OBJECTIVE: The aim of the study is to test the hypothesis that freehand human-figure drawing (HFD), can be used as a complementary screening instrument to differentiate between demented elderly people and healthy elderly controls in population based studies.

METHOD: HFD was examined in 668 elderly ( > or = 75 years of age) participants from an epidemiological study in Stockholm, who were asked to draw a human figure. The drawings were analysed on the content of body details and structural characteristics.

RESULT: The results show quite clearly that the body details and the height decrease with decreasing cognitive function, whereas the centredness (the distance in cm from the centre of the figure to the centre of the paper) increases with decreasing cognitive functioning. Demented people place their figures in the upper left corner of the sheet, compared with the mostly well centred figures of non-demented people. Age, on the other hand, has an influence on the HFD as after 90 years of age most of the variables show regressive changes.

CONCLUSION: The HFD can help to differentiate between demented and non-demented subjects as well as between dementia of different severity. The HFD does not help us, however, to discriminate between Alzheimer's disease and vascular dementia. Age has an influence on the HFD in the sense that after 90 years most of the variables regress to a smaller or more primitive form.