[Adaptation of a dementia test for use in visually handicapped persons exemplified by SIDAM (Structured Interview for the Diagnosis of Alzheimer disease, Multi-Infarct Dementia and Dementias of other Etiology)].
In epidemiological field studies on the prevalence and incidence of dementia, the problem of cognitive testing of visually impaired individuals is rarely discussed. In the Leipzig Longitudinal Study of the Aged (LEILA 75+) a version of the SIDAM (SIDAM-Structured Interview for the Diagnosis of dementia of the Alzheimer type, Multi-infarct dementia and dementias of other etiology) for the visually impaired was employed from which all items requiring image processing had been omitted (SIDAM-blind). In order to be able to interpret the test results, the final scores for the full SIDAM were estimated by linear transformation of the scores in the blind-version. The method of linear transformation is based on certain theoretical assumptions which are examined in this paper. Linear transformation of scores has proved to be a valid procedure only for individuals with high cognitive performances. Thus, an evaluation of estimated scores based on the norms for the original SIDAM has limitations. Consequently, age-specific norms for the SIDAM-blind are presented. Sensitivities and specificities for different cut-off points used to discriminate between demented and non-demented vision-impaired individuals are given.