See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/322250638

Simple Method of Screening for Frailty in Older Adults Using a Chronometer

and Tape Measure in Clinic

Article  in  Journal of the American Geriatrics Society · January 2018

DOI: 10.1111/jgs.15204

CITATIONS READS

5 107

7 authors, including:

Heewon Jung Min-Gu Kang

Asan Medical Center Chonnam National University Hospital
86 PUBLICATIONS   1,072 CITATIONS    17 PUBLICATIONS   76 CITATIONS   

SEE PROFILE SEE PROFILE

Choi Jung-Yeon Sun-wook Kim

Seoul National University Bundang Hospital Seoul National University Hospital
23 PUBLICATIONS   167 CITATIONS    32 PUBLICATIONS   675 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Developing strategies to improve frailty and disabilities in community-dwelling older adults View project

Gerontology View project

All content following this page was uploaded by Heewon Jung on 05 January 2018.

The user has requested enhancement of the downloaded file.

 BRIEF METHODOLOGICAL REPORT
Simple Method of Screening for Frailty in Older Adults Using a
Chronometer and Tape Measure in Clinic
Hee-Won Jung, MD, MSc,* Min-Gu Kang, MD,† Jung-Yeon Choi, MD,† Sol-Ji Yoon, MD,†
Sun-wook Kim, MD, MSc,† Kwang-il Kim, MD, PhD,†‡ and Cheol-Ho Kim, MD, PhD†‡
OBJECTIVES: Detecting frailty in older adults scheduled
for surgery is important to predict the occurrence of adverse
outcomes, but because of its complexity, frailty screening is
not commonly performed. The objective of the current
study was to assess whether frailty can be screened for using
automatically measured usual gait speed (UGS) and mid-
arm circumference (MAC) in the outpatient clinic.
DESIGN: Prospective, cross-sectional study.
SETTING: Geriatric center of a tertiary hospital.
PARTICIPANTS: Outpatients aged 65 and older
(N = 113).
MEASUREMENTS: Frailty status was evaluated accord-
ing to a multidimensional frailty score (MFS) using a com-
prehensive geriatric assessment, and participants were
classified into 5 categories. UGS was evaluated by having
participants walk through the clinic using an automated
laser-gated chronometer. MAC was recorded using a tape
measure on a blood pressure cuff. Correlations between
these two physical measurements and MFS were assessed.
RESULTS: The mean age of the 93 participants who suc-
cessfully underwent UGS evaluation was 75.8  4.7; 35
were male. In this population, the mean Charlson Comor-
bidity Index was 2.2  1.4, mean MFS was 4.1  2.0,
and 20 participants were considered to be at high risk of
experiencing adverse outcomes. Mean UGS was
0.75  0.16 m/s, and mean MAC was 31.2  1.9 cm);
both physical parameters were correlated with MFS (UGS:
standardized beta = 0.420, P < .001; MAC: standardized
beta = 0.457, P < .001). Using UGS and MFS, the area
under curve of receiver operating curve for determining
which participants were at high risk of experiencing
adverse outcomes was 0.809 (P < .001).
From the *Graduate School of Medical Science and Engineering, Korea
Advanced Institute of Science and Technology, Daejeon; †Department of
Internal Medicine, Seoul National University Bundang Hospital,
Seongnam; and ‡Seoul National University College of Medicine, Seoul,
Korea.
Address correspondence to Sun-Wook Kim, Department of Internal
Medicine, Seoul National University Bundang Hospital, 82, Gumi-ro
173 Beon-gil, Bundang-gu, Seongnam-si, Gyeonggi-do, 13620, Korea.
E-mail: swkim619@snubh.org
DOI: 10.1111/jgs.15204
JAGS 2017
© 2017, Copyright the Authors
Journal compilation © 2017, The American Geriatrics Society
CONCLUSION: UGS and MAC are viable methods of
clinically screening for frailty. J Am Geriatr Soc 2017.
Key words: screening; frail elderly; walking speed; arm
circumference
T here has been a rise in the number of older adults
undergoing surgical procedures ranging in severity
from minor outpatient surgeries to major surgeries requir-
ing concentrated post-operative care.1 Evidence suggests
that chronological age cannot precisely indicate the physio-
logical reserve of individual older adults.2 Previous studies
have shown that evaluating frailty and physical perfor-
mance using a comprehensive geriatric assessment (CGA)
in older adults undergoing surgery can help stratify post-
operative mortality and complications, providing better
predictability than traditional risk models.3–5 Preoperative
frailty assessment may be indispensable for older adults
undergoing surgical procedures,6 but assessing the frailty
and physical performance of every older adult in a CGA
requires enormous effort in most clinical situations,
because routine CGA examinations require trained person-
nel, established processes, and considerable time.7
There is a way for clinicians to objectively determine
frailty and physical performance without much effort dur-
ing outpatient clinic visits. For example, usual gait speed
(UGS) can predict life expectancy, functional dependency,
and institutionalization in older adults,8–10 and it has been
used as a marker for physical fitness in many geriatric
researches,11–14 but in most studies, trained personnel mea-
sured USG in a large, dedicated examination room. We
noticed that frail older adults tended to walk into the out-
patient clinic more slowly than fit individuals in our daily
clinical practice. We also noticed that frail participants
were generally skinner than their robust counterparts.15,16
Thus, this simple observation brought us to focus on mea-
suring UGS in individuals entering the outpatient clinic.
Mid-arm circumference (MAC) was also measured using a
tape measure while taking blood pressure (BP). We aimed
0002-8614/17/$15.00
2 JUNG ET AL.
to evaluate whether these easily obtainable values can
reflect frailty status in older adults undergoing surgery.
METHODS
Study Participants
We prospectively evaluated 113 individuals aged 65 and
older consecutively attending a geriatric clinic for a preop-
erative visit between November 28, 2016, and February
28, 2017. The purpose of the visits was to evaluate their
risk before undergoing elective surgery at Seoul National
University Bundang Hospital. Information on baseline par-
ticipant characteristics was collected from an electronic
medical records database and included age, sex, surgical
department, and type of disease (malignant or benign). To
confirm the predictive value of UGS and MAC, we
reviewed the in-hospital clinical courses of participants.
One of the authors blinded to screening results examined
postoperative complications including wound problems,
pneumonia, urinary tract infection, delirium, admission to
the intensive care unit, and discharge to a long-term care
hospital. The institutional review board of the Seoul
National University Bundang Hospital reviewed and
approved the study protocol.
Frailty Evaluation
From preoperative CGA data, we calculated a multidimen-
sional frailty score (MFS), which represents frailty on a
numeric scale and is known to predict the outcome of
older adults undergoing various surgical procedures.3,4
Detailed protocols for calculating MFS have been
described previously, and we categorized MFS (range 0–
15) into 5 categories for the purpose of this analysis (0–1,
2–3, 4–5, 6–7, ≥8).3 We considered individuals with a
MFS of 6 or greater to be at high risk of adverse out-
comes, in accordance with a previous report describing
surgical outcomes.3
Measurements for USG and MAC
We measured UGS by observing the gait of individuals
walking into the outpatient clinic, from the door to the
physician’s desk (a distance of approximately 2.7 m), using
an automated laser-gated chronometer attached to the
wall. To minimize any potential bias and random error,
individuals were not informed that they were being moni-
tored, and they were instructed to remove their winter
coats and bags before walking into the room. MAC was
recorded using a tape measure attached to a BP cuff during
routine BP measurement.
Statistical Analysis
Continuous variables were expressed as means with stan-
dard deviations (SDs), and discrete variables were
expressed as counts with percentages. We plotted the
mean and 2 SDs of UGS and MAC according to MFS
quintiles using bars and whiskers. The ability of UGS
and MAC to estimate frailty was evaluated using a linear
regression model. And then the logistic regression model
2017 JAGS
with input variables of UGS and MAC was developed
under the logistic assumption to predict outcome variable
of high MFS score (≥6 points). In the analysis of postop-
erative complications, logistic regression analyses with
unadjusted models and age- and sex-adjusted models
were used. Two-sided P < .05 was considered statistically
significant. All statistical analyses were performed using
SPSS for Windows version 22.0 (IBM, Armonk, NY).
RESULTS
Error during UGS measurement occurred in 15 of 113 par-
ticipants (e.g., take off coats, put bags on office floor, and
bow to doctor while they entered), and 5 participants
entered the office in a wheelchair. Accordingly, data from
93 individuals were used for the final analysis. MFS did
not differ between the 20 excluded and 93 included indi-
viduals (P = .15, t-test), and occurrence of postoperative
complications was not different (P = .63, logistic regres-
sion analysis).
Participants had a mean age of 75.8  4.7; 35 were
male, and 58 were female (Table 1). Mean Charlson
Comorbidity Index was 2.2  1.4; 4 participants were
dependent in 1 or more activities of daily living (ADLs),
and 6 were dependent in 1 or more instrumental ADLs
(IADLs). Seventy-eight participants (83.9%) were referred
from the department of surgery, 14 (15.1%) from gynecol-
ogy, and 1 (1.1%) from orthopedic surgery; 69 partici-
pants (74.2%) underwent surgery for cancer. Mean MFS
was 4.1  2.0, and 20 participants (21.5%) were consid-
ered to be at high risk of adverse outcomes using a MFS
cutoff value of 6 or higher.
Mean UGS was 0.75  0.16 m/s, and mean MAC
was 31.18  1.95 cm. UGS (coefficient of determination
(R2) = 0.177, standardized beta [b] = 0.420, P < .001)
and MAC (R2 = 0.209, b = 0.457, P < .001) were corre-
lated with MFS (Figure 1A,B). Furthermore, multivariable
linear regression analysis showed that combination of UGS
and MAC as independent variables was associated with
Table 1. Baseline Characteristics of Study Population
(N = 93)
Characteristic Value
Age, mean  SD 75.8  4.7
Sex, n
Male 35
Female 58
Charlson Comorbidity Index, mean  SD 2.2  1.4
Activity of daily living dependence, n (%) 4 (4.3)
Instrumental activity of daily living dependence, n (%) 6 (6.5)
Surgical department, n (%)
General surgery 78 (83.9)
Gynecology 14 (15.1)
Orthopedic surgery 1 (1.1)
Cancer, n (%) 69 (74.2)
MFS, mean  SD 4.1  2.0
High-risk surgical patients (MFS≥6) 20 (21.5)
Usual gait speed, m/s, mean  SD 0.75  0.16
Mid arm circumference, cm, mean  SD 31.2  1.9
SD = standard deviation; MFS = multidimensional frailty score.
JAGS 2017 SCREENING FRAILTY WITH GAIT SPEED AND ARM CIRCUMFERENCE 3

MFS (R2 = 0.345, b = 0.413, P < .001); this association

Table 2. Associations Between Various Physical Param-
persisted even after adjusting for age and sex. When a
eters and Postoperative Complications
joint UGS and MAC score was calculated using MFS ≥ 6
points as the dependent variables, (1/(1 + exp Odds Ratio (95% Confidence Interval)
( 7.739 + (4.150*UGS)+(0.531*MAC)))), this score was Parameter P-Value
correlated with MFS, as expected (Figure 1C). Moreover,
Usual gait speed, m/s (continuous)
the area under curve of receiver operating curve of this
Crude model 0.02 (0.00–0.39) .01
joint score in determining high-risk surgical patients was Age-, sex-adjusted 0.02 (0.00–0.57) .02
0.809 (P < .001). model
Thirty-two (34.4%) participants had at least 1 compli- Mid arm circumference, cm (continuous)
cation, including wound problems (n = 20), intensive care Crude model 0.83 (0.70–0.99) .04
unit admission (n = 9), delirium (n = 7), pneumonia Age-, sex-adjusted 0.85 (0.71–1.01) .07
(n = 3), urinary tract infection (n = 3), and discharge to model
long-term care hospitals (n = 5). Logistic regression analy- Joint score (continuous)
Crude model 1.84 (1.16–2.92) .01
sis revealed that UGS and a joint UGS and MAC score
Age-, sex-adjusted 1.77 (1.09–2.88) .02
were associated with the occurrence of at least 1 complica- model
tion after surgery, similarly to the association between Multidimensional frailty score ≥6
MFS and surgical complications (Table 2). Crude model 3.31 (1.12–9.83) .03
Age-, sex-adjusted 3.07 (1.02–9.23) .046
model
DISCUSSION
UGS and MAC, which are easily obtainable measurements
in an outpatient clinic, can be used to screen individuals frailty, are time consuming, requiring longer than 30 min-
for frailty. Moreover, UGS and MAC were associated with utes to complete. Furthermore, these objective methods
MFS, and the joint score derived from USG and MAC also require dedicated personnel and space. Alternatively,
could be used to identify individuals who were at high risk a combination of UGS and MAC could screen frail indi-
of adverse outcomes after surgery. viduals with high sensitivity and objectivity, with minimal
Although predictive ability of USG and MAC for sur- resources. Using this latter approach may allow physicians
vival have previously been evaluated in older adults, the to use the more resource-demanding CGA in frailer indi-
clinical advantages and significance of these measurements viduals, thereby maximizing cost effectiveness and ulti-
have not been fully elucidated.10–12,16 Measuring UGS and mately preventing adverse geriatric outcomes such as falls,
MAC in the office requires minimal effort. We were able delirium, institutionalization, and death.
to determine the frailty of older adults using these two UGS, as measured in a clinical entrance environment,
easily obtainable physical parameters and identify vulnera- can be measured automatically using a laser-gated
ble individuals who needed a more-detailed assessment, chronometer. It has an advantage over formal gait mea-
such as a CGA. surement method because it can be measured without the
There are many validated instruments for screening individual’s knowledge, minimizing any biases; if people
for frailty in older adults,17 but subjective factors, such as are made aware of the fact that their gait speed is being
the proficiency or judgement of the physician and the examined, they may alter their speed, which thus might no
patients’ prejudice on their health status, can influence longer be considered to be “usual” gait speed. Moreover,
results from these instruments, because these tools are usu- because people are usually distracted when entering the
ally measured using self-reported questionnaires. Alterna- clinic—preoccupied with thoughts of their symptoms,
tively, more-objective assessments, such as CGA, which is plans, and laboratory results—they may experience some
considered to be the criterion standard for evaluating cognitive stresses that usually occur in functional tests such

Figure 1. Distribution of (A) usual gait speed, (B) mid-arm circumference, and (C) joint score combining usual gait and mid-arm
circumference, according to categorized multidimensional frailty score. Horizontal bars represent means and standard deviations.
4 JUNG ET AL.
as dual task examination. Based on this contextual differ-
ence, more studies evaluating the mechanisms and clinical
implications of the differences between UGS measured in
the office and formal gait speed assessment are warranted.
Alternatively, not informing people that their gait
speed is being measured might prevent them from reaching
usual gait speed during clinic visit. A total of 15 partici-
pants in our study entered the clinic while taking off their
coats, putting bags on the floor, or bowing to the doctor.
Because previous studies have shown that inability to com-
plete performance-based tasks is associated with frailty, we
checked whether errors during measurement were associ-
ated with frailty or postoperative complication and found
that procedural error was not indicative of frailty.18,19
Nevertheless, to diminish the frequency of errors in mea-
surements, optimizations of the measurement protocol in
acquiring UGS of patients entering the clinic are
warranted.
Our study has several limitations. MAC, because it is
measured using a tape measure on a BP monitor cuff, may
be variable because the thickness of clothing can affect it.
Context-specific cutoff values may be needed. Neverthe-
less, despite these errors, MAC showed a linear relation-
ship with frailty status. In addition, our study included a
small number of participants from an outpatient clinic of a
single tertiary hospital. Furthermore, because UGS and
MAC are parameters not commonly measured clinically,
we could not cross-validate the joint score in an indepen-
dent dataset. Although we were able to show the ability of
UGS, MAC, and the joint score to predict postoperative
complications, a larger multicenter study is required to val-
idate this association for generalizability. Despite such lim-
itations, our prospective, exploratory study suggests that it
is possible to incorporate methods that are easier to per-
form to detect frailty in older adults undergoing, which
may be better than using chronological age.
In conclusion, UGS and MAC measured in an outpa-
tient clinic can be used to screen for frailty and aid in clinical
decision-making in older adults undergoing surgery.
ACKNOWLEDGMENTS
Conflict of interest: None.
Author Contributions: Study concept and design: Hee-
won Jung, Sun-Wook Kim. Acquisition of subjects and
data: Sun-Wook Kim. Analysis and interpretation of data:
Hee-won Jung, Sun-Wook Kim, Mun-gu Kang, Jung-Yeon
Choi, Sol-Ji Yoon, Kwang-il Kim, Cheol-Ho Kim. Prepara-
tion of manuscript: Hee-won Jung, Sun-Wook Kim.
Additional Contributions: Jongkab Kim designed and
provided the automated laser-gated chronometer, without
View publication stats
2017 JAGS
financial compensation. PANOPTICS Corp. and Dynamic
Physiology offer assistance to improve the automated
laser-gated chronometer, also without compensation.
Sponsor’s Role: None.
REFERENCES
1. Etzioni DA, Liu JH, Maggard MA, et al. The aging population and its
impact on the surgery workforce. Ann Surg 2003;238:170–177.
2. McIsaac DI, Bryson GL, van Walraven C. Association of frailty and 1-year
postoperative mortality following major elective noncardiac surgery: A
population-based cohort study. JAMA Surg 2016;151:538–545.
3. Kim SW, Han HS, Jung HW et al. Multidimensional frailty score for the
prediction of postoperative mortality risk. JAMA Surg 2014;149:633–640.
4. Choi JY, Yoon SJ, Kim SW et al. Prediction of postoperative complications
using multidimensional frailty score in older female cancer patients with
American Society of Anesthesiologists Physical Status Class 1 or 2. J Am
Coll Surg 2015;221:652–660.
5. Choi JY, Cho KJ, Kim SW et al. Prediction of mortality and postoperative
complications using the hip-multidimensional frailty score in elderly
patients with hip fracture. Sci Rep 2017;7:42966.
6. Oresanya LB, Lyons WL, Finlayson E. Preoperative assessment of the older
patient: A narrative review. JAMA 2014;311:2110–2120.
7. Puts MT, Hardt J, Monette J et al. Use of geriatric assessment for older
adults in the oncology setting: A systematic review. J Natl Cancer Inst
2012;104:1134–1164.
8. Woo J, Ho SC, Yu AL. Walking speed and stride length predicts
36 months dependency, mortality, and institutionalization in Chinese aged
70 and older. J Am Geriatr Soc 1999;47:1257–1260.
9. Shinkai S, Watanabe S, Kumagai S et al. Walking speed as a good predic-
tor for the onset of functional dependence in a Japanese rural community
population. Age Ageing 2000;29:441–446.
10. Studenski S, Perera S, Patel K et al. Gait speed and survival in older adults.
JAMA 2011;305:50–58.
11. Studenski S, Perera S, Wallace D et al. Physical performance measures in
the clinical setting. J Am Geriatr Soc 2003;51:314–322.
12. Afilalo J, Eisenberg MJ, Morin JF et al. Gait speed as an incremental pre-
dictor of mortality and major morbidity in elderly patients undergoing car-
diac surgery. J Am Coll Cardiol 2010;56:1668–1676.
13. Middleton A, Fritz SL, Lusardi M. Walking speed: The functional vital
sign. J Aging Phys Act 2015;23:314–322.
14. Alfredsson J, Stebbins A, Brennan JM et al. Gait speed predicts 30-day
mortality after transcatheter aortic valve replacement: Results from the
Society of Thoracic Surgeons/American College of Cardiology Tran-
scatheter Valve Therapy Registry. Circulation 2016;133:1351–1359.
15. Ishii S, Tanaka T, Shibasaki K et al. Development of a simple screening
test for sarcopenia in older adults. Geriatr Gerontol Int 2014;14(Suppl
1):93–101.
16. Tsai AC, Chang TL. The effectiveness of BMI, calf circumference and mid-
arm circumference in predicting subsequent mortality risk in elderly Tai-
wanese. Br J Nutr 2011;105:275–281.
17. Hamaker ME, Jonker JM, de Rooij SE et al. Frailty screening methods for
predicting outcome of a comprehensive geriatric assessment in elderly
patients with cancer: A systematic review. Lancet Oncol 2012;13:e437–
e444.
18. Rockwood K, Jones D, Wang Y et al. Failure to complete performance-
based measures is associated with poor health status and an increased risk
of death. Age Ageing 2007;36:225–228.
19. Brothers TD, Theou O, Rockwood K. Do performance-based health mea-
sures reflect differences in frailty among immigrants age 50 + in Europe?
Can Geriatr J 2014;17:103–107.