Estimating a preference-based index for a 5-dimensional health state classification for asthma derived from the asthma quality of life questionnaire.

BACKGROUND: This article presents a valuation study to estimate a preference-based index for a 5-dimensional health state classification for asthma (AQL-5D) derived from the Asthma Quality of Life Questionnaire (AQLQ). METHODS: A sample of 307 members of the UK general population valued 99 asthma health states selected from the AQL-5D using the time tradeoff technique. Models were estimated to predict all possible 3125 health states defined by the AQL-5D, and the models were compared in terms of their ability to predict mean values for the 99 states. RESULTS: Mean health state values ranged from 0.39 to 0.94 based on an average of 22 valuations per state. A main effects model estimated on mean health state values and adjusted for consistency had the best predictive ability (mean absolute error of 0.047 and only 9/98 states with errors >0.1) and the most logical consistency with levels of the AQL-5D. The low number of valuations per state may have resulted in unreliable estimates for the models. Preference-based condition specific measures are limited in their ability to make cross-disease comparisons. CONCLUSION: This is the first study to derive a condition-specific preference-based measure from an existing measure of health-related quality of life in asthma for use in economic evaluation.