Related Experiment Video
Updated: Jun 4, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Screening for cognitive impairment in an elderly veteran population: acceptability and results using different
J Riley McCarten1, Pauline Anderson, Michael A Kuskowski
1Geriatric Research, Education and Clinical Center, Minneapolis Veterans Affairs Medical Center, Minneapolis, Minnesota, USA.
Objectives:
To assess the feasibility of cognitive screening in older veterans presenting for routine primary care.
Design:
Quality improvement initiative.
Setting:
Seven Veterans Affairs Medical Centers.
Participants:
Veterans aged 70 and older without a prior diagnosis of cognitive impairment.
Measurements:
Scores from 16 different versions of the Mini-Cog, a 5-point cognitive screen consisting of three-word recall (0-3 points) with a clock-drawing interference task (0 or 2 points). Five word lists were paired with three clock times and randomly ordered for presentation with the original Mini-Cog version. The conventional dementia screening cut point was increased to maximize sensitivity; to pass, patients had to draw the clock correctly and recall at least two of three words (score 4 or 5/5).
Results:
Administering the Mini-Cog took 90 to 180 seconds. Of 8,342 veterans approached, 8,063 (96.7%) agreed to be screened; 2,081 (25.8%) scored less than 4 out of 5. Scores declined with age, but age did not predict pass or fail. Different word lists produced different screen failure rates, ranging from 21.2% to 33.4%. Five dementia specialists were unable to distinguish harder from easier lists. Different clock times accounted for 2% or less of the difference in failure rates.
Conclusion:
The Mini-Cog was quick and well accepted by older veterans. Many with no prior documentation of cognitive impairment failed the screen. Failure rates varied with the word list used, revealing that even apparently minor changes in test items affect screen results. Additional study is needed to establish the value of cognitive screening in shaping primary care of older veterans.
