| ObjectivesThe Comprehensive Frailty Assessment Instrument Plus(CFAI-Plus)developed and authorized by the team the Belgium scholars Daan Duppen and others work with was sinicized,and the measurement characteristics of the sinicized CFAI-Plus were evaluated in terms of reliability,validity and feasibility,so as to provide a more scientific and comprehensive assessment scale for the study of community senility in China.At the same time,the sinicized CFAI-Plus was used to predict the adverse outcomes of the frail elderly in the community,and to explore the predictive effect of this scale on the occurrence of adverse events such as falls,hospitalization and deaths,so as to provide a scientific evaluation tool for a comprehensive assessment of the frailty of the community elderly and a targeted frailty prevention and control strategy and prevent the occurrence of adverse events.Methods1. The sinicization and evaluation of the CFAI-Plus:this study was conducted in accordance with the guidelines recommended by the American Academy of Orthopaedic Surgeons Evidence Based Medicine(AAOS)and strictly with the principles of scale adjustment and revision.First,we obtained the CFAI-Plus scale authorization from the team the Belgium scholars Daan Duppen and others work with.Second,we carried out translation,back-translation and cultural adaptation of the English version of CFAI-Plus to have a sinicized CFAI-Plus.Finally,we used a cross-sectional survey method to evaluate the reliability,validity and feasibility of the sinicized CFAI-Plus.The stratified random sampling method was used From December 2018 to February 2019,5 administrative districts in Chengdu city(Jinniu,Chenghua,Jinjiang,Wuhou and Qingyang)were taken as the stratification factors.The subdistricts of the 5 districts were randomly sampled using the table of random number,and 5 subdistricts(Tianhuizhen subdistrict,Longtan subdistrict,Niushikou subdistrict,Jitouqiao subdistrict and Caotang streets)were selected;then one community from each subdistrict was selected as the investigated community by the convenience sampling,namely,Wansheng community,Baoping community,Huachenglu community,Banbianjie community,Caotanglu community;the study subjects were selected by using the convenience sampling in the five selected communities.According to the Kendall calculation method of sample content,a total of 300 subjects should be included,with at least 60 in each community.In the study,419 frail elderly in the community were included in strict accordance with the exclusion criteria and diagnostic criteria.First,Frailty phenotype(FP)was used as a diagnostic tool to screen the frail elderly in the community and as the gold standard of reference in order to facilitate the following criterion validity detection and prediction effectiveness research.Second,a questionnaire of the general situation was used to understand the demographic sociological characteristics and other features of the subjects.Finally,the sinicized CFAI-Plus was sent for investigation.One week after the initial investigation,a second investigation of 30 frail elderly in the community of the same study subjects was carried out,so as to facilitate the test-retest reliability evaluation.Correlation analysis,exploratory factor analysis and Receiver Operating Characteristic Curve(ROC)were used to evaluate the reliability,validity and feasibility of the sinicized CFAI-Plus.2.The prediction effect of adverse outcomes on frail elderly in the community with the sinicized CFAI-Plus:we used the scientific design of a prospective cohort study,and 419 frail elderly in the community were included in the first scale measurement as study subjects,using FP for diagnosis and as a reference standard for predictive validity.The sinicized CFAI-Plus was used to assess the frail severity of the community elderly in each dimension.With follow-up surveys in 6 and 12 months,adverse outcomes,including falls,hospitalization and deaths,were recorded.The Chi-square test or Fisher's exact test were used to explore the difference between the severity degree of frailty and the occurrence of adverse outcomes in 6 and 12 months in each dimension of the sinicized CFAI-Plus.Kaplan-Meier method was used to plot the temporal relationship between the severity of frailty and the occurrence of adverse outcome events in each dimension during the follow-up period.Log-rank method was used to test the difference of the curves of different severity of frailty.Linear-by-Linear Association method was used for the trend test.Cox proportional hazard regression model was used to explore the correlation between each dimension of frailty and the occurrence of adverse outcome events in 6 months and 12 months,and the correlation was further verified by correcting the age and the number of chronic diseases.The ROC curve was fitted with adverse outcome events as variables,and the area under the scoring curve of each dimension of the sinicized CFAI-Plus was analyzed.FP was used as the standard.They were compared to explore the prediction efficiency of the sinicized CFAI-Plus on adverse outcomes.Results1. The evaluation results of the sinicized CFAI-Plus:after translation and retranslation,an expert committee was formed to integrate and review the scale in order to ensure the acceptability of the sinicized CFAI-Plus and formulate the pre-survey version of CFAI-Plus.20 cases of frail elderly in the community who were qualified were selected to receive the pre-survey,and the study subjects'feedback on each item was recorded.The finally sinicized CFAI-Plus was formed through the adaptation according to the investigation situation.The finally sinicized CFAI-Plus contained 5 dimensions:cognitive frailty,physical frailty,psychological frailty,social frailty and environmental frailty,with a total of 25 items.The assessment results of this scale showed that:(1)Reliability:(1)test-retest reliability:the test-retest correlation coefficients of the 5 dimensions of the sinicized CFAI-Plus ranged from 0.600 to 0.817(P<0.001);(2)internal consistency and split-half reliability:Cronbach'alpha coefficient of the total instrument of the Sinicized CFAI-Plus was 0.826,and the split-half reliability was 0.890.Cronbach's alpha coefficients of the five dimensions ranged from 0.638 to0.878,and the split-half reliability ranged from 0.648 to 0.905.(2)Validity:(1)content validity:content validity index(CVI)was used for evaluate the content validity.The CVI of the each item level was 0.833?1,and the CVI of the scale level was 0.900;(2)construct validity:the item-dimension correlation analysis showed that the correlate on coefficients between each item of the sinicized CFAI-Plus and its dimension ranged from 0.518 to 0.808(P<0.05);exploratory factor analysis showed that the Bartlett sphericity test of the scale wasc~2=4912.475(P<0.001);the KMO value was 0.822,which meant the exploratory factor analysis was suitable.There were 7 principal component with an eigenvalue of above 1,and a total of 7 common factors were extracted.The cumulative variance contribution rate was 67.85%,the load coefficient of each factor ranged from 0.527 to0.914;(3)criterion validity:the sensitivity of each dimension was 0.230?0.803,and the specificity was 0.444?0.855.The results of ROC curve showed that the area under the curve of each dimension in the sinicized CFAI-Plus were ranging from 0.528 to0.707 using FP as the gold standard.(3)Feasibility:the recovery rate of the scale was 98.84%,the completion rate of the scale was 98.59%.On average,it takes 10?20 minutes for the frail elderly in the community to complete the scale.Most research objects thought that the meaning of the items was clear and they could easily finish it.2. The predictive effect of the sinicized CFAI-Plus on adverse outcomes of frail elderly in the community:389 frail elderly in the community completed a 6 months follow-up survey,366 completed a 12 months follow-up survey,53 cases were lost in total,and the lost rate was 12.65%.The results showed that:(1)The occurrence condition of adverse outcomes in 6 months:there were 26fell(6.21%)and 56 hospitalized(13.37%).The sinicized CFAI-Plus was used to assess the severity of frailty among the frail elderly in the community,and there were divided into 3 groups:no to low frailty group,low to mild frailty group and mild to high frailty group.The three severity of frailty in different dimensions and the incidence of adverse outcome events(falls and hospitalization)in 6 months were statistically analyzed by using the Chi-square test or Fisher's exact test,and the results showed that there was no statistically significant for the difference between the three severity degrees of frailty in different dimensions and the adverse outcomes of frail elderly in the community in 6 months(P>0.05).(2)The occurrence condition of adverse outcomes in 12 months:57 fell(13.60%),103 hospitalized(24.58%),and 1 died(0.24%).(1)The three severity degrees of frailty in different dimensions(no to low frailty group,low to mild frailty group,mild to high frailty group)and the occurrence of adverse outcome events in 12months(falls,hospitalization,deaths)were statistically analyzed through using the Chi-square test or Fisher's exact test.The results showed that the differences between the severity of frailty in the cognitive and physical frailty and the occurrence of falls in 12 months were statistically significant(P<0.05),there were statistically significant differences between the severity of frailty and hospitalization in the two dimensions of physical frailty and psychological frailty(P<0.05),and that there was no statistically significant difference between the rest severity of frailty and the occurrence of adverse outcome events(P>0.05).(2)According to the Kaplan-Meier survival curve,the higher the severity of frailty in the cognitive and physical frailty dimensions were,the steeper the survival curve and the higher the cumulative incidence of falls was.The Log-rank method indicaed that the three severity degrees of the cognitive frailty(c~2=6.744,P=0.034)and the three severity degrees of physical frailty(c~2=21.228,P=0.000)had significantly statisitic difference with the cumulative incidence rate of falls(P<0.05).Further,the Linear-by-Linear Association method was used for trend test,and the cumulative incidence rate of falls varied with the three severity degrees in the dimension of cognitive frailty(Z=6.874,P=0.009)and with the three severity degrees in the dimension of physical frailty(Z=16.177,P=0.000),and the difference was statistically significant(P<0.05).The higher the severity of frailty in the physical and psychological frailty dimension were,the steeper the decline of the survival curve and the higher the cumulative rate of hospitalization were.The Log-rank test showed that the three severity degrees of in the dimension physical frailty(?~2=14.678,P=0.001)and the three severity of in the dimension psychological frailty(?~2=51.297,P=0.000)were statistically significant different from the cumulative incidence rate of hospitalization(P<0.05).Further,the Linear-by-Linear Association method was used for trend test.The cumulative incidence rate of hospitalization varied with the three severity degrees of in the dimension physical frailty(Z=7.981,P=0.005)and the three severity degrees of in the dimension psychological frailty(Z=38.919,P=0.000),and the difference was statistically significant(P<0.05).(3)Cox proportional hazard regression model showed that the cognitive frailty(HR=3.524,95%CI 1.215?10.220,P=0.020),physical frailty(HR=2.442,95%CI 1.638?3.640,P=0.000)were associated with the falls of frail elderly in the community in 12 months(P<0.05),that the physical frailty(HR=1.888,95%CI 1.350?2.640,P=0.000)and the psychological frailty(HR=3.217,95%CI2.256?4.588,P=0.000)were associated with the hospitalization of frail elderly in the community in 12 months(P<0.05).After adjusting for age and the number of chronic diseases,the cognitive frailty(HR=3.554,95%CI 1.223?10.324,P=0.020)and physical frailty(HR=2.205,95%CI 1.445?3.363,P=0.000)were still associated with the falls of frail elderly in the community in 12 months(P<0.05).The physical frailty(HR=2.125,95%CI 1.484?3.042,P=0.000)and the psychological frailty(HR=3.175,95%CI 2.209?4.564,P=0.000)were still associated with the hospitalization of frail elderly in the community in 12 months(P<0.05).(4)ROC analysis results showed that when falls were taken as the state variable,the Area Under Curve(AUC)of the ROC in each dimension of the sinicized CFAI-Plus was 0.500?0.619.The area under curve of the physical frailty(AUC=0.619,95%CI 0.541?0.697,P=0.004)was larger than that of FP(AUC=0.544,95%CI 0.464?0.624,P=0.287),with no statistically significant difference(P>0.05),and the area under the curve of environmental frailty(AUC=0.544,95%CI 0.463?0.624,P=0.287)was equal to that of FP(AUC=0.544,95%CI 0.464?0.624,P=0.287),with no statistically significant difference(P>0.05).When hospitalization was taken as the state variable,the AUC of each dimension was0.464?0.624.The area under curve of the cognitive frailty(AUC=0.587,95%CI0.522?0.653,P=0.008)was similar to that of FP(AUC=0.580,95%CI 0.518?0.642,P=0.015),with no statistically significant difference(P>0.05),and the area under curve of the physical frailty(AUC=0.624,95%CI 0.562?0.687,P=0.000)was larger than that of FP(AUC=0.580,95%CI 0.518?0.642,P=0.015),with no statistically significant difference(P>0.05).Conclusions1.The sinicized CFAI-Plus had relatively good reliability,validity and feasibility,which could be used to evaluate the frail elderly in the Chinese community.2. The severity degress of frailty in each dimension of the sinicized CFAI-Plus was correlated with the occurrence of adverse outcome events in 12 months,and the severer the frailty was,the higher incidence rate of adverse outcome eventss was.The scale could predict the adverse outcome events of frailty,and the physical frailty dimension had a certain predictive value. |
PDF Full Text Request