[Japanese version of the Short-Memory Questionnaire and Mini-Mental State Examination in relation to demographic variables: community survey of elderly healthy residents in Nakayama, Japan].

ABSTRACT

BACKGROUND AND PURPOSE: The Short-Memory Questionnaire (SMQ) developed by Koss et al. is a standardized, validated reliable informant-based scale to assess everyday memory problems. In the previous study, we prepared its Japanese version and validated the reliability in assessing patients with Alzheimer Disease. In the present study, we examined the relations between the performance evaluated by the Japanese version of the SMQ combined with the Mini-Mental State Examination (MMSE) and demographic variables (age, sex, and education). SUBJECTS AND METHODS: Subjects were a total of 613 elderly healthy individuals living in the rural residency of Nakayama, Japan. They consisted of 287 males and 326 females and ranged from 65 to 94 years in age and from 0 to 16 years in education. The MMSE was conducted by neuropsychiatrists to assess the severity of cognitive impairment, and the SMQ was conducted by public health nurses, who asked a family member of each subject. The effects of age, sex and education on the MMSE and SMQ scores were analyzed by analysis of variance (ANOVA) using post-hoc Scheffé test. RESULTS: The mean MMSE and SMQ scores (+/- SD) were 27.6 (+/- 2.5) and 44.5 (+/- 2.1), respectively. The post-hoc Scheffé test revealed significantly higher MMSE scores for individuals with 8 years or more in education than for those with less than 8 education years. Between males, no significant differences were observed in MMSE score; however, among females, significantly higher scores were noted in those aged 69 or younger than in those aged 70-74 or those aged 80 or older. Of individuals aged 80 or older, males showed significantly higher scores than females, whereas no significant differences were observed between sex in other age groups. As to the SMQ, higher scores were shown by persons with longer education years and by females than males. CONCLUSION: Low MMSE scores were related to low education level and older females. Low SMQ scores were not related to aging, but linked to low education level and males. Both the MMSE and SMQ, which can be easily administered, are affected by education years and sex. In addition, MMSE score depends on age. Therefore, careful consideration should be paid to age, education level and sex for future execution of the MMSE and SMQ.