Claims-based machine learning screening model

Claims-based machine learning screening model decreases cognitive assessments in silico (from 100,000 to 2,924).

“decreased the total number of Mini‐Cog assessments from 100,000 to 2,924 while identifying 1,347 new dementia cases.”

In silicoin a hypothetical population of 100,000 older adults

Leveraging medicare claims to reduce false positives from 87% to 12%, a two‐stage approach to dementia screening

AAIC 2025 abstract·9 Jan 2026

Claims-based machine learning screening model decreases false positive rate in silico (from 84.3% to 12.7%).

“administering the Mini‐Cog only to those individuals flagged by the claims‐based model, reduced the false positive rate to 12.7%.”

In silicoin a hypothetical population of 100,000 older adults

Leveraging medicare claims to reduce false positives from 87% to 12%, a two‐stage approach to dementia screening

AAIC 2025 abstract·9 Jan 2026