The clock drawing test as a screening tool for detecting cognitive decline: an analysis in adults and elderly people...

The Clock Drawing Test (CDT) is widely used as a screening instrument for cognitive decline due to its simplicity and rapid administration. Despite its widespread clinical use, evidence regarding its diagnostic performance in the Brazilian Northeast region remains scarce.

This study examined CDT performance in a clinical sample of 113 adults and older adults assessed at a neuropsychology service in Northeastern Brazil, focusing on CDT ability to identify cognitive decline and the effects of age, educational level, and clinical diagnoses. The study tested the following hypotheses: (1) there is a significant association between age and CDT scores; (2) there are significant differences in CDT scores across different educational levels; (3) there are significant differences in CDT scores across different clinical conditions, indicating potential for differential diagnosis; and (4) the test would demonstrate high sensitivity, specificity, and accuracy in the overall sample, as well as high sensitivity in detecting each diagnostic condition. Using Shulman's scoring method, descriptive analyses, Spearman's correlation, Kruskal-Wallis tests and metrics of sensitivity, specificity, and accuracy were conducted.

The sample had a mean age of 65.19 years and was predominantly characterized by low educational attainment. A negative, albeit weak, correlation was observed between age and CDT scores, as well as significant differences across educational levels. Diagnostic group comparisons also revealed significant differences, most notably between cognitively unimpaired individuals and patients diagnosed with Major Neurocognitive Disorder due to Alzheimer's disease. Although the CDT demonstrated adequate specificity, its overall sensitivity and accuracy were low. Sensitivity was high for Major Neurocognitive Disorder Due to Alzheimer's Disease, moderate for Major Neurocognitive Disorder due to Non-Alzheimer's Disease (Major Vascular Neurocognitive Disorder, Parkinson's Disease, Mixed Dementia, Wernicke-Korsakoff syndrome and Major Frontotemporal Neurocognitive Disorder) and low for Mild Neurocognitive Disorder.

These findings demonstrate that Shulman's method of CDT is not suitable for assessing cognitive decline in the illiterate and low-education population and raise important concerns regarding its standalone clinical utility, especially in specific neurological conditions. The present study underscores the need for future research employing alternative scoring methods and more representative samples to refine the applicability and diagnostic value of the CDT in clinical practice.