Objective: To determine whether Alzheimer's disease (AD) associated with comorbidities or atypical features (possible AD) is marked by differences in clinical course and outcomes compared with uncomplicated AD (probable AD). Methods: Annual evaluations were made of patients with AD, for up to 11 years. Six hundred forty subjects with AD were clinically classified into two groups: 1) possible AD (n = 208), and 2) probable AD (n = 432). Data on demographics, Mini-Mental State Examination (MMSE), Short Blessed Test (SBT), psychometric performance, and Clinical Dementia Rating (CDR) were collected at baseline. Kaplan-Meier survival curves and Cox proportional hazards models were conducted to evaluate whether possible AD would have different outcomes on dementia progression, nursing home placement, and death compared with probable AD. Results: The possible AD group was slightly younger and less well educated than the probable AD group, but there were no group differences at baseline in MMSE, SBT, and CDR scores. Controlling for age and education, the possible AD group had poorer baseline psychometric performance (p = 0.022). There were no group differences, however, for rate of dementia progression, nursing home admission, and death. Conclusions: Comorbidities and atypical features in this sample of patients with Alzheimer's disease did not substantially affect dementia outcomes. The primary determinant of the clinical course of dementia in this sample was the presence of clinically diagnosed Alzheimer's disease, independent of the presence or absence of comorbidities or atypical features. Therefore, patients with possible Alzheimer's disease may be considered for inclusion in investigations of Alzheimer's disease, including clinical trials, to improve the generalizability of the findings.