In recent years, the standard for successful treatment of major depression has switched from response to remission; however, little is known about patients who have achieved remission, but still have some residual symptoms and whether they regain previous levels of functioning. In a large, nationwide, cross-sectional, naturalistic survey (VIVAL-D) of 907 patients with major depression treated with a new course of an antidepressant in 41 Italian community psychiatric centers, patients with a Hamilton Rating Scale for Depression, 17-item version (HAM-D17) score up to 14 were selected (n=499). Of these, 169 were considered to be in remission (HAM-D17 ≤ 7) and the other 330 to be mildly depressed. Their level of functioning was evaluated using the SF-12. Only a few (3%) patients in remission were completely symptom free; most were affected by residual symptoms. Patients in remission had better SF-12 scores than those with mild depression, but their functioning was significantly worse than general population norms. In the logistic regression analysis, the HAM-D17 total score and individual items were predictive of poor functioning. Analysis of sensitivity and specificity values showed that a lower cut-off score (4/5) of the HAM-D17 scale was best for predicting poor performance so that a reconsideration of the usual cut-off for remission of 7/8 for HAM-D17 seems overdue.