Abstract
BACKGROUND: The demand for more accurate and early diagnosis of mild cognitive impairment (MCI) patients due to Alzheimer's disease (AD) has increased after disease-modifying drugs were launched. Among these needs, there is a requirement for tools that can easily assess the ability to recall memories, which changes early in the disease. OBJECTIVES: We established Self Assessment Memory Scale (SAMS) method before, which includes 8-picture recall test and 16-word recognition test. We adopted this method to software that can be operated on a tablet computer so that participants can perform the method independently. The purpose of this study was to validate this method. DESIGN: Cross sectional research. SETTING: Some of the participants were recruited from hospitals for patients diagnosed with AD or MCI. The others were recruited from three regional cohorts of healthy older adults. PARTICIPANTS: The total number of participants was 304 (20 of whom had AD or MCI), and the mean age was 71.2 years. 64% of the participants were women. MEASUREMENTS: We used the logical memory subtest of the WMS-R as the standard for memory evaluation and assessed the relationship between this score and the SAMS score calculated by the software. RESULTS: The 2nd SAMS score were higher than the 1st SAMS score in some participants, on the other hand, the intraclass correlation coefficient was good. Since the number of false recognition in the 16-word recognition test was higher in participants with lower LM II scores, we developed a new score to reflect the ratio of false recognition, SAMS-R, and we observed it has good correlation with LM II. The mean SAMS-R score decreased gradually after the age of 65 years, indicating that age-related changes in memory recall can be detected. The ROC curve analysis was conducted to evaluate the detectability to determine whether if the WMS-R LM II score is above or below 10, showing that the AUC was greater than 0.9. CONCLUSION: SAMS-R, which can be performed on a tablet literally by himself/herself independently, shows a high correlation with the WMS-R Logical Memory II score, and has the advantage of being performed in a short time without the need for a clinical psychologist or other personnel.