2013 Volume 2 Issue 4 Pages 145-148
[Purpose] We evaluated the reproducibility and validity of our “cognitive function test using shiritori". [Subjects and Methods] The subjects consisted of 71 elderly outpatients (mean age, 82.4±8.7 years) visiting an outpatient rehabilitation facility. As a task, a word was presented using a word table we developed, and the subject said a word starting with the last syllable of the presented word, and this procedure was repeated. The reproducibility of this test was analyzed using the intraclass correlation coefficient (ICC),and its validity was evaluated using analysis of the correlation with MMSE. [Results] The reproducibility of the shiritori cognitive function test was markedly high (ICC=0.915), and a moderate-strong significant correlation was observed between this test and MMSE irrespective of sex. According to the severity of cognitive impairment, there was a moderate correlation between this test and MMSE in the groups with moderate or severe cognitive impairment but only a weak correlation in the group with mild impairment andthe healthy group. [Conclusions] The cognitive function test using shiritori has a high reproducibility, suggesting its clinical applicability as a simple cognitive function test, but its detection ability was limited in patients with mild cognitive impairment.