Study On The Consistency Of Gait Parameters And Scales In The Assessment Of Frailty In The Elderly

ObjectiveThree scales,the Physical Frailty Phenotype(PFP),the FRAIL scale(Fatigue Resistance Ambulation Illnesses and Loss of weight,FRAIL)and the Tilburg Frailty Indicator(TFI)to screen for frailty in community-dwelling older people aged≥60 years.Through the measurement and analysis of the gait parameters of the elderly walking at normal speed,the elderly are classified from the multi-dimensional gait parameters of the objective indicators,and whether this classification method can be consistent with the three scales for the frailty classification of the elderly is discussed.The consistency of these four methods in the assessment of frailty in the elderly is analyzed and compared,and the diagnostic performance of these three scales is compared in order to find a more suitable tool for assessing the frailty of the elderly in the community.MethodsFrom July to October 2021 and from September to December 2022,a questionnaire survey was conducted in two community hospitals in Xiangcheng District,Suzhou City,using a convenient sampling method.A total of 2429 community elderly people aged 60 and over were included in the physical examination.The general data(social demographic characteristics,physical examination,lifestyle)and laboratory indicators of the elderly were collected.Patient Health Questionnaire-9(PHQ-9)was used to screen for depression in the elderly,and PFP,FRAIL and TFI were used for frailty assessment.Zebris FDM gait analysis system was used to test the elderly walking at normal speed,and biomechanical characteristic parameters such as phase,time,space and plantar pressure center trajectory were collected.Questionnaire star and Zebris FDM supporting software were used to establish a database for data entry,and SPSS23.0 and Medcalc15.0 software were used for statistical analysis.The measurement data conforming to the normal distribution were expressed by x±s,the measurement data of non-normal distribution were expressed by M(P25,P75),and the count data were expressed by the number of cases and percentage(%).According to three scales,the elderly were divided into two groups:frailty and non-frailty.The independent sample T test was used to compare the measurement data with normal distribution.The Mann Whitney U non-parametric test was used to analyze and compare the measurement data without normal distribution.The samples were classified by 43 gait parameters and two-step cluster analysis,and compare whether there are differences in demographic characteristics,routine blood biochemistry and gait parameters between the different clustering categories of older people.Kappa test were used to analyze the consistency between the four evaluation methods and the difference between the methods.Receiver operating characteristic(ROC)curve and the area under the ROC curve(AUC)were used to analyze the accuracy of PFP,FRAIL and TFI in evaluating the frailty of the elderly in the community.Statistical significance was considered at P<0.05.Results1.Demographic characteristics and frailty status of the elderlyThere were 2429 elderly people aged 60 years and above in this survey,of whom 1178(48.50%)were male and 1251(51.50%)were female.The average age was 71.16±5.30 years,with a predominance of elderly people under 80 years old(92.38%).The number of people living alone was 174(7.16%).The PHQ-9 detected depression at a rate of 13.09%.The detection rate of frailty in older people was 10.17%for PFP,5.02%for FRAIL and 17.25%for TFI,based on the respective frailty diagnostic cut-offs of ≥3 for PFP,≥3 for FRAIL and≥5 for TFI.2.Characteristics of older people with different frailty statesTo compare differences based on the results of three scales,where the debilitating group was characterized by being female,≥80 years old,low literacy,no spouse,manual labour,low income,thinness and obesity,reduced grip strength,coexistence of multiple illnesses,depression,non-smoker,non-second-hand smoker,non-drinker,not regularly physically active,having sleep disorders,short sleep duration,living alone,and the differences in all these comparisons were statistically significant(P<0.05).3.Two-step cluster characteristics of the populationThe classification of 2429 community-dwelling older adults according to 43 gait parameters yielded two categories.Category Ⅰ had 537(22.11%)and category Ⅱ had 1892(77.89%).The quality of clustering was good.The proportions of females,advanced age,low literacy,no spouse,low income,thin and obese,reduced grip strength,sick,coexisting with multiple illnesses,depressed,non-smoker,non-smoker of second-hand smoke,non-drinker,sleep disorder,short sleeper,and living alone were significantly higher in category one than in category two,with statistically significant differences in all of these comparisons(P<0.05).Older people in category I have a high degree of overlap with the characteristics of older people in the frailty group as assessed by the three scales.4.Characteristics of gait parameters in different categories of older adultsThere were 16 gait parameters with a contribution greater than 0.6 in the clustered groupings.The centre-of-pressure trajectory parameter,phase parameter,spatial parameter and time parameter were included.Compared to category I older adults,the former had a shorter single-support gait line,longer load-response phase,pre-oscillation phase and double-support phase,shorter walk length and stride length,slower stride speed and frequency,and increased single-stride duration,all of which were statistically significant(P<0.001).5.Comparison of assessment consistency and diagnostic accuracyThe results of the two-step cluster analysis of gait parameters were used to define category one category two as debilitating and non-debilitating,and then the PFP,PRAIL and TFI assessment results respectively were tested for consistency with the clustering results.The obtained Kappa values respectively were 0.458,0.255 and 0.489(P<0.001 for all),and the gait parameter clustering results were moderately consistent with PFP and TFI,and relatively poorly consistent with FRAIL.Using the gait parameter classification as a criterion,the ROC was used to determine the effect of PFP,FRAIL and TFI on the assessment of frailty,and the results respectively showed that the AUC was 0.779,0.608 and 0.733 for PFP,FRAIL and TFI(P<0.001 for all).Conclusions1.The detection rate of frailty is high among older people in the community,with a wide range of variation in the detection rate of frailty assessed by different screening tools and a higher detection rate of frailty assessed by multidimensional tools than by unidimensional tools.2.Gait parameters in older people should focus on load response period,double support phase,stride length,stride speed,single stride duration and stride frequency.3.The classification of gait parameters is moderately consistent with the PFP and the TFI,and relatively poorly consistent with the FRAIL.The PFP and TFI are of moderate value in frailty assessment,and the TFI combined with relevant gait parameters is more appropriate for frailty screening and assessment of older people in the community.