Frequency and impact of informant replacement in Alzheimer disease research.

Informants serve an essential role in Alzheimer disease research. Were an informant to be replaced during a longitudinal study, this could have negative implications. We used data from the National Alzheimer's Coordinating Center Uniform Data Set to examine the frequency of informant replacement among Alzheimer disease dementia participants, whether patient and informant characteristics were associated with replacement, and how replacement affected research outcome measures. Informant replacement was common (15.5%) and typically occurred after the first or the second research visit. Adult child (24%) and other (38%) informants were more frequently replaced than spouse informants (10%). Older spouse informant age and younger adult child informant age were associated with replacement. The between-visit change in Functional Assessment Questionnaire scores was greater in patients who replaced informants than in those with stable informants. Clinical Dementia Rating-Sum of Boxes, Functional Assessment Questionnaire, and Neuropsychiatric Inventory scores showed greater variability in between-visit change in patients who replaced informants compared with those with stable informants. These findings suggest that informant replacement is relatively common, may have implications to study analyses, and warrant further examination in the setting of clinical trials.