Changing patient characteristics and survival experience in an Alzheimer's center patient cohort

Background: Large and diverse dementia patient cohorts can further a variety of research programs aimed at improving diagnosis, treatment, and meaningful survival in AD. Method: We recruited 1,502 dementia patients between 1989 and 2002, subclassified using standardized criteria and laboratory procedures, and treated according to established guidelines. Baseline clinical and psychometric measures were repeated annually, in person or by use of a multi-modal telephone follow-up program that included many of the measures obtained at in-person visits. We tracked vital status of all subjects at 6-month intervals and offered autopsies to all participants. We assessed for cohort effects in baseline characteristics by 2-year intervals, examined the characteristics and outcomes for those who remained active compared to those who were eventually lost to follow-up, examined survival times for demographic or diagnostic subgroups, and assessed the accuracy of clinical diagnoses versus neuropathology. Results: The average age at entry, average educational level, and baseline MMSE scores for subjects are increasing overtime, and probable AD diagnoses are also increasing. Most (80.6%) subjects have remained active in our Center; those who did not were more likely to have a non-AD diagnosis. Survival averages 5.2 years (Cl 4.98-5.37) and is influenced by age and gender, but not by diagnosis of probable versus possible AD. Our diagnostic accuracy is 89.6%, with high sensitivity to the presence of AD (96%). Conclusions: In a large and representative clinical cohort, the demographics of AD are changing over time. Careful analyses of those who continue and those who drop out from follow-up suggest that atypical diagnosis, rather than severity or demographic issues accounts for most of the attrition. Clinicians are likely to encounter increasingly older patients with milder disease, and these trends have implications for the design of clinical trials. Survival from the onset of first symptoms, similar for probable and possible AD cases, may be increasing over time.