Steps per day is an objectively monitored indicator of total daily accumulated ambulatory activity that has been collected with both pedometers and accelerometers to characterize older adults’ physical activity behavior (2,21,26). Both age and body mass index (BMI) are inversely associated with steps per day in older adults (26). However, a routine criticism of this simple volume indicator is that it fails to take into account intensity of this behavior (11,24,26). Commercially available accelerometers (and some pedometers) now have the capability of storing minute-by-minute step accumulations, which can be used to quantify stepping cadence (steps per minute). Originally investigated by Hoshikawa et al. (14,15), cadence is positively associated with ambulatory speed (23) and is a proxy indicator of intensity of ambulatory activity (1,6,17,22,35). In contrast to volume indicators depicting the quantity of daily ambulation (i.e., steps per day), cadence measures can be used to characterize the “quality” (i.e., intensity) of ambulatory activity.
Recently, we analyzed U.S. nationally representative data available from 20- to 85-yr-olds wearing time-stamped accelerometers to study daily minute-by-minute step accumulation patterns (i.e., steps per minute) (31) and in particular an index of best natural effort (i.e., intensity under free-living conditions), which we have called the peak 30-min cadence (defined as the steps per minute recorded for the 30 highest, but not necessarily consecutive, minutes in a day) (29). Peak 30-min cadence therefore is an index (i.e., indicator, marker, pointer, guide), which can be used to characterize an individual’s ambulatory intensity during the most active portions of the day. Steps per day and peak 30-min cadence demonstrate good test–retest reliability on 3-d monitoring periods (19). In a U.S. general population sample ranging in age from 20 to 85 yr or older, we documented age and BMI-related trends in peak 30-min cadence (29). We did not compare its performance with steps per day. Although peak 30-min cadence intuitively adds a new dimension to describing older adults’ daily ambulatory activity patterns using accelerometer-determined step-based variables, it is not clear if it offers any information beyond that which may be distilled from a simple indicator of steps per day. A study that delves more intently into this question in a sample that is more representative of older adults is needed.
The purposes of this analysis were to examine the relationship between steps per day and peak 30-min cadence in a community-dwelling older adult sample and to evaluate the strength of their relationships with both age and BMI.
This is a secondary analysis of cross-sectional data that were originally collected as part of a larger longitudinal study. One hundred and fifty community-dwelling, nondemented study participants, age 58–92 yr, were identified during a single 3-month window (January–March 2012) during ongoing recruitment for the Institute of Dementia Research and Prevention’s Louisiana Aging Brain Study. The Pennington Biomedical Research Center’s Institutional Review Board approved study protocols, and all participants provided written informed consent prior to any data collection.
Height and weight.
Height (to the nearest 0.1 cm) and weight (to the nearest 0.1 kg) were measured two times each, without shoes. Height was measured using a stadiometer, with the head held in the Frankfort Plane, and while the participant held a deep breath.
Ambulatory activity was measured using the GT3X+ accelerometer (ActiGraph, Pensacola, FL). Participants wore the accelerometer at the waist for seven consecutive days. They were asked to wear the accelerometer for 24 h·d−2, removing it only for showering/bathing or other water activities.
Average height and weight values were computed over the repeated measurements, and BMI was calculated as kilograms of body weight divided by height in meters squared (kg·m−2).
We used the accelerometers’ accompanying ActiLife 5 software to download the data in 60-s epochs (time recording intervals). Steps were identified from the raw data file using the manufacturer’s proprietary step filter (which we previously found was more in line with mean pedometer-determined physical activity in a similar older adult sample ). Data were subsequently treated using standard decision rules as imbedded in a National Cancer Institute (NCI)–supplied SAS macro (http://riskfactor.cancer.gov/tools/nhanes_pam/). This is the same data treatment process that has been used for National Health and Nutrition Examination Survey accelerometer data analysis (18,27). Specifically, accelerometer nonwear time was defined as a consecutive string of zeros for ≥60 min, and a valid day of monitored data was defined as having at least 600 min in a day, or 10 of 24 h of wear time.
Steps per day were derived by summing all steps taken during monitored wear time each day. Peak 30-min cadence was calculated by averaging the 30 highest steps per minute values (not necessarily consecutive) each day (29). Daily values for both accelerometer-determined step-based variables were averaged across days for each participant.
Only data from participants with at least three valid days of monitoring were retained for analysis (30). All but two participants had at least 3 d of valid data. The average number of valid days for this sample was 5.8 ± 0.9 d, and the average time worn per valid day was 1018.8 ± 147.0 min. Five individuals had BMI values more than 40 kg·m−2, so we conservatively removed them from this analysis with a partial focus on BMI. The final analysis sample comprised 43 men and 100 women. Relevant sample characteristics are presented in Table 1.
Descriptive data are presented as means (SD). Sex-related differences were evaluated using independent t-tests. The remainder of the analyses was stratified for sex. Data were assessed for normality, and having found inconsistent results for the assorted variables studied, we chose to consistently compute Spearman rank order coefficients for all bivariate correlations comparing 1) age with BMI, 2) steps per day with peak 30-min cadence, and 3) both accelerometer-determined step-based variables with age and BMI. We then computed Spearman partial correlations for steps per day and peak 30-min cadence versus 1) age, controlling for BMI, and 2) BMI, controlling for age. Partial correlations were then recomputed for peak 30-min cadence versus 1) age, controlling for BMI and steps per day, and 2) BMI, controlling for age and steps per day. We compared the significance and magnitude of correlations to interpret relative strength of any apparent relationship of steps per day and peak 30-min cadence with age and BMI.
There were no sex-related differences in participants’ age, steps per day, or peak 30-min cadence; only BMI differed significantly between men and women. The correlation between age and BMI for women was r = −0.236 (P < 0.01). The correlation between age and BMI for men was not significant (r = −0.144). Figures 1 and 2, respectively, present scatterplots depicting the relationship between steps per day and peak 30-min cadence in women (r = 0.881, P < 0.01) and men (r = 0.809, P < 0.01). Table 2 presents the bivariate and partial Spearman correlations between the two accelerometer-determined step-based variables, age, and BMI for older women and men.
Steps per day and peak 30-min cadence were both significantly and inversely correlated with age and BMI in women. The correlations were consistently of a stronger magnitude in relation to age, compared with BMI, in women. Controlling for BMI in women, the partial correlations of age with both accelerometer-determined step-based variables were strengthened relative to those not controlling for BMI. Controlling for age, the partial correlations of BMI with steps per day and peak 30-min cadence in women were also relatively strengthened. The partial correlation of age with peak 30-min cadence was weakened and fell to nonsignificance (r = 0.097, P > 0.05) after controlling for steps per day and BMI in women. Similarly, the partial correlation of BMI with peak 30-min cadence was weakened and nonsignificant (r = −0.098, P > 0.05) after controlling for steps per day and age.
For men, steps per day and peak 30-min cadence were both significantly related to age; no accelerometer-determined step-based variables were associated with BMI. Controlling for BMI in men, the partial correlations of age with steps per day and peak 30-min cadence were strengthened relative to those not controlling for BMI. Controlling for age, the partial correlation of BMI with steps per day in men was strengthened to a point of statistical significance. The partial correlation of BMI with peak 30-min cadence in men was also strengthened after controlling for age but remained statistically nonsignificant. The partial correlation of age with peak 30-min cadence was weakened but remained significant (r = 0.335, P < 0.05) after controlling for BMI and steps per day in men, whereas the partial correlation of BMI with peak 30-min cadence was further weakened and remained nonsignificant (r = 0.000, P > 0.05) after controlling for age and steps per day.
Peak 30-min cadence, an indicator of an individual’s natural best effort in daily life, has been introduced as an additional way of looking at objectively monitored ambulatory activity data (29,31). However, it should come as no surprise that there are strong relationships between steps per day and peak 30-min cadence. The more people move, the more steps they accumulate each minute. However, the focus herein was on the relationship of steps per day and peak 30-min cadence relative to age and BMI, factors that appear to have opposing influences on physical activity in older adults based on cross-sectional analyses. Specifically, these data confirm that, at least in older women, as mean age goes up, mean BMI goes down; or at any rate, those who survive into later years have lower BMI values (16).
Steps per day and peak 30-min cadence were both inversely associated with age in this sample of older men and women. Put another way, as mean age increases in older adults, both mean steps per day and peak 30-min cadence decrease. An age-related separation of steps per day and intensity may only be apparent in samples that are not restricted only to older adults. For example, Ayabe et al. (5) reported that intensity of daily activity decreased with age even though step-defined physical activity remained the same in adults ranging in age from 18 to 69 yr. Similarly, Ayabe et al. (3) observed no difference in daily steps between 21- to 29-yr-old and between 30- to 59-yr-old women, whereas time spent at cadences <100 steps per minute was significantly higher among 30- to 59-yr-olds in comparison with 21- to 29-yr-olds.
In addition to age-related associations with step-based variables, we observed significant associations with BMI, particularly in women. Previous work by Ayabe et al. (4) demonstrated that normal weight individuals spent more time at cadences ≥100 steps per minute and had higher daily stepping rates in comparison with overweight/obese individuals. Similarly, we (29) reported that BMI was inversely associated with peak 30-min cadence (r = −0.18), and even the highest single minute (r = −0.19) in daily living among a nationally representative sample of U.S. adults.
Findings presented here suggest age has a stronger relationship with steps per day and peak 30-min cadence than BMI. In fact, controlling for age strengthens these relationships with BMI in women and reveals the same in men. However, after controlling for steps per day, partial correlations for peak 30-min cadence and age or BMI were nonsignificant for women. Following a similar pattern to women, the partial correlations for peak 30-min cadence with age or BMI in men weakened after controlling for steps per day; however, the relationship with age remained significant. Steps per day generally appeared to be more strongly related to age and BMI than peak 30-min cadence. Indeed, for both women and men, the strongest correlations observed were between age and steps per day, controlling for BMI (r = −0.532 for women and r = −0.707 for men, both P < 0.001). Notable exceptions were the bivariate correlations between BMI and steps per day in women and men, which were weaker than the correlations between BMI and peak 30-min cadence.
Steps per day ranged from approximately 1000 to 15,000 steps per day in this sample; however, the mean values (i.e., approximately 5500 steps per day) were harmonious with those reported for other older adult samples (33). Although most reports indicate that men take more steps per day than women (33), we found that both older men and women were more uniformly inactive in this sample. In contrast, the peak 30-min cadence captures relative behavioral persistence, or at least habitual effort, executed in daily living (34). The peak 30-min cadence values herein were consistent with previously reported values for 60- to 69-yr-old adults (peak 30-min cadence = 65.2 steps per minute) (29). These consistent findings increase our confidence in the quality of these data.
Reported walking behavior has been associated with maintenance of mobility in older women; even small amounts of regular walking appear to confer some degree of protection against subsequent losses for 1 yr (25). Further, the frequency of “getting out and about” has been associated with older adults’ objectively determined steps per day and time spent in higher intensities of physical activity (9). Steps per day have also been associated with better performance on tests of function and muscular strength. These findings, coupled with the results obtained in this analysis, suggest that it is appropriate to simply focus on increasing older adults’ ambulatory activity.
By implementing a 24-h accelerometer wearing protocol, we were successful in getting our participants to wear the instruments for a much longer daily duration than conventional “waking time” protocols (approximately 17 h·d−1 herein vs approximately 14 h·d−1 in the National Health and Nutrition Examination Survey ). Although we applied an established data treatment decision rule for defining wear time as ≥60 min of consecutive zeros, recent evidence suggests that this may be better extended to 90 min or more in older adult samples (8). A definition of nonwear time was only used to help identify noncompliance to the study protocol. Because nonwear time primarily affects estimates of time spent in sedentary behaviors (18), which was not the concentration of this analysis, and because we only identified two participants who did not meet our protocol compliance criteria, this decision rule had little practical effect on the focal step-based variables of these analyses.
Several study limitations must be mentioned. This was a cross-sectional analysis of community-dwelling older adults who volunteered to engage in health-related research. As such, they can be described as a select sample. A proportion of the sample depicted herein may not be considered representative of older adults as the youngest participant was 58 yr old. The lack of a universally accepted age range for older adulthood makes any minimum age cutoff subjective. However, a previous review summarizing ambulatory activity among older adults (32) included studies with participants as young as 50 yr as long as the sample mean age was 65 yr or older. We chose to use a similar decision rule in defining this study’s sample as older adults (mean age = 71 yr) while still acknowledging this limitation. Further, more than half of the participants were women. Women live longer than men (13), and this simple fact somewhat shapes anticipated study enrollment statistics. Further, analysis is focused on a limited sample, with average values presented as descriptive statistics, and the findings do not necessarily apply to all individuals in a population. The small sample represented herein limited statistical power for several analyses (particularly in men) and may have contributed to the lack of statistically significant associations noted between BMI and step-based variables in men. Conclusions about the impact of age and BMI on all of these accelerometer-determined step-based indicators within individuals and over time may differ from what was apparent in this aggregate analysis. Finally, our focus was on accelerometer-determined step-based variables, but we acknowledge that the human movement repertoire includes behaviors other than simple ambulation. However, older adults tend to derive 100% of their leisure time physical activity energy expenditure from walking (7). Further, energy expenditure of nonexercise movement is primarily due to ambulatory behaviors in older adults (12). We have previously argued that ambulatory activity and, more specifically, walking is the single most important activity to measure well and promote effectively (34).
In summary, we considered both steps per day and peak 30-min cadence in the context of age and BMI in a community-dwelling older adult sample. Specifically, we sought to answer what peak 30-min cadence may (or may not) contribute to our understanding of physical activity, specifically ambulatory activity and the factors that shape it. What we discovered was that peak 30-min cadence did not really tell us much more, at least in terms of age and BMI, than steps per day in this older adult sample where average step-defined activity was, for the most part, uniformly low. In fact, the relationship between steps per day and peak 30-min cadence was so strong that steps per day could be feasibly used as a proxy indicator of the other. To be clear, peak 30-min cadence did not appear to offer any unique information over a simple indicator of steps per day relative to age or BMI. The usefulness of peak 30-min cadence needs to be evaluated in larger and more diverse samples and against other parameters of interest, including those more theoretically linked to intensity of effort. For example, the relative importance of quality versus quantity effort may be best examined with regards to function (e.g., gait speed ), fitness (e.g., time to complete a 400-m walk ), or leg strength (24). Ultimately, prospective and especially intervention studies may be necessary to tease out the relative importance of these different (but related) dimensions of ambulatory activity to older adults’ health.
This work was supported by the Hibernia National Bank/Edward G. Schlieder Chair, the Jo Lamar Dementia Fund, and the supporters of the Institute for Dementia Research and Prevention.
The authors declare they have no conflict of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Abel M, Hannon J, Mullineaux D, Beighle A. Determination of step rate thresholds corresponding to physical activity classifications in adults. J Phys Act Health
. 2011; 8 (1): 45–51.
2. Aoyagi Y, Shephard RJ. Steps per day: the road to senior health? Sports Med
. 2009; 39 (6): 423–38.
3. Ayabe M, Aoki J, Kumahara H, Tanaka H. Assessment of minute-by-minute stepping rate of physical activity under free-living conditions in female adults. Gait Posture
. 2011; 34 (2): 292–4.
4. Ayabe M, Aoki J, Kumahara H, et al.. Minute-by-minute stepping rate of daily physical activity in normal and overweight/obese adults. Obes Res Clin Pract
. 2011; 5 (2): e151–6.
5. Ayabe M, Yahiro T, Yoshioka M, Higuchi H, Higaki Y, Tanaka H. Objectively measured age-related changes in the intensity distribution of daily physical activity in adults. J Phys Act Health
. 2009; 6 (4): 419–25.
6. Beets MW, Agiovlasitis S, Fahs CA, Ranadive SM, Fernhall B. Adjusting step count recommendations for anthropometric variations in leg length. J Sci Med Sport
. 2010; 13 (5): 509–12.
7. Bryan SN, Katzmarzyk PT. Patterns and trends in walking
behaviour among Canadian adults. Can J Public Health
. 2009; 100 (4): 294–8.
8. Choi L, Ward SC, Schnelle JF, Buchowski MS. Assessment of wear/nonwear time classification algorithms for triaxial accelerometer
. Med Sci Sports Exerc
. 2012; 44: 2009–2016.
9. Davis MG, Fox KR, Hillsdon M, et al.. Getting out and about in older adults: the nature of daily trips and their association with objectively assessed physical activity. Int J Behav Nutr Phys Act
. 2011; 8: 116.
10. Fitzpatrick AL, Buchanan CK, Nahin RL, et al.. Associations of gait speed and other measures of physical function with cognition in a healthy cohort of elderly persons. J Gerontol A Bio Sci Med Sci
. 2007; 62 (11): 1244–51.
11. Foley S, Quinn S, Jones G. Pedometer
determined ambulatory activity and bone mass: a population-based longitudinal study in older adults. Osteo Int
. 2009; 21 (11): 1809–16.
12. Harris AM, Lanningham-Foster LM, McCrady SK, Levine JA. Nonexercise movement in elderly compared with young people. Am J Physio Endocrin Metab
. 2007; 292 (4): E1207–12.
13. Heron M, Hoyert DL, Murphy SL, et al.. Deaths: final data for 2006. Natl Vital Stat Rep
. 2009; 57 (14): 1–134.
14. Hoshikawa T, Toyoshima S, Ikegami Y, Mori S, Saito Y. Practical application of an actogram to physical fitness research: recording of physical activity pattern during daily life. Jpn J Phys Fit Sport
. 1992; 41 (2): 174–82.
15. Hoshikawa T, Toyoshima S, Mori S, Mori N, Ikegami Y. An application of an actogram to research of teaching method in physical education: a new device for recording physical activity pattern during class. Res Phys Ed
. 1992; 37 (1): 15–27.
16. Katzmarzyk PT, Reeder BA, Elliott S, et al.. Body mass index and risk of cardiovascular disease, cancer and all-cause mortality. Can J Pub Health
. 2012; 103 (2): 147–51.
17. Marshall SJ, Levy SS, Tudor-Locke CE, et al.. Translating physical activity recommendations into a pedometer
-based step goal: 3000 steps in 30 minutes. Am J Prev Med
. 2009; 36 (5): 410–5.
18. Masse LC, Fuemmeler BF, Anderson CB, et al.. Accelerometer
data reduction: a comparison of four reduction algorithms on select outcome variables. Med Sci Sports Exerc
. 2005; 37 (11 Suppl): S544–54.
19. Mudge S, Taylor D, Chang O, Wong R. Test–retest reliability of the StepWatch Activity Monitor outputs in healthy adults. J Phys Act Health
. 2010; 7 (5): 671–6.
20. Newman AB, Simonsick EM, Naydeck BL, et al.. Association of long-distance corridor walk performance with mortality, cardiovascular disease, mobility limitation, and disability. JAMA
. 2006; 295 (17): 2018–26.
21. Payn T, Pfeiffer KA, Hutto B, et al.. Daily steps in midlife and older adults: relationship with demographic, self-rated health, and self-reported physical activity. Res Q Exerc Sport
. 2008; 79 (2): 128–32.
22. Rowe DA, Welk GJ, Heil DP, et al.. Stride rate recommendations for moderate intensity walking
. Med Sci Sports Exerc
. 2011; 43 (2): 312–8.
23. Rowlands AV, Stone MR, Eston RG. Influence of speed and step frequency during walking
and running on motion sensor output. Med Sci Sports Exerc
. 2007; 39 (4): 716–27.
24. Scott D, Blizzard L, Fell J, Jones G. Ambulatory activity, body composition, and lower-limb muscle strength in older adults. Med Sci Sports Exerc
. 2009; 41 (2): 383–9.
25. Simonsick EM, Guralnik JM, Volpato S, Balfour J, Fried LP. Just get out the door! Importance of walking
outside the home for maintaining mobility: findings from the women’s health and aging study. J Am Geriatr Soc
. 2005; 53 (2): 198–203.
26. Strath SJ, Schwartz AM, Cashin SE. Ambulatory physical activity profiles of older adults. J Aging Phys Act
. 2009; 17 (1): 46–56.
27. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer
. Med Sci Sports Exerc
. 2008; 40 (1): 181–8.
28. Tudor-Locke C, Barreira TV, Brouillette RM, Foil HC, Keller JN. Preliminary comparison of clinical and free-living measures of stepping cadence in older adults. J Phys Act Health
. (in press).
29. Tudor-Locke C, Brashear MM, Katzmarzyk PT, Johnson WD. Peak stepping cadence in free-living adults: 2005–2006 NHANES. J Phys Act Health
. 2012; 9: 1125–1129.
30. Tudor-Locke C, Burkett L, Reis JP, Ainsworth BE, Macera CA, Wilson DK. How many days of pedometer
monitoring predict weekly physical activity in adults? Prev Med
. 2005; 40 (3): 293–8.
31. Tudor-Locke C, Camhi SM, Leonardi C, et al.. Patterns of adults stepping cadence in the 2005–2006 NHANES. Prev Med
. 2011; 53: 178–81.
32. Tudor-Locke C, Craig CL, Aoyagi Y, et al.. How many steps/day are enough? For older adults and special populations. Int J Behav Nutr Phys Act
. 2011; 8: 80.
33. Tudor-Locke C, Hart TL, Washington TL. Expected values for pedometer
-determined physical activity in older populations. Int J Behav Nutr Phys Act
. 2009; 6: 59.
34. Tudor-Locke C, Rowe DA. Using cadence to study free-living ambulatory behavior. Sports Med
. 2012; 42 (5): 381–98.
35. Tudor-Locke C, Sisson SB, Collova T, Lee SM, Swan PD. Pedometer
-determined step count guidelines for classifying walking
intensity in a young ostensibly healthy population. Can J Appl Physiol
. 2005; 30 (6): 666–76.