Cost-Effectiveness Modeling Using Health State Utility Values

Abstract

There is now widespread use of decision-analytic models as vehicles for assessing the cost-effectiveness of healthcare interventions in terms of the incremental cost per quality-adjusted life-years (QALY). Health state utility values (HSUVs) are used in the calculation of QALYs and are an important determinant of the output of a model. This article addresses a number of methodological issues around selection, analysis, and use of HSUVs for populating cost-effectiveness models. The first issue concerns the selection of measure for generating the HSUVs that meets the requirements of policy makers and best meets commonly accepted measurement criteria. There are a number of alternative approaches including those where health state descriptions are obtained from patients and then valued by a sample of the general population or patients. This approach is used by various generic preference-based measures (e.g., EQ-5D or Health Utilities Index 3), condition-specific preference-based measures (e.g., European Organisation for Research and Treatment of Cancer-8 dimensions and Asthma Quality of Life Questionnaire 5 dimensions) or bespoke condition and/or treatment specific vignettes. An alternative approach is to elicit values directly from patients (e.g., own time trade-off). The second issue is the source of data and whether to rely on the main clinical efficacy trials or to seek more relevant values for the population in the cost-effectiveness model from observational datasets, or to search, review, and synthesize an ever-growing literature. This article also considers situations where suitable utility data are not available from relevant studies and describes the use of regression techniques to map from various health or clinical measures to the selected utility measure. Finally, this article considers a number of technical problems in using HSUVs in cost-effectiveness models, including how to adjust values over time, estimate values for those not in the condition of interest, and combine the impact of conditions (comorbidities) or adverse events.