The aim of this study was to determine the diagnostic accuracy of the Turkish version of the revised Brief Patient Health Questionnaire (Brief PHQ-r) in depression, panic disorder and somatoform disorder, in primary care settings. This was a cross-sectional study with blinded psychiatric evaluation. The study was conducted at three sites which provide primary health services. Total number of the participants was 1556. Of these, 1387 comprised the analysis population. Diagnoses of depression, panic disorder and somatoform disorder made according to the patient's responses to the questions on Brief PHQ-r form were compared with the diagnoses made by psychiatrists using DSM-W. Diagnostic performance parameters of Brief PHQ-r were calculated. Main results are as follows: for any diagnosis k = 0.567, sensitivity 79.0%, specificity 82.9%; for major/minor depressive disorder k = 0.536, sensitivity 76.0%, specificity 85.3%; for panic disorder k = 0.640, sensitivity 74.4%, specificity 98.4% and for somatoform disorder k = 0.476, sensitivity 61.9% and specificity 92.5%. Diagnostic performance of Brief PHQ-r was found to be quite good in the diagnosis of major/minor depressive disorder, panic disorder and somatoform disorder, in primary healthcare settings. We recommend its use in routine clinical practice in order to help primary healthcare physicians, and also infield surveys on psychiatric disorders. Qnt] Psych Clin Pract 2004; 8: 11–18)