The study examines the prevalence of depressive syndromes among unselected primary care attenders, as well as doctors' recognition and treatment rates, in order to examine patient and doctor-related factors associated with poor recognition. This nationwide study included a total of 20421 patients (aged 15-99 years) attending their primary care doctors (n = 633) on the study's target day in April 1999. Patients received a self-report questionnaire, including the Depression Screening Questionnaire (DSQ), to provide diagnoses of depressive disorders according to the criteria of DSM-IV and ICD-10. Doctors completed: (1) a pre-study questionnaire assessing data on doctors' psychosocial, professional and training background, as well as current practices in patients with depression and (2) an evaluation form for each patient seen to assess his diagnostic decision, clinical severity and treatment choices. Taking the DSQ as a yardstick, 4.2% of all primary care attenders fulfilled criteria for a major depressive episode according to DSM-IV; considerably higher rates of 11.3% were obtained using the ICD-10 criteria for mild depressive episodes. Rates of depression were higher in females, increased by age, and were also elevated in those retired, unemployed as well as non-working housemen/wives. Taking the doctors' decision of definite or probable depression, 75% of all DSM and 59% of all ICD-10 diagnoses were recognized by the treating physician. However, doctors also assigned diagnoses of definite depression in an additional 11.7% of patients not meeting either ICD-10 nor DSM-IV criteria. Among correctly identified depression cases doctors decided to prescribe drug treatments in 72.7% (DSM) and 60.8% (ICD). Some 16.2% of DSM and 10.1% of ICD-cases were referred to mental health specialists; non-drug interventions were prescribed for 19.8% (DSM) and 24.9% (ICD), respectively. Multiple logistic regression analyses revealed that recognition is associated with prior treatment episodes, increasing number of depression symptoms, patients higher age, practice experience of treating physician greater 5 years and psychomotor retardation. These findings confirm the high prevalence of depressive syndromes in primary care settings and underline the particular challenge posed by a high proportion of with near-threshold symptomology patients. Although recognition rates among more severe major depressive patients as well as treatments prescribed appear to be more favourable than in previous studies, the situation in less severe cases, and the high proportion of doctors' definite depression diagnoses in patients with depression symptoms that are clearly below even the subthreshold level, raises significant concerns.