The relationship between pedometer-determined ambulatory activity and body composition variables

Abstract

OBJECTIVE: To examine the relationship between pedometer-determined ambulatory activity (steps/day) and body composition variables body mass index (BMI) and percentage body fat).

DESIGN: Secondary analysis of a cross-sectional objective activity monitoring study for up to 21 consecutive days.

SUBJECTS: A total of 109 apparently healthy adults (eight African American males, 23 African-American females, 33 Caucasian males, 45 Caucasian females), age 44.9±15.8 y, BMI=26.9±5.1 kg/m2.

MEASUREMENTS: Pedometer-assessed ambulatory activity (steps/day), height and weight, and percentage body fat by bioelectrical impedance.

RESULTS: Analyzed as both a continuous and a categorical variable (determined using 25th and 75th percentiles for distribution for steps/day), ambulatory activity was consistently related to body composition variables. Steps/day was inversely correlated with BMI and percentage body fat (r=−0.30, and r=−0.27, respectively, both P<0.01). The consistency of the relationship was also evident when examined using accepted BMI cut-off points for normal-weight, overweight, and obese categories.

CONCLUSIONS: Individuals in this small sample with values greater than approximately 9000 steps/day are more frequently classified as normal weight for height. Individuals with values less than approximately 5000 steps/day are more frequently classified as obese. These findings require further corroborative investigation but provide preliminary cutoff points for identifying individuals at risk who may benefit from appropriate physical activity intervention.

Introduction

The dramatic increase in overweight and obesity over the last few decades is considered to be of epidemic proportions in many countries around the world.1 The most recent statistics reported from NHANES III indicate that close to 55% of US adults are either overweight or obese.2 Paradoxically, participation in leisure-time physical activity has remained relatively stbl3 This finding suggests that traditional measurement approaches focused on leisure-time physical activity may be inadequately sensitive to changes in incidental, ubiquitous daily activity. As testament to the considered role of physical inactivity in obesity trends, the American College of Sports Medicine convened a February 1999 Roundtable of experts to examine the scientific evidence and deliberate pressing issues.4 One of the issues identified for physical activity assessment was an inability to quantify trends in physical activity using currently available data.5

There is increasing interest in objective monitoring of daily physical activity using electronic motion sensors, including accelerometers6,7 and pedometers.8,9,10 This shift was apparent in the June 2000 Research Quarterly for Exercise and Sport Measurement of Physical Activity Special Issue11 and again in the September 2000 Medicine and Science in Sports and Exercise Measurement of Moderate Physical Activity: Advances in Assessment Techniques Issue.12 In both these special journal supplements most articles discussed objective monitoring to some extent. The science and practice of objective monitoring of physical activity is still in its infancy. Researchers are slowly beginning to acknowledge that, in terms of practicality, however, pedometers currently offer the best solution for a low-cost, objective monitoring tool.13,14,15,16 Unfortunately, the cost of accelerometers ($50–400 per unit) is prohibitive for most practical and larger-scale applications including surveillance, screening, program evaluation and intervention.9

The new generation of electronic pedometers are battery-operated devices containing a spring-suspended, horizontal lever arm that moves up and down in response to vertical accelerations of the trunk that occur during ambulatory movement (eg walking, running) when worn at the waist. This movement opens and closes an electrical circuit; each cycle is registered as a step and the accumulated total is displayed digitally. Pedometers are particularly sensitive to walking behavior, or ambulatory activity.15,16,17,18,19,20 Walking is the most common manifestation of daily activity; engagement in other types of physical activities (ie structured and/or vigorous) is infrequent and intermittent.21,22 Worn over the course of the day, pedometers can potentially capture intentional as well as incidental walking behaviors.23 Pedometers are not designed to discriminate pattern or intensity of activity; their output is simply steps taken over the specified monitoring period. In a series of studies Bassett and colleagues23,24,25 have shown that pedometers are most accurate at measuring steps taken (within 1% of actual steps taken), less accurate when the output is manipulated to estimate distance traveled, and even less accurate when used to extrapolate to energy expenditure. The findings of Hendelman et al26 concur that the most accurate pedometer output is steps taken.

To be useful for assessment purposes (and for examining obesity-related trends), however, evidence of convergent validity of this objective monitoring tool must be presented. Specifically, pedometer output must be acceptably associated with indicators of body composition as well as other health consequences of physical activity. The purpose of this study is to describe the relationship between an objective measure of walking, assessed as steps/day using a pedometer, and two body composition variables, body mass index (BMI) and percentage body fat, among a group of African-American and Caucasian adults living in South Carolina. Additional purposes include: (1) to describe preliminary evidence of specific steps/day ‘cut-off points’ useful for identifying sedentary individuals who might benefit from an appropriate physical activity intervention; and (2) to provide preliminary evidence toward a minimum recommended level of steps/day associated with body composition benefits.

Methods

Subjects

Subjects were recruited by word of mouth and from posted announcements in the University of South Carolina and Columbia communities. Inclusion criteria were: (1) apparently healthy; (2) able to walk on a treadmill for a submaximal exercise test; and (3) willingness to comply with the research protocol. One hundred and thirty-nine subjects enrolled in the study. Subsequently, 10 dropped out due to self-reported time constraints and three were not compliant with the study protocol. Among the remaining 126 subjects, 49 were male (11 African-American and 38 Caucasian) and 77 were female (27 African-American and 50 Caucasian). Volunteers ranged in age from 17 to 79 y. All subjects read and signed a written informed consent form approved by the University of South Carolina Institutional Review Board.

Of the total sample, one subject had missing BMI data, nine had missing percentage body fat data, and seven had missing pedometer data. Analyses were therefore conducted on data obtained from 109 subjects (eight African-American males, 23 African-American females, 33 Caucasian males, 45 Caucasian females) with complete body composition (BMI and fat percentage) and pedometer-assessed ambulatory activity (steps/day) data. Nineteen or more days of pedometer data were collected from 96 of the subjects. The average number of days of pedometer data collected on the remaining subjects was 14 (range 4–18).

Descriptive data for age, BMI, percentage body fat and steps/day (calculated for the days collected) for each race–gender group and for the total sample are presented in Table 1. African-American subjects were significantly younger than Caucasian subjects regardless of gender (F=7.01, P<0.0002). Caucasian females had a significantly lower BMI than African-American males or females; Caucasian males did not differ significantly from either Caucasian females or African-American subjects (F=4.5, P=0.005). Females of both races had significantly higher body fat percentage compared with males of both races (F=15.9, P<0.0001). Steps/day did not different significantly between any race-gender groups (F=1.57, P=0.20).

Table 1 Descriptive characteristics for each race gender group and the total sample

Study protocol

Subjects were enrolled in the study for 21 consecutive days during which time they wore a pedometer (Yamax Digiwalker, Model DW-500, Accusplit, CA) and completed a daily physical activity log as direct measures of physical activity. Subjects also completed physical activity questionnaires, physical fitness and anthropometric assessments (height, weight, and percent body fat) as indirect measures of physical activity. To determine participation in vigorous activities, subjects were asked, ‘Last week, did you do vigorous activities continuously for at least 10 minutes that caused large increases in breathing or heart rate, such as running, swimming, aerobics, fast bicycling, competitive sports or heavy yard work?’ If subjects answered positively, they were queried about the total minutes spent doing vigorous activities per day. This report focuses primarily on the body composition and pedometer data, using the vigorous activities question only to classify individuals with regard to participation in vigorous activities (minutes of vigorous activities >10, n=61) or not (minutes of vigorous activities <10, n=48). The purpose of this procedure was to allow for testing of differences in steps/day between individuals so classified.

Height was measured without shoes using a wall-mounted tape measure and body mass was assessed using a digital scale (Seca, Shorr Products, Model 770, Onley, MD). Body mass index (BMI) was computed as weight in kg/height in meters squared. Percentage body fat was determined using an RJL bioelectrical impedance analyzer (Model 101, Clinton Twp., MI). Bioelectrical impedance is an acceptable adjunct to anthropometry for assessing body composition in field studies.27 Subjects were assessed over three consecutive trials conducted in the morning after a 12 h fast and within 30 min of voiding. Sex- and race-specific equations were used to predict percentage body fat from the average resistance and reactance in ohms.28,29

Subjects were instructed to wear the pedometer and to record the total number of steps taken each day in a physical activity log each night prior to retiring. If the pedometer was not worn on a day for any reason (eg forgot, chose not to) subjects were asked to leave the activity log blank on that day, indicating non-compliance. The pedometer was worn attached to the waistband of their clothing during waking hours, except when bathing or swimming. Subjects were encouraged not to alter their usual physical activity habits during the 3 week study. Subjects were contacted by telephone or face-to-face once a week during the 3 week study to check for compliance (ie ‘Have you been wearing the pedometer and recording in the log?’) with the protocol. At the end of the study, subjects returned the pedometers and the physical activity logs to the study center where the logs were checked by study staff for completeness and comprehension.

Data treatment and statistical analysis

Physical activity (steps/day) was examined as a continuous variable (range 1647–14 529) using correlation analysis (Pearson product–moment correlation coefficients) to quantify the relationship between steps/day and BMI or percentage body fat, after exploring for evidence of confounding. In both cases, regression analysis was used to predict the line of best fit against the data distribution. Physical activity also was examined as a categorical variable defined as low, moderate and high according to tertiles of steps/day using the 25th and 75th percentiles for distribution. Set categories for BMI were applied to represent obesity (BMI≥30), overweight (25≤BMI<30) and normal weight (BMI<25) status;30 there was no one in this study classified as underweight (BMI<18.5). Effect size of steps/day was calculated as the difference between means of the obese and normal weight groups, divided by the standard deviation for the total sample. Calculating effect size is a recommended technique for presenting the ‘meaningfulness’ of differences between groups.31 An effect size of 0.2 is considered small, 0.5 is moderate, and 0.8 is large.32

Comparison of mean steps/day between reported participation in vigorous activities (yes or no) was determined using independent t-tests, having satisfied relevant statistical assumptions. Chi-square analyses were used to examine frequency distributions of gender and race between pedometer-determined activity categories, and the frequency of BMI category classification between reported participation in vigorous activities. The relationship between steps/day and estimates of body fatness was further examined using one-way analysis of variance, statistical assumptions having been satisfied. In the first analysis, tertile of steps/day was the grouping variable and the dependent variables were age, BMI and percentage body fat. In the second analysis, BMI category was the grouping variable and dependent variables were age, percent body fat, and steps/day. Post-hoc analyses (Student–Newman–Keuls test) were used to compare differences between categories. Statistical analyses were conducted using SAS Version 6.12 Significance was set at an alpha level of P<0.05.

Results

Mean steps/day did not differ statistically between individuals reporting participation in vigorous activities or not (7036±2893 vs 7794±3285 steps/day, respectively). Analyzed as a continuous variable, steps/day was inversely correlated with both BMI (r=−0.30) and percentage body fat (r=−0.27), P<0.01. Age was not a confounder for these associations in this sample. Figure 1 presents the predicted regression line against the spread of steps/day vs BMI. A similar relationship is presented in Figure 2 for steps/day vs percentage body fat.

Figure 1
figure1

Regression of BMI and pedometer-determined steps/day.

Figure 2
figure2

Regression of percent body fat and pedometer-determined steps/day.

Tertiles for steps/day as a measure of physical activity were defined as low activity (below the 25th percentile of distribution, or ≤5267 steps/day), moderate activity (between the 25th and 75th percentiles of distribution, or 5268–9356 steps/day) and high activity (above the 75th percentile of distribution, or ≥9357 steps/day). Table 2 presents the distribution of BMI category by pedometer-determined activity tertile. Forty-one percent of individuals in the lowest activity tertile were also classified as obese, compared with 11% of those in the highest activity tertile. Similarly, 57% of subjects in the highest activity tertile were also classified as normal weight for height compared with 30% of those in the lowest activity tertile. There was no difference in BMI category distribution by reported participation in vigorous activities (data not shown).

Table 2 Subject classification by tertile of pedometer activity and BMI category

Table 3 presents a summary of the comparison of variables between pedometer-determined activity categories. There were no significant differences in frequency of race or gender across activity categories as determined by χ2 analyses. There were no significant differences in mean age across activity categories (F=1.49, P=0.23). Both BMI and percentage body fat were significantly different across activity categories (F=5.04, P=0.01, and F=3.90, P=0.02). Post-hoc analyses revealed significant differences in BMI between the low activity category vs the moderate and high activity categories. There were no differences in BMI between the moderate and high activity categories. Percentage body fat was significantly higher in the low activity category compared with the high activity category. There was no difference in percentage body fat between the moderate activity category and the low or high activity categories.

Table 3 Comparison of variables between categories of pedometer-assessed ambulatory activity

The comparisons of mean age, percentage body fat, and pedometer steps/day by BMI categories are presented in Table 4. Women and African-Americans (compared with men and Caucasians) were more frequently classified as obese according to this partitioning strategy. As expected, percentage body fat was significantly different across BMI categories (F=38.10, P<0.001) with percentage body fat increasing in higher BMI categories. There was a significant difference across BMI categories for steps/day with subjects classified as normal weight taking significantly more steps/ day than those classified as obese. There was no significant difference in steps/day between overweight subjects and those classified as normal weight or obese by BMI categories. Calculated effect size between the normal weight and obese BMI categories for steps/day was 0.75.

Table 4 Comparison of variables between categories of BMI-defined body composition

Discussion

This is the first study to examine the relationship between electronic pedometer-assessed ambulatory activity expressed as steps/day and two body composition variables in an adult population. Although Tryon et al33 previously used pedometers for a similar purpose, those authors expressed the pedometer output as a rate of distance traveled per hour. The authors concluded that pedometer-determined activity was inversely related to percentage of overweight in 127 women aged 19–55 y ranging from 14 to 99% overweight (r=−0.217, P<0.02).

We resolved not to convert raw pedometer outputs to an estimate of distance traveled during the day. The process of manipulating the raw step data in this manner introduces possible errors due to individual differences in stride length.8 The conversion of steps taken to distance walked is based on the assumption that all strides taken over the course of the day are similar in length. This makes little sense in a field study; steps taken while milling about the kitchen engaged in meal preparation probably differ in stride length from those taken while running an errand, walking the dog, dancing, or running for exercise. In a laboratory study of treadmill walking, Bassett et al24 demonstrated that error in estimating distance from pedometer stepping at fast speed was due to stride lengthening rather than to step miscounting, and while the pedometer is designed to be most sensitive to the vertical movements of walking, other movements throughout the day like hopping, kneeling and bending, will also cause a ‘step’ to register; this error will only be compounded if multiplied by an estimated stride length.34 Despite our different approaches to presenting pedometer output, however, our conclusions concur with those presented by Tryon et al;33 there was a distinct relationship, in the expected direction, between pedometer data and body composition variables.

This relationship is consistently reported in other populations as well. McClung et al35 reported that increased BMI was associated with reduced pedometer-assessed ambulatory activity (steps/day), adjusted for age and gender (P=0.05) in 209 subjects, including 151 subjects with hip or knee joint replacements. The mean pedometer values recorded for the present sample (7370±3080 steps/day) are similar to those reported by McClung et al35 for their subsample of 58 normal healthy adults aged 22–82 y (7781±2807 steps/day), although the brand of pedometer was not reported in that study. Rowlands et al36 reported that both the Yamax pedometer and the Tritrac accelerometer activity outputs were negatively correlated (both showing r=−0.42, P<0.05) with body fat (determined using skinfold measurements) in 8 to 12-y-old children. The present study confirms a similar inverse relationship (on the order of r=−0.30) between pedometer-assessed physical activity (expressed as steps/day) and BMI in an adult sample. Further, we document a similar relationship with percent body fat. The calculated effect size of the difference in steps/day taken by normal weight individuals compared with obese individuals (classified according to BMI) in this study can be considered moderate to large32 representing a meaningful difference between groups.31

Examination of the results from self-reported physical activity questionnaire validation studies37 shows that statistically significant negative correlations between self-reported physical activity and BMI range from −0.1238 to−0.13.39 Likewise, statistically significant negative correlations between self-reported physical activity and percent body fat range from −0.1340 to −0.51.41 The correlations we report agree favorably with these values.

Information about associations between accelerometers and body composition variables is more limited and variable depending on how the accelerometer data is presented (eg in the case of the Caltrac accelerometer, data has been presented as counts, kcal/day, and MET/min/day). Rutter42 reported a significant negative association between Caltrac accelerometer counts (an indicator of raw movement) and BMI (r=−0.47, P<0.01) in 39 young university students monitored for 4–6 days, but no correlation between Caltrac counts and any measures of body fat for the first 3-day period. The Survey of Activity, Fitness, and Exercise (SAFE) study reported a significantly positive association of r=0.52 between Caltrac accelerometer data (expressed as kcal/day) and percentage body fat determined by hydrostatic weighing,43 suggesting that individuals with greater adiposity expended more energy, but not necessarily due to increased movement. Expressed as MET minutes/day (summed minutes/day multiplied by the MET intensity level), the correlation between Caltrac accelerometer data and percent body fat determined by hydrostatic weighing was non-significant for either men (r=−0.10) or women (r=0.06).41 The body composition variables we report are similar to the SAFE data,41 a finding that increases our confidence in the validity of our data. There does not appear to be any other studies relating other brands of accelerometers to body composition variables in adults at this time.

Identification of risk groups, including obese and sedentary individuals, is a priority for appropriate targeting of public health approaches to increase physical activity. A specific ‘sedentary lifestyle index’44 would be helpful to identify those individuals who are not active in their daily lives and who would most likely benefit from an appropriate intervention.45 Based on up to 21 days of monitoring 109 adults in the present study, individuals with pedometer values lower than approximately 5000 steps/day (representing the 25th percentile of distribution of this sample's data) were more frequently classified as obese than either overweight or normal weight for height (see Table 2). These individuals could logically benefit from a physical activity intervention designed to increase steps/day.46 It is important to emphasize here, however, that not all individuals with values of approximately 5000 steps/day are obese; the mean value for obese individuals is closer to 6000 steps/day (see Table 4). Accordingly, pedometer values greater than approximately 9000 steps/day (representing the 75th percentile of this sample's data) were more frequently classified as normal weight by BMI category (see Table 2).

Japanese health promotion efforts recommend a goal of 10 000 steps/day.47,48 A review of the published literature49 indicates that this value seems a reasonable estimate for younger and/or otherwise healthy individuals, but there is currently little empirical evidence to support such a threshold. According to Hatano,47 walking 10 000 steps is approximately equivalent to energy expenditures of between 300 and 400 kcal/day (depending on walking speed and body size). This is substantially greater than the 150 kcal/day that accumulated evidence suggests is associated with health benefits in the US Surgeon General's Report on Physical Activity and Health50 and the average steps/day accumulated by those classified as normal weight according to their BMI (see Table 4). To be useful, a health promotion slogan for increasing physical activity using pedometers must be evidence-based.

The limitations of this study include its cross-sectional design of a convenience sample of adult volunteers. Such a design limits conclusions about causal inference and generalizability. In addition, pedometers do not distinguish speed or velocity of movement. Some activities will be missed or underestimated (eg swimming, bicycling). It has been suggested that a combination of pedometers and self-report methods (to eliminate or at least account for participation in vigorous and/or non-ambulatory activities) may extend the utility of a simple pedometer for physical activity assessment purposes.51 In the present study, however, pedometer-assessed steps/day did not differ between groups classified according to reported participation in vigorous activity or not. Further, there was no difference in BMI category distribution by reported participation in vigorous activities. Therefore, the differences observed in body composition variables related to steps/day could not be explained solely by reported participation in vigorous activities. Although the simple and inexpensive pedometer is admittedly not the best measure of overall physical activity, it appears to be useful for objectively describing daily incidental and intentional ambulatory activity.

A strength of this study is the extended length of monitoring frame; the ability to capture habitual activity is likely to be increased with more days of monitoring.52 Gretebeck and Montoye53 reported that 5–6 days (including weekend days) of pedometer data were needed to accurately (with less than 5% error) describe the physical activity patterns of young males purposefully recruited for exhibiting varied physical activity pursuits. The monitoring frame in the present study extends well beyond this minimal recommendation based on a sample likely to exhibit both extreme intra- and interindividual variability. Although it was not the purported objective of this study, it is important to determine the minimum monitoring frame necessary to achieve a reliable estimate of ambulatory activities; twenty-one days is not practical for clinical applications.

In summary, the findings of this study indicate that, regardless of analysis approach, there is a distinct inverse relationship between pedometer-assessed ambulatory activity and two body composition variables. The relationship is both statistically significant and meaningful. Further, similar relationships are consistently reported in other populations.33,35,36 The findings provide preliminary support of approximately >9000 steps/day associated with body composition benefits, and approximately <5000 indicative of an index for sedentarism related to unhealthy body composition. Caution is recommended when applying these cut-off points at this time. Further evidence of a confirmatory nature is required before any threshold values can be accepted as established guidelines for either health-related pedometer goals or for a ‘sedentary lifestyle index’. Given the practical appeal of the pedometer,9,46 we concur with others54 that the effectiveness of behavioral and environmental lifestyle physical activity interventions needs to be assessed using such simple and inexpensive approaches.

References

  1. 1

    Hill JO, Melanson EL . Overview of the determinants of overweight and obesity: current evidence and research issues Med Sci Sports Exerc 1999 31: (Suppl): S515–S521.

    Article  PubMed  Google Scholar 

  2. 2

    Flegal KM, Carrol MD, Kuczmarski RJ, Johnson CL . Overweight and obesity in the United States: prevalence and trends, 1960–1994 Int J Obes Relat Metab Disord 1998 22: 39–47.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  3. 3

    US Department of Health and Human Services . Physical activity and health: a report of the Surgeon General US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Promotion: Atlanta, GA 1996

  4. 4

    American College of Sports Medicine . American College of Sports Medicine Roundtable: physical activity in the prevention and treatment of obesity and its comorbidities Med Sci Sports Exerc 1999 31: (Suppl): S497–S667.

    Google Scholar 

  5. 5

    Pratt M, Macera CA, Blanton C . Levels of physical activity and inactivity in children and adults in the United States: current evidence and research issues Med Sci Sports Exerc 1999 31: (Suppl): S526–S533.

    Article  PubMed  Google Scholar 

  6. 6

    Westerterp KR . Assessment of physical activity level in relation to obesity: Current evidence and research issues Med Sci Sports Exerc 1999 31: (Suppl): S522–525.

    Article  PubMed  Google Scholar 

  7. 7

    Westerterp KR . Physical activity assessment with accelerometers Int J Obes Relat Metab Disord 1999 23: (Suppl 3): S45–S49.

    Article  Google Scholar 

  8. 8

    Rowlands AV, Eston RG, Ingledew EK . Measurement of physical activity in children with particular reference to the use of heart rate and pedometry Sports Med 1997 24: 258–272.

    CAS  Article  Google Scholar 

  9. 9

    Tudor-Locke CE, Myers AM . Challenges and opportunities for measuring physical activity in sedentary adults Sports Med 2001 31: 91–100.

    CAS  Article  Google Scholar 

  10. 10

    Sequeira MM, Rickenbach M, Wietlisbach V, Tullen B, Schutz Y . Physical activity assessment using a pedometer and its comparison with a questionnaire in a large population survey Am J Epidemiol 1995 142: 989–999.

    CAS  Article  Google Scholar 

  11. 11

    Mahar MT, Ainsworth BE . Measurement of physical activity: the Cooper Institute conference series Res Q Exerc Sport 2000 71: (Suppl 2): S1–S158.

    Article  Google Scholar 

  12. 12

    Raven PB, Perrel AW, Robertson DM, Kohl HW, Kibbe DL, Trost S . Measurement of moderate physical activity: advances in Assessment techniques Med Sci Sports Exerc 2000 33: (Suppl): S439–S516.

    Google Scholar 

  13. 13

    Welk GJ, Differding JA, Thompson RW, Blair SN, Dziura J, Hart P . The utility of the Digi-walker step counter to assess daily physical activity patterns Med Sci Sports Exerc 2000 32: (Suppl): S481–S488.

    Article  PubMed  Google Scholar 

  14. 14

    Welk GJ, Corbin CB, Dale D . Measurement issues in the assessment of physical activity in children Res Q Exerc Sport 2000 71: (Suppl 2): 59–73.

    Article  Google Scholar 

  15. 15

    Bassett DR . Validity and reliability issues in objective monitoring of physical activity Res Q Exerc Sport 2000 71: 30–36.

    Article  PubMed  Google Scholar 

  16. 16

    Freedson PS, Miller K . Objective monitoring of physical activity using motion sensors and heart rate Res Q Exerc Sport 2000 71: 21–29.

    Article  PubMed  Google Scholar 

  17. 17

    Bassey EJ, Dallosso HM, Fentem PH, Irving JM, Patrick JM . Validation of a simple walking mechanical accelerometer (pedometer) for the estimation of walking activity Eur J Appl Physiol 1987 56: 323–330.

    CAS  Article  Google Scholar 

  18. 18

    Kashiwazaki H, Inaoka T, Suzuki T, Kondo Y . Correlations of pedometer readings with energy expenditure in workers during free-living daily activities Eur J Appl Physiol 1986 54: 585–590.

    CAS  Article  Google Scholar 

  19. 19

    Saris WHM, Binkhorst RA . The use of pedometer and actometer in studying daily physical activity in man. Part I: reliability of pedometer and actometer Eur J Appl Physiol Occup Physiol 1977 37: 219–228.

    CAS  Article  Google Scholar 

  20. 20

    Shephard RJ . Assessment of physical activity and energy needs Am J Clin Nutr 1989 50: (Suppl): 1195–1200.

    Article  PubMed  Google Scholar 

  21. 21

    Blair SN . How to assess exercise habits and physical fitness. In: Matarazzo JD, Weiss SM, Herd JA, Miller NE, (eds). Behavioral health: a handbook of health enhancement and disease prevention Wiley: New York 1984 424–447.

    Google Scholar 

  22. 22

    Masse LC, Ainsworth BE, Tortolero S, Levin S, Fulton JE, Henderson KA, Mayo K . Measuring physical activity in midlife, older and minority women: issues from an expert panel J Women Health 1998 7: 57–67.

    CAS  Article  Google Scholar 

  23. 23

    Bassett DR, Cureton AL, Ainsworth BE . Measurement of daily walking distance—questionnaire versus pedometer Med Sci Sports Exerc 2000 32: 1018–1023.

    Article  Google Scholar 

  24. 24

    Bassett DR, Ainsworth BE, Leggett SR, Mathien CA, Main JA, Hunter DC, Duncan GE . Accuracy of five electronic pedometers for measuring distance walked Med Sci Sports Exerc 1996 28: 1071–1077.

    Article  Google Scholar 

  25. 25

    Bassett DR, Ainsworth BE, Swartz AM, Stratch SJ, O'Brien WL, King GA . Validity of four motion sensors in measuring moderate intensity physical activity Med Sci Sports Exerc 2000 32: (Suppl): S471–S480.

    Article  PubMed  Google Scholar 

  26. 26

    Hendelman D, Miller K, Baggett C, Debold E, Freedson P . Validity of accelerometry for the assessment of moderate intensity physical activity in the field Med Sci Sports Exerc 2000 32: (Suppl): S442–S450.

    Article  PubMed  Google Scholar 

  27. 27

    Baumgartner RN . Electrical impedance and total body electrical conductivity. In: Roche AF, Heymsfiled SB, Lohman TG (eds). Human body composition Human Kinetics: Champaign, IL 1996

    Google Scholar 

  28. 28

    Segal KR, Van Loan M, Fitzgerald PI, Hogdon JA, Van Itallie TB . Lean body mass estimation by bioelectrical impedance analysis: a four-site cross validation study Am J Clin Nutr 1988 47: 7–14.

    CAS  Article  Google Scholar 

  29. 29

    Ainsworth BE, Stolarczyk LM, Heyward VH, Berry CB, Irwin ML, Mussulman LM . Predictive accuracy of bioimpedance in estimating fat-free mass of African-American women Med Sci Sports Exerc 1997 29: 781–787.

    CAS  Article  PubMed  Google Scholar 

  30. 30

    National Institutes of Health . Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report Obes Res 1998 6: (Suppl 2): 51S–209S.

    Google Scholar 

  31. 31

    Thomas JR, Salazar W, Landers DM . What is missing in P&lt;0.05? Effect size Res Q Exerc Sport 1991 62: 344–348.

    CAS  Article  PubMed  Google Scholar 

  32. 32

    Cohen J . Statistical power analysis for the behavioral sciences, 3rd edn. Academic Press: New York 1988

  33. 33

    Tryon WW, Goldberg JL, Morrison DF . Activity decreases as percentage of overweight increases Int J Obes 1992 16: 591–595.

    CAS  Google Scholar 

  34. 34

    Tyron WW . Measurement of human activity. In: Tyron WW (ed). Behavioral assessment in behavioral medicine Springer: New York 1985

    Google Scholar 

  35. 35

    McClung CD, Zahiri CA, Higa JK, Amstutz HC, Schmalzried TP . Relationship between body mass index and activity in hip or knee arthorplasty patients J Orthoped Res 2000 18: 35–39.

    CAS  Article  Google Scholar 

  36. 36

    Rowlands AV, Eston RG, Ingledew DK . Relationship between activity levels, aerobic fitness, and body fat in 8- to 10-yr-old children J Appl Physiol 1999 86: 1428–1435.

    CAS  Article  Google Scholar 

  37. 37

    Kriska AM, Casperson CJ . A collection of physical activity questionnaires for health-related research Med Sci Sports Exerc 1997 29: (Suppl 6): S1–S205.

    Google Scholar 

  38. 38

    Lakka TA, Nyyssonene K, Salonen JT . Higher levels of conditioning leisure time physical activity are associated with reduced levels of stored iron in Finnish men Am J Epidemiol 1994 140: 148–160.

    CAS  Article  PubMed  Google Scholar 

  39. 39

    Washburn RA, Smith KW, Goldfield SR, McKinlay JR . Reliability and physiologic correlates of the Harvard Alumni Activity Survey in a general population J Clin Epidemiol 1991 44: 1319–1326.

    CAS  Article  PubMed  Google Scholar 

  40. 40

    Bouchard C, Tremblay A, Leblanc C, Lortie G, Savard R, Theriault G . A method to assess energy expenditure in children and adults Am J Clin Nutr 1983 37: 461–467.

    CAS  Article  Google Scholar 

  41. 41

    Richardson MT, Ainsworth BE, Wu H, Jacobs DR, Leon AS . Ability of the Atherosclerosis Risk in Communities (ARIC)/Baecke Questionnaire to assess leisure-time physical activity Int J Epidemiol 1995 24: 685–693.

    CAS  Article  PubMed  Google Scholar 

  42. 42

    Rutter S . Comparison of energy expenditure in normal-weight and overweight women using the Caltrac Personal Activity Computer Int J Eat Disord 1994 15: 37–42.

    CAS  Article  PubMed  Google Scholar 

  43. 43

    Jacobs DR, Ainsworth BE, Hartman TJ, Leon AS . A simultaneous evaluation of 10 commonly used physical activity questionaires Med Sci Sports Exerc 1992 25: 81–91.

    Article  Google Scholar 

  44. 44

    Jebb SA, Moore MS . Contribution of a sedentary lifestyle and inactivity to the etiology of overweight and obesity: Current evidence and research issues Med Sci Sports Exerc 1999 31: (Suppl): S534–S541.

    Article  PubMed  Google Scholar 

  45. 45

    Pate RR, Pratt M, Blair SN, Haskell WL, Macera CA, Bouchard C, Buchner D, Ettinger W, Health GW, King AC, Kriska A, Leon AS, Marcus BH, Morris J, Paffenbarger RS, Patrick K, Pollock ML, Rippe JM, Sallis J, Wilmore JH . Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine JAMA 1995 273: 402–407.

    CAS  Article  PubMed  Google Scholar 

  46. 46

    Tudor-Locke CE, Myers AM, Rodger NW . Development of a theory-based daily activity intervention for individuals with Type 2 diabetes Diabetes Educ 2001 27: 85–93.

    CAS  Article  Google Scholar 

  47. 47

    Hatano Y . Use of the pedometer for promoting daily walking exercise ICHPER 1993 29: 4–8.

    Google Scholar 

  48. 48

    Yamanouchi K, Takashi T, Chikada K, Nishikawa T, Ito K, Shimizu S, Ozawa N, Suzuki Y, Maeno H, Kato K, Oshida Y, Sato Y . Daily walking combined with diet therapy is a useful means for obese NIDDM patients not only to reduce body weight but also to improve insulin sensitivity Diabetes Care 1995 18: 775–778.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  49. 49

    Tudor-Locke CE, Myers AM . Methodological considerations for researchers and practitioners using pedometers to measure physical (ambulatory) activity Res Q Exerc Sport 2001 72: 1–12.

    CAS  Article  Google Scholar 

  50. 50

    US Surgeon General . Surgeon General's report on physical activity and health. From the Centers for Disease Control and Prevention JAMA 1996 276: 522.

    Google Scholar 

  51. 51

    Kriska A . Physical activity and the prevention of type 2 diabetes mellitus: How much for how long? Sports Med 2000 29: 147–151.

    CAS  Article  PubMed  Google Scholar 

  52. 52

    Rowlands AV, Ingledew DK, Eston RG . The effect of type of physical activity measure on the relationship between body fatness and habitual physical activity in children: a meta-analysis Ann Hum Biol 2000 27: 479–497.

    CAS  Article  Google Scholar 

  53. 53

    Gretebeck RJ, Montoye HJ . Variability of some objective measures of physical activity Med Sci Sports Exerc 1992 24: 1167–1172.

    CAS  Article  PubMed  Google Scholar 

  54. 54

    Dunn AL, Anderson RE, Jakicic JM . Lifestyle physical activity interventions. History, short-and long-term effects, and recommendations Am J Prev Med 1998 15: 398–412.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This study was supported by grant number U48/CCU409664 from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention. We wish to acknowledge Angie Morgan, Melinda Irwin and Bill Bartoli for their assistance with completion of this study.

Author information

Affiliations

Authors

Corresponding author

Correspondence to C Tudor-Locke.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Tudor-Locke, C., Ainsworth, B., Whitt, M. et al. The relationship between pedometer-determined ambulatory activity and body composition variables. Int J Obes 25, 1571–1578 (2001). https://doi.org/10.1038/sj.ijo.0801783

Download citation

Keywords

  • pedometer
  • walking

Further reading

Search

Quick links