Exploratory factor analysis of the WAIS-III thai version in patients with mild cognitive impairment

Abstract

Objective: To investigate the generalizability of the Wechsler Adult Intelligence Scale-III (WAIS-III) factors structure in a sample of mild cognitive impairment (MCI) patients. Materials and Methods: A retrospective study of patients older than 60 years receiving treatments at the Memory Clinic in Ramathibodi Hospital between January 2015 and July 2018 was conducted. The WAIS-III Thai version was used to measure the performance. All subtests except the Vocabulary and Object Assembly subtests were included for analyses. These 12 subtests were subjected to a principal axis factor analysis with oblique rotation and four factors were specified to be retained. Results: Out of 145 patients, 51% were female and 49%were male. The Full-Scale IQ ranged from 80 to 123 with the mean of 93.88 (SD 9.12). The mean of each subtest ranged from 7.82 (Similarities) to 10.45 (Digit Span), with the standard deviations ranging from 1.80 to 2.86. Based on the order of extraction and minimum loading criterion, results supported a four-factor solution composed of Verbal Comprehension, Perceptual Organization, Working Memory, and Processing Speed. The coefficients of congruence across groups on the four factors ranged from 0.92 for Perceptual Organization to 0.73 for Processing Speed. However, the Picture Completion subtest was found to have similar loadings between Perceptual Organization and Processing Speed, with the primary loading being on Perceptual Organization and secondary loading on Processing Speed. On the Working Memory factor, the MCI group showed areas of relatively greater divergence for Picture Arrangement, Block Design, and Digit Symbol: Coding subtests. Conclusion: Four factors from WAIS-III could be applied to patients with MCI. The decline of cognitive functions, particular in working memory, might be the explanation for the difference in loading factor among some subtests. Therefore, it might be possible to apply other models with other factors to the same patient group in the future.