What factors predict discrepancies between self and observer ratings of depression?

This study used a matched clinician and self-rating scale, the Inventory for Depressive Symptomatology (IDS; Rush et al., 1986), in the assessment of 48 patients meeting DSM-III-R criteria for non-psychotic major depressive disorder. Patients generally rated their symptoms as more severe than the clinician, but differences between the self and clinician rating (the delta-IDS) were significantly correlated with non-endogenous depressive subtype, higher levels of neuroticism and dysfunctional attitudes, and lower self-esteem. Multiple linear regression analysis which controlled for severity of depression demonstrated that whilst neuroticism, dysfunctional attitudes and self-esteem accounted for 48.5% of the variance in delta-IDS, the variable accounting for most of the variance was low self-esteem (47.9%).