A Practical Tool for the Detection of Limbic-type-memory Deficit: 5-Word-Test

Gül Kayserili; Zerrin Yıldırım; Özgül Vupa Çilengiroğlu; Görsev G. Yener

doi:10.4274/tnd.2019.39269

Gül Kayserili¹, Zerrin Yıldırım², Özgül Vupa Çilengiroğlu³, Görsev G. Yener⁴

¹Izmir Narlidere Nursing Home Elderly Care and Rehabilitation Center, Izmir, Turkey
²Bagcilar Training and Research Hospital, Clinic of Neurology; Istanbul University Aziz Sancar Experimental Medicine Research Institute, Department of Neuroscience, Istanbul, Turkey
³Dokuz Eylul University Faculty of Science, Department of Statistics, Izmir, Turkey
⁴Dokuz Eylul University Faculty of Medicine, Department of Neurology, Izmir, Turkey

Keywords: Five-Word-Test, Alzheimer’s disease, mild cognitive impairment, verbal memory

Abstract

Objective: The 5-Word-Test (5WT) is a quick and practical verbal memory assessment tool that was developed by Dubois et al. The objective of this study was to establish the validity and reliability of its Turkish version with the purpose of acquiring a practical memory testing instrument that would discriminate the stages of the normal aging - mild cognitive impairment (MCI) - early Alzheimer-type dementia (ATD) continuum with high sensitivity and specificity.

Materials and Methods: One hundred six healthy participants (HP), 22 with MCI and, 39 patients with ATD consented to participate in the study. Participants were evaluated using the 5WT, Standardized Mini-Mental State Examination, verbal fluency and 3 Words-3 Shapes tests. The simultaneous validity tests of the 5WT with other cognitive tests and test-retest validity of the test were investigated using Spearman’s rho correlation analysis. Parametric and non-parametric tests were used for group comparisons. For the discriminant ability of the 5WT, the Receiver Operating Characteristic (ROC) area under curve was measured.

Results: In 5WT cued recall score (CRS), all HPs, and in free recall score (FRS) 95.3% of them received 10 points. The 5WT score differences between HPs and the combined patient groups, and among MCI and ATD severity stage groups were significant (p<0.01). According to the ROC analysis, FRS and CRS had high sensitivity in discriminating HPs from patients, HPs from the MCI and MCI groups from the ATD. Inter-observer and test-retest correlations were quite high (r=0.85, r=0.95 respectively). The Cronbach alpha value of the test was found as 0.94.

Conclusion: In this study, we have shown that the Turkish version of 5WT is a valid and reliable test for HPs, and patients with ATD and MCI. Due to the ability of 5WT to discriminate among the stages of normal aging-ATD continuum, it may prove to be a practical tool to be used in specialized dementia out-patient clinics in our country.