Related Experiment Video
Updated: Jul 15, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Yield of a new method to detect cognitive impairment in general practice
Aaltje P D Jansen1, Hein P J van Hout, Giel Nijpels
1EMGO Institute, VU University Medical Center, Amsterdam, The Netherlands.
Objective:
To examine whether a new screening method that identified patients with cognitive impairment who needed further examination on the presence of dementia yielded patients who were not detected by their general practitioner (GP), and to identify factors associated with GPs' awareness of patients identified by the screening.
Methods:
Cross-sectional comparison between two methods used to identify dementia symptoms: (1) usual identification of dementia by GPs; (2) a two-stage screening to identify cognitive impairment. The two methods were implemented on the same older general practice population. The study was set in primary care practices in the Netherlands. The participants were 44 GPs and 2,101 general practice patients aged 75+ who lived at home. The following measurements were used: (1) yield of the screening; (2) determinants of GPs' awareness of patients identified by the screening.
Results:
The two-stage screening yielded 117 patients with cognitive impairment who needed further examination; in most cases (n = 82, 70.1%) their GP was unaware of the symptoms. Among patients identified by the screening, GPs' awareness was associated with co-morbidity of chronic diseases [odds ratio (OR) = 3.19; 95% Confidence Interval (CI) = 1.25 to 8.15], depressive symptoms (OR = 0.41; 95% CI = 0.17 to 0.99), and cognitive functioning (per point on the MMSE, OR = 0.88; 95% CI = 0.79 to 0.98).
Conclusion:
A two-stage screening method and increased alertness for cognitive impairment and dementia among patients with depressive symptoms may improve detection rate of dementia in general practice.
