Reliability and Validity of the Modified Mini Cog: A Measure for Screening Cognitive Functions in Literates and Non-literates

Abstract

Background: The original Mini Cog was modified to make it applicable to non-literates as well. However the reliability and validity of the modified mini cog (MMC) has not been examined. Therefore, we aimed to investigate the intra- and inter-rater reliability, criterion validity, sensitivity and specificity of the MMC.
Methods: In this methodological and repeated measures design, elderly individuals (>60 years) with no neurological diagnosis or adults (>18 years) with neurological diagnosis were recruited from a tertiary hospital and the local community using purposive and snowball sampling. One of the raters administered the MMC twice (one week apart). Another rater administered the MMC and the Rowland Universal Dementia Assessment Scale (RUDAS) once during the first assessment session.
Results: The ICC for consistency of a rater across the tests and absolute agreement between two raters ranged from 0.97-0.99. The MMC scores of two raters were not significantly different. The MMC was able to differentiate between elderly participants with no neurological diagnosis and adult participants with neurological diagnosis. A significant correlation (Coefficients: 0.52-0.68) was found between the MMC and RUDAS. The sensitivity and specificity of the MMC were 86% and 70% respectively. The cutoff score of the MMC was found ? 3.
Conclusions: We demonstrated an excellent intra- and inter-rater reliability, and adequate criterion validity, sensitivity and specificity of the MMC. The MMC was also able to discriminate various groups having varied level of cognitive dysfunction. However, further studies are warranted to get more insight into the value of this instrument.
Keywords: Modified mini cog; reliability; sensitivity; specificity; validity.