You are on page 1of 4


Comparison Between an Indoor and an Outdoor 6-Minute

Walk Test Among Individuals With Chronic Obstructive
Pulmonary Disease
Dina Brooks, PhD, Sherra Solway, MSc, Krisztina Weinacht, Dip(PT), David Wang, MSc, Scott Thomas, PhD
ABSTRACT. Brooks D, Solway S, Weinacht K, Wang D, NDIVIDUALS WITH CHRONIC obstructive pulmonary
Thomas S. Comparison between an indoor and an outdoor
6-minute walk test among individuals with chronic obstructive
disease (COPD) show variable exercise capacity and activity
that do not necessarily correlate with severity of airflow
pulmonary disease. Arch Phys Med Rehabil 2003;84:873-6. obstruction.1-4 Measures of functional status are commonly
used in this population and are essential for optimal clinical
Objectives: To investigate the feasibility of an outdoor management.2,5 In a national survey of pulmonary rehabilita-
6-minute walk test (6MWT) as a measure of functional status tion programs in Canada, 98% of respondents reported inclu-
among individuals with chronic obstructive pulmonary disease sion of a measure of functional exercise capacity, usually the
(COPD), and to examine the relationship between performance 6-minute walk test (6MWT) or the 12-minute walk test.6
on an indoor and an outdoor 6MWT. The 6MWT is a valid measure of functional status com-
Design: An experimental, repeated-measures crossover de- monly used for evaluating individuals with COPD.7 Typically,
sign. Subjects were studied on 2 separate days in the same this test is administered in an enclosed indoor corridor, free of
week. Two 6MWTs— one indoors and the other outdoors— distractions, and patients are asked to cover as much distance
were performed on each study day, with a rest in between. The as they can in 6 minutes.2 This environment is somewhat
test order was randomly selected on the first day and reversed artificial and may not reflect activities that individuals usually
on the second day. Outdoor tests were performed on days of do on a daily basis.
moderate weather conditions (mean temperature ⫾ standard There is no literature on the amount of time that individuals
deviation, 21°⫾3°C; mean wind speed, 15⫾7km/h; no precip- with advanced pulmonary disease spend indoors versus out-
itation) and on a flat surface (sidewalk). doors. Nevertheless, activities of daily living (ADLs) likely
Setting: Outpatient rehabilitation program in Ontario. require them to ambulate outdoors regularly (eg, walking from
Participants: Eighteen subjects with COPD (10 men, 8 their car to a doctor’s office, shopping mall, or grocery store).
women; age, 70⫾8y), 5 using supplemental oxygen at rest Many of these individuals will have experience with pulmo-
(forced expiratory volume in 1s, 1.0⫾0.3L; 42%⫾8% of pre- nary rehabilitation programs, in which they are encouraged to
dicted). maintain their exercise program by ambulating. From our clin-
Interventions: Not applicable. ical experience, individuals with COPD value the opportunity
Main Outcome Measures: Distance walked in 6 minutes to exercise outdoors. Furthermore, health care professionals in
(in meters), duration of rest (in seconds), and change in rate of pulmonary rehabilitation programs use the results of the indoor
perceived dyspnea. 6MWT to advise individuals with COPD about activities out-
Results: There was no significant effect of setting (indoors doors. Such recommendations are based on the health care
vs outdoors) on distance walked (394⫾86m vs 398⫾84m, professional’s clinical experience and not on any scientific
P⫽0.4), duration of rest (13⫾28s vs 9⫾20s, P⫽0.4), or change evidence. Finally, clinicians making home visits or those in
in rate of perceived dyspnea (2.3⫾1.7 vs 2.3⫾2.0, P⫽0.8). community clinics may not have access to the indoor space
Testing day had no significant effect on walk test performance required for the 6MWT, making an outdoor 6MWT a desirable
(all P⬎0.1). alternative for evaluating functional status in these environ-
Conclusions: The results indicate that the 6MWT performed ments.
outdoors within reasonable climatic parameters may be reflec- Thus, the purpose of this study was to investigate the feasi-
tive of 6MWT performance indoors. bility of an outdoor 6MWT as a measure of functional status
Key Words: Exercise test; Exercise tolerance; Pulmonary among individuals with COPD and to determine the relation-
disease, chronic obstructive; Rehabilitation; Walking ship between performance on an indoor and an outdoor
© 2003 by the American Congress of Rehabilitation Medi- 6MWT.
cine and the American Academy of Physical Medicine and
Rehabilitation METHODS

From the Department of Physical Therapy (Brooks, Solway, Wang, Thomas),
Subjects were recruited from the outpatient pulmonary re-
University of Toronto; Institute for Work & Health and Toronto Rehabilitation habilitation program at Toronto East General Hospital, in To-
Institute (Solway); and Toronto East General Hospital (Weinacht), Toronto, ON, ronto, ON, Canada. Subjects were considered eligible for the
Canada. study if they were clinically stable with a diagnosis of COPD,8
Supported by the Ontario Respiratory Care Society and Ontario Lung Association.
No commercial party having a direct financial interest in the results of the research
were between 55 and 85 years of age, had a forced expiratory
supporting this article has or will confer a benefit upon the author(s) or upon any volume in 1 second (FEV1) of less than 60% of predicted, and
organization with which the author(s) is/are associated. reported dyspnea or fatigue with ADLs. Exclusion criteria
Reprint requests to Dina Brooks, PhD, Dept of Physical Therapy, University of included the presence of associated medical conditions that
Toronto, 500 University Ave, Rm 848, Toronto, ON M5G 1V7, Canada, e-mail:
limited exercise tolerance (eg, symptomatic cardiovascular or
0003-9993/03/8406-7772$30.00/0 musculoskeletal conditions, recent surgery), inability to com-
doi:10.1016/S0003-9993(03)00011-X municate in English, or the use of a mobility aid. The rationale

Arch Phys Med Rehabil Vol 84, June 2003


for the last criteria was that the use of a mobility aid could alter from a list of possible factors (temperature, wind, pollution,
performance outdoors because of changes in the terrain and humidity, other).
that these alterations may be difficult to differentiate from Functional status was also evaluated by using the modified
differences unrelated to aids in a small sample. version of the Pulmonary Functional Status and Dyspnea Ques-
tionnaire (PSFDQ-M) on the first study day before the start of
testing. The PSFDQ is a 40-item, self-administered functional
The Research Ethics Boards at the University of Toronto and status and symptoms (dyspnea and fatigue) questionnaire that
Toronto East General Hospital approved the study. All subjects requires less than 10 minutes to complete.12,13 Psychometric
gave informed written consent. A randomized crossover design properties of this questionnaire (eg, internal consistency, test-
was used. Each subject was studied at the same time of the day retest reliability, construct validity) have been established in
on 2 separate days in the same week. Subjects were asked to adult patients with pulmonary disease.12-14
abstain from caffeine for 4 hours before each session and to
The questionnaire includes 3 domains: dyspnea, fatigue, and
administer their inhaled bronchodilator 30 minutes before each
session. Subjects wore the same footwear on both study days. activity level. For the activity domain, a list of 10 ADLs (eg,
On each study day, two 6MWTs were performed with a brushing/combing hair, walking) are included, and subjects
minimum of a 30-minute rest between tests (or when perceived were asked to indicate “involvement with the activity now as
dyspnea, heart rate, and oxygen saturation returned to baseline compared to before you developed lung problems.” A scale of
levels). One walk test was performed indoors and the other 0 (as active as I have ever been) to 10 (have omitted entirely)
outdoors. The test order was randomized for the first day and was used. The same activities were listed again, and the sub-
reversed on the second day. The indoor 6MWT was adminis- jects were asked to rate each according to the degree of short-
tered in a corridor, 30m in length. Pylons were placed at either ness of breath (dyspnea domain) and tiredness (fatigue domain)
end of the course. The tests were performed under quiet con- experienced. A scale of 0 (none) to 10 (very severe) was used.
ditions, with a minimum of distractions and corridor traffic. Within each domain, the values assigned for each activity were
The outdoor 6MWT was performed by using the same length summed and divided by the number of activities that applied.
of flat sidewalk, in a quiet neighborhood. Outdoor tests were Thus, for each domain, the score could range from 0 (minimal
performed on days in which the weather was “reasonable,” limitation) to 100 (maximal limitation).11
which was defined as an “apparent” temperature (a composite
of ambient temperature and humidity) of 10° to 25°C, no Statistical Analysis
precipitation, wind speed of less than 20km/h, and an air A sample size estimation using a 2-tailed paired t test with a
quality index of less than 32. The air quality index ranges from type I error of .05 and power of 90% determined that a
0 to more than 100, with 0 to 15 representing very good air clinically significant difference in 6MWT distance5 (ie, 54m)
quality; 16 to 31, good; 32 to 49, moderate; 50 to 99, poor; and would be detected with a minimum of 16 subjects (standard
100 or more, very poor.9 A value below 32 has no known deviation [SD], 86m).
health effects for the majority of the population.9 Weather Means and SDs were calculated for all outcomes. Descrip-
variables were not directly measured; this information was tive statistics (including frequencies) were also used for subject
taken from a weather report website corresponding to the characteristics (eg, age, FEV1) and subject preference. A total
location and time each outdoor walk test was administered. of four 6MWTs were performed; two 6MWTs were performed
All subjects performed at least 2 practice walks before data on each day, one indoors and the other outdoors. To determine
collection, to control for learning and practice effects.7 Stan- the effect of environment, analyses were performed by using
dardized instructions were provided to subjects and no encour- 2-way repeated-measures analysis of variance. The 2 factors
agement was offered during the tests.10 Subjects were re- were day (1, 2) and environment (indoors, outdoors). Linear
quested to cover as much ground as possible during the test regression was calculated to examine the relationship between
period, stopping only if they felt too tired or too breathless to the distance walked indoors and outdoors. Intraclass correla-
continue, and to resume walking as soon as they were able to tion coefficients (ICCs) were calculated to represent test-retest
do so. The tester accompanied the subjects for each walk test reliability for day and setting. To determine if there was an
and walked behind the subject to avoid pacing. A folding chair association between self-report level of functional status and
was used for sitting if the subject required a rest. Subjects using outdoor walk test performance, univariate regression was used
supplemental oxygen carried or pulled their tank during each to explore the relationship between distance walked outdoors
6MWT. and PFSDQ-M scores. To determine whether more function-
Measures ally disabled individuals had a greater deterioration in the
outdoor walk test as compared with the indoor walk test, we
For each 6MWT, we recorded distance walked and the also explored the relationship between the change in the dis-
number and duration of rests, and we monitored oxyhemoglo- tance walked indoors minus outdoors and PFSDQ-M scores.
bin saturation and heart rate continuously by using a pulse SigmaStat, version 2.03,b and SigmaPlot, version 5.0,b statis-
oximeter.a The oximeter provided a printout of values reflect- tical softwares were used for all analyses; a P value of .05 or
ing maximum and minimum recorded and the mean of the less was considered significant.
complete duration (value used in analysis). At the start and at
the end of each test, subjects rated their perceived rate of
dyspnea (“difficulty of breathing”) by using a modified Borg RESULTS
Scale.11 By using the same scale, subjects were also asked to Eighteen individuals with a medical diagnosis of COPD
rate their perceived leg effort. Furthermore, a self-report of participated in the study. The characteristics of these subjects
how the climate affected performance and preference between are presented in table 1.
the indoor and the outdoor 6MWT was recorded by means of The outdoor 6MWTs were performed on comparable days
a simple standardized questionnaire. This brief questionnaire with respect to climatic variables—that is, temperature, humi-
was piloted for suitability and clarity. Subjects were asked to dex, and wind speed (all P⬎0.2). The combined data for all
identify factors that influenced their ability to walk outdoors days are presented in table 2.

Arch Phys Med Rehabil Vol 84, June 2003


Table 1: Characteristics of the Subjects (Nⴝ18)

Mean ⫾ SD (Range)

Age (y) 70⫾8 (52–81)

FEV1 (L) 1.0⫾0.3 (0.5–1.5)
FEV1 (% predicted) 42⫾8 (28–42)
Sex 10 men, 8 women
Oxygen use (L) (n⫽5) 2.9⫾0.7 (2–4)

Figure 1 shows distance walked indoors and outdoors for

each subject. There was no effect of setting on distance walked
(indoors: 394⫾86m vs outdoors: 398⫾84m, P⫽0.4). Duration
of rest was 13⫾28 seconds for the indoor tests and 9⫾20
seconds for the outdoor tests (P⫽0.4). Furthermore, setting had
no effect on the change in rate of perceived dyspnea (indoors:
2.3⫾1.7 vs outdoors: 2.3⫾2.0, P⫽0.8) or perceived leg effort
(indoors: 1.1⫾1.0 vs outdoors: 0.9⫾0.8, P⫽0.6) (fig 2). Actual
values for rate of perceived exertion are presented in table 3.
Similarly, setting did not influence mean heart rate or oxygen
saturation (SpO2) during the test (P⬍0.3) (fig 3). Day of testing Fig 1. Total distance walked indoors and outdoors. Open circles
represent individual data points. Solid squares represent mean ⴞ SD.
had no significant effect on any of the variables examined (all
P⬎0.1), which indicates that the test was reliable on 2 separate
days (ICC⫽.94). Linear regression between distance walked
indoors from the 2 trials and the distance walked outdoors rial disease, surgical patients, and pediatric patients.7 A recent
revealed a significant correlation (r⫽.97, P⬍.001, slope⫽.99, systematic review of functional walk tests by our group con-
intercept⫽⫺.64), and an ICC of .95. cluded that the psychometric properties of the 6MWT have
Forty-two percent of the subjects had no preference for the been well researched and established, and we have recom-
environment of testing, whereas 36% preferred the indoor mended the 6MWT as the test of choice for clinical and
setting and 22% preferred the outdoor. When asked which research use.7
climatic variable influenced their performance, 30% of respon- An outdoor 6MWT test is more reflective of real life and is
dents identified temperature, 14% wind, 19% pollution, and not as artificial as the conventional 6MWT, in which an indoor
11% humidity. corridor, free of distractions, is used. One of the objectives of
The PFSDQ-M reflected minimal impairment for dyspnea this study was to compare performance on the indoor and on
(1.9⫾1.6; range, 0.1–5.0), fatigue (1.7⫾1.5; range, 0.1–5.6), the outdoor walk tests. The data indicate that the 6MWT
and activity (2.1⫾1.6; range, 0.1–5.0). There was no significant indoors is strongly correlated to the test performed outdoors.
correlation between distance walked outdoors and each of the This finding will enable health care professionals to use the
3 dimensions (r⬍0.2, P⬎0.6) or between the distance walked results of indoor performance to predict outdoor performance
indoors and the 3 dimensions (r⬍0.1, P⬎0.7). Similarly, there and vice versa. Additionally, the results of this study will allow
were no significant correlation between the difference in dis- health care professionals to use the 6MWT in the community
tance walked indoors and outdoors and the PFSDQ-M sub- (ie, during a home visit), to measure functional status and to
scales scores (r⬍0.2, P⬎0.5). monitor treatment effectiveness, when an indoor corridor is not
DISCUSSION The perception of individuals with COPD of the effect of the
The findings indicate that when performed within reasonable environment on functional capacity has not been investigated.
climatic conditions, the outdoor 6MWT is a feasible measure Despite no difference between performance indoors and out-
of functional status among individuals with COPD. Further- doors, slightly more subjects preferred the indoor setting (36%)
more, the outdoor 6MWT test was reproducible on 2 separate
days, and subjects’ performance did not differ significantly
from that indoors.
We were unable to locate any studies that included an
outdoor 6MWT. However, considerable attention has been
given to the indoor 6MWT, especially in recent years. The use
of the 6MWT has been studied in different populations, includ-
ing individuals with COPD, individuals with heart failure,
individuals with pacemakers, individuals with peripheral arte-

Table 2: Summary of Climatic Variables for the Outdoor Tests

Climatic Variables Mean ⫾ SD (Range)

Fig 2. Comparison of change in perceived dyspnea and leg effort
Temperature (°C) 21⫾3 (16–25.5)
ratings when walking indoors and outdoors. Open circles represent
“Apparent” temperature (°C) 25⫾4 (17–32) individual data points. Solid squares represent mean ⴞ SD. There
Wind speed (km/h) 15⫾7 (5–33) are several individual points at 0 that are reflected by 1 data point

Arch Phys Med Rehabil Vol 84, June 2003


Table 3: Actual Values for Modified Borg Ratings of Perceived Dyspnea and Leg Effort Before and After 6MWT Indoors and Outdoors

Day 1 (mean ⫾ SD) Day 2 (mean ⫾ SD)

Indoors Outdoors Indoors Outdoors
Rating Before After Before After Before After Before After

Dyspnea 0.5⫾0.8 2.7⫾1.9 0.8⫾0.8 3.0⫾2.3 0.5⫾0.5 2.9⫾1.8 0.5⫾0.7 3.0⫾2.2

Leg effort 0.5⫾0.9 1.6⫾1.3 0.3⫾0.4 1.4⫾1.3 0.4⫾0.5 1.4⫾1.3 0.4⫾0.6 1.1⫾0.9

to the outdoor setting (22%). Temperature was the main factor studies are needed to increase the generalizability of these
perceived as influencing ability to walk outside. findings.
The finding that the distance walked outdoors did not cor-
relate with PFSDQ-M scores was not surprising. The 2 mea- References
1. Guyatt GH, Sullivan MJ, Thompson PJ, et al. The 6-minute walk:
sures provide different information on functional capacity: the a new measure of exercise capacity in patients with chronic heart
PFSDQ-M examines various ADLs (1 component of which is failure. Can Med Assoc J 1985;132:919-23.
walking), and the 6MWT assesses the ability to walk. How- 2. Guyatt GH, Thompson PJ, Berman LB, et al. How should we
ever, our sample included individuals at a high level of func- measure function in patients with chronic heart and lung disease?
tion, as shown by the low scores on the PFSDQ-M and indoor J Chronic Dis 1985;38:517-24.
6MWT distances greater than 300m. Studies have indicated 3. Epstein SK, Celli BR. Cardiopulmonary exercise testing in pa-
that 300m on the 6MWT is a threshold for level of disability tients with chronic obstructive pulmonary disease. Cleve Clin
among individuals with respiratory disease, including those J Med 1993;60:119-28.
with COPD.15-17 Thus, it is possible that individuals with 4. Bittner V. Six-minute walk test in patients with cardiac dysfunc-
tion. Cardiologia 1997;42:897-902.
greater disability (eg, 6MWT distances ⬍300m) may be influ- 5. Redelmeier DA, Bayoumi AM, Goldstein RS, Guyatt GH. Inter-
enced by the outdoor environment. preting small differences in functional status: the six-minute walk
Two studies18,19 have reported distances walked in 6 minutes test in chronic lung disease patients. Am J Respir Crit Care Med
by healthy subjects. In the healthy elderly (aged 60 – 65y), the 1997;155:1278-82.
distances walked in 6 minutes ranged from 494 to 631m. These 6. Brooks D, Lacasse Y, Goldstein RS. Pulmonary rehabilitation
values are higher than those observed in our sample, in which programs in Canada: national survey. Can Respir J 1999;6:55-63.
the distances walked in 6 minutes were between 390 and 400m. 7. Solway S, Brooks D, Lacasse Y, Thomas S. A qualitative systematic
Therefore, although our sample was high functioning com- overview of the measurement properties of functional walk tests used
pared with others with COPD, they had more disability than a in the cardiorespiratory domain. Chest 2001;119:256-70.
8. American Thoracic Society. Standards for the diagnosis and care
healthy population. of patients with chronic obstructive pulmonary disease. Am J
Limited ranges of climatic parameters were used in this Respir Crit Care Med 1995;152:S77-121.
study. Future studies will need to investigate the effect of a 9. Ontario Ministry of Environment and Energy. Available at: http://
broader range of environmental factors (eg, more extreme Accessed June 10, 2002.
conditions, ie, colder temperature, higher winds, pollution hu- 10. Guyatt GH, Pugsley SO, Sullivan MJ, et al. Effect of encourage-
midity levels, and different terrains) and walking aids on out- ment on walking test performance. Thorax 1984;39:818-22.
door performance, to expand our understanding of the feasi- 11. Borg GA. Psychophysical bases of perceived exertion. Med Sci
bility and validity of using the 6MWT as a measure of Sports Exerc 1982;14:377-81.
functional status in an outdoor setting. 12. Lareau SC, Carrieri-Kohlman V, Janson-Bjerklie S, Roos PJ.
Development and testing of the Pulmonary Functional Status and
CONCLUSION Dyspnea Questionnaire (PFSDQ). Heart Lung 1994;23:242-50.
13. Lareau SC, Meek PM, Roos PJ. Development and testing of a
The data indicate that the 6MWT performed outdoors within modified version of the pulmonary status and dyspnea question-
reasonable climatic parameters may be reflective of 6MWT naire. Heart Lung 1998;27:159-68.
performance indoors. Therefore, an outdoor 6MWT may be an 14. Lareau SC, Meek PM. Testing of a modified version of the
alternative option to the traditional 6MWT for use in the pulmonary functional status and dyspnea questionnaire [abstract].
community when an indoor corridor is not available. Future Am J Respir Crit Care Med 1996;153:A421.
15. Solway S, Brooks D, Lau L, Goldstein R. The short-term effect of
a rollator on functional exercise capacity among individuals with
severe COPD. Chest 2002;122:56-65.
16. Honeyman P, Barr P, Stubbing DG. Effect of a walking aid on
disability, oxygenation, and breathlessness in patients with
chronic airflow limitation. J Cardiopulm Rehabil 1996;16:63-7.
17. Goldstein RS, Gort EH, Stubbing D, Avendano MA, Guyatt GH.
Randomized control trial of respiratory rehabilitation. Lancet
18. Troosters T, Gosselink R, Decramer M. Six minute walking
distance in healthy elderly subjects. Eur Respir J 1999;14:270-4.
19. Enright PL, Sherrill DL. Reference equations for the six-minute
walk in healthy adults. Am J Respir Crit Care Med 1998;158:

Fig 3. Mean oxygen saturation and heart rate during the 6MWT Suppliers
when walking indoors and outdoors. Open circles represent indi- a. Model 8800; Nonin Medical Inc, 2605 Fernbrook Ln N, Plymouth,
vidual data points. Solid squares represent mean ⴞ SD. Note that MN 55447-4755.
the y axis does not start at zero. b. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

Arch Phys Med Rehabil Vol 84, June 2003