Objectives: Impairments in the Boston Naming Test (BNT), which measures confrontational word retrieval, frequently accompanies Alzheimer's dementia (AD) and may predict a more rapid progression of illness. This study aims to validate the Thai version of the 15-item BNT (T-BNT) in participants with AD and amnestic mild cognitive impairment (aMCI) and to externally validate the T-BNT using clinical and biomarker measurements.
Methods: This cross-sectional study recruited patients with AD, diagnosed according to NINCDS-ADRDA criteria (n = 60), aMCI, diagnosed using the Petersen criteria (n = 60), and healthy controls (n = 62). We examined the internal consistency, concurrent and discriminant reliability of the T-BNT. We also assessed the Mini Mental State Examination (MMSE), the Verbal Fluency Test (VFT) and the Word List Memory (WLM) tests and measured apolipoprotein E polymorphism and serum levels of folic acid, high-density lipoprotein cholesterol (HDL) and triglycerides.
Results: This study validated a 10-item T-BNT (10T-BNT), which yielded good internal consistency (0.92), a one-factor unidimensional structure, and adequate concurrent and discriminant validity. Lower scores on the 10T-BNT highly significantly predict AD, but not aMCI, and are positively associated with VFT and WLM test scores. Furthermore, lowered 10T-BNT scores are significantly associated with the ApoE4 allele, lower folate levels and an increased triglyceride/HDL-cholesterol ratio.
Conclusions: This study validated the 10T-BNT and the total score on this scale is strongly associated with AD, impairments in semantic and episodic memory and biomarkers, which are known to modify memory via different mechanisms.
Keywords: ApoE4; cognitive; dysnomia; episodic memory; folate.