跳到正文
原文
Medical Xpress·· 2026-10-07AI 评分55

电子病历数据结合简短问卷,基层认知障碍筛查模型分类正确率74%与70%

Improving early detection of cognitive impairment in primary care

AI 导读

Regenstrief Institute 团队在印第安纳和佛罗里达的基层诊所测试一种结合电子病历数据与简短患者问卷的认知障碍筛查模型,合并模型对参与者的分类正确率分别为74%和70%。研究纳入321名65岁以上、此前无认知障碍诊断的成人,模型不用于诊断,而是帮助基层团队优先安排进一步评估或转诊。研究团队表示该模型未达到全部预设性能目标,广泛使用前还需在更多医疗系统中验证。

正文
screening healthcare
Credit: Unsplash/CC0 Public Domain

Early detection of cognitive impairment can connect patients with evaluation and treatment sooner, but many screening methods are difficult to incorporate into routine primary care. A study led by Regenstrief Institute research scientist Arthur Owora, Ph.D., MPH, evaluated a scalable approach that combines electronic health record data with a brief patient questionnaire to identify patients who may benefit from further cognitive assessment.

More than 90% of people with mild cognitive impairment and approximately half of Americans living with Alzheimer's disease and related dementias are not recognized by health care systems. When cognitive impairment is identified, diagnosis may occur two to five years after symptoms begin.

Primary care offers an important opportunity for earlier identification, but clinicians face limited time, reimbursement constraints and restricted access to specialty evaluation. Existing screening methods may also be expensive, invasive or difficult to integrate into routine care.

"Primary care clinicians need practical tools that can help them recognize patients who may be experiencing cognitive decline without creating an additional burden for patients or care teams," said Owora. "By combining information already available in the health record with the patient's perspective, we may be able to better identify who should receive a more comprehensive assessment."

The researchers studied 321 adults age 65 and older who did not have a previously documented diagnosis of cognitive impairment. Participants received care through five Eskenazi Health federally qualified health centers in Indianapolis and nine University of Miami primary care practices in South Florida. The findings are published in the journal Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring.

The team evaluated a multimodal screening model that combined:

  • A passive digital marker using routinely collected electronic health record data.
  • The Quick Dementia Rating System, a brief patient questionnaire about cognitive and functional changes.
  • Patient characteristics and health conditions associated with cognitive decline, including age, sex, race, cerebrovascular disease, diabetes, depression and hypertension.

An interdisciplinary team consisting of a neurologist, geriatrician and neuropsychologist reviewed comprehensive assessment results to determine whether each participant had normal cognition, mild cognitive impairment or dementia. Researchers then compared those determinations with the results produced by the screening approaches.

The electronic health record–based marker and patient questionnaire each demonstrated limited accuracy when used alone. Combining them with relevant patient and clinical characteristics improved the model's ability to distinguish between patients with and without cognitive impairment.

The combined model correctly classified 74% of participants in Indiana and 70% in Florida. The model is not intended to diagnose cognitive impairment. Instead, it could help primary care teams prioritize patients for comprehensive testing or referral to a specialist.

The approach uses information health systems already collect and requires only a brief additional questionnaire, making it potentially less burdensome than screening strategies that rely on extensive cognitive testing, imaging or biomarkers.

"This model is intended to support clinical decision-making, not replace a clinician or a comprehensive diagnostic assessment," said Owora. "The goal is to give primary care teams another source of evidence to help direct limited diagnostic and specialty resources to the patients most likely to benefit."

Differences between the Indiana and Florida results reinforced the importance of evaluating clinical models across different patient populations and health care environments. Documentation practices, patterns of health care use, access to specialty care and differences among patient populations can all affect model performance.

Although the combined approach performed better than either screening method alone, it did not meet all performance targets established by the researchers. Additional validation and recalibration across diverse health care systems will be necessary before the model can be broadly implemented.

More information

Arthur H. Owora et al, Optimizing scalable approaches for early detection of cognitive impairment in primary care, Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring (2026). DOI: 10.1002/dad2.70393

Key medical concepts

Cognitive Dysfunction

Who's behind this story?

Gaby Clark

Gaby Clark

MA in English, copy editor since 2021 with experience in higher education and health content. Dedicated to trustworthy science news. Full profile →

Andrew Zinin

Andrew Zinin

Master's in physics with research experience. Long-time science news enthusiast. Plays key role in Science X's editorial success. Full profile →

Citation: Improving early detection of cognitive impairment in primary care (2026, October 7) retrieved 7 October 2026 from https://medicalxpress.com/news/2026-10-early-cognitive-impairment-primary.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

来源:Medical Xpress · medicalxpress.com