A randomized clinical trial across more than 5,000 primary-care patients showed that combining the Quick Dementia Rating System (QDRS) with a Passive Digital Marker (PDM)—an EHR-based machine-learning natural language processing tool—lifted new diagnoses of Alzheimer’s disease and related dementias by roughly 33% within 12 months without adding clinician time or cost. The study was led by teams at Indiana University School of Medicine, Regenstrief Institute, Eskenazi Health and partner institutions. Investigators emphasized that embedding AI-driven passive markers into electronic health records can surface subtle documentation and symptom signals, enabling earlier referrals and diagnostic workups. Researchers framed the approach as scalable and equitable: it operates within existing workflows and avoids licensing fees or extra physician time, suggesting wide applicability across resource-constrained primary-care settings.