| Objective: The present study was to verify reliability and validity of the Chinese version of Alzheimer’s Disease Assessment Scale‐Cognitive Subscale(ADAS‐Cog)among the community old people in China.Methods: A total of 1378 individuals composed by 670 normal controls(NCs),602 people living with mild cognitive impairment(MCI),and 106 people living with dementia of Alzheimer’s type(DAT)were recruited for the current study.All of the participants underwent ADAS‐Cog,clinical interview and examinations and core neuropsychological test.Internal consistency reliability and retest reliability coefficient were used to evaluate the reliability of ADAS-Cog.Construct validity,concurrent validity and clinical validity was analyzed to assess the validation of ADAS-Cog.Furthermore,influence factors of the ADAS-Cog were explored via logistic regression analysis.Results:(1)Reliability: The Cronbach’s α coefficient of the ADAS-Cog was0.789 and the retest correlation coefficient of total score was 0.915(P < 0.001).In addition,retest correlation coefficients among 12 sub-items were between 0.247 and0.688(P < 0.001).(2)Construct validity: The correlation coefficient between the total score of ADAS-Cog and scores of each sub-item were between 0.483 and 0.709(P < 0.01),and the correlation coefficients between each item were between 0.247 and 0.688(P < 0.01).The model of 4 secondary-order factors were composed of memory and orientation,language,executive function and attention.The values of GFI,AFGI,CFI and TLI of the four-factor model were all greater than 0.9.Additionally,the values of RMSEA and RMR were 0.073 and 0.034 respectively.(3)Concurrent validity: The correlation coefficient of ADAS-Cog total score with MMSE、Mo CA、CDR、ADL and neuropsychological test were-0.706、-0.803、0.546、0.292、-0.530(P < 0.001),respectively.(4)Screening efficacy:(1)The optimal cut-off value of neurocognitive disorders(NCDs)was 10.0 points among the entire sample,with the sensitivity of66.20%,the specificity of 93.40% and the accuracy of 79.42%.The optimal cut-off value of DAT was 14.0 points,with the sensitivity of 80.20%,the specificity of89.20% and the accuracy of 88.51%.(2)The cut-off points were 10.0(Se: 76.60%,Sp: 89.00%)for NCDs and 15.0(Se: 89.30%,Sp: 83.60%)for DAT in the lower education group;and the the cut-off points were 10.0(Se: 59.90%,Sp: 94.10%)for NCDs and 10.5(Se: 84.00%,Sp: 78.40%)for DAT in the higher education group.The overall classification accuracy of NCDs was 79.45%,and that of DAT was80.31% among different educational groups.(3)The cut-off points were 10.0(Se:60.70%,Sp: 93.40%)for NCDs and 13.5(Se: 75.90%,Sp: 89.90%)for DAT in the younger elderly group;and the cut-off points were 10.0(Se: 76.50%,Sp: 93.50%)for NCDs and 16.0(Se: 84.60%,Sp: 92.30%)for DAT in the older elderly group.The overall classification accuracy of NCDs was 79.44%,and that of DAT was 89.78%among different age groups.(5)Influence factors: According to the result of liner regression analysis,age and educated years were the impact factors of ADAS-Cog scores among the entire sample and people with NCDs.Additionally,the ADAS-Cog scores were mainly influenced by the age and gender for the cognitive healthily old people.Conclusion: Chinese version of ADAS-Cog has good internal consistency,retest reliability,construct validity,concurrent validity,which meets the requirements of psychological testing.Besides,the scale is able to detect cognitive impairment of DAT patients in Chinese communities but has a lower efficacy for MCI. |
PDF Full Text Request