Validity and reliability of Mini-Mental State Examination in Older Adults in China: Inline Mini-Mental State Examination with cognitive functions

Abstract

The main aim of the study is to validate the factor structure of the Mini-Mental State Examination (MMSE) of China’s older population using the Chinese Longitudinal Healthy Longevity Survey. The validation process used the exploratory factor analysis (EFA) to determine the number of dimensions of MMSE, the confirmatory factor analysis (CFA) to confirm the factorial structure of MMSE, and the factorial invariance to conclude the factor structure does not differ between the young-old (aged 65 – 79) and old-old (aged 80 or older). The results of the EFAs suggested two possible factor structures: A six-factor and a seven-factor solution. The seven-factor confirmatory factor model turned out as the best fit by comparison to the four competing confirmatory models. Strict factorial invariance was attained for the two age groups, indicating a high level of measurement equality, a property of invariance was seldom achieved in the literature of factorial invariance studies. In comparison to the MMSE literature that focused solely on EFA that aims to establish a single summated score, the present study suggests using EFA, CFA, and factorial invariance that takes into consideration of measurement errors as the preferred procedure since it establishes the appropriate MMSE dimensionality that is in line with their respective cognitive functions.

Serum Vitamin D3 Concentration, Sleep, and Cognitive Impairment among Older Adults in China

“…The Chinese version of the Mini-Mental State Examination (MMSE) is adapted from Folstein's original scale to be more practical for older adults and patients with dementia in China, where the literacy level is relatively low [47,52]. The Chinese version of MMSE is composed of 23 questions from six domains (orientation, registration, attention and calculation, copy and drawing, recalling, and language) with a range of scores 0-30 [47].…”

“…The Chinese version of MMSE is composed of 23 questions from six domains (orientation, registration, attention and calculation, copy and drawing, recalling, and language) with a range of scores 0-30 [47]. The data quality of this indicator is relatively high [52]. Unlike objective questions that can be answered by informants, all MMSE questions have to be answered by the sampled older adults themselves.…”

“…The imputation assumes that answers for MMSE questions would be the same for older persons with missing values as those who did not have missing values on MMSE questions if the former group has the same demographics, socioeconomic status, family/social support, health behaviors, and physical health conditions as the latter group. Following the original criterion, participants with an MMSE score below 24 were considered cognitively impaired, while those with scores of 24 or higher were considered cognitively unimpaired [47,52].…”

Efficacy of multi-domain cognitive function training on cognitive function, working memory, attention, and coordination in older adults with mild cognitive impairment and mild dementia: A one-year prospective randomised controlled trial

“…Studies have noted that the MMSE is easy to use and has high reliability and validity. The sensitivity is 0.88 and 0.84, the ratio is 0.70 and 0.86, the area under the curve to distinguish mild cognitive impairment from mild dementia is 0.88 and 0.89, the positive predictive values are 0.94 and 0.80, the negative predictive values are 0.81 and 0.88, Cronbach's alpha is 0.68 to 0.96, the test-retest reliability is 0.74 to 0.99, and the interrater reliability is 0.83 [22][23][24].…”

Does Social Participation Modify the Association Between Depression and Cognitive Functioning among Older Adults in China? A Secondary Analysis Based on CHARLS

“…It is now widely used as a measure to assess older adults' cognitive functioning in research environments and performed in a systematic and routine manner as a screening tool in clinical and community settings to identify cognitive impairment in older, community-dwelling, hospitalized, and institutionalized adults [45]. The CHARLS dataset used the Chinese version of the MMSE, which shows good validity and reliability (Cronbach's α=0.86) [46,47,48]. In this study, Cronbach's α coefficient was .75.…”