Share this article on:

The 6MWT: Will Different Methods of Instruction and Measurement Affect Performance of Healthy Aging and Older Adults?

Southard, Veronica PT, DHSc; Gallagher, Rosemary PT, DPT

Journal of Geriatric Physical Therapy: April/June 2013 - Volume 36 - Issue 2 - p 68–73
doi: 10.1519/JPT.0b013e318264b5e8
Research Reports

Background and Purpose: The Six-Minute Walk Test (6MWT) is commonly used to assess the fitness level of healthy adults and of older adults with disabilities. It can also be used as an intervention to increase walking endurance. However, its use may be limited in certain rehabilitation settings due to space requirements. If it can be shown that the measured linear distance walked in the 6-minute walk is comparable to the distance walked as measured by a pedometer, the test may be more widely used in a variety of rehabilitation settings. In addition, questions exist as to whether the method of instruction (“walk as far as you can” vs “walk as fast as you can”) can impact the rate of perceived exertion of the person performing the test. The purposes of this study were to assess for differences in the measured linear distance and from the gender-based predicted value when compared to the pedometer measurement. In addition, we assessed the difference, if any, in the rate of perceived exertion (RPE) using the 2 different methods of administration. Furthermore, the distance in meters walked using the 2 different methods of instruction was compared; likewise, comparisons were made of these values to predicted values.

Methods: A group of 26 older adults participated in this descriptive study. After a practice trial, each person completed 2 linear trials using different methods of instruction, (“walk as fast as you can” or “walk as far as you can”) of the 6MWT while wearing a DIGI-WALKER SW-651 pedometer. Vital signs were taken before and after each trial. Linear distance, pedometer distance, and numeric value RPE were recorded.

Results: Paired t tests demonstrated no gender differences. An intraclass correlation coefficient (2,1) of 0.822 was calculated between all dependent variables. A repeated measures MANOVA was conducted to assess for differences between all variables resulting in no differences (F = 1.98; P = .13). Pairwise comparisons were also insignificant for the distance measurements except predicted value and pedometer fast P = .024. Paired t tests also demonstrated differences between RPE between trials (t = 2.15; P = .041).

Conclusion: There was good agreement between these distance measures for the 6MWT. The use of a pedometer was found to be a valid measure of walking distance during the 6MWT. It was also found that the method of instruction made no differences in walking distance. Although the change was minimal on the Borg scale, the RPE was found to be significantly different between far and fast trials in healthy adults. From this study, it appears that that either mode of instruction is valid in healthy community-dwelling populations. Future studies should include populations with impairments.

Department of Physical Therapy, School of Health Professions, New York Institute of Technology, Old Westbury

Address correspondence to: Veronica Southard, PT, DHSc, Program in Physical Therapy, New York Institute of Technology, Building 500, Room 501, Old Westbury, NY 11568 (

No financial support.

No conflicts of interest.

Written permission has been obtained from the second author and consent forms have been collected.

Back to Top | Article Outline


The Six-Minute Walk Test (6MWT) was adapted from the 12-minute walk test for those individuals with respiratory disease who could not complete the 12-minute walk test.1 While the 6MWT has been used for a wide variety of disorders, it was originally intended for use with patients with cardiac and pulmonary disorders.1 The test has also been used to assess functional capacity at submaximal levels.2,3 Kervio et al found that the 6MWT represented a submaximal exercise level that is 80% of

O2max.3 Potential clinical uses of the 6MWT include assessment of exercise endurance and as a possible indicator of a person's ability to walk safely in the community (by averaging gait velocity over the 6 minutes you can compare with functional gait speeds found in the literature). It may also be used to determine an individuals' progress following a rehabilitation intervention.4 Research has shown that patients walking a distance of less than 300 m have poor outcomes in terms of mortality.5 The 6MWT has also been found to highly correlate with exercise workloads, heart rate, oxygen saturation, and dyspnea response when compared to bicycle ergometry and treadmill exercises in middle age and older adults.6 The 6MWT distance has been estimated using a predicted distance formula via reference equations and assessed using a 30-m course with distance markers every 3 m. It has been recommended that as few turns as possible be included in the course to keep the pace without having to slow down for multiple turns.6

In this study, we applied the reference equation for healthy adults by gender, based on the far instruction, found in the Journal of Respiratory and Critical Care Medicine 6 where

  • for healthy adult males: 6MWT = 1140 m − (5.61 × body mass index) − (6.94 × age) and
  • for healthy adult females: 6MWT = 1017 m − (6.24 × body mass index) − (5.83 × age)

This reference study included 117 males and 173 female nonsmokers, who were less than 80 years of age, and had body mass index of less than 35, no history of peripheral vascular disease, cerebral vascular accident, or use of diuretics. The study strictly adhered to the American Thoracic Society (ATS) guidelines (cover as much ground as possible) to develop these equations. An alternate method of measuring the distance a person walks is with the use of a pedometer. Pedometers are motion sensors that are low cost, unobtrusive, and accurate.7 The pedometer is a simple, practical tool that can provide motivation to achieve a goal in the number of steps taken in a day. In addition, it is easy to use for self-exercise prescription and monitoring.8 Other important applications of pedometers include comparisons of the differences in the number of steps per day between individuals, also as an ongoing measurement for the effects of increased activity as an intervention, and as a measure of community activity.7

Elsworth et al9 have found pedometers to be reliable and accurate at measuring step counts in 13 healthy subjects with a mean age of 29 years. The SW-701 Digi walker has been shown to be one of the more reliable and valid pedometers to be used for research purposes, giving a value within 3% of actual step count.7 The mechanism of the Digi walker includes a spring-suspended horizontal mechanism that moves up and down based on the vertical movement of the hip during gait.7 Conversely, Le Masurier and Tudor-Locke10 found that those with walking speeds of less than 0.60 m/s the pedometer was only 75.4% accurate. This was because the forces causing vertical displacement were less than the 0.35 g necessary for a step to register. Gait speeds that are this low, however, generally are not indicative of an independent free-living adult.10

According to the ATS guidelines on the 6MWT, a subject is instructed to “walk as far as you can.”1 Other independent studies have also used this command.4, 11,12 Troosters et al13 and Rikli and Jones14 have given the instructions “walk as fast as you can.” ATS guidelines suggest caution in using the latter command, because it emphasizes initial speed at the potential expense of fatigue and cardiac stress later in the test.1 Guyatt et al15 also found that the addition of encouragement was associated with a significant increase in walking distance during the 6MWT. Standardized encouragement is used with the ATS instructions “walk as far as you can.” The ATS guidelines provide a clear description of dialog to use during the administration of the 6MWT.1

Rikli and Jones14 studied the 6MWT using the verbal command, “walk as fast as possible for 6 minutes,” in healthy adults and found it to be a moderately valid measure of physical endurance. In addition, it was reliable, especially when a practice trial was administered. They concluded that this test reflects an older adults' overall functional ability. Moreover, Rikli and Jones16 have developed a guide for use of the 6MWT with healthy older adults that contain norms based on age and gender. To our knowledge, no study has been executed which tests the use of pedometers in measuring distance ambulated in the 6MWT using different methods of instruction.

Back to Top | Article Outline


There were several purposes for this study: (1) Differences related to dissimilar instructions given during the 6MWT were assessed. (2) Values obtained under the far and fast conditions were compared to the derived predicted distance from the gender and Body mass index–based reference equation. (3) The validity of the use of a pedometer measurement was compared to the measured linear distance while performing the 6MWT under each type of instruction. (4) The RPE Borg 6-20 scale17 was assessed for any differences between the 2 different methods of instruction. The linear distances derived in this study were measures from the starting position, to the stopping point in meters, based on markings on the floor. Linear distance determination using the pedometer was that distance in meters at completion of the 6MWT that registered on the pedometer and was recorded at the place the subject stopped at after the trial. Stride length was adjusted on each pedometer for the subjects based on their height following the instructions from the manufacturer.7 The predicted distance was derived from the equations from the Journal of Respiratory and Critical Care Medicine.6

Back to Top | Article Outline


This study was approved by the Institutional Review Board at New York Institute of Technology. Healthy older subjects were recruited from the community via a church bulletin advertisement and a flyer sent to a local senior center. Inclusion criteria were that the subjects be between 50 and less than 86 years of age, have no acute illnesses, active orthopedic complications, or restrictions for walking. Resting vitals blood pressure <140/90, heart rate range 60 to 100 beats/min, respirations between 12 and 20 breaths/min, must be able to walk independently for 6 minutes. Exclusion criteria were pain or any active orthopedic or neurological condition that affects the ability to walk. Also those with active or chronic cardiopulmonary disease and those with resting vitals outside the acceptable ranges listed above were excluded from participating. Finally, any subject that could not tolerate 6 minutes of upright activity was excluded.

Twenty-six participants completed the study, which was composed of 8 men and 18 women. Eight subjects were lost to follow up after the practice trial and were not included in the data analysis. Reasons included sickness (1), transportation difficulties (2), forgot (2), and loss of interest in participating (3). The mean age was 62.09 years; the age range of these healthy aging adults was 50 to 87 years. The mean far instruction gait speed for the females was 1.57 m/s, and the males was 1.54 m/s. Descriptive information on subjects are shown in Table 1.

Table 1

Table 1

All subjects underwent a practice trial to reduce learning effects after which 2 trials were performed, one using “fast” and the other using “far” instructions. There was a minimum rest period of 3 hours between trials. The maximum time between trials was 3 days. Each participant wore a Digi Walker SW-651 pedometer placed midway between the umbilicus and iliac crest. The pedometer was affixed to the participants' pants line and taped onto their clothing, if loose at the waist, to avoid excessive motion of the pedometer. Pedometers were placed on either side of the patient's pants line randomly, in keeping with the findings from a study by Schneider that found no significant difference in recorded measurement between pedometers worn on the right or left side of the body.8

Heart rate, blood pressure, and respiratory rate were taken prior to and after each trial for all participants. Each subject was also shown the RPE Borg 6–20 scale17 before and after each trial; the numerical value was recorded for each.

Subjects were instructed to walk along a 30-m taped line on the floor with taped markers placed every 3 m. Subjects turned at each end of the course, resulting in an elliptical pattern. The subjects were allowed to stop if needed during the test, but were not allowed to sit. Stride length was calculated prior to the practice trial and was based on each subject's height. The pedometer was adjusted according to the instructions provided by the manufacturer. Groups of six subjects were scattered along the 30-m course and performed the test at the same time. Subjects were instructed not to run. Subjects were informed that this was an individual test not a competition, and that each person was to do their individual best during the trial. Subjects were randomly told either to “walk as far as you can in 6 minutes” strictly following the ATS recommendations or “walk as fast as you can in 6 minutes” (without running).14 A researcher acted as a counter and was assigned to each subject to record the number of laps a participant completed. At the end of 6 minutes, subjects were instructed to stop where they were. Both the linear distance and the distance from the pedometer were recorded for each subject. The RPE was also recorded. (See Table 2 for distance measures.)

Table 2

Table 2

Back to Top | Article Outline

Data Analysis

All statistics were completed using SPSS for Windows version 20.0. Descriptive statistics were completed on the subjects (Table 1). Table 2 contains the mean scores of all of the variables by gender and combined. A repeated measures MANOVA was used to assess 6MWT instruction fast and far and 6MWT measurement method both linear and pedometer and calculated distance. Intraclass correlation coefficients were applied to assess agreement between the method of instruction and linear distance, distance arrived at using the pedometer, and predicted distances. Paired t tests were used for pair-wise comparisons of all variables as seen in Table 3.

Table 3

Table 3

Back to Top | Article Outline


Prior to statistical analysis, independent t tests were completed to assess for gender differences. No differences were found at the .05 level of significance; therefore, the genders were combined.

The results of the repeated measures MANOVA used to compare the linear distance measures to the pedometer measures revealed Mauchleys' W was significant (W = 0.323 P = .002). This is indicative of a violation of sphericity, which is the assumption that the variance of the difference score in each level of the repeated measure is equal. This violation may increase the probability of a type I error.18 Subsequently, a conservative correction, the Greenhouse-Geisser, was applied to test within subjects effects for this factor. Based on the corrected degrees of freedom, the main effect of measure was not significant (F = 1.99, P = 0.130, partial η2 = 0.266). An intraclass correlation coefficients (ICC) (2,1) of 0.822 assessed agreement among the methods of linear distance; distance arrived at using the pedometer, and predicted distances. An ICC 0.822 was realized between all of the 6MWT measures.

Paired t tests were used for pairwise comparisons of combined gender results as seen in Table 3. There were no differences noted in any of the pairs compared except for RPE fast and RPE far: paired differences (1.27, t = 2.15; P = .041).

Back to Top | Article Outline


The purposes of this descriptive study were as follows: First, the repeated measures MANOVA was employed to compare the linear distance to the pedometer distance for both modes of instruction and the predicted formula. There was no significant difference in the distance walked between method of instruction, pedometer distance, and predicted formula.

The actual measures, regardless of method and instruction, were greater than the predicted values for both genders. However, on the fast trial the pedometer reading showed that the females performed a mean of 72 m greater than the predicted value and the males performed only 21 m better. There was a mean age difference between the genders with the males 7.49 years older. We posit that this was due to an age effect despite the calculated values accounting for this in our sample. The age effect evidenced is in agreement with Chui and Lusardi19 in their study with healthy community living adults, as well as with other studies that have also documented an overall reduction in gait speed with aging.11, 20

Our subjects older than 60 years (n = 15) demonstrated mean distances (533 m) similar to a meta-analysis conducted by Bohannon et al in which the mean value in 13 healthy adults older than 60 years was 499 m.21

Previous regression equations for predicted values used the far method only. On the basis of the Le Masurier and Tudor-Locke10 work that found a pedometer more accurate at faster gaits, we included the fast pedometer trial in the model and found there were differences between the predicted value and fast trial pedometer recorded measure. Dourado22 cites numerous formulas that calculate 6MWT using various approaches and reports that the intensity of a subject while participating in the 6 MWT is based on motivation, gender, and anthropometric characteristics. The mean gait speed for these healthy aging adults was in excess of 1.5 m/s. To our knowledge, this is the first study to apply fast instruction to Enright and Sherrill's6 formulas. We are confident that the application is acceptable for both linear and pedometer measurement based on the small differences in distances between the trials as seen in Table 2 and the results of the MANOVA.

ICC were applied to assess agreement between the methods of linear distance, distance arrived at using the pedometer, and predicted distances. An ICC (2,1) of 0.822 was calculated for all of the 6MWT measures, indicating good agreement between the variables assessed. The ICC uses the covariance to demonstrate agreement and relationships between ratings.18 Part of the reason for the large standard deviation within the groups may be the range of difference in age of the subjects. Other researchers such Steffen et al11 found an ICC of 0.95 when they defined their subjects by decade. Our healthy aging adult subjects ranged from 50 to 86 years.

In addition, no significant difference between instruction and pedometer distance was found, therefore indicating that use of the DIGI-WALKER-SW-651 to measure distance walked during the 6MWT is a valid and accurate tool to use with this population. Our research findings correspond to past research studies that demonstrated that the use of a pedometer is a valid and practical tool for distance measurement.7

Tudor-Locke et al23, who studied the strongest indicators of fitness, found that the strongest relationship of a valid indicator of fitness was the combination of 6MWT and pedometer measures, as opposed to pedometer measures and self- reported activity throughout an individual's day. Our results of accurate measurement of the pedometer relate to this finding. Cesari et al24 state that inadequate space as one reason why standardized assessments are not used. Therefore, using a pedometer might aid in reducing one reason why the 6MWT is not used. It would certainly be easier to have a patient dons a pedometer and walk for 6 minutes which would eliminate the need for the counter. Also home programs aimed to improve functional distances ambulated or fitness level could be more readily facilitated and accurately documented.

Third, we found there was a slight but significant difference in RPE between far and fast instruction. Between trials, the overall mean result for RPE during the far trial was 10.38 and 11.04 for the fast trial. As previously mentioned, the ATS recommends the 6MWT be administered using the far instruction because the subject may try too hard initially and lose the ability to complete their farthest distance. Although not significant, we did note the males' distance on the fast trial was 10 m less than the far trial. We have shown that there is no significant difference between the distance achieved after the repeated measures MANOVA, and a slight but statistically significant increase in RPE after the fast trial. On the Borg scale, this level of exertion is the difference between very light 10 and fairly light 11, levels of exercise. We were unable to find the minimal clinically important difference (MCID) for the 15-point RPE scale used in this study. However, Eaton and colleagues12 found the MCID for the Borg 10-point scale to be 1 unit. This is a 10% change. Their subjects were persons with pulmonary disease. Effect size changes in this study were based on the Borg score during trials with and without supplemental oxygen during the 6MWT. In addition, a retrospective review, Reis et al25 also reviewed MCID for the Borg and found that for large effect sizes of 0.8 or greater, improvements in the Borg were at 2 units or more when instructed to walk at a comfortable pace. In this study, subjects had less than a 10% change between the far and fast trials. There were no untoward responses in any of the subjects' vital signs after either trial. Moreover, since the mean 6MWT difference for the far versus fast trial is only 13 m the exertion should be comparable.

We believe that our findings indicate that because our older group was active and independent in the community may have accounted for the low score in exertion reported by our subjects. In addition, there were no remarkable changes in the subjects' vital signs on either trial. The typical reading on the Borg scale for the subjects was fairly light, approximately 11/20, regardless of the method of instruction. Based on our findings, the use of either verbal command should yield equivalent outcomes in distance ambulated during the 6MWT. According to Enright and Sherrill,6 older and heavier subjects typically have reduced muscle mass, and therefore, shorter 6MWT distances, as do those who are less motivated or have impaired cognition. Arthritis, and other musculoskeletal diseases, can also decrease the 6MWT distance.

Limitations include small sample size and the possible influence of the combination of healthy aging adults with healthy older adults. We also had a large variation in age of our participants with the men being older than the women. Although the subjects knew it was an individual effort and were well spaced, having others on the course may have affected their performance and therefore reduced the generalizability of the results.

Future studies should seek to test the accuracy of a pedometer in larger populations and those with impairments. Since the pedometer is valid for this test, it may have applications for other standardized linear measure tests. Also the validity of the pedometer over different course lengths would be beneficial. In the current health care climate, it is becoming increasingly important to provide accurate measures of exercise interventions and to aid in patient adherence. The effects of these interventions must also be quantifiable. Therefore, future studies should also be conducted to test the validity of physical activity assessment tools, in order that they may be implemented with confidence and provide accurate results. In the end, this will improve our level of care and benefit our patients.

Back to Top | Article Outline


By performing a study comparing the 6MWT to a variety of measures, we have shown that in aging adults, a carefully placed pedometer is an accurate measure of distance walked on a flat 30-m course. This held true regardless of the method of instruction and was valid based on the calculated values. A potential barrier of administration of the 6MWT is the required 30-m long course. In addition, this study provides supplementary evidence that, when used appropriately, pedometers are a low cost, practical, and accurate tool for measuring physical activity and functional capacity. However, one must cautiously accept the findings of this study when applying these results to actual patient functional endurance measurement, since all of these subjects were healthy and living independently in the community.

In conclusion, if the 6MWT can be used to assess a patient's abilities in terms of functional capacity, interventions can be derived from the measurement and will aid in exercise prescription, the development of independent home programs, and objective measurement of outcomes.

Back to Top | Article Outline


1. American Thoracic Society. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111–117.
2. Enright PL, McBurnie MA, Bittner V, et al. The 6-min walk test, a quick measure of functional status in elderly. Chest. 2003;123(2):387–398.
3. Kervio G, Carre F, Ville N. Reliability and intensity of the six-minute walk test in healthy elderly subjects. Med Sci Sports Ex. 2003;35(1):169–174.
4. Harada ND, Chiu C, Stewart AL. Mobility-related function in older adult: assessment with a 6-minute walk test. Arch Phys Med Rehabil. 1999;80(7):837–841.
5. Rostagno C, Olivo G, Comeglio M, et al. Prognostic value of 6 minute corridor walk test in patients with mild to moderate heart failure: comparison with other methods of functional evaluation. J Heart Fail. 2003;5(3):247–252.
6. Enright P, Sherrill D. Reference equations for the six-minute walk in healthy adults. Am J Respir Crit Care Med. 1998;158(5):1384–1387.
7. Digimax. Accessed May 27, 2012.
8. Schneider P, Crouter S, Bassett D. Pedometer measures of free living physical activity: comparison of 13 models. Med Sci Sports Med. 2004;36(2):331–335.
9. Elsworth C, Dawes H, Winward C, et al. Pedometer step counts in individuals with neurological conditions. Clin Rehabil. 2009;23(2):171–175.
10. Le Masurier G, Tudor-Lock C. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med Sci Sports Exerc. 2003;35(5):867–871.
11. Steffen T, hacker T, Mollinger L. Age-and gender–related test performance in community-dwelling elderly people: Six Minute Walk Test, Berg balance scale, timed up & go test, and gait speeds. Phys Ther. 2002;82(2):128–137.
12. Eaton T, Garrett JE, Young P, et al. Ambulatory oxygen improves quality of life of COPD patients: a randomised controlled study. Eur Respir J 2002;20(2):306–312.
13. Troosters T, Gosselik R, Decramer M. Six minute walking distance in healthy elderly subjects. Eur Respir J. 1999;14(2):270–274.
14. Rikli R, Jones J. The reliability and validity of a 6-Minute Walk Test as measure of physical endurance in older adults. J Aging Phys Act. 1998;6(4):363–375.
15. Guyatt GH, Pugsley SO, Sullivan MJ, et al. Effect of encouragement on walking test performance. Thorax. 1984;39(11):818–822.
16. Rikli R, Jones J. Senior Fitness Test Manual. Champaign, IL: Human Kinetics; 2001.
17. Borg G. Perceived exertion as an indicator of somatic stress. Scand Rehab Med. 1970;2(2):92–98.
18. Portney L, Watkins M. Foundations of Clinical Research Applications to Practice. 2nd ed. Upper Saddle River, NJ: Prentice Hall; 2000:577.
19. Chui KK, Lusardi MM. Spatial and temporal parameters of self-selected and fast walking speeds in healthy community–living adults aged 72–98. J Geriatr Phys Ther. 2010;33(4):173–183.
20. Bohannon R. Comfortable and maximal walking speed of adults aged 20–79 years: reference values and determinants. Age Aging. 1997;26(1):15–19.
21. Bohannon R. Six-Minute Walk Test: a meta-analysis of data from apparently healthy elders. Top Geriatr Rehabil 2007;23(2):155–160.
22. Dourado V. Reference equations for the 6-Minute Walk Test in healthy individuals. J Bras Pneumol. 2011;37:607–614.
23. Tudor-Locke C, Williams JE, Reis JP, Pluto D. Utility of pedometers for assessing physical activity: construct validity. Sports Med. 2004;34(5):281–91.
24. Cesari M, Kritchevsky SB, Penninx BW, et al. Prognostic value of usual gait speed of well functioning older people: results from the health, aging and body composition study. J Am Geriatr Soc. 2005;53(10):1675–1680.
25. Reis J, Ecternach J, Nof L, Gagnon Blodgett M. Test retest reliability and minimal detectable change scores for the Timed “Up and Go” Test, Six minute Walk Test, and gait speed in people with Alzheimer disease. Phys Ther, 2009;89(6):569–579.

healthy subjects; instruction; pedometer; 6MWT

Copyright © 2013 the Section on Geriatrics of the American Physical Therapy Association