The sensitivity and specificity of the Middlesex Elderly Assessment of Mental State (MEAMS) for detecting cognitive impairment after stroke.

The aim of the study was to assess the sensitivity and specificity of the MEAMS (Golding, 1989) for detecting cognitive impairment after stroke. Stroke patients admitted to hospital received a cognitive screening assessment, the MEAMS, and a detailed cognitive assessment. The information obtained from the detailed assessment was summarised in a structured written report. From the conclusions in these reports, patients were classified as "impaired" or "not impaired" in perception, memory, executive function and language. The sensitivity and specificity of the MEAMS subtests and the overall number of tests passed were determined in relation to the presence of impairment, as given in the overall conclusion of the written reports. There were 30 stroke patients, aged 58 to 92 (mean 75.80, SD 7.94) years. Of these, 17 were men and 13 were women. The sensitivity of the MEAMS subtests ranged from 11% to 100% and the specificity ranged from 69% to 100%. The sensitivity of the overall MEAMS score was 52% and the specificity was 100%, using a cut-off score of 3 or more fails to indicate impairment. Three subtests, Orientation, Naming and Unusual views had 81% sensitivity and 50% specificity for detecting problems in language, perception or memory. The MEAMS was not a sensitive screen for overall cognitive impairment or for memory, perceptual, language, or executive function problems after stroke, but it was specific. Although screening for cognitive impairment is important, the MEAMS is not recommended as the sole method, as it produces an unacceptably high false negative rate. Three subtests (Orientation, Naming and Unusual views) had 81% sensitivity and 50% specificity for detecting cognitive problems in language, perception or memory after stroke.