常用於中風病人臨床試驗研究之執行功能成效評估工具及其心理計量特性回顧

Executive function (EF) deficit is a common cognitive problem in stroke patients, which negatively influences patients' activities of daily-living functions and rehabilitations. EF outcome measurements with good psychometric properties and feasibilities help to identify patients' EF problems and to evaluate the effectiveness of treatments. This study aimed to review and appraise the psychometric properties (i.e., reliability, validity, responsiveness, and minimal important difference, MID) and feasibilities (e.g., measurement materials, assessment time, and culture/language barriers of measurements) of the frequently used (used at least 6 times) EF outcome measurements in empirical studies of stroke. The research consisted of empirical studies that were published from January 2010 to December 2017. According to the citation criteria, a total of 8 EF tests in 3 databases (Medline, PsycINFO, and Index to Taiwan Periodical Literature System) were included: Trail-making test A & B (TMT-A & TMT-B), Stroop test, category verbal fluency test, phonemic verbal fluency test (Phonemic VFT), digit-span test-backward, Wisconsin card sorting test (WCST), and Tower of London (TOL) test. However, the psychometric properties of these measurements were rarely examined. Only two studies examined tests' construct validities. They showed that WCST performance is associated with three EF abilities (mental flexibility, error correction, and maintenance) and that TOL performance is related to EF's planning ability. Regarding to the feasibilities of the eight measurements, TMT-A and Phonemic VFT had better feasibilities (short assessment time and a Chinese version). It is recommended that future studies should exam the psychometric properties of these eight tests in order to provide references for selecting a good EF measurement.