Premium
Detailed analysis of the J apanese version of the R apid D ementia S creening T est, revised version
Author(s) -
Moriyama Yasushi,
Yoshino Aihide,
Muramatsu Taro,
Mimura Masaru
Publication year - 2017
Publication title -
psychogeriatrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.647
H-Index - 32
eISSN - 1479-8301
pISSN - 1346-3500
DOI - 10.1111/psyg.12248
Subject(s) - clinical dementia rating , dementia , numeral system , frontotemporal dementia , task (project management) , psychology , medicine , audiology , disease , computer science , artificial intelligence , management , economics
Aim The number‐transcoding task on the J apanese version of the R apid D ementia S creening T est ( RDST‐J ) requires mutual conversion between A rabic and C hinese numerals (209 to 二百九, 4054 to 四千五十四, 六百八十一 to 681, 二千二十七 to 2027). In this task, question and answer styles of C hinese numerals are written horizontally. We investigated the impact of changing the task so that C hinese numerals are written vertically. Methods Subjects were 211 patients with very mild to severe A lzheimer's disease and 42 normal controls. M ini‐ M ental S tate E xamination scores ranged from 26 to 12, and C linical D ementia R ating scores ranged from 0.5 to 3. Results Scores of all four subtasks of the transcoding task significantly improved in the revised version compared with the original version. The sensitivity and specificity of total scores ≥9 on the RDST‐J original and revised versions for discriminating between controls and subjects with C linical D ementia R ating scores of 0.5 were 63.8% and 76.6% on the original and 60.1% and 85.8% on revised version. Conclusions The revised RDST‐J total score had low sensitivity and high specificity compared with the original RDST‐J for discriminating subjects with C linical D ementia R ating scores of 0.5 from controls.