Objective: To determine the validity of the Dementia Questionnaire (a semistructured informant interview) for the diagnosis of dementia.
Design: Comparison of dementia status determined by a telephone-administered informant questionnaire with the criterion standard of clinical diagnosis established by examination and laboratory studies.
Setting: Gerontology Research Center, the Baltimore Longitudinal Study of Aging.
Subjects: Volunteer cohort of 42 men and 32 women aged 68 to 97 years. Subjects were selected from strata defined by Blessed Information Memory Concentration Test scores, with oversampling of borderline scores (3 to 10).
Main outcome measures: Sensitivity and specificity of the Dementia Questionnaire in comparison with the criterion standard of clinical diagnosis.
Secondary outcome measure: Interrater reliability (kappa coefficient).
Results: Sensitivity and specificity for dementia were 100% and 90%, respectively. Most false-positive findings were from subjects with cognitive impairment that did not meet criteria for dementia (Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition. Interrater reliability was high (kappa = 0.83).
Conclusion: The Dementia Questionnaire can be used effectively in research studies to screen for dementia.