[Measuring preferences for depressive health states]

The purpose of this study was to analyse the construct validity and the concurrent validity of the EuroQol instrument (EQ-5D self-classifier, EQ VAS and EQ-5D indices), time trade-off (TTO) and contingent valuation (willingness-to-pay) for preference-based valuation of depressive health states.

A survey of a representative sample (N = 3522) of the non-institutionalised population aged 18 and above in Germany was conducted using the EQ-5D, TTO and contingent valuation as well as the SF-12 questionnaire. Construct validity was analysed by comparing 130 respondents in whom 12-month prevalence for major depression, minor depression, or dystimia according to DSM-IV was identified based on a version of the Composite International Diagnostic Interview (WMH-CIDI) to respondents without depressive disorders. Concurrent validity was analysed by assessing the correlation between these measures in the group with depressive disorders.

In the group with depressive disorders, problems in all EuroQol dimension except for mobility were significantly more frequent, and mean values of all analysed preference measures except for willingness-to-pay were less favourable than in the group without depressive disorders (EQ VAS score 67 vs. 78; EQ-5D index [VAS] 83 vs. 94; EQ-5D index [TTO] 0.81 vs. 0.91; TTO 0.89 vs. 0.95; always p < 0.001). In the group with depressive disorders, EQ VAS score and EQ-5D indices were significantly correlated one with another (ranging from r = 0.63 to r = 0.92) as well as with the mental and the physical component scale of the SF-12 (ranging from r = 0.26 to r = 0.45) and with willingness-to-pay (ranging from r = - 0.20 to r = - 0.22); besides, willingness-to-pay was significantly correlated with TTO (r = - 0.46).

The results of this study point to the construct validity and concurrent validity of preference measures in patients with depressive disorders.