ID: 7303
Meeting / Value in Health Info: ISPOR Second Asia-Pacific Conference
Shanghai, China
March, 2006
Code: PMC8
Disease: Multiple Diseases/No Specific Disease
Topic: Conceptual Papers (CP)
Topic Subcategory: Study & Methods Design-General (SD)
Title: AFFORDABLE BOOTSTRAPS? EVALUATING FREEWARE OPTIONS FOR ANALYZING INCREMENTAL COST EFFECTIVENESS DATA
Author(s):

William McGhan, PharmD, PhD, Professor of Pharmacy & Health Policy, Pranav Gandhi, BS, Research Assistant, Pooja Ruparel, BS, Research Assistant, Andrew Peterson, PharmD, Associate Professor University of the Sciences, Philadelphia, PA, USA

Content:

OBJECTIVE: To compare and evaluate freeware options for bootstrap analyses for incremental cost effectiveness data. METHODS:Obenchain's ICEplane software can be downloaded and installed from www.math.iupui.edu/~indyasa/bobodown.htm. This software was compared against the web-based analyses available through HDS at www.healthstrategy.com. Three datasets, from the Obenchain site, were used in the comparisons. The datasets are from published studies dealing with abciximab (ABX), tricyclic antidepressants (TCA), and pindolol (PIN). RESULTS: ICEplane must be installed on MS Windows operating systems and the HDS program runs online through any operating system with browsers such as Internet Explorer, Netscape or Firefox. Both software options provide output graphs such as scatter plots, confidence intervals and acceptability curves. ICEplane and HDS calculators both generated statistics on the initial raw data such as mean, median, standard deviation and standard error. Bootstrapped statistics include incremental cost- effectiveness ratio (ICER), and 95% Confidence Intervals. Respective mean ICERs and confidence intervals between ICEplane and HDS bootstraps are as follows: ABX: 3771 (-665, 20797) vs 3692 (-1054, 13303); PIN: -1880 (2665, -421) vs -1259 (16939,-260) and TCA: -16.48 (-169, 136) vs -19.96 (-221, 147). CONCLUSIONS: The Obenchain software has additional statistical and charting features that are not available in the HDS program. The analyses from the HDS site runs slowly on more than 1000 bootstrap replications, but the results obtained compare well to ICEplane. The analyses from the Obenchain site can perform up to 25,000 replications. ICEplane and HDS calculate Fieller's statistic on both raw and boostrapped data. ICEplane allows calculations with missing data, while HDS requires manual recoding. Both of these programs should make it easier for individuals to perform basic bootstrap analysis of their data, but certainly more powerful and user-friendly statistical software packages would be preferred.

Back to HDS Software List