Reliability and validity of the short form of the literacy-independent cognitive assessment in the elderly

Background and Purpose The Literacy-Independent Cognitive Assessment (LICA) has been developed for a diagnosis of dementia and is a useful neuropsychological test battery for illiterate populations as well as literate populations. The objective of this study was to develop the short form of the LICA (S-LICA) and to evaluate the reliability and validity of the S-LICA. Methods The subtests of the S-LICA were selected based on the factor analysis and validation study results of the LICA. Patients with dementia (n=101) and normal elderly controls (n=185) participated in this study. Results Cronbach's coefficient alpha of the S-LICA was 0.92 for illiterate subjects and 0.94 for literate subjects, and the item-total correlation ranged from 0.63 to 0.81 (p<0.01).The test-retest reliability of the S-LICA total score was high (r=0.94, p<0.001), and the subtests had high test-retest reliabilities (r=0.68-0.87, p<0.01). The correlation between the K-MMSE and S-LICA total scores were substantial in both the illiterate subjects (r=0.837, p<0.001) and the literate subjects(r=0.802, p<0.001). The correlation between the S-LICA and LICA was very high (r=0.989, p<0.001). The area under the curve of the receiver operating characteristic was 0.999 for the literate subjects and 0.985 for the illiterate subjects. The sensitivity and specificity of the S-LICA for a diagnosis of dementia were 97% and 96% at the cutoff point of 72 for the literate subjects, and 96% and 93% at the cutoff point of 68 for the illiterate subjects, respectively. Conclusions Our results indicate that the S-LICA is a reliable and valid instrument for quick evaluation of patients with dementia in both illiterate and literate elderly populations.