ArticlePDF Available

Development and Pilot Study of an iOS Smartphone Application for Perioperative Functional Capacity Assessment

Authors:

Abstract and Figures

Background: Functional capacity assessment plays a core role in the preoperative evaluation. The Duke Activity Status Index (DASI) and the 6-minute walk test (6MWT) are 2 methods that have demonstrated the ability to evaluate functional capacity and predict perioperative outcomes. Smartphones offer a novel method to facilitate functional capacity assessment as they can easily administer a survey and accelerometers can track patient activity during a 6MWT. We developed a smartphone application to administer a 6MWT and DASI survey and performed a pilot study to evaluate the accuracy of a smartphone-based functional capacity tool in our Anesthesia and Perioperative Medicine Clinic. Methods: Using the Apple ResearchKit software platform, we developed an application that administers a DASI survey and 6MWT on an iOS smartphone. The DASI was presented to the patient 1 question on the screen at a time and the application calculated the DASI score and estimated peak oxygen uptake (VO2). The 6MWT used the CMPedometer class from Apple's core motion facility to retrieve accelerometer data collected from the device's motion coprocessor to estimate steps walked. Smartphone estimated steps were compared to a research-grade pedometer using the intraclass correlation coefficient (ICC). Distance walked was directly measured during the 6MWT and we performed a multivariable linear regression with biometric variables to create a distance estimation algorithm to estimate distance walked from the number of steps recorded by the application. Results: Seventy-eight patients were enrolled in the study and completed the protocol. Steps measured by the smartphone application as compared to the pedometer demonstrated moderate agreement with an ICC (95% CI) of 0.87 (0.79-0.92; P = .0001). The variables in the distance estimation algorithm included (β coefficient [slope], 95% CI) steps walked (0.43, 0.29-0.57; P < .001), stride length (0.38, 0.22-0.53; P < .001), age in years (-1.90, -3.06 to -0.75; P = .002), and body mass index (-2.59, -5.13 to -0.06; P = .045). The overall model fit was R = 0.72, which indicates a moderate level of goodness of fit and explains 72% of the variation of distance walked during a 6MWT. Conclusions: Our pilot study demonstrated that a smartphone-based functional capacity assessment is feasible using the DASI and 6MWT. The DASI was easily completed by patients and the application clearly presented the results of the DASI to providers. Our application measured steps walked during a 6MWT moderately well in a preoperative patient population; however, future studies are needed to improve the smartphone application's step-counting accuracy and distance estimation algorithm.
Content may be subject to copyright.
Downloaded from https://journals.lww.com/anesthesia-analgesia by oacFA3WHJnsy4Y9pryI47b2JsX4/2a86P/ix7zW3ZK2DgAlVA7KRfx0/7EPqanwRGkwOO6yno7tWjsF77yZaVKGdqPD3+MEo9ntdyqRNCq4AjEqiaDI7bDOni1EHxj9n on 03/11/2020
Downloadedfromhttps://journals.lww.com/anesthesia-analgesia by oacFA3WHJnsy4Y9pryI47b2JsX4/2a86P/ix7zW3ZK2DgAlVA7KRfx0/7EPqanwRGkwOO6yno7tWjsF77yZaVKGdqPD3+MEo9ntdyqRNCq4AjEqiaDI7bDOni1EHxj9n on 03/11/2020
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
XXX XXX Volume XXX Number XXX www.anesthesia-analgesia.org 1
GLOSSARY
6MWT = 6-minute walk test; 95% CI = 95% condence interval; AUROC = area under receiver oper-
ating characteristic; APMC = Anesthesia Perioperative Medicine Clinic; CONSORT = Consolidated
Standards of Reporting Trials; DASI = Duke Activity Status Index; HIPPA = Health Insurance
Portability and Accountability Privacy Act; ICC = intraclass correlation coefcient; IQR = interquartile
range; LoA = limits of agreement; MACE = major adverse cardiac events; METs = metabolic equiva-
lents; METS = Measurement of Exercise Tolerance Before Surgery trial; SD = standard deviation;
TREND = Transparent Reporting of Evaluations with Nonrandomized Designs; Vo2 = oxygen uptake
KEY POINTS
Question: Can an Apple smartphone–based functional capacity assessment application that
includes a Duke Activity Status Index questionnaire and 6-minute walk test accurately assess
functional capacity assessment?
Findings: The Duke Activity Status Index was easy to implement and perform; however, the
smartphone was only moderately accurate when compared to a research-grade pedometer at
measuring steps walked and estimating the distance traveled during the 6-minute walk test,
especially at slower gait speeds.
Meaning: The results of this pilot study suggest a smartphone-based functional capacity
assessment platform can be used to implement a Duke Activity Status Index questionnaire,
but more research is required to improve the accuracy of a 6-minute walk test smartphone
application in preoperative patients with slower gait speeds.
BACKGROUND: Functional capacity assessment plays a core role in the preoperative evalua-
tion. The Duke Activity Status Index (DASI) and the 6-minute walk test (6MWT) are 2 methods
that have demonstrated the ability to evaluate functional capacity and predict perioperative out-
comes. Smartphones offer a novel method to facilitate functional capacity assessment as they
can easily administer a survey and accelerometers can track patient activity during a 6MWT.
We developed a smartphone application to administer a 6MWT and DASI survey and performed
a pilot study to evaluate the accuracy of a smartphone-based functional capacity tool in our
Anesthesia and Perioperative Medicine Clinic.
METHODS: Using the Apple ResearchKit software platform, we developed an application that admin-
isters a DASI survey and 6MWT on an iOS smartphone. The DASI was presented to the patient
1 question on the screen at a time and the application calculated the DASI score and estimated
peak oxygen uptake (Vo2). The 6MWT used the CMPedometer class from Apple’s core motion facil-
ity to retrieve accelerometer data collected from the device’s motion coprocessor to estimate steps
walked. Smartphone estimated steps were compared to a research-grade pedometer using the intra-
class correlation coefcient (ICC). Distance walked was directly measured during the 6MWT and we
performed a multivariable linear regression with biometric variables to create a distance estimation
algorithm to estimate distance walked from the number of steps recorded by the application.
RESULTS: Seventy-eight patients were enrolled in the study and completed the protocol. Steps
measured by the smartphone application as compared to the pedometer demonstrated mod-
erate agreement with an ICC (95% CI) of 0.87 (0.79–0.92; P = .0001). The variables in the
distance estimation algorithm included (β coefcient [slope], 95% CI) steps walked (0.43,
0.29–0.57; P < .001), stride length (0.38, 0.22–0.53; P < .001), age in years (−1.90, −3.06 to
−0.75; P = .002), and body mass index (−2.59, −5.13 to −0.06; P = .045). The overall model
t was R2 = 0.72, which indicates a moderate level of goodness of t and explains 72% of the
variation of distance walked during a 6MWT.
CONCLUSIONS: Our pilot study demonstrated that a smartphone-based functional capacity
assessment is feasible using the DASI and 6MWT. The DASI was easily completed by patients
and the application clearly presented the results of the DASI to providers. Our application mea-
sured steps walked during a 6MWT moderately well in a preoperative patient population; how-
ever, future studies are needed to improve the smartphone application’s step-counting accuracy
and distance estimation algorithm. (Anesth Analg XXX;XXX:00–00)
Development and Pilot Study of an iOS
Smartphone Application for Perioperative
Functional Capacity Assessment
Daniel S. Rubin, MD, MS,* Allison Dalton, MD,* Allyson Tank, BS, Mark Berkowitz, BS,
David E. Arnolds, MD, PhD,* Chuanhong Liao, MS,§ and Rebecca M. Gerlach, MD, FRCPC*
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
2 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Smartphone 6MWT Functional Capacity
Functional capacity assessment plays a critical role in
the preoperative evaluation of patients presenting
for noncardiac surgery. The 2014 American College
of Cardiology and American Heart Association Guidelines
on perioperative cardiovascular evaluation and manage-
ment recommend functional capacity should be assessed
on all patients with an elevated risk (1%) of major adverse
cardiac events (MACE) before noncardiac surgery.1 Two
common methods of assessment recommended by the
guidelines include subjective assessment by the physician
or utilizing a structured physical activity questionnaire,
such as the Duke Activity Status Index (DASI). Adequate
functional capacity as dened by the guidelines includes
performing activities associated with >4 metabolic equiva-
lents (METs), such as climbing a ight of stairs or walking
at a pace of 4 mph.1 However, the Measurement of Exercise
Tolerance before Surgery (METS) trial demonstrated that
subjective functional capacity assessment did not predict
MACE or accurately estimate a patient’s true functional
capacity when compared to peak oxygen uptake (V2) using
a cycle ergometer and expired gas analysis.2 In contrast to
subjective assessment, the DASI predicted MACE moder-
ately well in the same trial.
The 6-minute walk test (6MWT) is a submaximal exercise
test that is simple, safe, and cost-effective and can be used to
assess functional capacity.3 The distance covered during the
test is moderately correlated with peak V2 and ventilatory
threshold in preoperative patients.4,5 A recent substudy of the
METS trial demonstrated that the 6MWT predicts MACE in
patients having moderate risk surgery.4 Although smartphone
applications have been previously developed to administer a
6MWT, their use has been limited to patients with heart fail-
ure, pulmonary hypertension, and chronic obstructive pul-
monary disease.6,7 No 6MWT smartphone application has
been specically developed or tested in the preoperative pop-
ulation. A prior study demonstrated that a tablet (iPad; Apple
[Cupertino, CA])-based DASI questionnaire is easy to use by
patients.8 A smartphone application that combines the 6MWT
and DASI for preoperative evaluation may thus facilitate
6MWT testing and improve preoperative risk stratication.
We developed an Apple smartphone-based 6MWT and
DASI survey to assess preoperative functional capacity.
We hypothesized that a smartphone application can pro-
vide a good estimate of preoperative functional capacity
through the administration of a DASI survey and a 6MWT
that directly measures steps walked during the test and
estimates distance traveled. To test our hypothesis, we
conducted a prospective cohort study in our Anesthesia
Perioperative Medicine Clinic (APMC).
METHODS
Study Population
The study was approved by the University of Chicago
Institutional Review Board (IRB17-0191). This manuscript
adheres to the applicable Transparent Reporting of Evaluations
with Nonrandomized Designs (TREND) guidelines. Patients
with an appointment at University of Chicago’s APMC were
considered eligible for enrollment in the study. Patients at
elevated risk for perioperative complications are scheduled
for a preoperative evaluation in the APMC and functional
capacity is estimated in all patients during the visit. All adult
(age >18 years old) patients presenting for noncardiac surgery
were approached for enrollment unless they were nonambula-
tory or had a contraindication to performing a 6MWT such as
unstable angina, myocardial infarction within 30 days, resting
heart rate >120 bpm, a systolic blood pressure >180 mm Hg,
or a diastolic blood pressure >100 mm Hg.9 The use of assis-
tive walking devices such as canes, walkers, or other devices
and the use of supplemental oxygen were not contraindica-
tions for enrollment in the study. Informed written consent
was obtained from all participants before enrollment. Patients
were approached for enrollment in the study between June 12,
2017 and August 15, 2017 and between January 8, 2018 and
January 26, 2018. APMC and study protocols remained the
same in both time periods. The additional enrollment period
was added post hoc but before data analysis due to a malfunc-
tion in one of the pedometers used during data acquisition for
step counts. We approached 350 patients for enrollment during
the 2 time periods, and 86 patients were enrolled in the study.
To ensure that we had an adequate sample size to compare
the application to the pedometer, we enrolled 31 additional
patients in the second enrollment period.
Procedures
After enrollment, patients were assessed in APMC by an
APMC clinician: either a resident physician in anesthesiol-
ogy (together with a faculty anesthesiologist) or by a special-
ized advanced practice nurse. The routine APMC assessment
includes obtaining a complete medical and surgical history,
demographics, height, weight and vital signs, including heart
rate, blood pressure, and oxygen saturation. The patient’s
weight and height are obtained by a medical-grade scale and
stadiometer, respectively. As part of the review of systems,
APMC clinicians subjectively assess the patient’s functional
capacity and estimate whether the patient can achieve 4
METs. The APMC assessment of functional capacity occurred
independent of the study protocol. APMC clinicians did not
have access to the results of the DASI or 6MWT when mak-
ing their assessment of functional capacity. Similarly, study
investigators did not consider the APMC assessment of func-
tional capacity in conducting the 6MWT. Study investigators
obtained the patient’s medical history and procedure informa-
tion retrospectively from the chart. Study investigators addi-
tionally measured stride length that was averaged (cm) over a
10-step distance. Study investigators remeasured the patient’s
heart rate and oxygen saturation immediately before and after
the 6MWT using a Phillips SureSigns VS3 pulse oximeter
(Andover, MA).
From the *Department of Anesthesia and Critical Care, and Pritzker School
of Medicine, the University of Chicago, Chicago, Illinois; Lander College,
Kew Garden Hills, New York; and §Department of Public Health Sciences,
the University of Chicago, Chicago, Illinois.
Accepted for publication August 13, 2019.
Funding: The smartphone application was developed during the UChicago
App Challenge 2017 and won the grand prize. IT Services of the University of
Chicago provided programming support during the development.
Conicts of Interest: See Disclosures at the end of the article.
Reprints will not be available from the authors.
Address correspondence to Daniel S. Rubin, MD, MS, Department of Anes-
thesia and Critical Care, the University of Chicago, 5841 S Maryland Ave,
MC-4028, Chicago, IL 60637. Address e-mail to drubin@dacc.uchicago.edu.
Copyright © 2019 International Anesthesia Research Society
DOI: 10.1213/ANE.0000000000004440
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
XXX XXX Volume XXX Number XXX www.anesthesia-analgesia.org 3
Application Development
The Step Test application consisted of 2 primary components:
(1) a DASI questionnaire and (2) a timed 6MWT. The DASI
questionnaire was presented to the patient with every question
presented 1 at a time on the iOS screen and accompanied with
a yes/no response at the bottom of the screen (Figure1A). The
application stored the responses of the DASI questionnaire in
a comma-separated value le and calculated the DASI score
and peak V2 using the original formula described by Hlatky
peak V2 = (DASI score × 0.46) + 9.6.10 Estimated METs were
then calculated by dividing peak V2/3.5. The patient and
study investigator could view the results of the DASI after they
completed the questionnaire, as the application displayed the
DASI score, peak V2, and the responses to the DASI questions
(Figure1B). The patient was required to complete the survey
before engaging in the 6MWT. Once the DASI was completed,
the application directed the user to engage in a 6MWT.
The 6MWT was created using the Apple ResearchKit soft-
ware timed walk test feature. The mobile application used
the CMPedometer class from Apple’s core motion facility to
retrieve data collected from the device’s motion coproces-
sor. The coprocessor records data from the device’s sensors,
including the accelerometer and gyroscope, in raw form and
then uses these data to derive a calculated step count, ights
of stairs climbed, and other derived series. For this applica-
tion, we only recorded step counts during the 6MWT.
6-Minute Walk Test
The 6MWT was administered according to the American
Thoracic Society guidelines which included the script
detailing the instructions on how to perform a 6MWT
and voice prompts at each minute during the test.9 Only 2
research personnel (M.B./D.E.A.) administered the 6MWT
to all patients and both observed the conduct of a 6MWT by
physical therapists at our Adult Cardiac Rehabilitation cen-
ter before beginning enrollment. An indoor 30-m track was
created using orange cones to identify the beginning and
end of each lap and a line of cones was placed to the side
of the track to ensure that no pedestrians interfered with
the test. Patients were given the option to use their own
smartphone onto which our application was downloaded,
or they were provided with a study phone (iPhone 6, Apple,
Cupertino, CA) that contains the M8 motion coprocessor.
The smartphone was then placed in either the patient’s
front pants pocket or attached to a waist belt using a belt
clip (Stalion Secure Belt Clip Holster; Stalion Products,
Fresh Meadow, NY), depending on patient preference. Step
counts with the smartphone in the front pocket or hip/belt
location were considered equivalent.6 In addition, patients
were tted with a Yamax Digiwalker SW-701 Pedometer
(Yamasa Tokei Keiki Co, Ltd, Tokyo, Japan) that has been
previously validated to accurately identify step counts.11
After starting the 6MWT application, the patient placed
the smartphone in his or her pocket or hip holster. The
application would begin a 5-second countdown, the start
of the test was announced, and the patient would begin
walking 30 m to the next cone, making a 180° turn, and
returning to the rst cone to complete a 60-m lap. The track
was marked at 1-m intervals and the distance traveled was
measured to the nearest meter by the research personnel.
Figure 1. Sample screens of the iOS smart-
phone application. A, Example of a ques-
tion from the DASI questionnaire from our
smartphone application on an iOS device.
Each question was presented to the patient
individually and a yes or no response was
required to proceed to the next question.
B, Example of the results from the DASI
questionnaire for a sample patient. The
total DASI score, peak Vo2, and response
to each question can be presented to the
provider to review. A green square next to
the question indicates a “yes” response
by the patient to that question whereas a
red square indicates a “no” response. DASI
indicates Duke Activity Status Index; Vo2,
oxygen uptake.
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
4 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Smartphone 6MWT Functional Capacity
The patient walked until the phone stopped the test by
vibrating/ringing to signal that the 6MWT was completed.
On completion of the test, the research personnel would
directly measure the total distance walked, record the steps
walked on the pedometer, and conrm that the results of the
Step Test application had been successfully sent to a secure
University of Chicago server.
Statistical Analysis
Categorical variables are presented as the number and per-
centage; continuous variables are presented as a mean and
standard deviation or median and interquartile range (IQR)
if the values are not normally distributed. The Shapiro-
Wilk test, quantile-quantile plots, and histogram plots of
the variables were used to assess for normality. To validate
the step count of the smartphone application, we compared
the results of the application to the SW-701 Pedometer,
because an accurate measurement of steps walked dur-
ing the 6MWT is necessary to estimate the distance from
the multivariable regression. We used a Bland-Altman plot
to assess for agreement between steps counted by the iOS
smartphone and Yamax Pedometer. The systematic bias (ie,
mean difference) and the condence intervals for 95% limits
of agreement (LoA) were calculated with exact parametric
condence intervals.12 The intraclass correlation coefcient
(ICC) estimated agreement between the iOS application
and the pedometer for step counts in a 2-way mixed-effects
model, with the xed effects of the devices (smartphone
and pedometer) and individual patient as a random effect.
In addition, Lin's13 concordance correlation coefcient was
calculated between steps counted by the iOS smartphone
and Yamax pedometer. The best t and line of perfect con-
cordance were plotted to illustrate the agreement between
the 2 devices.14 The ICC and Lin's13 concordance correla-
tion coefcient were used to compare the accelerometers/
pedometers because the ICC is a standard measurement
of interrater reliability and Lin's13 concordance correlation
coefcient is a measure of concordance.
We then t a multivariable linear regression to estimate
6MWT distance using measured step count from the iOS
smartphone application and a priori determined biometric
variables as independent variables. All analysis was per-
formed with STATA/MP 15 (College Station, TX). Power
and sample size calculations were performed using PASS
(NCSS Statistical Software, Kaysville, UT). The statistically
signicant level was set as a 2-sided P value of <.05.
Sample Size/Power Analysis
A sample size of 50 subjects with 2 observations per sub-
ject achieves 80% power to detect an ICC of 0.75 under the
alternative hypothesis when the ICC under the null hypoth-
esis is 0.55 using an F test with a signicance level of .05.15
During initial patient enrollment, we encountered a faulty
Digimax pedometer that gave abnormally low step counts
(<30 steps for a >500-m walk) or failed to count steps dur-
ing the 6MWT. Those patients (n = 20) were not used to
compare the step count between the Step Test application
and the pedometer; however, they were included in the
multivariable distance estimation algorithm as the Step Test
application worked appropriately. We thus enrolled addi-
tional patients during the second enrollment period to meet
our enrollment criteria of 50 patients to compare the Step
Test application to the pedometer. Additional enrollment
was performed before data analysis.
RESULTS
Eighty-six patients were enrolled in the study during their
APMC visit (Figure2; Consolidated Standards of Reporting
Trials [CONSORT]). After enrollment, 8 patients were
dropped from the study, 4 patients declined to participate
in the 6MWT, 1 patient did not complete the 6MWT due
to lower extremity pain, 2 patients did not have successful
data transmission after the completed test, and 1 patient did
not ll out the DASI questionnaire appropriately. Patient
characteristics are shown in Table 1. The median age of
patients was 65 (IQR: 55–70) years, BMI 28 (IQR: 26–32), and
Figure 2. Consort diagram that indicates patient enrollment. An additional enrollment period was necessary to reach a sample size of 50
patients to be adequately powered for comparison between the smartphone and pedometer. *Patients enrolled who did not have pedometer
data (n = 20) were included in the multivariable regression analysis because we had smartphone step data and directly measured 6MWT
distance on those patients. 6MWT indicates 6-minute walk distance; DASI, Duke Activity Status Index.
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
XXX XXX Volume XXX Number XXX www.anesthesia-analgesia.org 5
stride length 60.0 ± 11.1 cm. The median DASI score was
50.7 (IQR: 30.2–58.2) which translated into an estimated 8
± 2 METs.
The mean distance achieved during the 6MWT was 432
± 116 m. 6MWT distance was normally distributed and the
relationship between DASI estimated METs and 6MWT is
shown in Figure 3. Pedometer data were available for 58
patients who completed the 6MWT. Mean steps recorded
were 599 ± 137 from iOS devices and 602 ± 156 for the pedom-
eter. Steps measured by the application as compared to the
pedometer were moderately accurate with an ICC (95% CI)
of 0.87 (0.79–0.92; P < .001) although the Lin's13 concordance
correlation coefcient was 0.73 (95% CI, 0.61–0.85; P < .001).
The Bland-Altman plot for the 2 devices in Figure4A illus-
trated that agreement between the 2 devices appears good
with a small mean difference which the mean observed dif-
ference ±1.96 standard deviations (SD) is equivalent to −3.3
steps (−214.7 to 208.0). Total step counts below the overall
mean (<600 steps) demonstrated greater variability and less
accuracy by the smartphone as compared to the pedometer
(Figure4B). For step counts <600 steps, the mean difference
(SD) was 13.8 (151.2) and the Bland-Altman 95% LoA were
(−283.6 to 310.2). For step counts >600 steps, the mean dif-
ference was −13.8 (70.1) steps and the Bland-Altman 95%
LoA were (−151.2 to 124.6).
Seventy-eight patients were included in the distance esti-
mation regression. For the distance estimation regression,
we included biometric variables that a priori impact 6MWT
distance. Variables included in model development were
stride length, measured steps, body mass index, and age.
The results from the multivariable regression can be seen in
Table2. The variables in the distance estimation algorithm
include (β coefcient, 95% CI) steps walked (0.43: 0.29–0.57;
P < .001), stride length in cm (0.38: 0.22–0.53; P < .001), age
in years (−1.90: −3.06 to −0.75; P = .002), and body mass
index (−2.59: −5.13 to −0.06; P = .045). The mean estimated
distance from the distance estimation regression is 432 ± 99
m as compared to the mean measured distance of 432 ± 112
m. The overall model t was R2 = 0.72, which indicates that
our regression equation accounted for 72% of the variation
in 6MWT distance as estimated by the coefcients in our
regression model, which suggests it performed moderately
well.
DISCUSSION
We found that, in a preoperative population, a functional
capacity smartphone application that combined a DASI
survey and 6MWT can be implemented in a preoperative
medicine clinic to facilitate functional capacity assessment.
The 6MWT smartphone application was moderately accu-
rate at measuring steps walked as compared to a validated
research pedometer and our distance estimation algorithm
accounted for roughly 72% of the variation seen in 6MWT
distance. Further research is required to improve and rene
step-counting accuracy and 6MWT distance prediction in a
larger population and ensure that the application can accu-
rately estimate distance walked in a nonclinic setting.
Our study is novel in that a preoperative smartphone
application may facilitate greater use of the 6MWT and
DASI before noncardiac surgery. Functional capacity assess-
ment is a core component of the preoperative evaluation;
however, the optimal method of functional capacity assess-
ment remains uncertain. The DASI was recently shown in
the METS trial to be superior to cardiopulmonary exercise
testing and subjective functional capacity assessment in a
2018 prospective cohort.2 In a subsequent substudy of the
METS trial, the 6MWT distance had a superior prediction of
30-day mortality or MI (6MWT distance area under receiver
Table 1. Patient Demographics and Medical
History
Characteristics N = 78
Age, median (IQR), y 65 (55–70)
Sex, n (%)
Male 40 (51)
Female 38 (49)
BMI, median (IQR) 28 (26–32)
Comorbidities, n (%)
Coronary arter y disease 17 (22)
Heart failure 4 (5)
Valvular disease 8 (11)
Atrial brillation 4 (6)
Hypertension 74 (62)
COPD 3 (4)
Asthma 10 (13)
PVD 1 (1)
Diabetes mellitus 26 (13)
History of stroke/TIA 10 (2)
Arthritis 17 (22)
Surgical procedure, n (%)
Urology 18 (23.4%)
General 16 (20.7%)
Orthopedic 13 (16.9%)
Plastics 10 (13%)
Gynecology 9 (11.7%)
Vascular 5 (6.5%)
Otolaryngology 3 (3.9%)
Othera4 (5.2%)
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary
disease; IQR, interquartile range; PVD, peripheral vascular disease; TIA,
transient ischemic attack.
aOther: thoracic (n = 2), transplant (n = 1), ophthalmology (n = 1).
Figure 3. A box-plot graph illustrates the relationship between
estimated METs as calculated from the DASI and the 6MWT. As
expected, there is an increase in 6MWT distance with predicted
METs from the DASI questionnaire. METs were calculated using the
formula: (DASI × 0.43) + 9.6)/3.5 = METs. 6MWT indicates 6-min-
ute walk distance; DASI, Duke Activity Status Index; METs, metabolic
equivalents.
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
6 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Smartphone 6MWT Functional Capacity
operating characteristic [AUROC], 0.70 [0.52–0.70; P = .03])
as compared to the DASI, while the DASI had a superior
prediction for disability-free survival.4 Our application com-
bines both the DASI and 6MWT in an easy-to-use smart-
phone platform, providing the anesthesiologist with access
to both measures of functional capacity. The application
can easily guide the patient to complete the DASI survey
and clearly present the results to the provider in an easy-
to-visualize format that concisely presents the DASI score,
peak V2 and displays responses to the questions (Figure2).
Figure 4. A, Bland-Altman plot comparing steps measured from the Yamax Digiwalker SW-701 Pedometer and smartphone during 6MWT.
At lower step counts (<600 steps), there is increased variability between the 2 devices and undercounting of steps by the iOS device which
is consistent with previous studies. One outlier is secondary to a signicant undercounting of steps by the iOS smartphone. B, Agreement
between the Smartphone and Pedometer: The dashed line represents the line of perfect concordance and the solid line represents the “best
t” for the observed data. The data suggest that the device is less accurate at lower step counts (<600) with increased variability and scatter
away from the line of perfect concordance at lower step counts. For step counts <600 steps, the mean difference (SD) was 13.8 (151.2) and
the Bland-Altman 95% LoA (−283.6 to 310.2). For step counts >600, the mean difference was −13.8 (70.1) steps and the Bland-Altman 95%
LoA (−151.2 to 124.6). 6MWT indicates 6-minute walk distance; CI, condence interval; LoA, limits of agreement; SD, standard deviation.
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
XXX XXX Volume XXX Number XXX www.anesthesia-analgesia.org 7
In addition, the 6MWT can be performed independently
by the patient without tying up ofce staff. The use of this
application may facilitate a transition away from subjec-
tive functional capacity assessment, which is unreliable for
perioperative risk stratication.2 Future studies will evalu-
ate whether the application can be used at home before the
clinic visit, because a previous study demonstrated patients
can complete a 6MWT at home without any prior education.6
Our results demonstrate that an iOS smartphone at either
the hip-belt or front hip-pocket position was moderately accu-
rate at measuring steps during a timed walk test in a preop-
erative patient population. The proliferation of smartphones
and other wearable technology has increased the opportunity
to easily measure objective physical activity in preoperative
patients. A primary concern of this new technology has been
the validity of the measurements because the proprietary
step-counting algorithms created by the smartphone manu-
facturer cannot be independently validated.16 In addition,
the majority of validation studies on wearables have focused
on healthy individuals walking on a motorized treadmill,
or required patients to alter their normal walking speed (eg,
slow walking pace), which can alter gait mechanics.17–22 Our
study was performed using standardized placement of the
smartphone on the hip (front pant pocket or belt clip) and self-
paced walking with a wide range of 6-minute walk distances.
The wide LoA as demonstrated in the Bland-Altman plot are
concerning for the accuracy and assessment of patients with a
poor functional capacity, because the majority of the variation
was seen in patients with <600 total steps during the walk.
This nding is consistent with previous studies that identi-
ed decreased accuracy of smartphone step measurements in
patients walking at slower (1 m/s) speeds.19,20 To improve
the accuracy, we have recently developed an updated ver-
sion of the application that uses an open-source algorithm to
measure steps walked, which has been validated in older and
slower populations.23,24 Future studies are required to validate
this method in preoperative patients.
Our application utilized a distance estimation algo-
rithm to generate distance walked rather than the global
positioning system or estimated distance from the device
itself. In contrast, the MyHeart Counts study used a pedom-
eter calibrated by the global positioning system and noted
decreased accuracy in the distance identied during the
6MWT distance.25 Using the global positioning system
method, distance traveled is derived in part from the length
of the device owners’ stride, which the device determines in
the rst few days of use by comparing the owner’s unique
gait and pace to global positioning data. This approach
limits the utility of the application to the owner of the
phone and cannot be deployed in a clinic setting with other
patients. In addition, we did not utilize the global position-
ing system to identify the distance traveled because the
signal in a dense urban area, even when connected to a wire-
less server, is poor and would have high error rates.26 We
believe these concerns outweigh the requirement that stride
length be measured in our distance estimation algorithm.
Our approach also mirrored a previous study to estimate
the distance from an iOS-administered 6MWT in patients
with heart failure and pulmonary hypertension.6 However,
our application did not vary the default sensitivity of the
application based on gait characteristics of the target popu-
lation as in previous studies. In effect, the inclusion of stride
length as a variable in our distance estimation algorithm
makes our application more generalizable to the preopera-
tive population. A follow-up study is thus needed to verify
the accuracy in a much larger patient population.
Patient-facing smartphone applications pose unique
challenges to effectively integrate into the normal workow
of a clinic, including distribution of the application, inte-
gration of the data collected, and compliance with Apple’s
privacy policy and the Health Insurance Portability and
Accountability Privacy Act (HIPPA). Our study distrib-
uted the application to patients using the Apple Business
Manager platform that allows selective distribution of
applications using a unique redemption code. Patients who
chose to use their own smartphones were sent a link to a
unique code that would automatically download the appli-
cation when the code was redeemed. Patients without an
Apple smartphone were offered the use of a study phone
but not all clinics may have this resource available. For
applications that require patients to wear the smartphone,
such as during the 6MWT, the patient’s attire can pose limi-
tations, such as with a dress or pants pockets that cannot
accommodate a smartphone. Patients need to wear clothes
that facilitate performance of a 6MWT and securely hold the
device to ensure optimal performance and data collection.9
This limitation will likely disproportionately impact female
patients. The data generated from the application need to be
integrated into the medical record if this application is to be
used effectively by clinicians. Our application displayed the
results of the DASI and 6MWT on the smartphone only, but
integration into the electronic medical record will greatly
facilitate adoption and smooth clinic workow. In addition,
Apple’s terms of use for HealthKit requires a privacy policy
that patients consent to before using any information gen-
erated by the smartphone. Our study required patients to
consent to this agreement on opening the application and
data transmitted to our research server were encrypted and
did not contain protected health information to ensure com-
pliance with HIPPA.
Table 2. Multivariable Linear Regression Equation
to Estimate Distance Walked During the 6MWT
Variable Units
β Coefficient
(95% Confidence
Interval) P
Steps iPhone Number of steps
measured by
smartphone
application
0.43 (0.29–0.57) <.001
Stride Length Centimeters 0.38 (0.22–0.53) <.001
Age Years −1.90 (−3.06 to −0.75) .002
BMI kg/m2−2.59 (−5.12 to −0.06) .045
Data from 78 patients were used to generate the distance estimation
algorithm. A multivariable linear regression was performed to estimate the
distance walked during the 6MWT from the number of steps measured by the
smartphone application. Variables included in the regression were identied
a priori and include: Steps measured from the Step Test application (Steps
iPhone), average stride length of the patient (stride length in cm), age of
the patient (age in years), and BMI (BMI in kg/m2). A patient in our cohort,
who walked 583 steps, an average stride length of 60 cm, an age of 62 y,
and a BMI of 30 would have an estimated distance walked of 428 m (95%
condence interval, 414–442). 95% condence intervals were generated
using the following formula: β coefcient ± 1.96 × standard error.
Abbreviations: 6MWT, 6-minute walk test; BMI, body mass index.
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
8 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Smartphone 6MWT Functional Capacity
LIMITATIONS
Our study has limitations. Because we limited smart-
phone position to the front pocket or hip-belt position,
our results may not be generalizable to other positions on
the patient’s body (coat pocket, front shirt pocket, purse,
skirt pocket). Such changes in position may produce dif-
ferent acceleration patterns from normal walking and
make measured steps inaccurate. Also, in our study, we
used an iPhone 6 as the study phone. Other phone sizes,
hardware, and operating systems may affect step count-
ing. While newer smartphones contain different motion
coprocessors, this modication is unlikely to worsen
device performance. As mentioned earlier, the propri-
etary nature of the step-counting algorithm by Apple can
be updated/changed without notication to the user or
researcher. While it is likely that the accuracy of the algo-
rithm will improve with each iteration for patients with a
normal gait pattern, the results may change for individu-
als with abnormal gait pathology. In light of this possibil-
ity, our future research will utilize open-source algorithms
to interpret raw accelerometry data rather than relying on
proprietary algorithms.16,23 In addition, our application
cannot be generalized to include the Apple Watch as wrist
and hip walking patterns are fundamentally different. We
chose to focus on a smartphone-based approach as they
are much more prevalent and hip accelerometry is more
reliable over a variety of different pathologies for identi-
fying steps.27 Finally, because this was a pilot study, it con-
tained a relatively small sample size and thus our results
may be statistically fragile.
CONCLUSIONS
We demonstrated that our smartphone application can be
easily implemented into a perioperative medicine clinic to
perform a DASI survey and 6MWT to preoperative patients.
As the role of functional capacity assessment continues to
evolve, the combination of a smartphone-based DASI ques-
tionnaire and 6MWT offers an easy-to-use and ubiquitous
platform to improve perioperative risk stratication. Further
research is needed to continue to improve the 6MWT smart-
phone application step counting and distance estimation
to ensure that patients with slow gait speeds are accurately
assessed because these patients are the most likely to be clas-
sied as having a poor functional capacity. E
ACKNOWLEDGMENTS
The authors are grateful to Ms Cornelia Bailey, MS, Strategic
Innovation Consultant IT Services, and Fritz Anderson, JD,
Senior iOS Developer IT Services, for their assistance in
developing and programming the Step Test application.
DISCLOSURES
Name: Daniel S. Rubin, MD, MS.
Contribution: This author helped with the conception, acquisition,
analysis, drafting, and revising for content, nal approval, held
accountable for all aspects of the study.
Conicts of Interest: D. S. Rubin is the president of DRDR Mobile
Health, a company that creates mobile applications for healthcare.
He has engaged in consulting for mobile applications as well.
Name: Allison Dalton, MD.
Contribution: This author helped with the conception, analysis,
drafting, and revising for nal content and held accountable.
Conicts of Interest: None.
Name: Allyson Tank, BS.
Contribution: This author helped with the analysis, drafting, and
revising for nal content, held accountable for all aspects of the
study.
Conicts of Interest: None.
Name: Mark Berkowitz, BS.
Contribution: This author helped with the acquisition, revising for
nal content, held accountable for all aspects of the study.
Conicts of Interest: None.
Name: David E. Arnolds, MD, PhD.
Contribution: This author helped with the acquisition, revising for
nal content, held accountable for all aspects of the study.
Conicts of Interest: None.
Name: Chuanhong Liao, MS.
Contribution: This author helped with the statistical analysis,
drafting and revising for nal content, held accountable for all
aspects of the study.
Conicts of Interest: None.
Name: Rebecca M. Gerlach, MD, FRCPC.
Contribution: This author helped with the conception, acquisition,
analysis, drafting, and revising for content, nal approval, held
accountable for all aspects of the study.
Conicts of Interest: None.
This manuscript was handled by: Tong J. Gan, MD.
REFERENCES
1. Fleisher LA, Fleischmann KE, Auerbach AD, et al; American
College of Cardiology; American Heart Association. 2014 ACC/
AHA guideline on perioperative cardiovascular evaluation and
management of patients undergoing noncardiac surgery: a
report of the American College of Cardiology/American Heart
Association Task Force on practice guidelines. J Am Coll Cardiol.
2014;64:e77–e137.
2. Wijeysundera DN, Pearse RM, Shulman MA, et al; METS Study
Investigators. Assessment of functional capacity before major
non-cardiac surgery: an international, prospective cohort study.
Lancet. 2018;391:2631–2640.
3. Arena R, Myers J, Williams MA, et al; American Heart Association
Committee on Exercise, Rehabilitation, and Prevention of the
Council on Clinical Cardiology; American Heart Association
Council on Cardiovascular Nursing. Assessment of functional
capacity in clinical and research settings: a scientic statement from
the American Heart Association Committee on exercise, rehabilita-
tion, and Prevention of the Council on Clinical Cardiology and the
Council on Cardiovascular Nursing. Circulation. 2007;116:329–343.
4. Shulman MA, Cuthbertson BH, Wijeysundera DN, et al;
Measurement of Exercise Tolerance for Surgery Study
Investigators. Using the 6-minute walk test to predict disability-
free survival after major surgery. Br J Anaesth. 2019;122:111–119.
5. Sinclair RC, Batterham AM, Davies S, Cawthorn L, Danjoux
GR. Validity of the 6 min walk test in prediction of the anaero-
bic threshold before major non-cardiac surgery. Br J Anaesth.
2012;108:30–35.
6. Brooks GC, Vittinghoff E, Iyer S, et al. Accuracy and usability of
a self-administered 6-minute walk test smartphone application.
Circ Heart Fail. 2015;8:905–913.
7. Annegarn J, Spruit MA, Savelberg HH, et al. Differences in
walking pattern during 6-min walk test between patients with
COPD and healthy subjects. PLoS One. 2012;7:e37329.
8. Stokes JW, Wanderer JP, McEvoy MD. Signicant discrepancies
exist between clinician assessment and patient self-assessment
of functional capacity by validated scoring tools during preop-
erative evaluation. Perioper Med (Lond). 2016;5:18.
9. ATS Committee on Prociency Standards for Clinical Pulmonary
Function Laboratories. ATS statement: guidelines for the six-
minute walk test. Am J Respird Critl Care Med. 2002;166:111–117.
10. Hlatky MA, Boineau RE, Higginbotham MB, et al. A brief self-
administered questionnaire to determine functional capacity
(the duke activity status index). Am J Cardiol. 1989;64:651–654.
11. Lee JA, Williams SM, Brown DD, Laurson KR. Concurrent vali-
dation of the actigraph gt3x+, polar active accelerometer, omron
HJ-720 and yamax digiwalker SW-701 pedometer step counts in
lab-based and free-living settings. J Sports Sci. 2015;33:991–1000.
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
XXX XXX Volume XXX Number XXX www.anesthesia-analgesia.org 9
12. Carkeet A. Exact parametric condence intervals for Bland-
Altman limits of agreement. Optom Vis Sci. 2015;92:e71–e80.
13. Lin LI. A concordance correlation coefcient to evaluate repro-
ducibility. Biometrics. 1989;45:255–268.
14. Lin L, Hedayat AS, Sinha B, Yang M. Statistical methods in
assessing agreement. J Am Stat Assoc. 2002;97:257–270.
15. Walter SD, Eliasziw M, Donner A. Sample size and optimal
designs for reliability studies. Stat Med. 1998;17:101–110.
16. Karas M, Bai J, Strączkiewicz M, et al. Accelerometry data in
health research: challenges and opportunities. review and
examples. bioRxiv. 2018;11:276154.
17. Case MA, Burwick HA, Volpp KG, Patel MS. Accuracy of
smartphone applications and wearable devices for tracking
physical activity data. JAMA. 2015;313:625–626.
18. Duncan MJ, Wunderlich K, Zhao Y, Faulkner G. Walk this
way: validity evidence of iPhone health application step
count in laboratory and free-living conditions. J Sports Sci.
2018;36:1695–1704.
19. Major MJ, Alford M. Validity of the iPhone M7 motion co-pro-
cessor as a pedometer for able-bodied ambulation. J Sports Sci.
2016;34:2160–2164.
20. Bergman RJ, Spellman JW, Hall ME, Bergman SM. Is there a
valid app for that? Validity of a free pedometer iPhone applica-
tion. J Phys Act Health. 2012;9:670–676.
21. Hurt CP, Lein DH Jr, Smith CR, et al. Assessing a novel way
to measure step count while walking using a custom mobile
phone application. PLoS One. 2018;13:e0206828.
22. Amagasa S, Kamada M, Sasai H, et al. How well iPhones mea-
sure steps in free-living conditions: cross-sectional validation
study. JMIR Mhealth Uhealth. 2019;7:e10418.
23. Urbanek JK, Zipunnikov V, Harris T, et al. Prediction of sus-
tained harmonic walking in the free-living environment using
raw accelerometry data. Physiol Meas. 2018;39:02NT02.
24. Urbanek JK, Zipunnikov V, Harris T, Crainiceanu C, Harezlak
J, Glynn NW. Validation of gait characteristics extracted from
raw accelerometry during walking against measures of physi-
cal function, mobility, fatigability, and tness. J Gerontol A Biol
Sci Med Sci. 2018;73:676–681.
25. McConnell MV, Shcherbina A, Pavlovic A, et al. Feasibility of
obtaining measures of lifestyle from a smartphone app: the
MyHeart Counts Cardiovascular Health Study. JAMA Cardiol.
2017;2:67–76.
26. Ofcial US Government Information about the Global
Positioning System (GPS) and Related Topics. GPS Accuracy.
Available at: https://www.gps.gov/systems/gps/perfor-
mance/accuracy/. Accessed November 2, 2018.
27. Meyns P, Bruijn SM, Duysens J. The how and why of arm swing
during human walking. Gait Posture. 2013;38:555–562.
... Figure 4 reports the different brands used for smartphones, sensors, and IMUs. Regarding smartphones, Samsung [41,45,51,68,69,77,86,103] and iPhone [40,42,69,76,89] are the most represented, certainly because of their health applications made for gait recording. Actigraph is the most commonly used brand for sensors [38,40,48,49,67,71,74,85,96,103]. ...
... Regarding the positioning of sensors and/or devices (Table 4), 60% of the studies placed them on an inferior part of the body [35][36][37]40,[47][48][49][52][53][54][55][56][57][58]60,62,63,67,70,71,[73][74][75][76][78][79][80][81]83,85,87,88,[90][91][92][95][96][97][98][99][100]102,103], generally on the feet (N = 14) or on the hips (N = 6). The chest was also widely used (49%) [34,[37][38][39]44,48,50,[54][55][56][59][60][61]64,65,67,70,72,73,75,77,79,83,84,89,90,[92][93][94][95]97,99,101,102]; 17% of the studies carried out sensor positioning on the hands and arms [38,40,46,48,52,63,66,67,77,80,82,90,102], while the other 17% used a trouser or jacket pocket [42,43,45,50,51,59,68,70,77,86,89,103] Table 4. Frequency of sensor locations reported on the patient from the included studies. These different locations were classified into the four categories described in Section 2.6. ...
... Regarding the positioning of sensors and/or devices (Table 4), 60% of the studies placed them on an inferior part of the body [35][36][37]40,[47][48][49][52][53][54][55][56][57][58]60,62,63,67,70,71,[73][74][75][76][78][79][80][81]83,85,87,88,[90][91][92][95][96][97][98][99][100]102,103], generally on the feet (N = 14) or on the hips (N = 6). The chest was also widely used (49%) [34,[37][38][39]44,48,50,[54][55][56][59][60][61]64,65,67,70,72,73,75,77,79,83,84,89,90,[92][93][94][95]97,99,101,102]; 17% of the studies carried out sensor positioning on the hands and arms [38,40,46,48,52,63,66,67,77,80,82,90,102], while the other 17% used a trouser or jacket pocket [42,43,45,50,51,59,68,70,77,86,89,103] Table 4. Frequency of sensor locations reported on the patient from the included studies. These different locations were classified into the four categories described in Section 2.6. ...
Article
Full-text available
Gait, balance, and coordination are important in the development of chronic disease, but the ability to accurately assess these in the daily lives of patients may be limited by traditional biased assessment tools. Wearable sensors offer the possibility of minimizing the main limitations of traditional assessment tools by generating quantitative data on a regular basis, which can greatly improve the home monitoring of patients. However, these commercial sensors must be validated in this context with rigorous validation methods. This scoping review summarizes the state-of-the-art between 2010 and 2020 in terms of the use of commercial wearable devices for gait monitoring in patients. For this specific period, 10 databases were searched and 564 records were retrieved from the associated search. This scoping review included 70 studies investigating one or more wearable sensors used to automatically track patient gait in the field. The majority of studies (95%) utilized accelerometers either by itself (N = 17 of 70) or embedded into a device (N = 57 of 70) and/or gyroscopes (51%) to automatically monitor gait via wearable sensors. All of the studies (N = 70) used one or more validation methods in which “ground truth” data were reported. Regarding the validation of wearable sensors, studies using machine learning have become more numerous since 2010, at 17% of included studies. This scoping review highlights the current state of the ability of commercial sensors to enhance traditional methods of gait assessment by passively monitoring gait in daily life, over long periods of time, and with minimal user interaction. Considering our review of the last 10 years in this field, machine learning approaches are algorithms to be considered for the future. These are in fact data-based approaches which, as long as the data collected are numerous, annotated, and representative, allow for the training of an effective model. In this context, commercial wearable sensors allowing for increased data collection and good patient adherence through efforts of miniaturization, energy consumption, and comfort will contribute to its future success.
... Therefore, we only report in Table II.1 ranges of acquisition times which go from hour to year. Among the selected studies, 33% (N = 25) focused on neurodegenerative diseases [8,9,18,24,54,72,86,88,97,100,109,126,146,169,172,192,206,209,213,252,269,280,283,313,346], 24% (N = 18) on orthopedic disorders [5,62,75,107,149,170,174,185,186,203,221,243,263,277,302,303,314,322], 24% (N = 18) on diseases of vascular origin [29,52,59,63,64,74,75,77,134,147,155,159,160,256,277,291,314,335], 8% (N = 6) on aging and associated pathologies [21,92,154,259,277,315], and 4% (N = 3) on diseases associated with poor lifestyle [48,134,177]. Finally, 5 studies were classified as "others" [33,60,219,227,279] because they could not be grouped together in an existing group. ...
Thesis
With the development of the Internet of Things (IoT), smartphones and sensors are now able to provide information about the user's activity and even their physiology. This has led to a growing interest from the scientific community, particularly in the field of e-health, with applications in the monitoring of patients undergoing rehabilitation in order to offer more personalised follow-up. However, in addition to guiding the rehabilitation process, the generation and transmission of IoT data is also vulnerable to privacy breaches. Indeed, the complex processing chain of the IoT application in healthcare multiplies the risk of privacy threats throughout the life cycle of IoT data, including collection, transmission and storage, by an adversary who can retrieve the data and re-identify or reveal sensitive patient information. This thesis focuses on the following questions: Is the data collected sufficiently protected so that no one can misuse it to re-identify the owner or infer sensitive information? Is the protected data still accurate enough for healthcare applications such as rehabilitation? Achieving balance between data utility and privacy protection is an important challenge that we explore in this thesis from different angles. More specifically, the first part focuses on the problem of data anonymisation through minimisation, while the second part focuses on preventing the inference of sensitive attributes through a Generative Adversarial Network to sanitise sensor data and an approach exploiting private layers in Federated Learning.
Article
Rigorous clinical evaluation of the physiological performance is currently performed with complex and long procedures which need expensive technology and skilled operators. In a wide range of situations (frail patients, daily clinical practice, etc.), these approaches are difficult to be applied and simpler tests, with a lack of scientific background, are mandatory. To avoid these problems, we propose a test (test of physiological performance (TOPP)) to evaluate the physiological behavior of a subject, in a really easy and safe clinical setting, measuring only the heart rate. The subject is submitted to an active standing-up test and then two submaximal exercises (with a low power load) on a cycle-ergometer. The heart rate modifications due to each submaximal step are analyzed by exponential interpolation to calculate the ascending and descending time constants and evaluate the way each subject adapts his heart rate to work. The standard deviation of the RR for each stationary phase (warm-up, load, recovery) was calculated as an index of short-term variability. Then a standard Fourier analysis of the stationary periods of the standing-up procedures allows to quickly and easily evaluate the autonomic nervous activation. We tested the protocol on five healthy subjects to verify the feasibility and the acceptance of the procedure. The five subjects demonstrated a good tolerance of the entire procedure. The standing-up showed a behavior of the autonomic system consistent with the physiology (with an increase in sympathetic activation in the passage to standing position). The analysis of the two submaximal steps highlights how younger and trained subjects present lower heart rates (both in the ascending phase and in the recovery) with a quicker adaptation ability (smaller time constants) consistent with what is expected. The short-term variability of heart rate is greater in young and trained subjects, thus confirming how the sympatho-vagal balance, in these subjects, is more dynamic. The proposed test is well tolerated by the subjects and the results, albeit in a small cohort of healthy volunteers, are consistent with what is expected from physiology and is already present in the literature. Our work aims to be a proposal with a feasibility check of a method for evaluating performance. The work to be done for the clinical validation of the TOPP is still long, but we are aware that it can give important results and that the TOPP can become an effective tool for the assessment of the physiological performance even of fragile subjects.
Article
Background: The increasing use of smartphones by providers and patients alike demonstrates that digital health utilizing mobile applications has the potential to transform perioperative care and education in anesthesia. Objective: This literature review describes the current scope of the use of mobile applications in anesthesiology. Methods: Literature was searched using PubMed, Scopus, and clinicaltrials.gov for articles published from January 1, 2010, through April 1, 2020. Only English language studies were included. Articles were included if they examined the use of a mobile health application in the setting of anesthesia or the perioperative (immediate preoperative, intraoperative, and postoperative) period. Studies were excluded if they explored video interventions or did not examine the feasibility or efficacy of the mobile app. Results: We included 29 articles, and three areas of clinical functionality were identified: patient-centered care (preoperative, intraoperative, and postoperative), systems-based improvement, and medical education. Several studies demonstrate the feasibility and reliability of mobile apps in these areas, but many are only tested for efficacy in simulated environments or with small patient samples. Conclusions: Mobile health applications show promise in improving communication between anesthesiologists, improving workflow efficiency, enhancing medical education, and reducing hospital costs. However, there is a need for validation and improvement before full implementation by the provider, patients, and hospital systems. Future studies are needed to demonstrate meaningful health outcomes to create guidelines and recommendations specific to the application of mobile technology to health care.
Article
Dry eye disease (DED) is a chronic, multifactorial ocular surface disorder with multiple etiologies that results in tear film instability. Globally, the prevalence of DED is expected to increase with an aging society and daily use of digital devices. Unfortunately, the medical field is currently unprepared to meet the medical needs of patients with DED. Noninvasive, reliable, and readily reproducible biomarkers have not yet been identified, and the current mainstay treatment for DED relies on symptom alleviation using eye drops with no effective preventative therapies available. Medical big data analyses, mining information from multiomics studies and mobile health applications, may offer a solution for managing chronic conditions such as DED. Omics-based data on individual physiologic status may be leveraged to prevent high-risk diseases, accurately diagnose illness, and improve patient prognosis. Mobile health applications enable the portable collection of real-world medical data and biosignals through personal devices. Together, these data lay a robust foundation for personalized treatments for various ocular surface diseases and other pathologies that currently lack the components of precision medicine. To fully implement personalized and precision medicine, traditional aggregate medical data should not be applied directly to individuals without adjustments for personal etiology, phenotype, presentation, and symptoms.
Article
Full-text available
Allergic conjunctivitis affects approximately 15–20% of the global population and can permanently deteriorate one's quality of life (QoL) and work productivity, leading to societal work force costs. Although not fully understood, allergic conjunctivitis is a multifactorial disease with a complex network of environmental, lifestyle, and host contributory risk factors. To effectively enhance the quality of treatment for patients with allergic conjunctivitis, as well as other allergic diseases, the field must first comprehend the pathology underlying various individualized subjective symptoms and stratify the disease according to risk factors and presentations. Such competent stratification and societal reconstruction that targets the alleviation of the damage due to allergic diseases would greatly help ramify personalized treatments and prevent the projected increase in societal costs imposed by allergic diseases. Owing to the rapid advancements in the information and technology sector, medical big data are greatly accessible and useful to decipher the pathophysiology of many diseases. Such data collected through multi-omics and mobile health have been effective for research on chronic diseases including allergic and immune-mediated diseases. Novel big data containing vast and continuous information on individuals with allergic conjunctivitis and other allergic symptoms are being used to search for causative genes of diseases, gain insights into new biomarkers, prevent disease progression, and, ultimately, improve QoL. The individualized and holistic data accrued from new angles using technological innovations are helping the field realize the principles of P4 medicine: predictive, preventive, personalized, and participatory medicine.
Article
Full-text available
Introduction: Walking speed has been associated with many clinical outcomes (e.g., frailty, mortality, joint replacement need, etc.). Accurately measuring walking speed (stride length x step count/time) typically requires significant clinician/staff time or a gait lab with specialized equipment (i.e., electronic timers or motion capture). In the present study, our goal was to measure "step count" via smartphones through novel software and to compare with step tracking software that come standard with iOS and Android smartphones as a first step in walking speed measurement. Methods: A separate calibration and validation data collection was performed. Individuals in the calibration collection (n = 5) walked 20m at normal and slow speed (<1.0 m/s). Appropriate settings for the novel mobile application were chosen to measure step count. Individuals in the validation (n = 52) collection walked at 6m, 10m, and 20m at normal and slow walking speeds. We compared step difference (absolute difference) from observed step counts to native step tracking software and our novel software derived step counts. We used generalized estimated equation adjusted (participant level) negative binomial regression models of absolute step difference from observed step counts, to determine optimal settings (calibration) and subsequently to gauge performance of the shake algorithm settings and native step tracking software across different distances and speeds (validation). Results: For iOS/iPhone 6, when compared to observed step count, the shake service (software driven approach) significantly outperformed the embedded native step tracking software across all distances at slow speed, and for short distance (6m) at normal speed. On the Android phone, the shake service outperformed the native step tracking software at slow speed at 6 meters and 20 meters, while its performance eclipsed the native step tracking software only at 20 meters at normal speed. Discussion: Our software based approach outperformed native step tracking software across various speeds and distances and carries the advantage of having adjustable measurement parameters that can be further optimized for specific medical conditions. Such software applications will provide an effective way to capture standardized data across multiple commercial smartphone devices, facilitating the future capture of walking speed and other clinically important performance parameters that will influence clinical and home care in the era of value based care.
Article
Full-text available
Background: Smartphones have great potential for monitoring physical activity. Although a previous laboratory-based study reported that smartphone apps were accurate for tracking step counts, little evidence on their accuracy in free-living conditions currently exists. Objective: We aimed to investigate the accuracy of step counts measured using iPhone in the real world. Methods: We recruited a convenience sample of 54 adults (mean age 31 [SD 10] years) who owned an iPhone and analyzed data collected in 2016 and 2017. Step count was simultaneously measured using a validated pedometer (Kenz Lifecorder) and the iPhone. Participants were asked to carry and use their own iPhones as they typically would while wearing a pedometer on the waist for 7 consecutive days during waking hours. To assess the agreement between the two measurements, we calculated Spearman correlation coefficients and prepared a Bland-Altman plot. Results: The mean step count measured using the iPhone was 9253 (3787) steps per day, significantly lower by 12% (1277/10,530) than that measured using the pedometer, 10,530 (3490) steps per day (P<.001). The Spearman correlation coefficient between devices was 0.78 (P<.001). The largest underestimation of steps by the iPhone was observed among those who reported to have seldom carried their iPhones (seldom carry: mean −3036, SD 2990, steps/day; sometimes carry: mean −1424, SD 2619, steps/day; and almost always carry: mean −929, SD 1443, steps/day; P for linear trend=.08). Conclusions: Smartphones may be of practical use to individuals, clinicians, and researchers for monitoring physical activity. However, their data on step counts should be interpreted cautiously because of the possibility of underestimation due to noncarrying time.
Article
Wearable accelerometers provide detailed, objective, and continuous measurements of physical activity (PA). Recent advances in technology and the decreasing cost of wearable devices led to an explosion in the popularity of wearable technology in health research. An ever-increasing number of studies collect high-throughput, sub-second level raw acceleration data. In this paper, we discuss problems related to the collection and analysis of raw accelerometry data and refer to published solutions. In particular, we describe the size and complexity of the data, the within- and between-subject variability, and the effects of sensor location on the body. We also discuss challenges related to sampling frequency, device calibration, data labeling, and multiple PA monitors synchronization. We illustrate these points using the Developmental Epidemiological Cohort Study (DECOS), which collected raw accelerometry data on individuals both in a controlled and the free-living environment.
Article
Background: The 6-min walk test (6MWT) is a common means of functional assessment. Its relationship to disability-free survival (DFS) is uncertain. Methods: This sub-study of the Measurement of Exercise Tolerance for Surgery study had co-primary outcome measures: correlation of the preoperative 6MWT distance with 30 day quality of recovery (15-item quality of recovery) and 12 month WHO Disability Assessment Schedule scores. The prognostic utility of the 6MWT and other risk assessment tools for 12 month DFS was assessed with logistic regression and receiver-operating-characteristic-curve analysis. Results: Of 574 patients recruited, 567 (99%) completed the 6MWT. Twelve months after surgery, 16 (2.9%) patients had died and 444 (77%) had DFS. The 6MWT correlated weakly with 30 day 15-item quality of recovery (ρ=0.14; P=0.001) and 12 month WHO Disability Assessment Schedule (ρ=-0.23; P<0.0005) scores. When the cohort was split into 6MWT distance tertiles, the adjusted odds ratio of low vs high tertiles for DFS was 3.13 [95% confidence interval (CI): 1.54-6.35]. The only independent variable predictive of DFS was the Duke Activity Status Index (DASI) score (adjusted odds ratio: 1.06; P<0.0005). The area under the receiver-operating-characteristic curve for DFS was 0.63 (95% CI: 0.57-0.70) for the 6MWT, 0.60 (95% CI: 0.53-0.67) for cardiopulmonary-exercise-testing-derived peak oxygen consumption, and 0.70 (95% CI: 0.64-0.76) for the DASI score. Conclusions: Of the risk assessment tools analysed, the DASI was the most predictive of DFS. The 6MWT was safe and comparable with cardiopulmonary exercise testing for all predictive assessments. Future research should aim to determine the optimal 6MWT distance thresholds for risk prediction.
Article
Background: Functional capacity is an important component of risk assessment for major surgery. Doctors' clinical subjective assessment of patients' functional capacity has uncertain accuracy. We did a study to compare preoperative subjective assessment with alternative markers of fitness (cardiopulmonary exercise testing [CPET], scores on the Duke Activity Status Index [DASI] questionnaire, and serum N-terminal pro-B-type natriuretic peptide [NT pro-BNP] concentrations) for predicting death or complications after major elective non-cardiac surgery. Methods: We did a multicentre, international, prospective cohort study at 25 hospitals: five in Canada, seven in the UK, ten in Australia, and three in New Zealand. We recruited adults aged at least 40 years who were scheduled for major non-cardiac surgery and deemed to have one or more risk factors for cardiac complications (eg, a history of heart failure, stroke, or diabetes) or coronary artery disease. Functional capacity was subjectively assessed in units of metabolic equivalents of tasks by the responsible anaesthesiologists in the preoperative assessment clinic, graded as poor (<4), moderate (4-10), or good (>10). All participants also completed the DASI questionnaire, underwent CPET to measure peak oxygen consumption, and had blood tests for measurement of NT pro-BNP concentrations. After surgery, patients had daily electrocardiograms and blood tests to measure troponin and creatinine concentrations until the third postoperative day or hospital discharge. The primary outcome was death or myocardial infarction within 30 days after surgery, assessed in all participants who underwent both CPET and surgery. Prognostic accuracy was assessed using logistic regression, receiver-operating-characteristic curves, and net risk reclassification. Findings: Between March 1, 2013, and March 25, 2016, we included 1401 patients in the study. 28 (2%) of 1401 patients died or had a myocardial infarction within 30 days of surgery. Subjective assessment had 19·2% sensitivity (95% CI 14·2-25) and 94·7% specificity (93·2-95·9) for identifying the inability to attain four metabolic equivalents during CPET. Only DASI scores were associated with predicting the primary outcome (adjusted odds ratio 0·96, 95% CI 0·83-0·99; p=0·03). Interpretation: Subjectively assessed functional capacity should not be used for preoperative risk evaluation. Clinicians could instead consider a measure such as DASI for cardiac risk assessment. Funding: Canadian Institutes of Health Research, Heart and Stroke Foundation of Canada, Ontario Ministry of Health and Long-Term Care, Ontario Ministry of Research, Innovation and Science, UK National Institute of Academic Anaesthesia, UK Clinical Research Collaboration, Australian and New Zealand College of Anaesthetists, and Monash University.
Article
Several attempts have been made to demonstrate the accuracy of the iPhone pedometer function in laboratory test conditions. However, no studies have attempted to evaluate evidence of convergent validity of the iPhone step counts as a surveillance tool in the field. This study takes a pragmatic approach to evaluating Health application derived iPhone step counts by measuring accuracy of a standardized criterion iPhone SE and a heterogeneous sample of participant owned iPhones (6 or newer) in a laboratory condition, as well as comparing personal iPhones to accelerometer derived steps in a free-living test. During lab tests, criterion and personal iPhones differed from manually counted steps by a mean bias of less than ±5% when walking at 5km/h, 7.5km/h and 10km/h on a treadmill, which is generally considered acceptable for pedometers. In the free-living condition steps differed by a mean bias of 21.5% or 1340 steps/day when averaged across observation days. Researchers should be cautioned in considering the use of iPhone models as a research grade pedometer for physical activity surveillance or evaluation, likely due to the iPhone not being continually carried by participants; if compliance can be maximized then the iPhone might be suitable.
Article
A method is developed to calculate the required number of subjects k in a reliability study, where reliability is measured using the intraclass correlation ρ. The method is based on a functional approximation to earlier exact results. The approximation is shown to have excellent agreement with the exact results and one can use it easily without intensive numerical computation. Optimal design configurations are also discussed; for reliability values of about 40 per cent or higher, use of two or three observations per subject will minimize the total number of observations required. © 1998 John Wiley & Sons, Ltd.
Article
Background Data collected by wearable accelerometry devices can be used to identify periods of sustained harmonic walking. This report aims to establish whether the features of walking identified in the laboratory and free-living environments are associated with each other as well as measures of physical function, mobility, fatigability, and fitness. Methods Fifty-one older adults (mean age 78.31) enrolled in the Developmental Epidemiologic Cohort Study were included in the analyses. The study included an “in-the-lab” component as well as 7 days of monitoring “in-the-wild” (free-living). Participants were equipped with hip-worn Actigraph GT3X+ activity monitors, which collect raw accelerometry data. We applied a walking identification algorithm and defined features of walking, including participant-specific walking acceleration and cadence. The association between these walking features and physical function, mobility, fatigability, and fitness was quantified using linear regression analysis. Results Acceleration and cadence estimated from “in-the-lab” and “in-the-wild” data were significantly associated with each other (p < 0.05). However, walking acceleration “in-the-lab” were on average 96% higher than “in-the-wild”, whereas cadence “in-the-lab” was on average 20% higher than “in-the-wild”. Acceleration and cadence were associated with measures of physical function, mobility, fatigability, and fitness (p < 0.05) in both “in-the-lab” and “in-the-wild” settings. Additionally, “in-the-wild” daily walking time was associated with fitness (p < 0.05). Conclusions The quantitative difference in proposed walking features indicates that participants may over-perform when observed “in-the-lab”. Also, proposed features of walking were significantly associated with measures of physical function, mobility, fatigability, and fitness, which provides evidence of convergent validity.
Article
Importance: Studies have established the importance of physical activity and fitness, yet limited data exist on the associations between objective, real-world physical activity patterns, fitness, sleep, and cardiovascular health. Objectives: To assess the feasibility of obtaining measures of physical activity, fitness, and sleep from smartphones and to gain insights into activity patterns associated with life satisfaction and self-reported disease. Design, setting, and participants: The MyHeart Counts smartphone app was made available in March 2015, and prospective participants downloaded the free app between March and October 2015. In this smartphone-based study of cardiovascular health, participants recorded physical activity, filled out health questionnaires, and completed a 6-minute walk test. The app was available to download within the United States. Main outcomes and measures: The feasibility of consent and data collection entirely on a smartphone, the use of machine learning to cluster participants, and the associations between activity patterns, life satisfaction, and self-reported disease. Results: From the launch to the time of the data freeze for this study (March to October 2015), the number of individuals (self-selected) who consented to participate was 48 968, representing all 50 states and the District of Columbia. Their median age was 36 years (interquartile range, 27-50 years), and 82.2% (30 338 male, 6556 female, 10 other, and 3115 unknown) were male. In total, 40 017 (81.7% of those who consented) uploaded data. Among those who consented, 20 345 individuals (41.5%) completed 4 of the 7 days of motion data collection, and 4552 individuals (9.3%) completed all 7 days. Among those who consented, 40 017 (81.7%) filled out some portion of the questionnaires, and 4990 (10.2%) completed the 6-minute walk test, made available only at the end of 7 days. The Heart Age Questionnaire, also available after 7 days, required entering lipid values and age 40 to 79 years (among 17 245 individuals, 43.1% of participants). Consequently, 1334 (2.7%) of those who consented completed all fields needed to compute heart age and a 10-year risk score. Physical activity was detected for a mean (SD) of 14.5% (8.0%) of individuals' total recorded time. Physical activity patterns were identified by cluster analysis. A pattern of lower overall activity but more frequent transitions between active and inactive states was associated with equivalent self-reported cardiovascular disease as a pattern of higher overall activity with fewer transitions. Individuals' perception of their activity and risk bore little relation to sensor-estimated activity or calculated cardiovascular risk. Conclusions and relevance: A smartphone-based study of cardiovascular health is feasible, and improvements in participant diversity and engagement will maximize yield from consented participants. Large-scale, real-world assessment of physical activity, fitness, and sleep using mobile devices may be a useful addition to future population health studies.
Article
Background This study examined the validity of a selected free pedometer application (iPedometer; IP) for the iPhone that could be used to assess physical activity. Methods Twenty college students (10 men, 10 women; mean age: 21.85 ± 1.57 yrs) wore an iPhone at 3 locations (pocket, waist, arm) and a StepWatch 3 Step Activity Monitor (SW) on their right ankle while walking on a treadmill at 5 different speeds (54, 67, 80, 94, 107 m·min ⁻¹ ). A research assistant counted steps with a tally counter (TC). Results Statistical significance between the TC, SW, and IP was found during every condition except IP in the pocket at 107 m·min ⁻¹ ( F 2,38 = .64, P = .54). Correlations involving the IP revealed only 1 positive correlation (IP on arm at 54 m·min ⁻¹ ) for any of the conditions ( r = .46, P = .05). Conclusion The IP application was not accurate in counting steps and recorded significantly lower step counts than the SW and TC. Thus, the free pedometer application used is not a valid instrument for monitoring activity during treadmill walking.