| Background:With the rapid development of aging population,frailty as an age-related syndrome has gradually attracted sustained and widespread attention of international society.Frailty would not only increase the care burden of the family and society,but also greatly increase the risk of older adults' adverse outcomes such as falls,hospitalization and even mortality.Therefore,it is of great practical significance to identify the frail older adults as early as possible,find the key intervention factors that affect frailty,design targeted interventions,to delay or even reverse the frailty process,and to achieve healthy aging.Nursing home residents are a group with high incidence of frailty,which is more prevalent than community-dwelling older adults.However,most measurements are developed for community-dwelling older adults and/or hospitalized individuals that may be less applicable to nursing home residents.The FRAIL-NH scale is a specialized measurement for use among older adults in nursing homes.It is a brief,quick-to-complete,and user-friendly measurement,with good validity and reliability.However,it has not been used in China,and further cross-cultural adaptation and validation need to be undertaken.Additionally,a good measurement shall not only identify frail older adults,but also serve as a risk prediction measurement to timely screen out older adults at high risk of adverse outcomes.Previous research in developed countries such as Australia and Spain showed that the FRAIL-NH scale have been validated for the prediction of adverse outcomes,including hospitalization,and mortality.However,due to the heterogeneity of older adults'health condition in nursing homes among regions,the prediction of the FRAIL-NH scale on adverse outcomes still requires to be further studied.Besides the FRAIL-NH scale,are there other frailty measurements that may be applicable to older adults among nursing homes in China?Compared with these measurements,what is the predictive effect of the FRAIL-NH scale on adverse outcome?Further research is also needed.Meanwhile,existing research on influencing factors mainly designed with the cross-sectional design from the perspective of individual and social factors.The effects of certain factors,such as chewing function,pain,anxiety,social networks and other relevant factors are divergent and required to be further systematically explored.In addition,after older adults move into nursing homes,the facilities,service processes,care measures and medical conditions of nursing homes and other factors related to nursing homes may also affect the occurrence and development of frailty,which has not been reported in previous studies and further research is urgently needed.Objective:To cross-culturally adapt and validate the FRAIL-NH scale among nursing home residents in China.Then from the existing frailty measurements preliminarily screen out the measurements that be suitable for Chinese nursing home residents.Further comparison of the ability to predict adverse outcomes of frailty measurements was conducted to determine the most appropriate frailty measurement and to provide the standard instrument for Chinese nursing home residents and the following research.Based on this measurement,describing frailty level and dynamic changes,analyzing the influencing factors from individual,social,and institutional aspects,exploring the key factors in the progress of frailty,as well as further analyzing the mechnism of the key factors to frailty were conducted,respectively.Therefore,it can provide empirical evidence for designing targeted interventions to improve frailty in nursing homes,reduce or avoid adverse health outcomes and improve the care quality of nursing homes.Methods:1.Translation and cross-cultural adaptation of the FRAIL-NH scale:in strict accordance with the internationally established guideline,the Chinese FRAIL-NH scale version was generated using the translation,backward translation,cross-cultural adaptation,and pilot survet.2.Screening measurements:online searching was conducted in the following online platform:CNKI,Wanfang,CBM,VIP Data-base,PubMed.Web of science and Embase from inception to July 2017(the time when the research project started).The search aimed to find the studies of frailty measurements and psychometric properties to measure frailty.Two trained researchers independently screened literature,extracted data according to the purpose of the study,the inclusion and exclusion criteria.Psychometric quality of included studies using COnsensus-based Standards for the selection of health Measurement Instruments(COSMIN)was also assessed.3.Emperical data gathering:from March to June in 2018.three hundred and fifty-three older adults residing across 27 nursing homes were enrolled from 5 districts in Jinan.China.And two follow-ups were conducted at intervals of 6 months and 12 months,respectively.Well-trained and eligible research assistants collected data through face-to-face interviews.Standardized scales were used to collect the information of frailty,physical health,mental health and social function of older adults.The self-designed general information questionnaire was used to collect the socio-demographic characteristics of older adults and the relevant information of nursing homes.Adverse outcomes were assessed using falls,hospitalization,mortality,and quality of life.4.Data analysis:the following softwares,SPSS22.0.Stata15.1 and Mplus 7.4 were used to analyze the data.Baseline data were used to evaluate the reliability and validity of the FRAIL-NH scale.Evaluation indicators included:internal consistency,test-retest reliability,convergent validity,criterion validity,and diagnosis accuracy,meanwhile,Kappa test.The following statistical methods included:descriptive analysis,Cronbach's alpha coefficient,intraclass correlation coefficients,Spearman's correlation analysis,Pearson's correlation analysis,receiver operating characteristic(ROC)curve analysis and kappa statistic.Using longitudinal data,we performed ROC curve to compare the ability among frailty measurements in predicting the 6-month and 12-month adverse outcomes,and the corresponding Area under Curve(AUC)and 95%Confidence Interval(CI)were calculated.The logistic regression was used to examine the independent association between frailty and adverse outcomes.The odds ratio(OR)and 95%CI were reported.Using longitudinal data,the trend of frailty was described by multi-level model,the factors affecting frailty and its changes were analyzed,meanwhile the key factors were explored.Using longitudinal data,correlation analysis,multiple linear regression.and Bias-corrected nonparametric percentile Bootstrap methods were used to explore the mechanism of physical performance,which plays a key role in the occurrence and development of frailty,to predict frailty,that is,the mediating role of active leisure activities and passive leisure activities.Results:1.The Cronbach's alpha of the FRAIL-NH scale was 0.67.The test-retest reliability of 38 individuals within a 1-to 2-week interval was 0.84.As expected,a statistically significant correlation was observed between the FRAIL-NH scale and the comparable constructs related to frailty,namely,the SOF index,FI-35,Mini Nutritional Assessment-Short Form,and self-rated health(r=-0.29-0.60),presenting convergent validity.Using the frailty phenotype as a reference criterion,the AUC was 0.79.The optimal cutoff point for frailty was 2 in Chinese nursing homes(sensitivity:69.90%and 77.33%).The FRAIL-NH scale was significantly associated with the frailty phenotype(correlation coefficient?0.61,P<0.001),but showed fair agreement with it(kappa=0.46,P<0.001).2.Four frailty instruments were preliminarily selected that maybe suitable for Chinese nursing home residents:Phenotype of Frailty,Study of Osteoporotic Fractures index(SOF index),FRAIL-NH Scale,the Elderly Frailty Scale(FI-35).Among them,the FRAIL-NH scale had the best prediction on adverse outcomes,which mainly reflected in two aspects.Firstly,the FRAIL-NH scale had the highest discriminatory power for outcome prediction at the first follow-up(T1,AUC=0.83,95%CI=0.78,0.88)and at the second follow-up(T2,AUC=0.75,95%CI=0.69,0.81),respectively.Except for SOF index and FI-35(only predicted adverse outcomes at T2),the AUC of other frailty measurements for predicting adverse outcomes were all above 0.70(AUC=0.73-0.83).Then the FRAIL-NH scale was also found to be most predictive of adverse outcomes at T1 and T2(T1:OR=7.68,95%CI=3.70,15.97,T2:OR=3.04,95%CI=1.81-5.12),adjusted for age,gender,education time and other socio-demographic characteristics,as well as baseline adverse outcomes.SOF index,and T2 FI-35 could not predict adverse outcomes.3.In the baseline survey,the prevalence of frailty for Chinese nursing home residents was 49.7%,and the frailty gradually became severe over time.4.Among individual factors,age(?=-0.04,P=0.001),self-rated health status(?=-0.16,P=0.003),physical performance(?=-0.23,P<0.001),moderate limited chewing ability(?=0.41,?=0:002),extremely limited chewing ability(?=0.75,P<0.001),malnutrition(?=0.68,P<0.001),anxiety(?=0.45,P=0.007)and loneliness(?=0.13,P=0.006)were the influencing factors of frailty for nursing home residents.Among social factors,leisure activities(?=-0.06,P<0.001)were the influencing factor of frailty.Among institutional factors,fitness sites were influencing factors of frailty.The influencing factor of frailty deterioration was nutritional status.5.Individual,social,and institutional factors accounted for 56.3%of the fiailty variation.Each of significant factors and the frailty variation it could independently explained were listed as following:physical performance 19.0%,leisure activities 5.5%,loneliness 3.7%age 2.7%,the availability of fitness places in nursing homes 2.7%,self-rated health status 13%,chewing function 1.3%,nutritional status 1.2%,anxiety 1.1%,the interactions of nutritional status and time 0.5%,respectively.Among all these factors,physical performance was the core and significant effect on frailty,followed by leisure activities.6.The results of linear regression analysis showed that physical performance at TO was negatively associated with frailty at T2(?=-0.11,P<0.05);Physical performance at TO was significantly associated with active leisure activities at T1(?=0.33 P<0.001)and passive leisure activities at T1(?=0.15,P<0.01);With the inclusion of leisure activities,active leisure activities at T1 were significantly associated with frailty at T2(?=-0.16,P<0.05).However,the predictive effect of physical performance at TO on frailty disappeared at T2(?=-0.06,P>0.05).Simultaneously,the predictive effect of passive leisure activities at T1 on frailty disappeared at T2(?=-0.08,P>0.05).Further Bootstrap test confirmed that the indirect mediation effect of active leisure activities was statistically significant(Bootstrap 95%CI=-0.09,-0.02)and explained 41.7%of the association of physical performance to frailty.Conclusions:1.The FRAIL-NH scale is simple and easy to understand with acceptable validity and reliability,which is the best measurement for assessing the frailty among nursing homes in China.2.Frailty tend to more severe over time for one year.This indicated frailty is not optimistic and deserves attention innursing homes.3.Frailty of older adults in nursing homes is affected by individual,social and institutional factors,among which physical function is the core critical factor of the frailty.The frailty of the malnourished older adults is obviously worsened over time.Staff of nursing homes should identify and intervene the factors affecting the frailty and trends as early as possible,so as to delay or even reverse frailty,avoid adverse health outcomes and achieve healthy aging.4.Physical performance has an indirect effect on frailty through active leisure activities.Therefore,preventing frailty by employing interventions that promote older adults' active leisure activities may improve nursing home residents'quality of life. |
PDF Full Text Request