The screening characteristics (sensitivity and specificity) of a ten-item self-rated depression scale, the Brief Depression Scale (BDS), were assessed according to two different scoring mechanisms in a hospital sample of 275 patients with depressive illness and anxiety disorders and a non-clinical sample of 50 subjects. The first scoring method was 'conventional' in using the BDS total score and the second scoring method was to count the number of items on the BDS where the individual item scores were greater than a certain 'threshold' score. There was a very high likelihood of fulfilling the closest corresponding ICD-10 (International Statistical Classification of Mental and Behavioural Disorders, 10th revision) or DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, 4th edition) diagnostic criterion to a specific BDS item at an item score of 4 points. The sensitivity and specificity of the BDS in detecting a case of ICD-10 or DSM-IV depression were both 85% if four or more items scored above this level (4 points per item). These values compared favourably to those obtained at the conventional BDS 'cut-off' score of 19 points (sensitivity 87% and specificity 90%). The novel scoring mechanism might offer the advantages of a reduced assessment time after screening.

A new, ten-item, self-rated depression scale, the Brief Depression Scale (BDS), was validated in 275 psychiatric in-patients and out-patients with depressive illness and anxiety disorders and a non-clinical sample of 50 subjects. The BDS showed satisfactory convergence with the interview-rated Montgomery Asberg Depression Rating Scale (r(s) = 0.72 and p < 0.001) and Hamilton Depression Rating Scales (r(s) = 0.70 and p < 0.001) and with the depression subscale of the self-rated Hospital Anxiety and Depression Scale (r(s) = 0.89 and P < 0.0001). The BDS had satisfactory internal consistency with a Cronbach's a-value of 0.86. The construct validity of the scale was assessed by principal components analysis and a two-factor solution was found in the clinical sample which explained 54% of the variance in the BDS total score. The sensitivity of the scale to clinical change in patients treated over an 8 week period was found to be satisfactory las judged by the biserial correlation coefficient of change in BDS score with changes in other depression measures). The BDS had a sensitivity of 87% and specificity of 90% at a cut-off score of 19. The BDS is a brief, reliable and valid measure which may be useful in primary and secondary care settings as both a screening and outcome measure for depression. Primary Care Psychiatry 2000; 6:111-118, Copyright (C) 2000 by LibraPharm Limited.