1979
ORIGINAL ARTICLE
Activity Monitoring for Assessment of Physical Activities in
Daily Life in Patients With Chronic Obstructive Pulmonary
Disease
Fabio Pitta, PT, MSc, Thierry Troosters, PT, PhD, Martijn A. Spruit, PT, PhD, Marc Decramer, MD, PhD,
Rik Gosselink, PT, PhD
ABSTRACT. Pitta F, Troosters T, Spruit MA, Decramer M,
Gosselink R. Activity monitoring for assessment of physical
activities in daily life in patients with chronic obstructive
pulmonary disease. Arch Phys Med Rehabil 2005;86:1979-85.
Objective: To investigate the degree of agreement between
different methods of assessing physical activities in daily life in
patients with chronic obstructive pulmonary disease (COPD):
video recordings (criterion standard), the DynaPort Activity
Monitor (DAM), and patient self-report.
Design: Study A: outcomes from video recordings were
compared with DAM outcomes and with patient estimation of
time spent on each activity after a 1-hour protocol including
walking, cycling, standing, sitting, and lying. Study B: DAM
outcomes and patient self-report were compared during 1 day
in real life.
Setting: Outpatient clinic in a university hospital.
Participants: Study A: 10 patients with COPD (mean age,
62⫾6y; forced expiratory volume in the first second
[FEV1]⫽40%⫾16% of predicted). Study B: 13 patients with
COPD (mean age, 61⫾8y; FEV1⫽33%⫾10% of predicted).
Interventions: Not applicable.
Main Outcome Measures: Time spent on different activities and movement intensity during walking and cycling.
Results: Study A: time estimated by the patients in the
sitting position was significantly lower than the time showed by
the video recordings and the DAM (both P⬍.001). For the
other variables, there were no statistically significant differences (all P⬎.05). However, Bland and Altman plots and
intraclass correlation coefficients showed large disagreement
between video recordings and patients’ estimations, in contrast
to the high degree of agreement between video recordings and
DAM. Changes in walking speed correlated highly to changes
in DAM movement intensity (r⫽.81, P⬍.01). Study B: patients
significantly overestimated walking time (22⫾47min, P⫽.04)
and underestimated standing time (– 45⫾71min, P⫽.04).
Conclusions: The DAM showed high accuracy in objectively assessing time spent on different activities and changes
From the Respiratory Rehabilitation and Respiratory Division, University Hospitals, Katholieke Universiteit Leuven, Leuven, Belgium (Pitta, Troosters, Spruit,
Decramer, Gosselink); Faculty of Kinesiology and Rehabilitation Sciences, Katholieke Universiteit Leuven, Leuven, Belgium (Pitta, Troosters, Spruit, Decramer,
Gosselink); and Departamento de Fisioterapia, Universidade Estadual de Londrina,
Londrina, Brazil (Pitta)
Supported in part by CAPES/Brasil, the Fonds voor Wetenschappelijk Onderzoek
(Levenslijn grant nos. 7.0007.00, G.0237.01), Katholieke Universiteit Leuven (grant
no. PDM/04/230), and the Foundation Van Goethem-Brichant.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the authors or upon any
organization with which the authors are associated.
Reprint requests to Rik Gosselink, PT, PhD, Respiratory Rehabilitation and Respiratory Division, University Hospital Gasthuisberg, Herestraat 49, B-3000 Leuven,
Belgium, e-mail: [email protected].
0003-9993/05/8610-9883$30.00/0
doi:10.1016/j.apmr.2005.04.016
in walking speed in patients with COPD. Patients’ estimations
of time spent on physical activities in daily life disagreed with
objective assessment.
Key Words: Activities of daily living; Physical effort; Pulmonary disease, chronic obstructive; Rehabilitation; Validity
of results.
© 2005 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and
Rehabilitation
ECENT STUDIES HAVE SHOWN that physical activities in daily life are an important predictor of risk of
R
hospital readmission and mortality in patients with chronic
obstructive pulmonary disease (COPD).1-3 In addition, the
World Health Organization’s Global Initiative for Chronic
Obstructive Lung Disease (GOLD) report4 states that increasing physical participation in everyday activities is among the
important goals of treatment in patients with COPD. Guidelines recommend that a minimum of 30 minutes of daily
physical activity of moderate intensity, such as walking, is
necessary to maintain fitness, and those not meeting this standard are considered insufficiently active.5 Therefore, time spent
in physical activity during daily life is an important issue in the
characterization of physical activity patterns.
Physical activities in daily life in patients with COPD are
mostly investigated with subjective methods like questionnaires and diaries. Subjective methods are useful to provide a
patient’s personal perception of functional status, effort, and
difficulties in performing activities,6,7 although physical activity questionnaires have shown limited validity and reliability.8
It is uncertain whether patients with COPD can report accurately in logbooks or diaries their time spent in activity. Studies
performed in different populations suggest that self-reported
data should be treated with caution because subjects may
misreport activity time when compared with automated measures.9,10 This misinterpretation in itself may be interesting,
because it gives information on how active people perceive
they are. However, self-report methods may fail to provide
clinicians with accurate data on daily activity patterns. Furthermore, because of impaired cognitive capacity linked to hypoxemia,11 activity recollection in patients with COPD may be
even less accurate than in other populations. Hence, different
devices have been used, aiming to objectively “count” activities, both in healthy subjects under controlled conditions12 and
in COPD patients’ home environments.13-15 These devices,
however, do not allow classification of different activities or
body positions and do not provide descriptions of time spent in
these activities and positions during real life. In addition, they
do not report on the intensity with which subjects carry out
activities such as walking, which is also an important feature in
the recommendations of physical activity to maintain health.16
Therefore, amounts of time spent daily in walking or in other
activities, as well as movement intensity, are not available for
Arch Phys Med Rehabil Vol 86, October 2005
1980
ACTIVITY MONITORING IN PATIENTS WITH COPD, Pitta
patients with COPD because of the lack of tools that could
objectively measure these outcomes.
Recently, an accelerometer-based activity monitor (DynaPort Activity Monitor [DAM]a) (fig 1) was developed to provide real-life assessment of activities in a patient’s environment. It distinguishes movement patterns (walking, cycling)
and body positions (standing, sitting, lying) during long-term
measurements, and it measures the intensity (acceleration) at
which movements such as walking are performed. However,
the populations in which the DAM was investigated (eg, children,17 people in working situations18) are assumed to have
different activity patterns than COPD patients. Our objective
was to investigate the accuracy of this activity monitor in
objectively measuring time spent walking, cycling, standing,
sitting, and lying in patients with COPD. In addition, we also
investigated the accuracy of patients’ self-reports of time spent
on these activities in real life.
METHODS
Participants and Design
All patients who volunteered to take part in the study were
recruited from the outpatient rehabilitation program at the
University Hospital Gasthuisberg, Leuven, Belgium. All of
them gave informed consent to participate in the study. Patients
all were retired or disabled and no longer employed. The
inclusion criteria were (1) clinically stable condition, with no
infection or exacerbation in the last 3 months before inclusion;
(2) no recent cardiac complaints; and (3) absence of other
Fig 1. The DAM device is worn under the clothes.
Arch Phys Med Rehabil Vol 86, October 2005
Table 1: Patient Characteristics
Characteristics
Study A
Study B
Age (y)
Sex (male/female)
BMI (kg/m2)
FEV1 (%pred)
FVC (%pred)
Quadriceps force (%pred)
PImax (%pred)
PEmax (%pred)
6MWD (%pred)
62⫾6
6/4
21⫾4
40⫾16
84⫾13
89⫾23
81⫾16
101⫾24
74⫾9
61⫾8
10/3
23⫾5
33⫾10
83⫾15
70⫾21
75⫾24
89⫾23
58⫾13
NOTE. Values are mean ⫾ standard deviation (SD) or n. The percentage of predicted for all variables was calculated based on the
reference values provided in the methods section (see Materials,
Other Measures).
Abbreviations: BMI, body mass index; FVC, forced vital capacity;
6MWD, 6-minute walking distance; PEmax, maximal expiratory pressure; PImax, maximal inspiratory pressure; %pred, percentage of the
predicted value.
pathologic conditions that could impair physical activity performance (eg, cerebrovascular diseases, rheumatism, arthritis).
Study A. Two different studies were performed. In study
A, a group of 10 patients with COPD in GOLD stages II
through IV4 (6 men; mean age, 62⫾6y; forced expiratory
volume in the first second [FEV1]⫽40%⫾16% of predicted
value) (table 1) underwent a 1-hour standardized protocol in
which they performed different activities (walking, cycling)
and stayed in different positions (standing, sitting, lying) during a given time, which was not disclosed to the patients.
Patients wore the DAM during the protocol, and video recordings were made simultaneously. Time spent on the various
activities and postures obtained through careful digital chronometer analysis of the video recording was used as the criterion standard. In addition, walking and cycling were performed
twice during the protocol at 2 different self-selected speeds: 1
time at a speed considered comfortable by each patient
(“slow”) and 1 time at a speed that each patient considered fast
(“fast”). Walking was performed in a corridor previously measured in meters and the mean speed performed by each patient
was determined by analysis of the video (distance covered
divided by time spent to cover the distance). For cycling,
patients were asked to keep the chosen number of revolutions
per minute (rpm) constant during each 1 of the 2 self-selected
speeds (slow, fast). In both conditions (slow, fast), walking and
cycling speed were compared with the movement intensity
given by the DAM. After the protocol, patients reported how
many minutes they thought were spent in each activity (walking, cycling) and position (standing, sitting, lying) during the
1-hour protocol (see Materials, Patients’ Reports). This estimation was also compared with the digital chronometer results
obtained via video recordings.
Study B. In study B, physical activities in daily life were
measured with the DAM on a representative day (from waking
time until 12h after that) in a different group of 13 consecutive
COPD patients who were also in GOLD stages II through IV
(10 men; mean age, 61⫾8y; FEV1⫽33%⫾10 % of predicted
value) (see table 1). Patients were strongly encouraged to keep
their normal pattern of daily activities. During the measurement period, patients also filled out the logbook, reporting
hourly the time spent in each activity or body position (see
Materials, Patients’ Reports). Time spent in the different activities and body positions measured with the DAM was compared afterward with each patient’s report. Patients also filled
ACTIVITY MONITORING IN PATIENTS WITH COPD, Pitta
Table 2: Study A: Data Obtained from the Analysis of Video
Recordings, DAM, and Patients’ Estimation After the 1-Hour
Standardized Protocol
Variables
Walking time
Mean ⫾ SD (min)
95% CI (min)
SE (min)
Cycling time
Mean ⫾ SD (min)
95% CI (min)
SE (min)
Standing time
Mean ⫾ SD (min)
95% CI (min)
SE (min)
Sitting time
Mean ⫾ SD (min)
95% CI (min)
SE (min)
Lying time
Mean ⫾ SD (min)
95% CI (min)
SE (min)
Video
Recordings
DAM
Patients’
Estimation
7⫾2
(5–8)
0.7
7⫾2
(5–8)
0.7
7⫾3
(4–9)
1
6⫾1
(5–7)
0.4
6⫾1
(5–7)
0.4
7⫾2†
(5–9)
0.8
9⫾1
(8–10)
0.4
9⫾1
(8–10)
0.4
11⫾4†
(8–14)
1.4
22⫾3
(19–25)
1.2
23⫾5
(19–27)
1.7
16⫾4*
(12–20)
1.4
12⫾3
(10–15)
0.9
11⫾4
(8–15)
1.6
12⫾4
(9–16)
1.4
Abbreviation: SE, standard error.
*P⬍.001 vs video recordings and vs DAM.
†
P⫽0.2 vs video recordings and vs DAM.
out a checklist verifying whether the day was representative
and describing any possible hindrance of the activity monitor.
Materials
Assessment of physical activities with the activity monitor. The accelerometer-based activity monitor used was the
DAM. It consists of a small, light-weight box enclosed in a belt
worn on the frontal part of the waist and a leg sensor (total
weight, 375g) (fig 1). The device records signals of 3 unidimensional piezoresistive acceleration sensors: 2 placed in the
belt around the waist and 1 placed in the upper part of the left
leg, measuring gravitational and body segment accelerations.
The sensors have a range of ⫾8g (78.5m/s2) and a resolution of
.02g. The signals captured by the sensors are recorded and
stored in 10-, 16-, or 32-Mb memory cards at a sampling rate
of 32Hz. The data collected were downloaded to a computer,
and the positions and accelerations of the 3 sensors were
analyzed with the DynaScope software.a The software translates the recorded acceleration signals into basic activities
(walking, cycling) and postures (standing, sitting, lying) and
quantifies the time spent in each of these activities or postures.
Movement intensity during walking or cycling (in m/s2) is
determined based on accelerations recorded during these activities. Therefore, unlike other devices,13-15 the DAM does not
assess vector magnitude units (VMU) or number of counts but
provides a detailed analysis of time and intensity of activities
performed. In pilot experiments we investigated the sensitivity
of the DAM to 10-minute periods of car trip, wheelchair riding,
and taking an elevator. These are activities that might affect the
outcomes of accelerometry when VMU or counts are used.19 It
was shown that 100% of the time during car trips and wheelchair riding were correctly interpreted as sitting time; taking an
elevator was correctly interpreted as standing time 100% of the
time. Hence, outcomes of the DAM seem not to be influenced
by externally imposed acceleration. This is explained by the
1981
fact that these outcomes (ie, time spent in a given position or
activity) are based in the position of the sensors, and not on
acceleration units.
Video recordings. As performed in previous validation
studies of activity monitoring,17,18 video recordings were used
as the criterion standard in study A. Patients were videotaped
during the performance of the protocol. After the measurements, the tapes were analyzed by the same examiner (FP),
who was blinded to the values obtained by the DAM and
patients’ reports. The outcomes were the same as the DAM and
patients’ reports—that is, time spent in each activity or position. Walking speed (in m/s) and cycling speed (in rpm) were
compared to the movement intensity measured by the DAM
(see Participants and Design, Study A).
Patients’ reports (logbook). The report of activities and
body positions filled out by the patients was based on the diary
developed by Follick et al.20 The category “pain rating,” included in the original diary, was left out, because in this study
we were interested mainly in the time spent by the patients on
5 physical activities in daily life: walking, cycling, standing,
sitting, and lying.
In study A, patients reported the number of minutes they
thought were spent in each different activity (walking, cycling)
and body position (standing, sitting, lying) during the protocol.
In study B, patients reported hourly the number of minutes they
thought were spent in each different activity and body position
in the last hour. Use of the logbook was carefully explained to
the patients during instruction sessions, and study staff was
available at all times during the measurement days to respond
to possible queries by telephone. Patients were asked to be as
accurate as possible in their reports, and anonymity was
warranted.
Other measures. Pulmonary function,21 maximal inspiratory and expiratory pressures,22 quadriceps force,23 and
6-minute walking distance24 were assessed to characterize the
population. All methods and equipment have been previously
described.25
Statistical Analysis
Statistical analysis was performed using the SAS, version 8,
statistical packageb and GraphPad Prism 3.c Data are expressed
as mean ⫾ standard deviation (SD).
In study A, comparison of outcomes from video recordings, DAM, and patients’ estimations was performed with
the repeated-measures analysis of variance (ANOVA) procedure. Standard errors and 95% confidence intervals (CIs)
are also shown. In study B, comparison of time obtained
from the DAM and patients’ estimations was performed with
the paired t test. Correlations were analyzed with the Pearson correlation coefficient. Level of significance was set at
P less than .05.
To compare the assessment techniques, the Bland and Altman method26 was used in both protocols A and B. In this
graphic method, the closer the points and the mean are to the 0
line in the y axis, the higher the agreement between the techniques. The Bland and Altman plot is a recommended method
for comparing techniques because it provides a detailed view
on the degree of agreement between outcomes of these techniques.27 In addition, the intraclass correlation coefficient
(ICC) and its CI between outcomes from DAM and video
recordings and patients’ estimations and video recordings were
also calculated.
RESULTS
Patients reported that the DAM did not hinder them notably
in their activities. In study A, 2 patients reported being unable
Arch Phys Med Rehabil Vol 86, October 2005
1982
ACTIVITY MONITORING IN PATIENTS WITH COPD, Pitta
Fig 2. Bland and Altman plots for comparison of (A) walking time, (B) standing time, and (C) sitting time assessed by video recordings
(criterion standard), DAM, and patients’ reports. Left panels: comparison between video recordings and DAM. Right panels: comparison
between video recordings and patients’ estimations. Abbreviations: LL, lower limit (–1.96 SDs); UL, upper limit (ⴙ1.96 SDs).
Arch Phys Med Rehabil Vol 86, October 2005
ACTIVITY MONITORING IN PATIENTS WITH COPD, Pitta
Table 3: Protocol B: Data Obtained from DAM and Patients’
Estimation (logbook) After 12 Hours of Physical Activities’
Assessment in Daily Life
Variables
DAM
Patients’
Estimation
Walking time (min)
Cycling time (min)
Standing time (min)
Sitting time (min)
Lying time (min)
45⫾20
4⫾9
191⫾85
390⫾161
88⫾141
66⫾47*
3⫾9
146⫾71*
375⫾134
83⫾83
NOTE. Values are mean ⫾ SD.
*P⬍.05.
to differentiate the time spent among the different activities
performed, choosing not to fill out the patient’s estimation
report. These patients were excluded from the ANOVA procedure, which was performed with 8 patients. In study B, several
patients reported having difficulties filling out the logbook at
home, mostly because it was time consuming. Baseline characteristics of both groups are found in table 1.
Study A
ANOVA results performed comparing outcomes from video
recordings, DAM, and patients’ estimations are shown in table
2. The time that patients estimated spending in the sitting
position was significantly lower than the time shown by the
video recordings and the DAM (both P⬍.001). For the other
variables, there were no statistically significant differences (all
P⬎.05). However, in the Bland and Altman plots, a large
disagreement between video recordings and patients’ estimations (fig 2, right panels) was shown, in contrast to the very
small disagreement between the video recordings and DAM
(see fig 2, left panels). ICCs (CIs) between outcomes from
DAM and video recordings were as follows: .999 (.998 –1) for
walking time, .990 (.979 –1) for cycling time, .998 (.996 –1) for
standing time, .77 (.510 –1) for sitting time, and .75 (.480–1)
for lying time. ICCs (CIs) between outcomes from patients’
estimations and video recordings were as follows: –.21 (–.86 to
.44) for walking time, –.07 (–.72 to .58) for standing time, –.06
(–.73 to .61) for sitting time, and .55 (.03–1) for lying time.
ICC for cycling time between patients’ estimations and video
recordings could not be calculated because the procedure did
not converge.
The DAM had a very small but systematic overestimation of
walking time compared with the video recordings (3.7⫾2.4s in
420s [7min] of walking, P⬍.05) and cycling time (7.9⫾3.4s in
360s [6min] of cycling, P⬍.05).
Movement intensity correlated significantly with walking
speed during fast (1.3⫾0.1m/s) walking (r⫽.72, P⬍.05) but
not during slow (0.9⫾0.1m/s) walking (r⫽.21, not significant
[NS]). Increase in walking speed highly correlated with increase in movement intensity measured by the DAM (r⫽.81,
P⬍.01). On the other hand, DAM movement intensity did not
correlate either with slow (37⫾4rpm) cycling (r⫽–.25, NS) or
with fast (63⫾8 rpm) cycling (r⫽.02, NS). Increase in cycling
speed did not correlate with increase in DAM movement intensity (r⫽–.04, NS).
Study B
Mean values of the DAM outcomes and logbook outcomes
during 12 hours of measurement in each patient’s environment
are summarized in table 3. Statistically significant differences
were found for walking time, which was overestimated by the
patients, and standing time, which was underestimated (both
1983
P⫽.04). The Bland and Altman plot for walking time assessed
by the 2 methods showed a significant relation between the
degree of overestimation and walking time (r⫽–.71, P⬍.01),
indicating that overestimation of walking time was especially
present in the more active patients (fig 3).
DISCUSSION
The results of our study show 2 main findings. First, the
DAM is an accurate method for measuring the time spent in
walking, cycling, standing, sitting, and lying in patients with
COPD, as well as changes in movement intensity during walking. Important aspects of the device’s validity were shown: its
high degree of agreement with the criterion standard and its
stability, with high ICCs and narrow CIs. Second, patients’
estimations of time spent on these activities show important
inaccuracy compared with the objectively assessed values.
Because the use of video recordings was not feasible during
long-term assessment in the patients’ environments and because the DAM was highly accurate in protocol A, the DAM
was used as a surrogate for video recordings in study B. It
appears that patients significantly overestimate their walking
time in daily life (see fig 3). This warrants caution when using
self reports by patients with COPD of time spent in activity or
inactivity in daily life.
Time spent in physical activity during daily life (eg, walking), together with intensity and frequency, are fundamental
parts of recommendations of physical activity for health maintenance.5 Assessment of habitual physical activity has been
performed frequently with questionnaires or activity diaries.
Our study has shown that individual patients were unable to
report accurately the amount of activities performed over 12
hours and even over 1 hour. Although mean values of outcomes from patients’ estimations and video recordings in study
A (see table 2) did not differ significantly (except for sitting
time), the analysis of the Bland and Altman plot shows that the
degree of agreement is not strong (fig 2, right panel), compared
with the degree of agreement between the DAM and video
recordings (fig 2, left panel). Accuracy can be defined as the
degree of agreement of a measured value with the true value.28
Therefore, in study A, the DAM was shown to be accurate,
whereas patients’ estimations were not. The very small overestimation observed in walking and cycling time for the DAM
when compared with video recordings in study A has no
clinical relevance, and it was probably due to the very few
seconds missed by clocking analysis of the video during transitions of activities.
In study B, 9 of 13 patients (69%) overestimated their
walking time compared with the objectively assessed values.
This lack of accuracy in patients with COPD may be due to
natural shortcomings of the human memory,8 lower cognitive
function in patients with COPD when compared with healthy
subjects,11 difficulties in distinguishing time spent on walking
and standing (table 3), and the “social desirability” to report
high time spent in physical activity, because benefits of exercise are well known. The fact that this inactive29 population
tends to overestimate their walking time is in agreement with
findings in healthy subjects, both adults30 and elderly.31 Furthermore, the subjective perceptions of patients with COPD
may influence their capacity to report time spent in different
activities. For example, a “tiring” (exhaustive) activity, hard to
perform for these patients, can be interpreted by them as taking
longer time than it really did. However, whatever the explanation, discrepancy between objective and subjective methods of
assessment of physical activities was found in this population.
Therefore, objective measures to obtain accurate data from real
life are indicated.
Arch Phys Med Rehabil Vol 86, October 2005
1984
ACTIVITY MONITORING IN PATIENTS WITH COPD, Pitta
table 2). Therefore, caution may be necessary when using the
DAM in obese subjects. Because the comparison between
outcomes of the DAM and video recordings was performed in
only a limited number of situations, further studies should
focus on more varied situations than the ones used in study A
(eg, walking in bumpy terrain, moving around in a furnished
room, cycling on a nonstationary bike, sitting in different kinds
of chairs and sofas).
CONCLUSIONS
The DAM accurately assessed the time spent in different
postures and activities in patients with COPD. Therefore, it is
a promising tool for providing accurate data on the amount of
daily activity in these patients. The estimations by patients with
COPD of time spent in different physical activities in daily life
had large disagreement compared with objective assessment.
Fig 3. Agreement in walking time between DAM and patients’ estimations, assessed at home for 12 hours, according to the Bland
and Altman method (rⴝ–.71, P<.01). Points below the zero line as
difference between tools (y axis) mean overestimation by the patients’ reports (logbook), whereas points above the zero line mean
underestimation. Abbreviations: LL, lower limit (–1.96 SDs); UL,
upper limit (ⴙ1.96 SDs).
Acknowledgments: We thank the physiotherapists Iris Coosemans, Veronica Barbier, Luk Herremans, Vanessa Probst, Sofie
Mortier, Anja Bocken, and Pau Villela, from the UZ Gasthuisberg’s
Respiratory Rehabilitation Division, for the great help in the assessments; Silvia Cecere, from the Biostatistical Center (KU Leuven), for
the statistical assistance; and Joris Snaet, for drawing figure 1.
The results of our study do not minimize the importance of
questionnaires as a measure of the patients’ subjective perceptions of their limitations and difficulties. Patients can report
their perceived limitations and symptoms in daily life6 or even
to give an appropriate classification to their disability.7 However, in our study, patients with COPD did not report accurately the amount of time spent in activities compared with
outcomes from the objective methods. Hence, although objective and subjective methods are useful and assess complementary aspects reflecting physical activity,32 they do not provide
similar estimates of time spent actively (or inactively) in daily
life.33
The activity-monitoring technique used in our study has
some limitations. These limitations include the inability to
assess movements of the upper extremities and a lower accuracy in the assessment of the movement intensity during cycling and slow walking, despite measuring acceleration in the
vertical and horizontal axis. The low accuracy during cycling
occurs because, in the current software program, the movement
intensity analysis does not include accelerations captured by
the leg sensor but only by the other 2 accelerometers placed at
the waist. Therefore, cycling speed cannot be properly assessed. The lower accuracy of the measurements during slow
walking may occur because of the smooth movement of the
body’s center of gravity, which may not be properly detected
by the accelerometer. This issue deserves further evaluation in
subsequent studies. The device, however, was sensitive to
increases in acceleration during walking, and this measurement
highly correlated with increase in walking speed. Although it is
clear that speed and acceleration are not the same, an increase
(or decrease) in speed is generally linked to an increase (or
decrease) in acceleration. Because the DAM was sensitive to
acceleration and deceleration during walking, we believe it
allows us to use this outcome as a surrogate for changes in
walking speed in daily life. Furthermore, the device is relatively unbiased, with the only important misinterpretation of
the device occurring in a single patient with a prominent
abdominal volume (body mass index, 28kg/m2), where portions of the lying time were interpreted as sitting time. This
isolated observation led to the small and nonsignificant differences in sitting and lying time between DAM and video (see
References
1. Garcia-Aymerich J, Farrero E, Felez MA, et al. Risk factors of
readmission to hospital for a COPD exacerbation: a prospective
study. Thorax 2003;58:100-5.
2. Kessler R, Faller M, Fourgaut G, Mennecier B, Weitzenblum E.
Predictive factors of hospitalization for acute exacerbation in a
series of 64 patients with chronic obstructive pulmonary disease.
Am J Respir Crit Care Med 1999;159:158-64.
3. Yohannes AM, Baldwin RC, Connolly M. Mortality predictors in
disabling chronic obstructive pulmonary disease in old age. Age
Ageing 2002;31:137-40.
4. Global Initiative for Chronic Obstructive Lung Disease. Global
strategy for the diagnosis, management and prevention of chronic
obstructive pulmonary disease— executive summary. Heart,
Lung, and Blood Institute, World Health Organization; 2003.
5. Pate RR, Pratt M, Blair SN, et al. Physical activity and public
health. A recommendation from the Centers for Disease Control
and Prevention and the American College of Sports Medicine.
JAMA 1995;273:402-7.
6. Lareau SC, Meek PM, Roos PJ. Development and testing of the
modified version of the pulmonary functional status and dyspnea
questionnaire (PFSDQ-M). Heart Lung 1998;27:159-68.
7. Bestall JC, Paul EA, Garrod R, et al. Usefulness of the Medical
Research Council (MRC) dyspnoea scale as a measure of disability in patients with chronic obstructive pulmonary disease. Thorax
1999;54:581-6.
8. Shephard RJ. Limits to the measurement of habitual physical
activity by questionnaires. Br J Sports Med 2003;37:197-206.
9. Giantomaso T, Makowsky L, Ashworth NL, Sankaran R. The
validity of patient and physician estimates of walking distance.
Clin Rehabil 2003;17:394-401.
10. Sanders SH. Automated versus self-monitoring of up-time in
chronic low-back-pain patients—a comparative study. Pain 1983;
15:399-405.
11. Antonelli Incalzi R, Marra C, Giordano A, et al. Cognitive impairment in chronic obstructive pulmonary disease—a neuropsychological and SPECT study. J Neurol 2003;250:325-32.
12. Leidy NK, Abbott RD, Fedenko KM. Sensitivity and reproducibility of the dual-mode actigraph under controlled levels of activity intensity. Nurs Res 1997;46:5-11.
13. Steele BG, Holt L, Belza B, Ferris S, Lakshminaryan S, Buchner
DM. Quantitating physical activity in COPD using a triaxial
accelerometer. Chest 2000;117:1359-67.
Arch Phys Med Rehabil Vol 86, October 2005
ACTIVITY MONITORING IN PATIENTS WITH COPD, Pitta
14. Schonhofer B, Ardes P, Geibel M, Kohler D, Jones PW. Evaluation of a movement detector to measure daily activity in patients
with chronic lung disease. Eur Respir J 1997;10:2814-9.
15. Singh S, Morgan MD. Activity monitors can detect brisk walking
in patients with chronic obstructive pulmonary disease. J Cardiopulm Rehabil 2001;21:143-8.
16. American College of Sports Medicine Position Stand. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in
healthy adults. Med Sci Sports Exerc 1998;30:975-91.
17. Busser HJ, Ott J, van Lummel RC, Uiterwaal M, Blank R. Ambulatory monitoring of children’s activity. Med Eng Phys 1997;
19:440-5.
18. Uiterwaal M, Glerum EB, Busser HJ, van Lummel RC. Ambulatory monitoring of physical activity in working situations, a validation study. J Med Eng Technol 1998;22:168-72.
19. Steele BG, Belza B, Cain K, Warms C, Coppersmith J, Howard J.
Bodies in motion: monitoring daily activity and exercise with
motion sensors in people with chronic pulmonary disease. J Rehabil Res Dev 2003;40(5 Suppl 2):45-58.
20. Follick MJ, Ahern DK, Laser-Wolston N. Evaluation of a daily
activity diary for chronic pain patients. Pain 1984;19:373-82.
21. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R,
Yernault JC. Lung volumes and forced ventilatory flows. Report
Working Party Standardization of Lung Function Tests, European
Community for Steel and Coal. Official Statement of the European
Respiratory Society. Eur Respir J Suppl 1993;16:5-40.
22. Black LF, Hyatt RE. Maximal respiratory pressures: normal values and relationship to age and sex. Am Rev Respir Dis 1969;99:
696-702.
23. Decramer M, Lacquet LM, Fagard R, Rogiers P. Corticosteroids
contribute to muscle weakness in chronic airflow obstruction.
Am J Respir Crit Care Med 1994;150:11-6.
24. Troosters T, Gosselink R, Decramer M. Six minute walking
distance in healthy elderly subjects. Eur Respir J 1999;14:270-4.
1985
25. Troosters T, Gosselink R, Decramer M. Short- and long-term
effects of outpatient rehabilitation in patients with chronic obstructive pulmonary disease: a randomized trial. Am J Med 2000;
109:207-12.
26. Bland JM, Altman DG. Statistical methods for assessing agreement between 2 methods of clinical measurement. Lancet 1986;
1:307-10.
27. Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med 1998;26:217-38.
28. Taylor JK. Quality assurance of chemical measurements. Chelsea:
Lewis Publishers; 1987.
29. Pitta F, Troosters T, Spruit MA, Probst VS, Decramer M, Gosselink R. Characteristics of physical activities in daily life in
chronic obstructive pulmonary disease. Am J Respir Crit Care
Med 2005;171:972-7.
30. Strath SJ, Bassett DR Jr, Ham SA, Swartz AM. Assessment of
physical activity by telephone interview versus objective monitoring. Med Sci Sports Exerc 2003;35:2112-8.
31. Starling RD, Matthews DE, Ades PA, Poehlman ET. Assessment
of physical activity in older individuals: a doubly labeled water
study. J Appl Physiol 1999;86:2090-6.
32. Schutz Y, Weinsier RL, Hunter GR. Assessment of free-living
physical activity in humans: an overview of currently available
and proposed new measures. Obes Res 2001;9:368-79.
33. Ainsworth BE, Bassett DR Jr, Strath SJ, et al. Comparison of three
methods for measuring the time spent in physical activity. Med
Sci Sports Exerc 2000;32(9 Suppl):S457-64.
Suppliers
a. DynaPort Activity Monitor; McRoberts BV, Raamweg 43, 2596
The Hague, The Netherlands.
b. SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513.
c. GraphPad Software Inc, S 11452 El Camino Real, #215, San Diego,
CA 92130.
Arch Phys Med Rehabil Vol 86, October 2005