Physiotherapy 99 (2013) 21–26
Activity level predicts 6-minute walk distance in healthy older females: an observational study Daniel Steffens a , Paula R. Beckenkamp a,∗ , Mark Hancock b , Dulciane Nunes Paiva c , Jennifer A. Alison b , Sergio Saldanha Menna-Barreto d a
The George Institute for Global Health, Sydney Medical School, The University of Sydney, Sydney, NSW, Australia b Discipline of Physiotherapy, Faculty of Health Sciences, The University of Sydney, Lidcombe, NSW, Australia c Universidade de Santa Cruz do Sul, Departamento de Educac ¸ ão Fisica e Saude, Curso de Fisioterapia, Santa Cruz do Sul, RS, Brazil d Universidade Federal do Rio Grande do Sul, Faculdade de Medicina, Departamento de Medicina Interna, Porto Alegre, RS, Brazil
Abstract Background The 6-minute walk test (6MWT) is widely used in clinical practice and research. Few studies have investigated activity level as a predictor of 6-minute walk distance (6MWD), and existing predictive models do not allow for activity level. Objectives To evaluate if knowledge of the level of physical activity enhanced the ability to predict 6MWD, and if the inclusion of activity level added to the predictive accuracy of existing models for the 6MWT in healthy older women; and to validate existent predictive models for 6MWD in a new sample. Design Cross-sectional, observational study. Setting Four elderly communities. Participants A convenience sample of healthy active and sedentary older non-smoking females with no musculoskeletal or lung disorders. Main outcome measures Age, height, weight, spirometric values and 6MWD. Results Seventy-seven out of 154 females met the inclusion criteria [mean age 66 (standard deviation 6.5) years]: 46 were active and composed the active group and 31 were sedentary and composed the sedentary group. The active group had significantly greater 6MWD than the sedentary group (mean 44 m; 95% confidence interval 14 to 73 m; P < 0.01). Previous published models that did not allow for activity level either over or underestimated the 6MWD in this sample. The activity level was shown to be an important independent predictor of 6MWD. Conclusion This study demonstrates the importance of considering the level of physical activity when predicting 6MWD in older women. © 2012 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved. Keywords: Older people; Physical activity; 6-minute walk test
Introduction The 6-minute walk test (6MWT) is a quick and inexpensive performance-based measurement that reflects the capacity to undertake day-to-day activities [1,2]. It was first introduced in 1986 to assess patients with chronic obstructive pulmonary disease , and has been used extensively in research in different populations [1,3–20]. The 6MWT can be safely performed by older people and frail patients who may ∗ Corresponding author at: The George Institute for Global Health, PO Box M201, Misenden Road, NSW 2050, Australia. Tel.: +65 02 9657 0318; fax: +65 02 8580 6242. E-mail address: email@example.com (P.R. Beckenkamp).
not be able to undertake a standard maximal cycle ergometry test [2,3,9,11,13,19,21]. Many previous studies have investigated predictors of the 6MWT [1,3,7–9,11–13,19–22]. Some of these studies have developed predictive equations to help generate normative values. These are important to aid interpretation of an individual’s 6MWT. Factors found to be important predictors of 6-minute walk distance (6MWD) in healthy older people include age, height, weight and body mass index [1,7–9,12,13,19–23]. Surprisingly, few studies to date have investigated the relationship between reported activity level and 6MWD in healthy individuals [8,9,12,13]. Several previous studies of the 6MWT in older populations around the globe have created different predictive models for
0031-9406/$ – see front matter © 2012 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.physio.2011.11.004
D. Steffens et al. / Physiotherapy 99 (2013) 21–26
6MWD [1,8,13,21,22]. Most of these have not been validated in new samples of older populations to assess their external validity. These predictive models have not included information about level of physical activity. If activity level is shown to be an important predictor of the 6MWD, its inclusion in predictive models should improve the predictive accuracy of these models. Therefore, the aims of this study were: (1) to evaluate if activity was an independent predictor of 6MWD in a convenience sample of older females; (2) to evaluate if the inclusion of activity level improved the predictive accuracy of existing models for the 6MWT in healthy older women; and (3) to validate existing predictive models for the 6MWT in healthy older women. Methods Study population One hundred and fifty-four healthy Brazilian females who volunteered to participate in this study were recruited from a convenience sample of four community groups. Two of these groups were active and were combined to form the ‘active group’, and the other two groups were combined to form the ‘sedentary group’. The active group included individuals who had been participating in exercise programmes, water-based exercise or aerobics for a minimum of 6 months for 1 hour, three times per week. Exercise sessions included five phases of exercise: warm-up, stretching, aerobic and strengthening exercises and relaxation. The sedentary group included older women from two community centres who reported limited levels of activity, and had not been involved in any type of daily activity or regular exercise for the previous 6 months. Potential participants were excluded from the study if they had musculoskeletal or neurological problems that limited walking, were current smokers, had a body mass index >30 kg/m2 or had abnormal lung function, indicated by forced expiratory volume in 1 second (FEV1 ) <80% predicted, forced vital capacity (FVC) <80% predicted or FEV1 /FVC ratio <70%. Lung function Spirometry was performed using a portable spirometer (EasyOne, Model 2001, Diagnostic Spirometer ndd Medical Technologies, Andover, USA), according to the guidelines of the American Thoracic Society . The predicted values for FEV1 , FVC and FEV1 /FVC ratio were calculated from regression equations developed for the local population . Anthropometric values The weight and height of the participants were measured and expressed in centimetres (cm) and kilograms (kg), respectively. Body mass index was calculated as weight divided by height squared (kg/m2 ).
Six-minute walk test Two 6MWTs were conducted in accordance with the protocol of the American Thoracic Society , with a minimum interval of 30 minutes between the tests. Participants were instructed to walk as quickly as possible for 6 minutes up and down a 30-m straight indoor corridor, and were informed that they could slow down or rest if necessary. Standardised encouragement was given each minute during the tests. The distance walked was recorded and the better of two 6MWDs was used for analysis. Participants were instructed not to undertake any vigorous activity or eat in the 2 hours preceding the 6MWTs, and were previously directed to wear appropriate clothes and shoes. Prior to the 6MWTs, oxygen saturation, heart rate, and systolic and diastolic blood pressure were measured when the subject had rested for 10 minutes. The same measures were collected immediately after the tests. Medline was searched for studies that assessed different populations using the 6MWT in order to compare the present findings with previously reported 6MWT outcomes in healthy populations. Previous studies with similar sample characteristics to the present study, such as a cohort of female participants and older age, were identified. Only studies that presented an equation for a female population were considered. In addition, the protocol used for measuring the 6MWT was taken into account. Statistical analysis All statistical analyses were performed using PAWS Statistics 18 (Chicago, USA). The baseline characteristics for the two groups were compared, including potential predictors of 6MWD. The primary analysis involved a comparison of 6MWD for the active and sedentary groups using a two-tailed independent group t-test. In order to externally validate previously published models for the prediction of 6MWD, linear regression was used to compare each participant’s predicted score using these models with the participant’s actual score. Adjusted R2 values were used to determine the amount of variability in the 6MWT explained by the predictive models. A second step was then performed where the variable ‘group’ (active or sedentary) was added to the model to assess if the activity/sedentary variable added predictive value to the existing models, reflected by an increase in the adjusted R2 value. To further investigate the importance of activity level in combination with other known predictors of 6MWD, a stepwise regression was used to build a predictive model based on the data from participants in the current study. Predictor variables included those previously shown to predict 6MWD (age, weight, height and body mass index) and active/sedentary group. Prior to building the model, correlations between predictors were assessed. If any variables were highly correlated with each other, the least predictive variable was excluded.
D. Steffens et al. / Physiotherapy 99 (2013) 21–26
Results Sixty-three of the 154 participants initially assessed were excluded due to abnormal lung function, five had body mass index >30 kg/m2 , and nine did not perform a second 6MWT. Therefore, the final group consisted of 77 healthy females (46 in the active group and 31 in the sedentary group). The participants’ characteristics are summarised in Table 1. Participants from the two groups were similar for all anthropometric and physiological characteristics that have previously been shown to influence 6MWD (Table 1). The mean 6MWD was 502 [standard deviation (SD) 67] m. There was substantial variability in 6MWD, with a range of 327 to 629 m. The distance walked was significantly greater in the
active group compared with the sedentary group (mean 44 m, 95% confidence interval 14 to 73 m; P < 0.01). Four previous studies reporting predictive models for the 6MWT met the criteria for selection and are detailed in Table 2 [12,13,21,22]. Table 3 [12,13,21,22] presents 6MWDs achieved by the two groups in this study, and the predicted distances according to the four previous studies. The percentage of the predicted distance is also shown in Table 3. It can be seen that the previous published models over or underestimated 6MWD in the present study population, especially for the active group. The four previously published models explained between 17% and 24% of the variance in 6MWD in the current study
Table 1 Characteristics of the study participants. Parameters
Active group n = 46 Mean (SD)
Sedentary group n = 31 Mean (SD)
Age (years) Height (cm) Weight (kg) Body mass index (kg/m2 ) FEV1 (% predicted) FVC (% predicted) FEV1 /FVC (%predicted) Basal heart rate (beats/minute) Basal oxygen saturation (%) Basal systolic blood pressure (mmHg) Basal diastolic blood pressure (mmHg) 6-minute walk distance (m)
66 (6) 161 (6) 67 (10) 26 (3) 94 (14) 94 (12) 99 (7) 78 (12) 97 (3) 133 (18) 82 (11) 520 (64)
67 (6) 161 (5) 68 (9) 26 (3) 94 (11) 92 (11) 102 (7) 77 (11) 97 (1) 125 (19) 79 (13) 476 (64)
0.85 0.82 0.74 0.65 0.95 0.40 0.11 0.83 0.31 0.10 0.21 <0.01*
FEV1 , forced expiratory volume in 1 second; FVC, forced vital capacity; SD, standard deviation. Table 2 Characteristics of the selected studies. Studies
Age range (years)
Regression equation (female)
Enright and Sherrill (1998) Troosters et al. (1999)
173 healthy Americans 22 healthy sedentary English 61 healthy Australians 73 healthy Brazilians
45 to 79
One 6MWT 30-m corridor Two 6MWTs 50-m corridor
6MWD = (2.11 × height) − (2.29 × weight) − (5.78 × age) + 667
6MWD = 218 + (5.14 × height) − (5.32 × age) − (1.80 × weight)
Two 6MWTs 45-m corridor Two 6MWTs 30-m corridor
6MWD = 525 − (2.86 × age) + (2.71 × height) − (6.22 × BMI)
6MWD = 622.461 − (1.846 × age)
Jenkins et al. (2009) Iwama et al. (2009)
50 to 85
45 to 85 24 to 52
6MWT, 6-minute walk test; 6MWD, 6-minute walk distance; BMI, body mass index. Table 3 Comparison of 6-minute walk distance (6MWD) for the active and sedentary groups, and their predicted distances based on previously published equations. Group
Active group (m) Mean (SD) [% predicted]
Sedentary group (m) Mean (SD) [% predicted]
6MWD Enright and Sherrill (1998) Troosters et al. (1999) Jenkins et al. (2009) Iwama et al. (2009)
520 (64) 470 (43)  572 (50)  611 (37)  500 (12) 
476 (64) 470 (40)  571 (42)  609 (32)  501 (12) 
SD, standard deviation.
D. Steffens et al. / Physiotherapy 99 (2013) 21–26
Table 4 Correlation between actual 6-minute walk distance (6MWD) and predicted 6MWD, and correlation when group factor (active/sedentary) was included. Model Enright and Sherrill (1998) Troosters et al. (1999) Jenkins et al. (2009) Iwama et al. (2009)
Adjusted R2 group factor
24% 18% 17% 17%
33% 29% 27% 28%
P-value <0.01 <0.01 <0.01 <0.01
Table 5 Variables retained in the model for predicting 6-minute walk distance in active and sedentary healthy female participants. Variable
Constant Age (years) Groupa Body mass index (kg/m2 )
963.04 −4.71 43.25 −5.07
<0.01 <0.01 <0.01 <0.01
Group: sedentary = 0, active = 1.
population. Adding a participant’s group factor (active or sedentary) to the model significantly increased the predictive value for all four existing equations (Table 4) [12,13,21,22]. Table 5 presents the results for the predictive model developed in the current study. Predictors remaining in the final model included age, body mass index and group (active or sedentary). In this model, allowing for age and body mass index, activity level independently predicted a 43 m greater 6MWD in active participants.
Discussion The primary finding of this study was that activity level was an important independent predictor of 6MWD. The active group walked significantly further (mean 44 m further) than the sedentary group, despite being matched for all other known important predictors, including body mass index (Table 1). Knowledge of activity level added significantly to the predictive value of all four of the previously published equations for the 6MWT that were evaluated [12,13,21,22]. Finally, in the predictive model developed in the current study, activity level remained an important independent predictor, explaining a 43 m difference in 6MWD after allowing for age and body mass index. There is a strong biological rationale why activity level would be an important predictor of 6MWD. It is known that older populations have reduced skeletal mass and strength which, combined with declining cardiorespiratory function, can lead to lower functional status. Regular physical activity can minimise the effects of ageing [27,28], which could explain the superior performance achieved on the 6MWT by the active group. A few previous studies have investigated the influence of activity level on 6MWD [8,9,12,13]. Only one study showed a correlation between reported physical activity and 6MWD in a population of healthy Brazilians . However, this correlation was weak (r = 0.25; P < 0.01), and the female population studied was younger than the subjects in the present study (mean age 35 years) . Two additional studies that
reported physical activity [9,13] and a study that assessed minutes walked in the previous week  found no significant correlation between activity and 6MWD. One possible reason why the study by Casanova et al.  found no relationship between 6MWD and activity level is that more than 50% of the total sample was between 40 and 60 years of age. A ceiling effect, in which the distance walked in 6 minutes is limited by participants reaching maximum stride length, may be more likely in these younger participants, and may have masked the impact of physical activity on 6MWD. In the study by Camarri et al. , the age of the study population was similar to that in the present study, but the criteria to identify activity level differed from the present study. Most of the subjects were classified as undertaking a ‘sufficient level of physical activity’ (66%) as they walked for a mean of 177.8 (SD 146.3) minutes in the previous week . Compared with the present sample, this population appears to be less active, which could explain the non-significant results found in the study. For the present study population, each additional year of age reduced the distance walked by more than 4 m. Furthermore, each additional 1 kg/m2 of body mass index reduced the distance walked by 5 m. It has been proposed that increased weight also increases the workload of walking, resulting in a shorter distance walked . Predicted 6MWD varied greatly between equations [12,13,21,22]. Using the equation from Jenkins et al. , both the active and sedentary groups in the present study performed less well than expected, with the sedentary group achieving less than 80% of the predicted distance, despite the fact that the populations appeared to be similar in terms of age and body mass index, and the same 6MWT protocol was used. The equation published by Enright and Sherrill  better reflected the results achieved by the sedentary group in the present study. However, this equation underestimated the distance walked by the active group (Table 3). It is important to highlight that neither the study by Jenkins et al.  nor the study by Enright and Sherrill  took into account the activity level of the study population when developing the equations. The predictive model developed
D. Steffens et al. / Physiotherapy 99 (2013) 21–26
in the current study and presented in Table 5 includes activity level. However, this new model needs testing in new populations. Health professionals are becoming more involved in primary prevention, particularly in the use of exercise training to reduce risk factors for disease . Accurate prediction of 6MWD from variables of physical activity, age and body mass index may aid in exercise prescription for walking training programmes for the older population. Walking speed for training could be calculated as a percentage of the predicted 6MWT speed, rather than needing to perform a 6MWT; this may be useful for individual exercise prescription when managing large exercise groups. The inclusion of activity level in the predictive model may lead to more accurate exercise prescription. There were a number of limitations in the present study. Only female participants were included as this was a convenience sample. There is no clear rationale why the influence of activity level should be different between the genders, but this requires future investigation before the findings can be generalised to men. There was no standardised questionnaire to assess activity levels; however, those in the active group regularly attended supervised exercise classes and so were observed to be active, whereas those in the sedentary group reported that they had not maintained any type of regular physical activity for the previous 6 months and so were observed to be relatively more sedentary. There was no assessor blinding; however, standardised instructions and encouragement were used for all 6MWTs which would have helped to reduce assessor bias.
Conclusion In summary, this study found that 6MWD differed significantly between older active females and matched sedentary females. The three variables in the final model to predict 6MWD were physical activity, age and body mass index. Previously published equations under or overestimated values for both the active and sedentary groups. Knowledge of activity level increased the predictive ability of all four previously published equations. This study demonstrates the importance of considering the level of physical activity when predicting 6MWD in older women.
Acknowledgements The study was carried out by the University of Santa Cruz do Sul - UNISC. The authors wish to thank Zelia Coletti Ohlweiler, Caroline Bottlender Machado, Priscila Zingler, Camila Hammes, Juliana Franceschette and Mr. João Dutra for assistance with the study. Ethical approval: This study was approved by CEP-UNISC (Ethics in Research Committee) of the University of Santa
Cruz do Sul - UNISC, under the protocol number 1706/06. Funding: FIPE (Research Incentive Fund), Federal University of Rio Grande do Sul - UFRGS. Conﬂict of interest: None declared.
References  Poh H, Eastwood PR, Ho KT, Cecins NM, Jenkins SC. Six-minute walk distance in healthy Singaporean adults cannot be predicted using reference equations derived from Caucasian populations. Respirology 2006;11:211–6.  Salzman SH. The 6-min walk test: clinical and research role, technique, coding, and reimbursement. Chest 2009;135:1345–52.  Bautmans I, Lambert M, Mets T. The six-minute walk test in community dwelling elderly: influence of health status. BMC Geriatr 2004;4:6.  Alameri HF, Sanai FM, Al Dukhayil M, Azzam NA, Al-Swat KA, Hersi AS, et al. Six minute walk test to assess functional capacity in chronic liver disease patients. World J Gastroenterol 2007;13:3996–4001.  Araújo CO, Makdisse MRP, Peres PAT, Tebexreni AS, Ramos LR, Matsushita AM, et al. Different patterns for the 6-minute walk test to measure exercise ability in elderly with and without clinically evident cardiopathy. Arq Bras Cardiol 2006;86:198–205.  Arslan S, Erol MK, Gundogdu F, Sevimli S, Aksakal E, Senocak H, et al. Prognostic value of 6-minute walk test in stable outpatients with heart failure. Tex Heart Inst J 2007;34:166–9.  Beriault K, Carpentier AC, Gagnon C, Menard J, Baillargeon JP, Ardilouze JL, et al. Reproducibility of the 6-minute walk test in obese adults. Int J Sports Med 2009;30:725–7.  Camarri B, Eastwood PR, Cecins NM, Thompson PJ, Jenkins S. Six minute walk distance in healthy subjects aged 55–75 years. Respir Med 2006;100:658–65.  Casanova C, Celli BR, Barria P, Casas A, Cote C, de Torres JP, et al. The 6-min walk distance in healthy subjects: reference standards from seven countries. Eur Respir J 2011;37:150–6.  Dourado VZ, Antunes LCD, Tanni SE, de Paiva SAR, Padovani CR, Godoy I. Relationship of upper-limb and thoracic muscle strength to 6-min walk distance in COPD patients. Chest 2006;129:551–7.  Enright PL, McBurnie MA, Bittner V, Tracy RP, McNamara R, Arnold A, et al. The 6-min walk test – a quick measure of functional status in elderly adults. Chest 2003;123:387–98.  Iwama AM, Andrade GN, Shima P, Tanni SE, Godoy I, Dourado VZ. The six-minute walk test and body weight-walk distance product in healthy Brazilian subjects. Braz J Med Biol Res 2009;42:1080–5.  Jenkins S, Cecins N, Camarri B, Williams C, Thompson P, Eastwood P. Regression equations to predict 6-minute walk distance in middle-aged and elderly adults. Physiother Theory Pract 2009;25:516–22.  Maher CA, Williams MT, Olds TS. The six-minute walk test for children with cerebral palsy. Int J Rehabil Res 2008;31:185–8.  Montes J, McDermott MP, Martens WB, Dunaway S, Glanzman AM, Riley S, et al. Six-minute walk test demonstrates motor fatigue in spinal muscular atrophy. Neurology 2010;74:833–8.  Pires SR, Oliveira AC, Parreira VF, Britto RR. Six-minute walk test at different ages and body mass index. Rev Bras Fisioter 2007;11:147–51.  Priesnitz CV, Rodrigues GH, Stumpf CD, Viapiana G, Cabral CP, Stein RT, et al. Reference values for the 6-min walk test in healthy children aged 6–12 years. Pediatr Pulmonol 2009;44:1174–9.  Rubim VSM, Drumond Neto C, Romeo JLM, Montera MW. Prognostic value of the six-minute walk test in heart failure. Arq Bras Cardiol 2006;86:120–5.  Wanderley FAC, Oliveira J, Mota J, Carvalho MJ. Six-minute walk distance (6MWD) is associated with body fat, systolic blood pressure, and rate-pressure product in community dwelling elderly subjects. Arch Gerontol Geriatr 2011;52:206–10.
D. Steffens et al. / Physiotherapy 99 (2013) 21–26
 Chetta A, Zanini A, Pisi G, Aiello M, Tzani P, Neri M, et al. Reference values for the 6-min walk test in healthy subjects 20–50 years old. Respir Med 2006;100:1573–8.  Enright PL, Sherrill DL. Reference equations for the six-minute walk in healthy adults. Am J Respir Crit Care Med 1998;158:1384–7.  Troosters T, Gosselink R, Decramer M. Six minute walk distance in healthy elderly subjects. Eur Respir J 1999;14:270–4.  Dourado VZ. Reference equations for the 6-minute walk test in healthy individuals. Arq Bras Cardiol 2011;96:128–38.  Miller M, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. Standardization of spirometry. Eur Respir J 2002;26:319–38.  Pereira C, Barreto S, Simöes J, Pereira F, Gerstler J, Nakatani J. Reference values for spirometry in Brazilian adults. J Bras Pneumol 1992;18:10–22.
 Crapo RO, Casaburi R, Coates AL, Enright PL, MacIntyre NR, McKay RT, et al. ATS statement: guidelines for the sixminute walk test. Am J Respir Crit Care Med 2002;166: 111–7.  Mechling H, Netz Y. Aging and inactivity – capitalizing on the protective effect of planned physical activity in old age. Eur Rev Aging Phys Act 2009;6:89–97.  Tuna HD, Edeer AO, Malkoc M, Aksakoglu G. Effect of age and physical activity level on functional fitness in older adults. Eur Rev Aging Phys Act 2009;6:99–106.  Teixeira-Lemos E, Nunes S, Teixeira F, Reis F. Regular physical exercise training assists in preventing type 2 diabetes development: focus on its antioxidant and anti-inflammatory properties. Cardiovasc Diabetol 2011;10:12.
Available online at www.sciencedirect.com