Related Experiment Video
Updated: May 10, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
The Introduction of a New Screening Tool for the Identification of Cognitively Impaired Medically At-Risk Drivers:
Bonnie M Dobbs1, Donald Schopflocher
1Department of Family Medicine, University of Alberta, Edmonton, Alberta, Canada.
Unlabelled:
The number of drivers with a cognitive impairment due to dementia or other age-associated pathologies will increase significantly over the next 3 decades. Physicians are well placed to identify medically at-risk drivers, but are hampered by the lack of a valid, easy to administer screening tool. This research develops and validates a brief screening tool for use in the primary care setting to identify drivers with cognitive impairment with or without dementia. Initial Study Participants: A cohort of 146 consecutive referrals from community-based family physicians, diagnosed with an undifferentiated cognitive impairment or dementia, as well as 35 community dwelling healthy controls. Validation Study: A cohort of 192 consecutive referrals carrying the same diagnosis as above and 52 community dwelling healthy controls. Criterion Measure: Pass/fail on an On-Road evaluation. Predictor Measures: Subtests of the DemTect, a screening test for cognitive impairment or dementia developed by Kalbe and colleagues.(1) Initial Study: Three of the DemTect measures predicted On-Road outcomes (R(2) = .262). Regression results were used to develop a simple scoring algorithm, with cut-points then derived by identifying those most at risk for failing and passing the On-Road assessment, and those needing a driving assessment for determination of driving competency. 89 individuals scored in the indeterminate range, with 49 and 43 predicted to fail and pass, respectively-86% and 84% of those predicted to fail and pass did subsequently fail and pass. Validation Study: 123 individuals scored in the indeterminate range, with 66 and 55 predicted to fail and pass, respectively-80% and 87% of those predicted to fail and pass did subsequently fail and pass.
Conclusions:
The SIMARD A Modification of the DemTect ( S creen for the I dentification of cognitively impaired M edically A t- R isk D rivers) is a brief paper and pencil screening tool with a high degree of accuracy that can be used for immediate decisions in the clinical setting.
