CLINICAL-FEATURES DISTINGUISHING LARGE COHORTS WITH POSSIBLE AD, PROBABLE AD, AND MIXED DEMENTIA

Objective: To determine whether clinical features and rate of cognitive and functional decline differed in cohorts of possible AD (poAD), probable AD (prAD), and mixed dementia (MIX) patients. Design: Cohort study with 1-year follow-up examination, comparing three groups of subjects. Setting: Outpatient evaluation at nine California Alzheimer's Disease Diagnostic and Treatment Centers (ADDTC). Patients: There were 1701 elderly patients who presented for evaluation of memory complaints. Measurements: Historical, physical, and neurological variables for cross-sectional comparisons and 1-year rate of change on the Mini-Mental State Examination (MMSE), Blessed Information-Memory-Concentration test (BIMC), and Blessed Dementia Scale (BDS). Results: Mean initial MMSE scores for poAD (n = 279), prAD (n = 928) and MIX (n = 430) were 17.9 (+/-7.4), 13.9 (+/-7.5), and 15.4 (+/-7.1). Delusions and psychosis occurred in about one-third of each group, most often in those with moderate dementia (MMSE 11-20). PoAD were distinguished from prAD by significantly more alcohol abuse, physical health problems, and focal motor or sensory findings. MIX differed from AD alone by increased prevalence of cardiovascular disease, hypertension, stroke, TIA, and exposure to general anesthesia, and by a greater frequency on exam of depressed mood, focal motor or sensory findings, and gait disorder. All groups declined by about 2.8 points on the BIMC, 2.9 points on the MMSE, and 1.8 points on the BDS, a functional scale, over 1 year. Neither extrapyramidal signs nor psychosis predicted a more rapid rate of decline. Conclusions: Various features help to distinguish poAD, prAD, and MIX in a large cohort of patients, but do not predict the rate of progression.