Psychometric Characteristics of the Mini-Mental State Examination among Older Chinese Immigrants

Abstract

Backgrounds: The recent growth of older adult immigrants and the increased burden of Alzheimer’s disease and related dementias (ADRD) call for validation of the assessment tools and timely detection of cognitive impairment. In particular, older Chinese Americans are among the fastest growing populations in the U.S., who are at high risk for delayed diagnosis and deficient management of ADRD due to limited knowledge of dementias and barriers to healthcare access. The value of the MMSE as a screening tool in detecting ADRD is well documented in western countries. Yet, little is known about the psychometric characteristics of a Chinese version of the MMSE (CMMSE) in older Chinese immigrants, whose cognitive test performances may relate to cultural and linguistic differences. Purpose: The study aims to assess factor structure and psychometric properties of CMMSE among older Chinese immigrants in the United States. Methods: Participants were recruited from the greater Chicago area in the Population-based Study of Chinese Elderly (PINE). The sample included 3,126 respondents aged 60 or over, who selected Chinese dialects in completing face-to-face interviews. A bifactor analysis was conducted to test unidimensionality of the CMMSE. Rasch modeling was used to assess scale properties. Age, sex, and education were tested for differential item functioning (DIF). Results: A bifactor analysis identified a unidimensional general factor with 30 items, reflecting a single underlying construct of cognitive capacity. We also found five concurrent subdomains with 19 items. The general factor explained 71 percent of the variance in the CMMSE. The Rasch model provided evidence of construct validity with acceptable item fit statistics. The DIF analysis showed that item biases were generally negligible. Conclusion: The CMMSE appeared to be a valid, reliable screening instrument for cognitive impairment in the US Chinese older adults. Findings provided support for the use of the CMMSE in detecting cognitive impairment among older Chinese immigrants. Although the five-factor solution has been replicated, item compositions of subdomains remain controversial. Future efforts are needed to improve neuropsychological assessment accuracy by developing a culturally and linguistically appropriate assessment tool in minority older populations.