A short screening test for visual neglect in stroke patients.

Objective: to extend a preliminary study of the internal structure of six measures comprising the 'conventional' subtests of the Behavioural Inattention Test (BIT) in order to develop a short screening test for visual neglect.

Design: discriminant function analysis of the 15 tests constituting BIT.

Setting: rehabilitation centre.

Patients: 59 selected from consecutive stroke patients entering the centre.

Selection criteria: absent prior history of major CNS disorder, at least 1 week post-stroke, right-handed for writing, absent significant visual impairment other than visual field deficit, no psychiatric overlay or generalized intellectual deterioration, and ability to comprehend and respond to the visuomotor tasks administered. COMPARISON GROUP: 50 non-brain-damaged subjects (hospital employees, members of University subject panel, and local community volunteers).

Assessments: the 15 tests constituting the BIT.

Results: two subtests, letter cancellation and star cancellation, were the most sensitive measure, identifying 74% of neglect patients with no false positives.

Conclusions: letter and star cancellation offer an adequate yet brief screening test for determining which patients might benefit from administration of the complete neglect test battery.