Related Experiment Video
Updated: Oct 1, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
A comparison of scoring protocols on the Clock Drawing Test in relation to ease of use, diagnostic group, and
Hayley E Richardson1, John N Glass
1Psychogeriatric Service, Taranaki Healthcare Ltd, New Plymouth, New Zealand.
Objectives:
To compare five scoring protocols for the Clock Drawing Test (CDT).
Design:
A retrospective study in which four formal and one informal scoring methods were used to rate clock drawings. Correlations between CDT scores and the abbreviated version of the Mini-Mental State Examination (MMSE-brief) total and memory scores were compared for the total sample and for three diagnostic groups. Protocols were also compared for content and ease of use.
Setting:
A comprehensive, community mental health service for older adults.
Participants:
Sixty-three patients with dementia of the Alzheimer's type, multi-infarct dementia, or a mixed diagnosis.
Measurements:
The CDT, scored under five different scoring protocols, and the MMSE-brief.
Results:
For the total sample, significant correlations were obtained between all five scoring methods and MMSE scores. Two scoring methods stood out. By diagnostic group, all correlations except one between scoring methods were significant, whereas relationships between the MMSE and the CDT varied.
Conclusions:
Deterioration in clock drawing scored under five different scoring protocols correlates with severity of global cognitive impairment, as assessed by the MMSE-brief. Although two scoring methods were easier to use, the type of dementia may dictate which scoring protocol is most suitable. Because clinicians now also use the CDT to screen for executive functioning, future scoring methods may need to be selected in terms of what the CDT is screening.
