A tablet-based cognitive test completed by patients themselves has proven more accurate than primary care physicians at identifying who needs further testing for Alzheimer’s disease, according to new research that could reshape how the condition is diagnosed in doctor’s offices.
The Swedish study, published in Nature Medicine, found that their BioCog digital test achieved 85% accuracy in detecting cognitive impairment among 403 primary care patients with memory concerns. By comparison, doctors using traditional methods reached only 73% accuracy.
“The vast majority of people who experience memory loss will first seek treatment at their health centre. Our new digital test provides a first objective picture – at an earlier stage and with greater precision – of which patients have cognitive impairment indicative of Alzheimer’s disease.”
The test takes about 11 minutes to complete on a tablet and measures memory, processing speed, attention, and orientation. Patients memorize ten words, recall them later, recognize them among distractors, and answer questions about the date and time. The system captures not just correct answers but also response times and search patterns – data impossible to gather with paper tests.
Filling a Primary Care Gap
The timing couldn’t be more critical. New Alzheimer’s drugs like Leqembi and Aduhelm require early detection of the disease, but primary care doctors often lack the time and specialized training needed for comprehensive cognitive assessments. A typical appointment lasts 15-20 minutes, barely enough time to conduct thorough memory testing.
Researchers at Lund University developed BioCog specifically for this challenge. Unlike existing digital tests designed for specialist clinics, BioCog was tested and validated in real primary care settings across 19 health centers in southern Sweden.
The results were striking. When researchers combined BioCog with a blood test measuring tau protein – a key Alzheimer’s biomarker – the accuracy jumped to 90% for diagnosing clinical Alzheimer’s disease. This two-step approach significantly outperformed current standard care, which achieved only 70% accuracy.
Beyond Traditional Testing
The digital approach also beat established cognitive tests. BioCog outperformed the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and Mini-Cog – the paper-and-pencil tests commonly used in primary care. It even surpassed CANTAB, an FDA-approved digital test already in clinical use.
“Primary care does not have the resources, time or specialist knowledge to investigate possible Alzheimer’s disease in the same way as specialised memory clinics. And this is where a digital cognitive test can make the biggest difference.”
The study included patients whose doctors suspected possible neurodegenerative disease but excluded those with obvious alternative explanations like depression or sleep disorders. About 57% of participants had objectively verified cognitive impairment when tested with comprehensive neuropsychological batteries.
Importantly, the test maintained high accuracy across different patient populations. When researchers used a two-cutoff approach – identifying high-confidence positive and negative results while flagging uncertain cases for further evaluation – accuracy rose to 90% in the primary care group.
The research addresses a growing clinical need. As Alzheimer’s treatments become available, healthcare systems face pressure to identify eligible patients efficiently. Blood tests for Alzheimer’s pathology now exist, but using them on everyone with memory complaints would be costly and potentially misleading, since many cognitive problems stem from other causes.
The BioCog approach offers a solution: screen patients digitally first, then use blood tests only on those who show objective cognitive impairment. This workflow could prevent false-positive diagnoses while ensuring patients who need treatment aren’t missed.
Still, questions remain about real-world implementation. The study was conducted in Sweden with Swedish-speaking patients, though translations into English, Finnish, Dutch, and Spanish are underway. Researchers also acknowledge they haven’t yet studied how the test performs when used repeatedly over time or how doctors respond to having this tool available.
The work represents a significant step toward making Alzheimer’s diagnosis more accessible and accurate in primary care – potentially transforming how millions of patients with memory concerns are evaluated and treated.
Nature Medicine: 10.1038/s41591-025-03965-4
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.