The validity of informant reports in assessing the severity of dementia: evidence from the CAMDEX interview

The evaluation of a patient's mental state, overall clinical profile and behavioural disturbance in the process of diagnosing dementia requires at least two sources of information: the patient and the informant. Since the severity of the dementia may interfere with the subjective perception of these disorders, it is important to evaluate the consistency between these two sources of information and the clinical and psychometric evaluation made by the physician. Accordingly, in this study five behavioural areas, derived from the semi-structured interview schedule provided by the Cambridge Examination for Mental Disorders of the Elderly (CAMDEX, i.e., Sleep, Depressed Mood, Everyday Activity, Memory and Global Mental Functioning) have been tested on the patient and his/her informant. Eighty dementia patients (mean age = 74 years) and their informants participated in the study. The dementia group was subdivided into two levels of severity according to DSM-IIIR criteria: 41 with mild dementia and 39 with moderate dementia, respectively, matched for age and schooling. The rating of impairment was found to increase along with the severity of dementia in all the above-mentioned areas, except for sleep. However, the source of information per se significantly influenced the evaluation of memory functioning. Moreover, the significant interaction between the two factors considered indicates that memory functioning is evaluated quite differently by the patients and the informants, as only in the assessment made by the latter group did the impairment increase in parallel with severity of dementia. Finally, whereas none of the subjective measures recorded in the patients were significantly correlated with their test scores, the correlations between the informant memory appraisals and patient test results proved to be significant. The present findings confirm the validity of informant reports in assessing cognitive and memory disorders in early-stage dementia, as well as in distinguishing patients with mild from those with moderate dementia.