Heterogeneity in diagnostic trajectories of ADNI participants

Understanding Alzheimer’s Disease (AD) progression requires careful consideration of the substantial diversity observed across participants and their clinical trajectories. Using comprehensive data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), this study characterizes heterogeneity in participant demographics, cognitive performance, visit patterns, and diagnostic trajectories across 2430 individuals followed for up to 15 years. We identified significant demographic variation across diagnostic groups, with education showing particularly strong associations with diagnostic status. Cognitive performance showed very large differences across diagnostic categories, though substantial score overlap between groups helps explain the diagnostic instability observed over time. The number of diagnostic transitions strongly correlated with visit frequency, an important consideration when interpreting apparent trajectory complexity in longitudinal research. Remarkably, nearly half of all diagnostic transitions represented improvements rather than progression, challenging assumptions of unidirectional disease course. These findings clearly demonstrate that participants do not exhibits equivalent diagnostic trajectories. Therefore, we recommend that all studies utilizing DNA data for further analysis including progression modeling, clinical trial design, and interpretation of longitudinal AD studies, first conduct a trajectory analysis as described in this work. • ADNI provides a rich longitudinal data that support a range of Alzheimer’s disease (AD) studies. • ADNI participants exhibit heterogeneous diagnostic outcomes in the progression of AD. • Diagnostic trajectories include stable, increasing, decreasing and instable or oscillating transitions. • Transitions occur between three core stages of the disease: Normal, MCI, and Dementia.