Strengthening early detection of dementia with a dual approach that includes AI and patient input

Few primary care practices are designed to detect Alzheimer's disease and related dementias in a timely manner. The limited time that primary care clinicians have with their patients, the need to focus on the health problem that brought the patient to the clinic, and the stigma surrounding Alzheimer's disease and dementia are the main reasons why this condition goes unrecognized.

Researchers have demonstrated that a fully digital artificial intelligence, zero-cost method for detecting dementia can be scaled up across primary care clinics without requiring additional time for physicians.

In a pragmatic, real-world, randomized clinical trial involving more than 5,000 patients in primary care practices, researchers from Regenstrief Institute, Indiana University School of Medicine, Eskenazi Health, University of Miami School of Medicine, and Lamar University tested a dual approach that combined the Quick Dementia Rating System (QDRS), a 10-question patient-reported tool, with an artificial intelligence (AI) tool called Passive Digital Markers. The combination of these tools increased the diagnosis rate of new Alzheimer's disease and related dementias by 31% compared to usual care, without the need for additional clinician time or expensive tests.

The AI ​​tool, developed for more than a decade by research scientist Malaz Boustani, MD, MPH, and his team at Regenstrief, is a machine learning algorithm that uses natural language processing to analyze data from electronic health records (EHRs). Identify information such as memory problems, blood vessel problems, and other factors associated with dementia.

“Building on more than 50 years of innovation in digital health data science and machine learning, this passive digital marker developed at the Regenstrief Institute is now open source,” said Dr. Boustani, a Regenstrief and IU School of Medicine faculty member.

In keeping with the tradition of Regenstref's open medical records approach, there are no licensing fees. Just like deploying any other app, there are only base deployment costs. With electronic health records and the right personnel, it can be implemented in any healthcare system. It costs zero and requires no clinician time. ”

Maraz Boustani, Regenstrief Institute

The combination of digital approaches not only improved detection rates, but also increased follow-up diagnostic evaluations, such as neuroimaging and cognitive tests, by 41%. This suggests that we can provide earlier and more accessible dementia care to people who have traditionally been underserved by the health care system.

“This is the most scalable approach to early detection that I know of,” said Dr. Boostani, lead author of Digital Detection of Dementia in Primary Care Clinical Trials. “Most early detection methods require at least five minutes of a clinician's time and often come with a licensing fee. In contrast, our dual approach requires no clinician time or money.”

The study was conducted at Eskenazi Health Center, one of nine federally qualified health centers in Indianapolis, and embedded QDRS and passive digital markers directly into the Epic electronic health record. The system automatically invites patients aged 65 and older to complete a brief QDRS survey through a patient portal, while passive digital marker algorithms continuously analyze existing clinical data to flag patients at risk. Results automatically appear in clinicians' EHR inboxes, prompting further evaluation only when necessary, without requiring additional time, personnel, or manual screening.

“The power of this approach is that it helps level the playing field,” said Dr. Gina Ben-Mired, a Regenstrief affiliate scientist and professor at Lamar University. He is the person who developed the passive digital marker tool in collaboration with Dr. Boostani. “By embedding these tools directly into the electronic health record, we can reach patients who are often overlooked and ensure that everyone has the same opportunity for early detection and care, regardless of background or resources.”

“The Dementia Quick Assessment System was designed to allow patients and families to easily and quickly report cognitive changes,” said James E. Galvin, MD, MPH, professor of neurology and director of the Center for Comprehensive Brain Health at the University of Miami Miller School of Medicine. “When used in conjunction with digital tools like Regentrief passive digital markers, early detection can be carried out efficiently and effectively at scale.”

This breakthrough represents a major step forward in integrating AI and patient-reported outcomes into daily clinical care. The research team demonstrated how technology can enhance early detection, reduce the burden on primary care teams, and improve outcomes for older adults by integrating scalable digital tools that work seamlessly within existing healthcare systems.

“This initiative represents the next step in Regenstrief's half-century legacy of leveraging data, innovation and compassion to transform healthcare delivery,” said Dr. Boustani. “We have shown that the power of AI and patient-reported outcomes can be brought directly into the clinic, seamlessly, affordably, and at scale.”