The Quiet Revolution in Digital Cognitive Assessment

In 2022, a nationally representative study by Manly and colleagues from Columbia University, published in JAMA Neurology, estimated that among 3,496 US adults aged 65 and older, 10% had dementia and 22% had mild cognitive impairment. The study found a disproportionate burden among older Black and Hispanic adults and those with lower education. If the people most at risk are being systematically missed by tools validated in narrow populations, the entire screening enterprise fails its purpose.

This is the environment in which ReCOGnAIze was published in Alzheimer's and Dementia in 2026, demonstrating an AUC of 0.90 for detecting vascular cognitive impairment in a Singapore cohort. Validated in an Asian population with a specific focus on vascular cognitive impairment, it addresses exactly the representational gap the Manly et al. study identified.

A meta-analysis by Malek-Ahmadi and colleagues from Banner Alzheimer's Institute, published in JAMA Neurology in 2026, synthesised data from 7,834 participants and found plasma p-tau217 achieved an AUC of 0.87 in identifying amyloid-positive cognitively unimpaired individuals. Digital cognitive assessment and blood-based biomarkers are converging on the same goal: making early detection accessible at scale.

The revised VasCog-2-WSO diagnostic criteria published by the Sachdev-led international consortium in JAMA Neurology in 2025, involving 70 experts from over 30 countries, confirmed that vascular cognitive impairment is a distinct, diagnosable condition that requires its own validated tools. To understand your own cognitive baseline, book an assessment with one of our clinical partners.