Physical frailty and gait speed in community elderly : a systematic review

www.ee.usp.br/reeusp Rev Esc Enferm USP · 2018;52:e03392 1 Universidade Federal do Parana, Programa de Pos-Graduacao em Enfermagem, Curitiba, PR, Brazil. 2 Universidade Estadual do Centro-Oeste, Departamento de Educação Física, Guarapuava, PR, Brazil. 3 Universidade de Málaga, Málaga, Spain. ABSTRACT Objective: To identify the outcomes of studies on gait speed and its use as a marker of physical frailty in community elderly. Method: Systematic review of the literature performed in the following databases: LILACS, SciELO, MEDLINE/PubMed, ScienceDirect, Scopus and ProQuest. The studies were evaluated by STROBE statement, and the PRISMA recommendations were adopted. Results: There were 6,303 studies, and 49 of them met the inclusion criteria. Of the total number of studies, 91.8% described the way of measuring gait speed. Of these, 28.6% used the distance of 4.6 meters, and 34.7% adopted values below 20% as cutoff points for reduced gait speed, procedures in accordance with the frailty phenotype. Regarding the outcomes, in 30.6% of studies, there was an association between gait speed and variables of disability, frailty, sedentary lifestyle, falls, muscular weakness, diseases, body fat, cognitive impairment, mortality, stress, lower life satisfaction, lower quality of life, napping duration, and poor performance in quantitative parameters of gait in community elderly. Conclusion: The results reinforce the association between gait speed, physical frailty and health indicator variables in community elderly.


INTRODUCTION
Physiological changes in aging, sometimes aggravated by the presence of chronic diseases, result in geriatric conditions arising in advanced ages and are amenable to prevention and treatment.Frailty is an example of a severe adverse outcome in the elderly.It increases substantially after the age of 75-80 years, and identifies a subgroup with low resistance and high risk of dependence, falls and mortality (1) .This condition has been recognized as a geriatric syndrome because of its complex symptoms, high prevalence in the elderly, and for being a result of several diseases and multiple risk factors (2)(3) .Therefore, it represents a priority for public health (4) .
Conceptually, physical frailty is defined as "a medical syndrome with multiple causes characterized by decrease of strength, endurance, and reduction of physiological functions that increase the individual's vulnerability for development, and greater dependence and/or death" (5) .It is associated with outcomes such as falls, dependence, hospitalization, institutionalization, death (3,(5)(6) , risk of compromised recovery after illness, surgery and worse response to treatment (1) .
The functional aspects affected by the condition of frailty are those dependent on energy and speed of performance, and affect tasks that require mobility (6) .From this perspective, one of the frailty phenotype markers is gait speed (GS).Reduced GS is the main indicator of physical frailty in the elderly (12)(13) .Besides being one of the pillars of the frailty phenotype, GS is strongly related to sarcopenia (14) .
GS can be influenced by individuals' health status, neuromuscular control, cardiorespiratory condition, physical activity level, sensorial and perceptual functions, as well as by characteristics of the environment where they walk (15) .Over time, these combined processes lead to scarcity of available energy, including that for the body's homeostatic balance.Thus, the elderly may develop adaptive behaviors, such as reduced GS (16) .The gait is a sequence of repeated movements of the lower limbs in order to move the body forward, while simultaneously holding the posture steady.For the harmonious performance of these movements, there must be a perfect balance between external forces acting on the body and the response of internal forces from muscles, tendons, bones, ligaments and joint capsules (17) .
GS measurement is an indicator of the elderly's health status and wellbeing (18) .It is easily measurable, clinically interpretable and a potentially modifiable risk factor (15) .GS has been recognized as a vital sign, and a valid, reliable and sensitive measure for assessing and monitoring the elderly's functional status and health conditions (19)(20)(21) .In addition, GS is a parameter of impairment of physical and cognitive functions, and a strong clinical indicator of the presence of frailty (22) .In a cross-sectional study, was investigated the prevalence of frailty and gait speed, and the relationship between these two indicators was analyzed in a sample of 1,327 individuals aged 65 years or older residing in Northern Madrid, Spain.The results showed that 32.1% of the elderly aged 75 years or older presented reduced GS (<0.8 m/s) and high risk of frailty (23) .
Gait speed is an important indicator of health conditions and physical frailty in the elderly, and like other physical changes, it suffers a decline with aging (24) .In a study, were analyzed data from seven studies conducted in the United States and Italy with the objective of estimating the incidence of disability and risk of mortality in 27,220 elderly people (≥65 years old) living in the community and monitored for three years.The results showed GS as a predictor of disability and mortality in the elderly (21) .A study was developed with the objective of investigating the pre-frail condition and its associated factors, and were considered the GS measurements of 195 elderly (≥60 years) users of a Basic Health Unit of Curitiba/PR/Brazil.The condition of prefrailty for GS was 27.3% and associated with the following: age group between 60 and 69 years old, low schooling, not feeling lonely, use of antihypertensive medication, presence of cardiovascular disease and overweight (25) .
This systematic review of the literature is justified by the academic-scientific contribution of a set of knowledge on an emerging topic in geriatrics and gerontology, and more specifically on GS, which has appeared as an important measure in gerontological evaluation.
In view of the above, the objective of the systematic review was to identify the outcomes of studies on gait speed and its use as a marker of physical frailty in community elderly.

METHOD
For the selection of studies and writing of the systematic review, were used the items proposed by the Checklist Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (26) .The STROBE statement (Strengthening the reporting of observational studies in epidemiology) (27) was used for evaluation of the availability of information and methodological procedures adopted in the selected articles.The 22 checklist items refer to cohort, case-control, and sectional studies (28) .For each of the STROBE items, was assigned a score (integral -1.0 point, partial -0.5 points or nonexistent -0 point) according to availability of information and/or adoption of the inquired procedure in the item, and the maximum value of 22 points.Higher scores represent greater availability of information and methodological procedures adopted by the studies.The checklist was applied individually for each study.
The research question (29) was: what are the results presented in studies regarding gait speed and its use as a marker of physical frailty in community elderly?
Studies considered as eligible for this systematic review were those that met the following inclusion criteria: a) published as an original article in scientific journals; b) publication period between January 2010 and October 2016; c) available in full in Portuguese, English or Spanish; d) indexed in the selected databases; e) indicating the evaluation of frailty by means of the frailty phenotype (6) ; f ) involving community elderly aged ≥60 years.The exclusion criteria adopted were: a) repeated in the databases; b) included as editorials, reviews, reports of experience, abstracts published in events, monographs, dissertations or theses, review studies and meta-analysis; c) conducted in hospital institutions or with long-term institution residents; d) involving elderly people with a specific disease (hypertension, diabetes, arthritis/arthrosis, cardiovascular diseases, Alzheimer's, Parkinson's).
The selection of studies and gathering of information were performed by two reviewers independently with the aid of a standardized instrument.Firstly, the following information was collected from the selected studies: location (country) where the study was conducted, year and journal of publication, study design, number and characteristic of the sample involved.Secondly, were extracted the objective, prevalence of categorization of physical frailty (frail, prefrail and non-frail), form of GS measurement, prevalence of reduced GS as a marker of physical frailty and outcomes of the GS.
The procedures for selection of eligible studies involved reading the titles, abstracts and the studies in full.Studies that did not meet the inclusion criteria or did not address the research question were excluded.When applying the search strategies, were found 6,303 studies in the seven databases consulted.After screening, 49 studies were eligible for the systematic review.Figure 1 illustrates these steps, according to PRISMA methodological recommendations (26) .

RESULTS
The characteristics of the selected studies indicated there were more articles published in the years of 2015 (n=14; 28.6%), 2014 (n=12; 24.5%) and 2013 (n=12; 24.5%).Studies conducted in developed countries were predominant (n=25; 51%).Among developing countries, Brazil presented  a significant number of studies on the subject (n=15; 30.6%).The majority of studies (n=38; 77.5%) were published in 34 different international journals.Regarding sample size, there were variations in quantity, ranging from 51 elderly subjects in a cross-sectional study to 13,924 participants in a cohort study.There was a predominance of cross-sectional or sectional studies (n=37; 75.5%), and participants' mean age ranged from 68.7±6.9 to 86.0±4.9 years (Table 1).
Table 1 shows the study designs, characteristics of the sample, study objectives, prevalence of frail, pre-frail and non-frail elderly, form of gait speed measurement, distribution of reduced gait speed, outcomes of gait speed in community elderly, and the STROBE score.
The GS measurement protocol was described in 45 (91.8%) studies, which demonstrates different ways of measuring this variable.The distance of GS was described in 34 (69.5%)studies and ranged from 2.4 to 20 meters.In 14 (28.6%)studies, was adopted a distance of 4.6 meters.Regarding the cutoff points for reduced GS, seven (14.3%) studies did not describe the values and adjustment variables, and of those where this information was reported, nine (18.4%) mentioned only adjustment variables (gender, height, body mass index -BMI).Out of the total number of studies, 30 (61.3%) reported cutoff points for reduced GS, of which 13 (26.5%)considered the values below 20% (quintile).In studies where the percentage of reduced GS (n=28; 57.2%) was described, there was a variation between 2.7% and 83.9%.The prevalence of reduced GS as a marker of physical frailty was not described in large part of the studies (n=21; 42.8%).In studies that described the prevalence of reduced GS (n=5; 10.2%) in frail and pre-frail groups, there was variation of 4.7% -89% and 9.9% -86.5%, respectively.Fifteen (30.6%) studies emphasized as outcomes of GS, its association with the variables of disability, frailty, sedentary lifestyle, falls, muscular weakness, diseases, body fat, cognitive impairment, mortality, stress, lower life satisfaction, lower quality of life, napping duration and low performance in quantitative parameters of gait in community elderly.In 14 (28.6%)studies, the results referred to GS by sex.The classification of studies according to the STROBE statement resulted in a score ranging from 15 to 19.5 points with a prevalence of 18 points (n=10; 22.7%).The cutoff point for frail was 0.7 m/s and for pre-frail 1.1 m/s.

Missing information
The slowest GS was found in frail elderly, advanced age, and women.

19.90%
The percentages of elderly who scored for slow gait were statistically comparable between the percentages of frail, pre-frail and non-frail.

18%
There was an association between napping duration and the decrease in GS (p=0.1770).

19.5/22
Han, Lee, Kim, 2014 (7) Cross-sectional In order to measure slow GS, was calculated the speed to walk 10 m, and slow GS was defined as <1.0 m/s.

Missing information
There was a significant difference between GS and the frailty groups (frail, pre-frail and non-frail).
As observed in the different studies, even though there is no established consensus for GS measurement, there was a predominance of distance of 4.6 meters (23,34,38,(39)(40)42,(44)(45)50,(58)(59)(60)(73)(74) , and cutoff points for reduced GS were defined by the lowest quintile value (7,34,40,42,44,49,54,57,59,65,69,72,75) . These studies follw the procedure/protocol of the study (6) .Given the importance of GS in clinical practice, there must be a consensus, a standardization of the measurement of this marker of physical frailty. The use of GS in clinical practice is rcommended as a key tool in geriatric assessment given its simplicity, speed, objective parameter and sensitivity to changes caused by the aging process (17) .It is noteworthy that the studies included in this systematic review evaluated gait speed in different ways and presented diverse interventions and designs.
The distance for calculation of GS used in the different studies varied greatly.A recent systematic review revealed that gait speed distance ranged from four to six meters in 83% of the studies analyzed, and four meters was the most used distance (79) .In another study (14) , was observed that most researchers used distances between four and six meters and the distance of the course should allow the test application in the clinical setting as a routine examination.
Reduced GS as a marker of physical frailty was higher in women compared to men (32,34,40,(57)(58)60,(62)(63)69,76) . This is confirmd by the results of nine studies, which totaled a sample of 26,625 elderly (≥ 65 years) living in the community.Researchers found a significantly lower percentage of GS (≤ 0.8 m/s) in women (31%), while it was 10% in men (80) .In contrast, in a study with the objective of investigating sex differences in gait patterns in elderly participants of the Baltimore Longitudinal Study of Aging, no difference between sexes for GS (p=0,185) was found after adjusting for age, height, and body mass (81) .
The prevalence of reduced GS varied greatly in the studies, and values ranged from 2.7% (31) to 83.9% (36) .However, in most of them, the percentages of reduced GS as a marker of physical frailty were not described.In some studies (8,42,59,(72)(73) , the prevalence of reduced GS was reported in frail and prefrail groups.This demonstrates that GS and other markers of physical frailty (fatigue/exhaustion, weight loss, physical activity, muscle strength) were poorly explored in the studies analyzed.
National surveys describe percentage values of reduced GS in the elderly close to those found in the present review.The study "Frailty in Brazilian Elderly (FIBRA)" with a sample of 5,532 community elderly (> 65 years) found that 20.9% of elderly people had slow gait.The markers with greatest odds for development of frailty were slow gait and muscle weakness (82) .In a study conducted with elderly (≥ 60 years) and the aim to associate physical frailty with the quality of life of elderly users of basic health care in Curitiba/PR/ Brazil, 25.6% of the 203 elderly individuals participating showed reduced GS (83) .
Studies (31,33,35,55,59,(65)(66) showed that frailty is associated with reduced GS, and these findings are in agreement with another study.Data from the English Longitudinal Study of Aging (ELSA) showed that 90% of elderly classified as fragile had reduced GS (84) .
The limitations of this systematic review study are related to information deficits found in some studies, which can impair the analyzes.As for the method, the definition of the search period, languages and databases consulted may have delimited the search and, consequently, relevant studies on the subject may have not been selected.

CONCLUSION
The studies evaluating physical frailty in community elderly indicated the association of the outcome of GS with disabilities, frailty, sedentary lifestyle, falls, muscle weakness, diseases, body fat, cognitive impairment, mortality, stress, lower life satisfaction, lower quality of life, napping duration, and poor performance in quantitative parameters of gait in community elderly.
The GS measurement protocol varied among the studies.The distance and cutoff points for reduced GS defined by the frailty phenotype were adopted in some studies.Efforts are needed in order to standardize the way of measuring this variable, mainly because of its importance in clinical practice.
The findings of this systematic review reinforce the association between GS and physical frailty and health indicator variables in community elderly.The studies demonstrate the importance of GS measurement in gerontological evaluations.Randomized studies are recommended in order to validate and establish a consensus regarding the way of measuring GS as a tool for gerontological evaluation.

Figure 1 -
Figure 1 -PRISMA flowchart with information on phases of the selection process of studies for this systematic review.

Table 1 -
Description of the design, sample, objective (s), prevalence of physical frailty, GS measurement, prevalence of reduced GS, outcomes of GS and STROBE score in the studies selected for this systematic review.