The accuracy study of the Thai word list learning test

บทคัดย่อ

Objective  To develop the Thai word list learning test for screening dementia in the elderly.

Materials and methods  This study had 2 steps. 1) To develop the word list learning test (WLL) Thai version and try out among normal elderly  2) To study the learning scores, forgetting rate, and words recognition differentiation between dementia patients and normal people in order to find the appropriate cut-off point by using the receiver operating characteristic curve (ROC). The case group was the elderly who came for treatment at Srithanya Hospital. This group was diagnosed as dementia according to the DSM-4 criteria. The control group was the non-dementia elderly recruitted by a neurologist. The statistics used were t-test, Mann-Whitney U test, and ROC. The correlation coefficient between MMSE-Thai 2002 and word list learning test Thai version were compared by using Pearson’s correlation and Spearman rank correlation coefficients. November 2012 – December 2014.

Results The sample was 77 elderly. There were female about 60%, an average aged was 72.8 years and 7 years of education. There were 44 elderly in the control group and 33 elderly with dementia. The rate of forgetting in delayed recall had highest accurately prediction in separating between the control group and the dementia group. The appropriate cut-off point was ≤-45 with the specific sensitivity were 79.3% and 73.8% respectively. The scores derived from MMSE-Thai 2002 and  word list learning test Thai version, had statistical significant correlation.

Conclusion  The Thai word list learning test, can be used to screen dementia in elderly with the optimal cut-off point was the rate of forgetting at ≤-45.