If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
School of Physiotherapy, Curtin University of Technology, Perth, Western Australia, AustraliaDepartment of Pulmonary Physiology, Sir Charles Gairdner Hospital, Perth, Western Australia, AustraliaSchool of Anatomy and Human Biology, University of Western Australia, Perth, Western Australia, Australia
School of Physiotherapy, Curtin University of Technology, Perth, Western Australia, AustraliaDepartment of Physiotherapy, Sir Charles Gairdner Hospital, Perth, Western Australia, Australia
Centre for Asthma, Allergy & Respiratory Research, University of Western Australia, Perth, Western Australia, AustraliaSchool of Medicine & Pharmacology, University of Western Australia, Perth, Western Australia, Australia
Corresponding author. School of Physiotherapy, Curtin University of Technology, Perth, Western Australia, Australia. Tel.: +61892663639; fax: +61892663699.
School of Physiotherapy, Curtin University of Technology, Perth, Western Australia, AustraliaDepartment of Physiotherapy, Sir Charles Gairdner Hospital, Perth, Western Australia, AustraliaCentre for Asthma, Allergy & Respiratory Research, University of Western Australia, Perth, Western Australia, Australia
The six minute walk test (6MWT) is the most commonly used exercise test in pulmonary rehabilitation; however, the paucity of six minute walk distance (6MWD) reference values from population-based samples limits data interpretation in patients. This study was undertaken to determine 6MWD in a population-based sample of healthy subjects and to identify predictors of 6MWD in this group.
Methods
Seventy Caucasian subjects (33 males) aged 55–75 years performed three tests using a standardised protocol. 6MWD was defined as the greatest distance achieved from the three tests. Other measurements included height, leg length, weight, forced expiratory volume in 1s (FEV1), exhaled carbon monoxide and self-reported physical activity including habitual walking.
Results
The average 6MWD was 659±62m (range 484–820m). Males walked 59±13m further than females (). Height (, ), weight (, ) and FEV1 (, ) were significantly correlated with 6MWD. Forwards stepwise multiple regression showed height () and FEV1 (R2 change=0.045) to be independent predictors of 6MWD (), explaining 33.9% of the variance.
Conclusions
6MWD in this healthy population-based sample of males and females exceeds values previously reported. Height and FEV1 were identified as significant independent predictors of 6MWD in this group.
The six minute walk test (6MWT) is the most frequently used exercise test in pulmonary rehabilitation and is considered to provide a reliable and valid measure of functional exercise capacity.
Despite the popularity of the 6MWT there is limited data relating six minute walk distance (6MWD) in patients with lung disease to that of healthy subjects. The paucity of 6MWD reference values from population-based healthy subjects limits the interpretation of 6MWD in patients and poses problems for clinicians wishing to provide patients with a measure of their expected 6MWD in the absence of disease.
To date, four studies have determined regression equations to predict 6MWD in healthy populations.
will underestimate 6MWD, as repeated testing is required to account for the effect of familiarisation on the distance walked by both healthy subjects and patients with chronic obstructive pulmonary disease (COPD).
Course length, pre-test instructions and the frequency of encouragement also differed between the studies. Taken together, these differences highlight the need to determine 6MWD in a healthy population-based sample following a standardised protocol.
This study sought to determine 6MWD in a healthy population-based sample of males and females. As the majority of patients with COPD undergoing pulmonary rehabilitation are aged 55–75 years, only individuals falling within this age range were studied. The study also aimed to identify factors which could predict 6MWD. Most previous studies have shown that age, gender, height and weight account for between 19% and 42% of the variability in 6MWD.
forced expiratory volume in 1s (FEV1), and habitual walking activity as potential predictors of 6MWD.
Methods
Subjects
A population-based sample of healthy males and females were recruited from a database of research volunteers established by the Centre for Asthma, Allergy and Respiratory Research (Perth, Western Australia) for genetic studies. Volunteers were included in the database following postal requests mailed to randomly selected individuals (using the program Marketing Pro, DtMS, Blackburn, Victoria, Australia) listed in the telephone directory of a state capital city (Perth, Western Australia, population 1.3 million). All volunteers underwent an interview to confirm their healthy status and spirometry testing at the time of recruitment to the database.
Screening
Subjects were considered for the study if they were aged 55–75 years and had no documented cardiopulmonary disease. Exclusion criteria were abnormal lung function (FEV1<80% predicted [pred] or FEV1/forced vital capacity [FVC]<70%); medications affecting exercise capacity; hypertension (blood pressure>150/100mmHg); diabetes mellitus; recent illness including upper respiratory tract infection or the presence of any factor that may limit ability to participate in the 6MWT (e.g. impaired cognitive function, neuromuscular disease, claudication, severe musculoskeletal problems affecting the lower extremity or spine or use of a walking aid).
An information sheet was mailed to 168 consecutive non-asthmatic individuals listed on the database. Twenty-five individuals were unable to be contacted and 37 declined to participate. The remaining 106 individuals agreed to participate in the study and were contacted by telephone to determine suitability for inclusion. Thirty-three were subsequently excluded due to the presence of cardiovascular disease (), osteoarthritis (), respiratory disease () and other illness (). Seventy-three subjects attended for assessment, of whom a further two were excluded due to abnormal lung function and one subject failed to comply with the 6MWT protocol. Thus, the study population comprised 70 healthy subjects, all were of Caucasian origin and 33 were males. Approval for the study was granted by the Human Research Ethics Committees at Sir Charles Gairdner Hospital and Curtin University of Technology. Written, informed consent was obtained from each subject prior to testing.
Procedure
Data collection took place over a 4-month period with each subject attending on a single occasion. Subjects were asked to avoid caffeine, alcohol and consumption of a heavy meal for at least 2h prior to testing and strenuous physical exercise in the previous 24h.
Smoking history and blood pressure were recorded. Weight and height were measured and used to calculate body mass index (BMI). Leg length for the right and left leg was measured
and mean leg length was used in subsequent analyses. Lung function was measured in accordance with the recommendations of the American Thoracic Society
using a Vitalograph wedge bellows spirometer (Model S, Vitalograph Ltd., Buckingham, UK). Exhaled carbon monoxide was measured in duplicate using an EC50 Micro Carbon Monoxide Monitor (Micro Medical Ltd., Southampton, UK). This monitor provides a conversion to percentage of carboxyhaemoglobin (COHb) and the higher of the two measurements was recorded.
Subjects completed a Physical Activity Questionnaire
Bull F, Milligan R, Rosenberg M, MacGowan H. Physical activity levels of Western Australian Adults 1999. Perth, Western Australia: Health Department of Western Australia and Sport and Recreation Way2Go, Western Australian Government; 2000.
to determine the amount of physical activity performed in the previous week, or another recent week if the last week was atypical. Questionnaire responses were used to classify subjects as undertaking sufficient physical activity, insufficient physical activity for health benefits, or inactive using the definitions from the Survey of Physical Activity Levels of Western Australian Adults 1999.
Bull F, Milligan R, Rosenberg M, MacGowan H. Physical activity levels of Western Australian Adults 1999. Perth, Western Australia: Health Department of Western Australia and Sport and Recreation Way2Go, Western Australian Government; 2000.
The 6MWT was performed over a 45m long straight course within an enclosed level corridor. To allow familiarisation with the protocol, subjects performed the test three times separated by 20min rest intervals. The same investigator (BC) supervised all tests. The temperature of the corridor was 25.8±1.5°C and humidity was below 55% during all tests.
Taped standardised instructions (modified from Sciurba and Slivka
) were played to subjects prior to each 6MWT. Subjects were told to “walk as quickly as you can for six minutes so that you cover as much ground as possible.” They were informed that they could slow down or rest if necessary. At the end of each minute subjects were given feedback on the elapsed time and standardised encouragement in the form of statements such as “you’re doing well, keep it up” and “do your best.
Heart rate (HR) was recorded at rest and at the end of each minute during the 6MWT using a lightweight telemetric HR monitor (Polar Electro Oy, Kempele, Finland). Oxygen saturation (SpO2) was recorded prior to and upon completion of the 6MWT using an oximeter and finger sensor (Ohmeda Biox 3700e, Ohmeda, Colorado, USA). At the end of the test, the distance was measured and, where applicable, leg fatigue was rated using a modified Borg 0–10 scale.
During the 20min rest between consecutive tests subjects were provided with a glass of water and asked to remain seated. HR was re-measured at the end of the rest period to ensure that it was within 10 beats per minute (bpm) of the previous resting value and where necessary the rest period was extended.
Statistical analyses
Data were checked for distribution and means and standard deviations calculated. All data were normally distributed with the exception of FEV1/FVC, COHb and minutes walked in the previous week. Acceptable normalisation of these variables was achieved with a square root transformation.
A one-way repeated measures ANOVA with contrasts was used to compare the three 6MWD's. Reproducibility of the 6MWD was examined using intraclass correlation coefficient (ICC) and coefficient of variation (CV).
Maximum 6MWD from the three tests was identified for each subject and used in the following analyses. Unpaired t-tests were performed to determine differences in subject characteristics and 6MWD between males and females. Pearson's univariate correlation coefficients (r) were computed to examine the association between subject characteristics and 6MWD. Forward stepwise multiple regression was used on the variables height, age, weight, gender, leg length, FEV1 (L) and minutes walked in the previous week to determine their contribution to 6MWD.
The 6MWD measured in our study was compared with predicted 6MWD derived from three published equations
using scatter plots and paired t-tests. Comparisons were made of maximum 6MWD obtained from an equivalent number of tests where possible. Specifically, when comparing our data to that of Enright and Sherrill
we used the 6MWD from the first test and the best of the first and second tests, respectively. The maximum 6MWD from the three tests in each subject was compared to the data of Gibbons et al.
Analyses were performed using the SPSS computer package (Version 11). An alpha value of 0.05 was used to determine significance and data are reported as mean±standard deviation (sd).
Results
Subject characteristics
The characteristics of the 70 subjects (33 males) are summarised in Table 1. The majority of subjects (63%) were overweight or obese. Sixty-three subjects (90%) were classified as undertaking sufficient physical activity for health benefits.
Bull F, Milligan R, Rosenberg M, MacGowan H. Physical activity levels of Western Australian Adults 1999. Perth, Western Australia: Health Department of Western Australia and Sport and Recreation Way2Go, Western Australian Government; 2000.
The subjects had a high level of education, with 37 subjects (53%) having attended a vocational training college or university. In eight subjects COHb levels were above the normal range. Males were taller by 10.1±1.5cm (); however, there was no significant difference in leg length between males and females (88.0±6.5cm vs 85.3±7.0cm, ). There were also no significant differences between the males and females in BMI, minutes walked in the last week, physical activity levels and smoking history.
Table 1Characteristics of the 70 healthy subjects.
Bull F, Milligan R, Rosenberg M, MacGowan H. Physical activity levels of Western Australian Adults 1999. Perth, Western Australia: Health Department of Western Australia and Sport and Recreation Way2Go, Western Australian Government; 2000.
No test was terminated prematurely by the investigator and no subject required a rest during the 6MWT. The resting HR prior to commencing successive tests was not different, being 75.3±10.8, 76.6±11.0 and 77.7±10.8bpm.
6MWD during tests 1, 2 and 3 was 626±63, 645±58 and 655±61m, respectively. There was a significant increase in 6MWD between consecutive tests with subjects walking 18±30m (a 3.0±5.0% increase, maximum 20.7%) further on the second test () and 10±17m (a 1.6±2.7% increase, maximum 13.0%) further on the third test (). Distances walked on tests 1, 2 and 3 were 652±58, 669±49 and 685±49m in the males, and 603±60, 623±57 and 628±59m in females. Analysis by gender showed the increase between the second and third test was significant for males only (males 16±20m, ; females 5±12m, ). The number of subjects achieving their maximum distance on the first, second and third test was 4, 15 and 51, respectively. The ICC for the three tests was 0.84 and the CV was 2.9% (range 0.23–11.0%).
The overall maximum 6MWD (best of the three tests) was 659±62m (range, 484–820m) and was greater in males than females by 59±13m (), being 690±53 and 631±57m in males and females, respectively.
Heart rate, oxygen saturation and leg symptoms
Peak HR was 135±20bpm (87±13% predicted maximum HR [220-age]) on the best of the three tests with 39 subjects exceeding 85% of predicted maximum HR (HRmax). There were no significant differences in pre- and post-test SpO2 (96.7±1.1% vs. 96.6±1.34%, ). Fifteen subjects reported leg fatigue (score >0) at the end of the 6MWT.
Associations with six minute walk distance
Univariate correlation coefficients between 6MWD and subject variables are presented in Table 2. Significant correlations were observed with height, weight and FEV1 but not with age, leg length, BMI, COHb or minutes walked in the previous week.
Table 2Univariate correlation coefficients (r) for 6MWD and subject variables ().
Age (years)
Height (cm)
Leg length (cm)
Weight (kg)
BMI (kg/m2)
COHb (%)
FEV1 (L)
Mins walked
6MWD (m)
−0.03
0.54
0.05
0.25
−0.07
−0.09
0.48
0.13
P value
NS
<0.01
NS
<0.05
NS
NS
<0.001
NS
NS: not significant at ; Mins walked: minutes walked in previous week.
Forward stepwise multiple regression analysis revealed that height and FEV1 were significant independent predictors () and together explained 33.9% of the variability in 6MWD. The regression equation for this relationship is: 6MWD(m)=64.69+3.12 (height, cm)+23.29 (FEV1, L). Height alone accounted for 29.4% of the variance and FEV1 explained an additional 4.5% of the variance in 6MWD. Gender was not a significant predictor and explained only 2.7% of the variance in 6MWD.
Multiple regression analysis on height, age, weight and gender, shown to be predictors of 6MWD in other studies,
explained 36% of the variance in 6MWD. The regression equation for this relationship is: 6MWD(m)=216.90+4.12 (height, cm)−1.75 (age, years)−1.15 (weight, kg)−34.04 (gender, where males=0 and females=1).
Figure 1 shows the individual data points for these comparisons and demonstrates that, regardless of which equation was used, the predicted 6MWD tended to underestimate our measured 6MWD. Group mean 6MWD±sd (95% CI) was underestimated by 119±65m (103–134m)
This study demonstrated that the average 6MWD in healthy subjects aged 55–75 years recruited from our database in Western Australia was 659m. Height and FEV1 were significant independent predictors of 6MWD and together explained 33.9% of the variance in 6MWD.
Small but statistically significant improvements in 6MWD were observed between consecutive tests and may be explained by test familiarisation which has been shown to influence intra-subject variability in 6MWD in both healthy subjects and COPD patients.
Data comparing the magnitude of any learning effects in males and females is absent in COPD patients and scarce in healthy subjects. In our healthy females, the distance plateaued by the second test but in males 6MWD significantly increased from test 2 to test 3. The latter finding is consistent with Wu et al.
who showed healthy males required more practice tests than females to reach a maximal 6MWD. Further improvements in 6MWD may have been achieved in male subjects with a fourth test; however, the magnitude of any such increase is likely to be small.
While a statistically significant gender difference was noted for 6MWD in our study, the mean increase between the first and third test for males and females was small (6.0% and 4.8%, respectively), suggesting that in studies with large numbers of healthy subjects a single test may provide a useful representative measure of 6MWD. It was notable, however, that the magnitude of increase was large in some individuals being 23% in both one male and one female subject. There were no obvious identifying characteristics in these individuals.
Comparison of 6MWD with other studies
Compared to estimated 6MWD using regression equations from other studies,
and the subjects’ age and anthropometric characteristics were similar in the two studies. It is possible that the effort levels of the subjects studied by Enright and Sherrill
were submaximal, as HR during their 6MWT (only one test performed) reached <65% age-predicted HRmax. This contrasts to a HR of 81% of predicted maximum attained in our subjects on their first 6MWT, 80% of predicted HRmax achieved on the first 6MWT in the study by Kervio et al.
Significant positive correlations were found between 6MWD with height, weight and FEV1 but not with age, leg length, BMI, COHb or minutes walked in the previous week. The observed association between height and weight with 6MWD is consistent with previous findings.
recently found an association between FEV1 and the distance walked on the incremental shuttle walking test in healthy males.
Height and FEV1 were the only significant independent predictors of 6MWD, explaining 33.9% of the between-subject variance in 6MWD. This is comparable to the variance explained by most other studies in healthy subjects (19–42%).
who found that age, height, weight and gender explained 66% of the variance in 6MWD. Using the same four variables we were still only able to account for 36% of the variance in 6MWD. The reason for the difference is not clear but may relate to differences in subject characteristics and the greater range of 6MWD in Troosters et al.
study (383–820m) compared to ours (484–820m). Verification of our regression equation and 6MWD achieved by subjects in our study should be undertaken in a prospective cohort of healthy subjects recruited using identical inclusion criteria.
Age was not a significant predictor of 6MWD, a finding that contrasts with previous studies.
Stringent exclusion criteria prohibited the participation of subjects with age-related health problems that may have adversely affected walking speed. Some studies have shown that age is not a significant predictor of walking speed when plantar flexor strength and the presence of health problems are taken into consideration,
suggesting that pathology associated with ageing rather than chronological age may account for a greater proportion of the variation in walking speed during the 6MWT.
However, our findings did not support this. This is most likely because only nine of our subjects had a BMI>30, a value which has been identified as the upper threshold for a low 6MWD in healthy subjects.
Inclusion of subjects with a wider range of BMI may be necessary to more clearly define the relationship between BMI and 6MWD. Alternatively, it may be that other anthropometric variables such as waist size could account for a greater proportion of the variability in 6MWD.
who found that after correcting for height, males did not walk further than females. We chose not to measure quadriceps strength as it does not appear to be a predictor of 6MWD in healthy subjects after correcting for height, weight, age and gender.
In an attempt to explain a greater proportion of the variability in 6MWD we assessed the contribution of leg length, minutes walked in the previous week and COHb. We hypothesised that leg length may be a better predictor of 6MWD than height as it is a primary determinant of stride length.
However, this was not supported by our data. We hypothesised that 6MWD would be greater in individuals who walked for prolonged periods on a regular basis. Of particular interest was the time spent walking for continuous periods of at least 10min either for recreation, exercise or transport. However, minutes walked in the previous week was not a predictor of 6MWD. This finding is consistent with other studies that also failed to demonstrate an association between self-reported physical activity and 6MWD.
It is possible that the speed of habitual walking, which was not assessed in this study, may be a more relevant predictor of 6MWD than time spent walking. Since high levels of carbon monoxide in arterial blood reduce the oxygen carrying capacity of haemoglobin and may reduce exercise capacity
, we hypothesised that elevated levels of COHb may contribute to a lower 6MWD. However, we were unable to identify an association between raised COHb levels and decreased 6MWD. This is most likely because the selection criteria required subjects to have normal lung function and the majority of subjects (89%) had a level of COHb within the normal range. To more specifically examine the effect of COHb on 6MWD, a study with larger sample size, including greater numbers of current smokers would be required. A greater proportion of the variability in 6MWD may also be explained by inclusion of factors such as mood, attitude, motivation and race, which have also been shown to influence 6MWD.
In conclusion, this study has shown that 6MWD in a healthy population-based sample of 55–75-year-old males and females from Western Australia exceeds values reported in previous studies. A prediction equation containing height and FEV1 explained 33.9% of the variance in 6MWD.
Acknowledgements
The authors thank the Centre for Asthma, Allergy and Respiratory Research for assistance in subject recruitment and Dr. Marie Blackmore for methodological and statistical assistance. We also thank the Physiotherapy Department at Sir Charles Gairdner Hospital for allowing us to use their facilities and each subject for their generous participation in this research. This study was funded, in part, by a Sir Charles Gairdner Hospital Research Grant and NHMRC (Australia) Grant no. 212016. Dr. Peter Eastwood was supported by a NHMRC R. Douglas Wright Fellowship (No. 294404).
References
Guyatt G.H.
Thompson P.J.
Berman L.B.
et al.
How should we measure function in patients with chronic heart and lung disease?.
Bull F, Milligan R, Rosenberg M, MacGowan H. Physical activity levels of Western Australian Adults 1999. Perth, Western Australia: Health Department of Western Australia and Sport and Recreation Way2Go, Western Australian Government; 2000.