Abstract
OBJECTIVE: Understanding how well older individuals with suspected cognitive impairment are functioning within the real-world environment can have important implications for diagnosis and treatment. To evaluate whether an individual is experiencing functional limitations suggesting the presence of mild cognitive impairment (MCI) or dementia, we establish diagnostic cutoff scores for the informant version of the Instrumental Activities of Daily Living-Compensation (IADL-C) scale. METHOD: Informants of research (n = 488) and clinical (n = 119) samples of participants designated as healthy older controls, MCI, or dementia completed the IADL-C. Receiver operating characteristic curve analyses and diagnostic statistics were used to determine optimal cutoffs on the IADL-C for both the 27-item IADL-C and an 11-item short form created using item-level analysis. RESULTS: The optimal cutoff scores that maximized the Youden Index for the research sample long-form were 1.41 in distinguishing cognitively healthy versus MCI participants, and 3.60 in distinguishing dementia from MCI participants, favoring specificity for the clinical sample, the optimal cutoffs were 1.32 and 3.06, yielding higher sensitivity. CONCLUSIONS: These cutoff scores, when used as a screening measure or combined with other clinical and cognitive measures, may be useful for understanding whether an individual may be experiencing functional difficulties in everyday life consistent with a diagnosis of MCI or dementia.