Translation and concurrent validity, sensitivity and specificity of Chinese version of Short Orientation Memory...

This study aimed to translate the English version of the Short Orientation-Memory-Concentration (SOMC) test into a Chinese version, denoted the C-SOMC test, and to investigate the concurrent validity, sensitivity, and specificity of the C-SOMC test against a longer and widely used screening instrument in people with a first cerebral infarction.

An expert group translated the SOMC test into Chinese using a forward-backward procedure. Eighty-six participants (67 men and 19 women, mean age = 59.31 ± 11.57 years) with a first cerebral infarction were enrolled in this study. The validity of the C-SOMC test was determined using the Chinese version of Mini Mental State Examination (C-MMSE) as the comparator. Concurrent validity was determined using Spearman's rank correlation coefficients. Univariate linear regression was used to analyze items' abilities to predict the total score on the C-SOMC test and the C-MMSE score. The area under the receiver operating characteristic curve (AUC) was used to demonstrate the sensitivity and specificity of the C-SOMC test at various cut-off values distinguishing cognitive impairment from normal cognition.

The total score for the C-SOMC test and the score for item 1 on this test exhibited moderate-to-good correlations with the C-MMSE score, with respective ρ-values of 0.636 and 0.565 (P < 0.001). The scores for each of items 2, 4, 5, 6, and 7 yielded fair correlations with C-MMSE score, with ρ-value from 0.272 to 0.495 (P < 0.05). The total score on the C-SOMC test and the item score were good predictors (adjusted R² = 0.049 to 0.615) of the C-MMSE score, and six items were good predictors (adjusted R² = 0.134 to 0.795) of the total score. The AUC was 0.92 for the C-SOMC test. A cut-off of 17/18 on the C-SOMC test gave optimal performance: correct classification of 75% of participants, with 75% sensitivity and 87.9% specificity.

The C-SOMC test demonstrated good concurrent validity, sensitivity and specificity in a sample of people with a first cerebral infarction, demonstrating that it could be used to screen for cognitive impairment in stroke patients.