Objective: The EQ-5D, a short generic health-related quality of life (HRQOL) questionnaire, can derive preference-based index scores (e.g. EQ VAS, British and German EQ-5D indices) for economic evaluation. This study examined if the EQ-5D could be a valid measure describing and valuing HRQOL in alcohol dependent individuals.
Methods: In a sample of 103 individuals diagnosed with alcohol dependence (ICD-10 F10.2), we compared the EQ-5D against a quality of life measure (WHOQoL-BREF), a utility scale (TTO), measures of psychopathology (SCL-90R, CGI-S) and measures of social functioning (GAF, GARF, SOFAS, HoNOS).
Results: The response rate to EQ-5D dimensions was > 98%. Twelve percent of the individuals indicated "extreme problems" in at least one dimension. Different response levels in the dimension "anxiety/depression" were associated with largest effect sizes (absolute value(d)) of differences in mean scores of the WHOQoL-BREF domain "mental" (absolute value(d) = 1.17), the SCL-90R scales "obsessive-compulsive" (absolute value(d) = 1.15), "depression" (absolute value(d) = 1.16), "anxiety" (absolute value(d) = 1.10) and "GSI" (absolute value(d) = 1.09) indicating a similarity between the underlying constructs; concerning the dimensions "self-care", "usual activities", "pain/discomfort" and "mobility" effect sizes were generally lower (0.74 < absolute value(d) < 1.07) or insignificant. The EQ-5D VAS score (mean 58.0), the British EQ-5D index (mean 0.74) and the German EQ-5D index (mean 0.85) showed moderate correlations with other scales (0.28 < r < 0.60).
Conclusion: Seventeen out of 30 hypothesized relationships between similar constructs of the EQ-5D and measures used for comparison were confirmed, possibly favoring EQ-5D's validity. However, the EQ-5D showed a moderate ceiling effect. Further confirmatory research is needed to support the EQ-5D suitability for economic evaluation in alcohol dependent populations.