(Photo by pics five on Shutterstock)
The BioCog assessment, validated in hundreds of primary care patients, proved more accurate than traditional screening and physician evaluation.
In A Nutshell
- A digital, 11-minute iPad test (BioCog) identified cognitive impairment with 85–90% accuracy in primary care patients.
- BioCog outperformed traditional tests (MMSE, MoCA, Mini-Cog, CANTAB) and even physician judgment (73% accuracy).
- When paired with the PrecivityAD2 blood test, accuracy for Alzheimer’s diagnosis rose to 90%.
- The study highlights potential for earlier, more reliable detection in primary care but notes limitations, including validation beyond Swedish populations and reimbursement hurdles.
LUND, Sweden — An 11-minute tablet test can detect cognitive impairment with up to 90% accuracy, and when paired with a blood test for Alzheimer’s biomarkers, it reached 90% accuracy in diagnosing Alzheimer’s disease. In primary care settings, this approach outperformed the traditional methods family doctors currently use.
Currently, diagnosing Alzheimer’s in primary care is notoriously unreliable, with doctors missing or misdiagnosing the condition about 40% of the time. This breakthrough could transform how the most common form of dementia is detected, potentially connecting patients to newly available treatments faster than ever before.
How the iPad Alzheimer’s Test Works
BioCog solves major problems with current cognitive testing in primary care. Traditional paper-and-pencil tests like the Mini-Mental State Examination require healthcare staff to administer them, leading to inconsistent results depending on who gives the test and how they score it. Many family doctors also lack the time or specialized training needed for thorough cognitive evaluations.
The tablet-based test eliminates these variables because patients take it independently. They complete three main tasks: memorizing and recalling a list of words, completing cognitive speed exercises, and answering questions about the current date and time. The entire process takes about 11 minutes and requires no staff involvement.
Dr. Pontus Tideman and his research team at Lund University in Sweden developed the test using 223 patients from specialized memory clinics, then tested its real-world performance with 403 patients from 19 primary care centers where most people first report memory problems.

Digital Assessment Beats Traditional Cognitive Tests
BioCog consistently outperformed every comparison test. When measured against widely used cognitive assessments, the digital tool beat the Mini-Mental State Examination, Montreal Cognitive Assessment, and Mini-Cog tests that family doctors typically rely on. It also exceeded the performance of CANTAB, another digital assessment already approved for clinical use.
More importantly, BioCog proved more accurate than doctors’ clinical judgment. Primary care physicians using their standard evaluation methods, including brief cognitive tests, brain scans, and clinical assessments, achieved 73% accuracy in identifying cognitive impairment (when brain function declines enough to affect daily activities). BioCog’s 85% accuracy using a single threshold represents a substantial improvement that could prevent both missed diagnoses and false alarms.
The research team also tested a refined approach that creates three categories instead of a simple pass-fail result: clearly unimpaired, clearly impaired, and uncertain cases requiring further evaluation. Using this method, BioCog achieved 90% accuracy while identifying 18% of patients as needing additional assessment.
Blood Test Combination Achieves 90% Accuracy for Alzheimer’s Detection
The most compelling results emerged when researchers paired BioCog with a blood test measuring proteins associated with Alzheimer’s disease. This two-step process first screens for cognitive impairment with the digital test, then tests for Alzheimer’s biomarkers (biological indicators of disease) in those who screened positive.
The blood test, called PrecivityAD2, measures levels of specific tau and amyloid proteins that accumulate in the brains of people with Alzheimer’s disease. When used alone, this blood test achieved 80% accuracy in diagnosing Alzheimer’s. However, when preceded by the cognitive screening step, the combined approach reached 90% accuracy.
This precision matters enormously because false positive results can cause unnecessary anxiety and lead to inappropriate treatment decisions. Blood tests for Alzheimer’s work best when applied to people who already show signs of cognitive decline, rather than those with only subjective memory concerns.
Challenges in Bringing Digital Alzheimer’s Testing to Primary Care
The study’s design reflects realistic primary care scenarios. Participating patients were those whom doctors suspected might have neurodegenerative disease after initial evaluation, not random screening of healthy individuals. People whose symptoms were clearly attributable to other causes like depression or sleep disorders received appropriate treatment for those conditions instead.
Several limitations require consideration. The research, published in Nature Medicine, was conducted entirely in Swedish populations, and while the test has been translated into multiple languages including English, broader validation remains necessary. The study focused on detecting current cognitive problems rather than tracking disease progression over time.
Healthcare systems also face practical hurdles before widespread use. Providers need training on interpreting results, and insurance coverage for combining digital cognitive tests with blood biomarker analysis remains uncertain. The newly available Alzheimer’s treatments that make early detection more valuable, including monoclonal antibody therapies like aducanumab and lecanemab, require specialized monitoring and carry significant side effect risks.
A simple tablet test that takes 11 minutes could help ensure people with Alzheimer’s are diagnosed early enough to access treatments that may slow cognitive decline, rather than missing that critical window.
Test Still Experimental, Not Yet Available to Patients
BioCog remains in the research phase and is not available for clinical use. The authors describe their work as a “proof-of-concept study” that requires several additional steps before the test can be offered to patients in healthcare settings.
The research team identifies multiple hurdles that must be cleared first: validation studies in more diverse populations beyond Swedish participants, integration protocols for different healthcare systems, provider training programs, cost-effectiveness analyses, and insurance coverage determinations. The test has been translated into English, Finnish, Dutch, and Spanish, but researchers emphasize that broader validation across different cultures and healthcare systems remains necessary.
The authors state that “further validation in additional research studies and integration into structured care pathways is essential before deployment.” They also call for “future implementation-based research in different healthcare systems” to evaluate how the test would work in practice.
No timeline for clinical availability has been provided. The path from promising research to widely available medical tool typically involves additional clinical trials, regulatory review, and healthcare system adoption processes that can take several years.
Paper Summary
Methodology
Researchers developed BioCog, a self-administered digital cognitive test requiring approximately 11 minutes to complete on a tablet. The test includes word list memory tasks, cognitive processing speed exercises, and time orientation questions. They first established the test’s parameters using 223 patients from specialized memory clinics, then validated performance in 403 patients from 19 primary care centers in Sweden. All participants underwent thorough neuropsychological testing to determine actual cognitive status, and a subset received blood biomarker analysis for Alzheimer’s disease proteins.
Results
BioCog achieved 85% accuracy in detecting cognitive impairment in primary care patients, outperforming standard physician assessment (73% accuracy) and traditional paper-and-pencil cognitive tests. When combined with blood biomarker testing, the approach identified clinical Alzheimer’s disease with 90% accuracy, substantially better than physician evaluation alone (70%) or blood testing alone (80%). A two-cutoff approach further improved accuracy to 90% for cognitive impairment detection while identifying 18% of patients needing additional evaluation.
Limitations
The study was conducted exclusively in Swedish populations, limiting generalizability to other demographics and healthcare systems. Researchers used different cognitive assessment methods between the development and validation groups, creating some methodological inconsistency. The test focused on detecting current cognitive impairment rather than predicting future decline or monitoring progression over time. Real-world implementation challenges including cost-effectiveness, provider training, and integration with existing healthcare workflows were not addressed.
Funding and Disclosures
Research was supported by multiple organizations including the National Institute of Aging, European Research Council, Alzheimer’s Association, and various Swedish foundations. Several authors disclosed consulting relationships with pharmaceutical companies developing Alzheimer’s treatments, and two authors are employees of C2N Diagnostics, which developed the blood biomarker test used in the study.
Publication Information
Published in Nature Medicine on September 15, 2025, authored by Pontus Tideman, Oskar Hansson, and colleagues from Lund University and other institutions. The study was conducted between February 2022 and December 2024 as part of the ongoing BioFINDER-2 and BioFINDER-Primary Care studies registered under clinical trial numbers NCT03174938 and NCT06120361.







