Cognitive-based measures screening for depression in the medically ill: the DMI-10 and the DMI-18


Professor G. Parker Euroa Unit, Prince of Wales Hospital, Randwick 2031, Sydney, Australia E-mail:


Objective: We suggest that the identification of depression in the medically ill (DMI) might be improved by focussing on cognitive features.

Method: We recruited 302 patients to complete our provisional cognitive-based measure. Subsets also completed one of two comparator screening measures, either the Hospital and Anxiety Depression Scale (HADS) or the Beck Depression Inventory for Primary Care (BDI-PC). One hundred and sixty patients were then assessed by a psychiatrist who estimated whether they were `clinically depressed' and who also administered a standardized interview for depression (the CIDI).

Results: Analyses identified items discriminating clinically depressed and non-depressed individuals, allowing development of brief (10-item) and extended (18-item) measures. The two new measures were compared with the HADS and the BDI-PC in discriminating depressed and non-depressed medically ill patients.

Conclusion: A cognitive construct-based approach to assessing depression in the medically ill appears strongly supported. We provide brief (DMI-10) and extended (DMI-18) measures that appear to have utility as screening instruments. Consideration of the discriminating items may also assist clinical decision making.