATOR AND PLOTTER FOR POWER ANALYSIS AND SAMPLE SIZE ESTIMATION FOR COST EFFECTIVENESS STUDIES
Author(s)
William F McGhan, PharmD, PhD, Professor of Pharmacy & Health Policy, Vincent J Willey, PharmD, Associate Professor of Pharmacy, Andrew M. Peterson, PharmD, Chair, Department of Pharmacy Practice and Pharmacy AdministrationUniversity of the Sciences, Philadelphia, PA, USA
OBJECTIVES: To evaluate an online, freeware calculator that generates and plots sample size estimations and power analyses for cost effectiveness studies. METHODS: Online software was developed and results were compared with a published analytical formula for power analysis and sample size calculations for cost and effectiveness data. The web-based, cost effectiveness calculator formulas, data and evaluation were based on published articles by Briggs A, Gray AM and Tambour M. The online calculator required data inputs include: probability of Type I and Type II error, standard deviation of costs and effects, mean cost and effect differences, correlation between differences in cost and effects, as well as willingness to pay (WTP) for additional health effects. This Internet tool outputs results for sample size in each study arm that would be required versus WTP threshold ranges, and power versus sample size. RESULTS: Compared to the published manuscripts for a power of 0.90 and effectiveness only, the online calculated sample size results were identical (N=536). For the published examples with correlation differences in effect and cost of minus 1.0, the estimated sample sizes based on WTP compared as follows: WTP=$7500:(1400 vs 1387), WTP=$10,000:(1150 vs 1096), WTP=$15,000:(850 vs 865), WTP=$20,000:( 790 vs 769), WTP=$30,000:(700 vs 682). The Briggs et al articles include additional results and sensitivity analyses based on additional correlations and power, which have to be run one at a time with the online software. CONCLUSIONS: With this online freeware calculator, the user can enter their own data to estimate sample size and power in planned or published cost effectiveness studies. This web-based software has potential benefit as a basic tool for students, health professionals, and decision makers.
Conference/Value in Health Info
2009-05, ISPOR 2009, Orlando, FL, USA
Value in Health, Vol. 12, No. 3 (May 2009)
Code
PMC16
Topic
Economic Evaluation
Topic Subcategory
Cost/Cost of Illness/Resource Use Studies
Disease
Multiple Diseases