EQ-5D Scores for Diabetes-Related Comorbidities

Background: A single-source catalogue of nationally representative EuroQol five-dimensional questionnaire (EQ-5D) scores for chronic conditions in the United States and the United Kingdom from the Medical Expenditure Panel Survey (MEPS) has been published and has been used extensively for public health and cost-effectiveness modeling. Objectives: To estimate EQ-5D scores for diabetes-related chronic conditions that are consistent with the previous catalogue. Methods: The MEPS is a nationally representative survey of the US civilian population. EQ-5D-3L questionnaire responses were mapped from short-form 12 health survey responses in 2000 to 2011 MEPS data using multinomial logistic regression. Country-specific tariffs were applied to the mapped EQ-5D responses for the United Kingdom, the United States, France, and Spain. EQ-5D scores were regressed on diabetes-related comorbidities controlling for sociodemographic characteristics, body mass index, and comorbidity using robust (median) regression. Methods were consistent with the previously published catalogue of EQ-5D scores for the United States. Results: There were 20,705 individuals with diabetes and a valid mapped EQ-5D score in the 2000 to 2011 MEPS data. Unadjusted mapped EQ-5D scores for individuals with diabetes varied by country-specific tariff from 0.70 (France) to 0.79 (United States). Regression results reflecting marginal disutility estimates for EQ-5D scores are provided for 17 diabetes-related comorbidities and body mass index categories for US, UK, French, and Spanish tariffs. Conclusions: The estimates provided in this research may be useful for analysts attempting to model the impact of diabetes and diabetes-related comorbid conditions on utility. They are derived from a single-source, nationally representative population and augment the existing off-the-shelf catalogue of EQ-5D scores for chronic conditions.