Application of Montreal cognitive assessment cut-off in screening mild cognitive impairment

Objective To study the application of Montreal cognitive assessment(MoCA) in screening mild cognitive impairment(MCI) and its optimal cut-off value.Methods One hundred and fifty -three MCI patients were divided into control group(n= 69),MCI group(n= 60),and Alzheimer's disease(AD) group(n= 24) according to its diagnostic criteria and evaluated using MoCA and mini-mental state examination(MMSE).Correlation between the MoCA and MMSE scores for the patients was analyzed.Sensitivity,specificity,Kappa value and Youden index of MoCA in screening MCI patients were calculated to select its optimal cut-off value.Results The MMSE and MoCA scores were significantly lower in MCI and AD groups than in control group(P 0.05).The MMSE score was closely related with the MoCA score(r=0.847,P0.01).The sensitivity, specificity and Kappa value of MoCA for the diagnosis of MCI were 98.3%,85.5%and 0.830 respectively when its cut-off value was 26.ROC curve showed that the sensitivity,specificity and Kappa value of MoCA for the diagnosis of MCI were 93.3%,97.1%and 0.906 when its cut-off value was 25.Conclusion MMSE and MoCA scores are closely related in MCI patients and consistent with their clinical diagnosis.The cut-off value of MoCA at 25 is recommended for the diagnosis of MCI.