To examine whether informant-based assessments of patients with Alzheimer's disease (AD) can be used longitudinally to track patient functioning, the authors followed AD patients (N=153) and their caregivers over 1 year with the Relative's Assessment of Global Symptomatology–Elderly (RAGS–E) and the Alzheimer's Disease Assessment Scale–Cognitive Subscale (ADASc). Factor analysis of the RAGS–E yielded two subscales, Cognitive Functioning and Mood Disturbance. The cognitive subscale and ADASc correlated at all visits, whereas the mood subscale did not. After 12 months (n=62), the cognitive scale worsened at a rate similar to the ADASc, suggesting concurrent validity. Therefore, informant-based measures appear to be reliable and valid methods of identifying cognitive change in AD patients.