Related Experiment Video
Updated: Aug 18, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Simplifying detection of cognitive impairment: comparison of the Mini-Cog and Mini-Mental State Examination in a
Soo Borson1, James M Scanlan, Jill Watanabe
1Alzheimer's Disease Research Center, University of Washington, Seattle, Washington, USA. soob@u.washington.edu
Objectives:
To compare detection of cognitive impairment using the Mini-Cog and Mini-Mental State Examination (MMSE) and to identify sociodemographic variables that influence detection in an ethnoculturally diverse sample.
Design:
Cross-sectional.
Setting:
A registry of the University of Washington Alzheimer's Disease Research Center Satellite.
Participants:
A heterogeneous community sample (n=371) of predominantly ethnic minority elderly assessed using a standardized research protocol, 231 of whom met criteria for dementia or mild cognitive impairment (MCI).
Measurements:
Demographic data, a standardized research protocol for cognitive assessment and dementia diagnosis, MMSE, and Mini-Cog.
Results:
Both screens effectively detected cognitive impairment, the Mini-Cog slightly better than the MMSE (P<.01). Overall accuracy of classification was 83% for the Mini-Cog and 81% for the MMSE. The Mini-Cog was superior in recognizing patients with Alzheimer-type dementias (P=.05). Low education negatively affected detection using the MMSE (P<.001), whereas education did not affect the Mini-Cog, and low literacy minimally affected it.
Conclusion:
The Mini-Cog detects clinically significant cognitive impairment as well as or better than the MMSE in multiethnic elderly individuals, is easier to administer to non-English speakers, and is less biased by low education and literacy.
