The Use of Disease-Specific Outcome Measures in Cost-Utility Analysis

Abstract

The Functional Assessment of Cancer Therapy-Lung (FACT-L) is a validated, sensitive and reliable patient questionnaire that evaluates and quantifies quality of life (QOL) across several domains, including lung cancer-related symptoms. The FACT-L was not designed for use in economic evaluation and does not incorporate preferences into its scoring system. To derive a set of Dutch preference weights for FACT-L health states that can be used to convert FACT-L into a single value that can be used in cost-utility analyses. A representative sample of the Dutch population (n = 1076) directly valued an orthogonal set of eight FACT-L health states on a 100-point rating scale with the anchor points 'worst imaginable health state' and 'best imaginable health state'. Eleven FACT-L items were selected to describe the FACT-L health states that were directly valued. Regression analysis was used to interpolate values for all other possible health states. Scores were transformed into values on a scale where 0 indicated dead and 1 indicated full health. The estimated values for FACT-L health states ranged from 0.08 to 0.93. The estimated value sets were applied to FACT-L data of lung cancer patients participating in a clinical study. Significant differences in the mean value and mean gain of 0.12 and 0.07, respectively, were found between patients in remission and patients with progressive disease at 4 weeks' follow-up. Our results reaffirmed that the methodology used here is a feasible option to convert data collected with a disease-specific outcome measure into preferences. We concluded that the sensitivity of the derived set of societal preferences to capture differences and changes in clinical health states is an indication of its construct validity.