Validity, Reliability and Turkish Norm Values of the Clock Drawing Test for Two Different Scoring Systems

Derya Durusu Emek Savaş; Deniz Yerlikaya; Görsev G. Yener

doi:10.4274/tnd.26504

Derya Durusu Emek Savaş¹, Deniz Yerlikaya², Görsev G. Yener³

¹Dokuz Eylul University Faculty of Letters, Division of Psychology, Izmir, Turkey
²Dokuz Eylul University Institute of Health Sciences, Department of Neurosciences, Izmir, Turkey
³Dokuz Eylul University Faculty of Medicine, Department of Neurology, Izmir, Turkey

Keywords: Clock drawing test, normative data, validity-reliability, Alzheimer’s disease, mild cognitive impairment

Abstract

Objective: The clock drawing test (CDT) version with the pre-drawn circle has been widely used in research and clinical practice without standardized Turkish norms. The present study aimed to standardize CDT scores according to the most frequently used scoring methods in the literature (Manos-Wu and Shulman) and to estimate the validity and reliability of both methods.

Materials and Methods: The norm determination phase of the study was performed with 244 healthy individuals in the age range of 50-92 years. The effects of age, education, and sex on the CDT scores were examined and normative data stratified by age and education were derived. Test-retest reliability, inter-rater reliability, and concurrent validity and criterion validity of the two scoring methods were tested. Criterion validity was assessed using the ROC analysis to examine the extent to which the CDT scores could distinguish among healthy individuals, patients with mild cognitive impairment (MCI), and patients with Alzheimer’s disease (AD).

Results: In the Manos-Wu method, age and education had significant effects on CDT scores of healthy individuals; however, in the Shulman method, only an education effect was found. Both methods had high test-retest reliability (Manos-Wu, 0.81; Shulman, 0.72) and inter-rater reliability (Manos-Wu, 0.98; Shulman, 0.96) and showed high correlations with each other and with other cognitive screening tests. The area under the ROC curve had high values only in discriminating between healthy individuals and AD, and healthy individuals and MCI.

Conclusion: In the present study, detailed normative data for individuals aged 50 and older were established. Age and education levels should be taken into account when interpreting CDT scores. Both scoring methods provided similar results in terms of high test-retest and inter-rater reliability, as well as high concurrent and criterion validity.