Related Experiment Video
Updated: Aug 6, 2026
12:18
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Adaptive pathways for multimodal community-based detection of cognitive impairment: the CogScreen I study
Carolin Kurz1,2, Nikola Clara-Sophie Wüsten3, Paulina Tegethoff3
1Technical University of Munich, TUM School of Medicine and Health, Department of Psychiatry and Psychotherapy, TUM University Hospital, Munich, Ismaninger Str. 22, 81675, Munich, Germany. carolin.kurz@mri.tum.de.
Scientific Reports
|July 22, 2026
Summary
Early dementia detection is feasible and acceptable in community settings using a tiered approach. Combining subjective, digital, and biological measures helps identify cognitive decline risk profiles in older adults.
Area of Science:
- Gerontology
- Neurology
- Public Health
Background:
- Early detection of cognitive impairment is crucial for dementia prevention and care.
- Current implementation in primary and community settings is limited.
- Dementia-prepared health systems need scalable, adaptive detection pathways.
Purpose of the Study:
- To assess the feasibility and acceptability of a tiered, community-based cognitive decline detection framework.
- To explore the integration of subjective, digital, and biological indicators for dementia risk profiling.
- To develop a hypothesis-generating adaptive detection model for cognitive impairment.
Main Methods:
- A cluster-randomized trial (CogScreen I) involving 473 community-dwelling older adults (≥60 years) with subjective cognitive concerns.
- Randomization to three arms: Subjective Cognitive Decline Questionnaire (SCD-Q) only, SCD-Q + digital testing, or SCD-Q + digital testing + blood biomarkers.
- Feasibility and acceptability assessed via questionnaires and interviews; secondary analysis of cognitive structures and biomarker associations.
Main Results:
- High feasibility and acceptability reported by participants and general practitioners.
- Digital testing and biomarkers identified distinct cognitive dimensions (memory, attention) and pathologies (glial, amyloid).
- Multimodal clustering revealed three dementia risk profiles, informing an adaptive, tiered detection model. SCD-Q strongly correlated with subjective symptom burden.
Conclusions:
- Community-based, tiered cognitive decline detection is feasible, acceptable, and meaningful for older adults.
- Integrating subjective, digital, and biological measures supports individualized assessment strategies.
- Enhanced primary care integration is needed for downstream diagnostics and prevention; the framework requires prospective validation.
