OBJECTIVES: Utility scores are needed for computing Quality-Adjusted Life Years (QALYs), commonly used as outcomes in economic evaluations. Understanding patient preferences for depression outcomes is important for economic evaluations of new antidepressants, as well as for understanding patient behaviour and compliance to antidepressant regimens. METHODS: A cross-sectional survey was carried out in order to evaluate EuroQol scores in three profiles of Swedish outpatients suffering from Major Depressive Disorder (MDD): patients experiencing a new episode of MDD (“Non-treated” group), patients successfully treated for eight weeks (“Responders” group) and patients treated with an antidepressant for eight weeks and currently in relapse (“Relapsing” group). Patients were asked to fill in the EuroQol while the physicians were asked to fill a standardised form. After applying the original social tariff proposed by the EuroQol Group, patients' sociodemographics were compared to reveal potential confounders between stratification groups, and adjusted mean EuroQol scores were estimated and compared using analysis of covariance. RESULTS: Nine psychiatrists and five general practitioners recruited a total of 205 patients (“Non-treated”: 61; “Responders”: 91; “Relapsing”: 53). No statistically significant differences between groups were found in terms of gender (Chi-square test: p=0.156) or age (ANOVA: p=0.093). The overall EuroQoL score was 0.584 ± 0.304 (mean ± standard deviation), ranging between -0.239 and 0.919. The “Responders” patients got significantly higher EuroQol scores (mean ± standard error) than the “Non-treated” or the “Relapsing” patients (0.702 ± 0.033 vs. 0.482 ± 0.038 vs. 0.467 ± 0.043, respectively; p<0.001 for both comparisons), with differences clearly greater than the minimal clinically important difference of 0.05. The “Non-treated” patients were not statistically significantly different from the “Relapsing” patients (p=0.961). CONCLUSIONS: Relapse in MDD appears to be a very debilitating situation in terms of patient preferences, since relapsing patients reported EuroQol scores slightly lower than those reported by non-treated patients.