In this study we have examined four parts of the 5th revision of the World Health Organization schedule for Standardized Assessment of Depressive Disorders (WHO/SADD-5): (I) items that cover the present depressive state; (II) items that cover the psychiatric history; (III) a global assessment scale for the severity of depression, and (IV) the current ICD-9 diagnosis. Our analysis was based on a comparison of the interobserver reliability of item combinations leading to DSM-III, RDC, Newcastle and Melancholia Scale classifications of patients with depressive disorders. To facilitate these combinations we had added 4 items to SADD: (a) quality of depression, (b) persistence of depression, (c) reactivity of symptoms, and (d) accusations of others. Our results showed that WHO/SADD-5 has an acceptable degree of interobserver reliability both at the levels of global assessment of severity of depression and ICD-9 diagnosis, whereas the item combinations obtained lower intraclass coefficients. However, the item analysis focused on two SADD subscales of acceptable interobserver reliability: a severity scale of 16 items selected from our Melancholia Scale, and a diagnostic scale of another 10 items selected from the two Newcastle Scales.