Depression is both over- and underdiagnosed, especially in primary care. The ICD-10 criteria fail to provide a precise diagnosis because of depression heterogeneity and clinician rating biases. Better understanding of depression origin may result from combining different approaches to psychometric evaluation of depression.

To find out whether an approach to depression evaluation – clinician-rated scale, self-rated scale, behavioural test – affects determination of its core features.

85 primary care patients with ICD-10 depressive episode were examined. Depressive syndrome was evaluated with Clinical Interview for Depression and Related Syndromes (CIDRS; with items of HAMD-17), Beck Depression Inventory (BDI) and MMPI test. Data from different instruments were independently analyzed with factor analysis. General factors of each analysis were compared for concordance. Patients groups were formed in accordance with factor scores; first groups in each approach were compared for identity of participating patients.

General factors derived from analyses of different psychometric instruments data were found to be formed by distinct psychopathological features. For CIDRS these were lowered mood and diminished interest, for BDI – guilt and self-dislike, for MMPI – depressive tendencies and lack of social contacts. An approach to evaluation also determined what patients entered the nuclear depressives group.

Depressive syndrome represents a multidimensional clinical construct with its nuclear components being substantially dependent on an approach to its evaluation. Clinician-rated scales, patient-rated scales and behavioural tests should be regarded as complementary instruments in evaluating depressive syndromes in order to provide more precise diagnosis and treatment of depression.