Journal of Physical Activity and Health, 2014, 11, 1367 -1372 http://dx.doi.org/10.1123/jpah.2012-0333 © 2014 Human Kinetics, Inc.
Official Journal of ISPAH www.JPAH-Journal.com ORIGINAL RESEARCH
Steps Per Day Required for Meeting Physical Activity Guidelines in Japanese Adults Zhen-Bo Cao, Taewoong Oh, Nobuyuki Miyatake, Kazuyo Tsushita, Mitsuru Higuchi, and Izumi Tabata Background: This study aimed to determine the optimal number of steps per day needed to meet the current physical activity guidelines in a large population sample of Japanese adults. Methods: An accelerometer-based activity monitor (Kenz Lifecorder) was used to simultaneously measure moderate-to-vigorous intensity physical activity (MVPA) and step counts in 940 Japanese adults (480 women) aged 20 to 69 years. The step count per day equivalents to 2 different physical activity recommendations (23 MET-h/wk and 150 min/wk of MVPA) were derived using linear regression analysis and receiver operating characteristic (ROC) methodology. Results: Linear regression analysis showed that daily step counts correlated with weekly PAEE (r = .83) and daily minutes of MVPA (r = .83). Linear regression analysis also showed that 23 MET-h/wk of MVPA is equivalent to 11,160 steps/d, and 150 min/wk of MVPA is equivalent to 7716 steps/d. ROC analysis yielded similar findings: 10,225 steps/d are required to accumulate ≥ 23 MET-h/wk of MVPA and 7857 steps/d are needed to meet the recommendation of ≥ 150 min/wk of MVPA. Conclusions: The findings suggest that 10,000 to 11,000 and 7700 to 8000 steps/d represent the optimal thresholds for accumulating ≥ 23 MET-h/wk of MVPA and ≥ 150 min/wk of MVPA, respectively, for Japanese adults. Keywords: accelerometer, energy expenditure, pedometer Physical inactivity is a well-documented risk factor for coronary heart disease and is associated with increased risk for other chronic diseases, including obesity, Type II diabetes mellitus, hypertension, colon cancer, depression, and osteoporosis.1 Accordingly, promoting physical activity has become a public health priority worldwide,2 and numerous physical activity guidelines have been published by government agencies, professional organizations, and associations to provide information and guidance on the types and amounts of physical activity sufficient to offer substantial health benefits.3–8 Usually, physical activity guidelines are expressed in terms of time or energy expenditure (metabolic equivalent, MET). A number of international and national health authorities have produced similar guidelines recommending that adults accumulate at least 150 min of moderate-to-vigorous intensity physical activity (MVPA) per week.3–7 In 2006, the Ministry of Health, Labor and Welfare of Japan revised its previous physical activity guidelines and recommended that Japanese adults should engage in a minimum of 23 MET-h per week of MVPA,8 which is more than twice the volume of activity in the recommendation of 150 min/wk of MVPA.3 Recently, steps per day, a simple, useful indicator of physical activity volume, widely accepted by researchers, practitioners, and the general public, has been used in large, nationally representative surveillance studies in Japan,9 Canada,10 and the USA11 and in physical activity intervention research.12,13 If current physical activity guidelines can be translated into this indicator, steps per day would
be a useful additional measurement for surveillance of adherence to these recommendations. Some studies have tried to determine how many steps per day equate to current MVPA guidelines in freeliving samples.14–22 However, most of the studies conducted to date did not use 2 classic analytic approaches, linear regression analysis and receiver operating characteristic (ROC) methodology, nor did they assess classification accuracy of the step count thresholds. Furthermore, many of these prior studies14–19 have used relatively small and/or selected samples such as college students, postmenopausal women, or workplace employees. Only 2 studies used large population samples.20,21 To the best of our knowledge, no study has used both linear regression analysis and ROC methodology to determine the step-based translations of 23 MET-h/wk of MVPA and 150 min/wk of MVPA in Japanese adults. The purposes of this study, therefore, were to examine the relationship between accelerometer-assessed step count per day and MVPA and to determine the optimal number of steps per day needed to meet current public health recommendations for weekly physical activity (23 MET-h/wk of MVPA and 150 min/wk of MVPA) by using linear regression analysis and ROC methodology in a large population sample of Japanese adults.
Cao ([email protected]) is with School of Kinesiology, Shanghai University of Sport, Shanghai, China. Higuchi is with the Faculty of Sport Sciences, Waseda University, Tokorozawa, Japan. Oh is with the Faculty of Human Health Science, Matsumoto University, Matsumoto, Japan. Miyatake is with the Dept of Hygiene, Faculty of Medicine, Kagawa University, Miki, Japan. Tsushita is with the Comprehensive Health Science Center, Aichi Health Promotion Foundation, Higashiura, Japan. Tabata is with the Faculty of Sport and Health Science, Ritsumeikan University, Kusatsu, Japan.
This cross-sectional study included 940 Japanese adults (480 women) aged 20 to 69 years who participated in the Exercise and Physical Activity Reference and Guide for Health Promotion Study (EPARGHP), which was designed to investigate Japanese levels of physical activity and cardiorespiratory fitness and the association of these levels with other risk factors for lifestyle-related diseases. Data were collected from June 2007 to November 2009 in 4 Japanese cities (Tokyo, Chita-gun, Okayama, and Matsumoto). Study
Methods Subjects
1367
1368 Cao et al
procedures were approved by the Ethical Committee of the National Institutes of Health and Nutrition. Written informed consent was obtained for all participants.
Anthropometrics Body mass was measured to the nearest 0.1 kg and height was measured to the nearest 0.1 cm. Body mass index (BMI) was calculated according to the formula [body mass (kg)/height squared (m2)]. Waist circumference (WC) was measured at the umbilical level with a nonelastic tape to the nearest 0.1 cm.23
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Physical Activity The Kenz Lifecorder PLUS (LC; Suzuken Co. Ltd., Nagoya, Japan), an accelerometer-based activity monitor, was used to record participants’ step count data along with time and physical activity-related energy expenditure (PAEE) spent in MVPA for 7 consecutive days. All participants were asked to engage in their normal daily pattern of physical activity. The average steps per day (steps/d), accumulated MVPA per day (min/d), and PAEE per week (MET-h/wk) were calculated using data from at least 10 h per day for 7 days (including 2 weekend days), without consideration of MVPA bout length. Previous investigations24–26 have demonstrated that the LC provides a valid assessment of step count, as well as physical activity intensity and PAEE. Additional details of the methodology and equipment have been published elsewhere.23,26
Statistical Analyses Gender differences were evaluated with a Student’s t test (normal distribution) or Mann-Whitney U test (nonnormal distribution). Linear regression analysis was used to estimate relationships between step count per day and MVPA and PAEE. The optimal number of steps/d required to meet the different physical activity recommendations was determined using 2 analytic approaches:
linear regression analysis and ROC curves. In the ROC analysis, we considered meeting or not meeting the different MVPA recommendations (23 MET-h/wk or 150 min/wk) as an outcome variable and step count per day as a testing variable. The distance between each point on the ROC curve and the upper left corner was calculated as the square root of [(1 – sensitivity)2 + (1 – specificity)2], and the point with the shortest distance was judged to correspond to the optimal cutoff value.27,28 Furthermore, we calculated the area under the curve (AUC) and its 95% confidence interval (CI). The optimal step counts thresholds derived from linear regression analysis or ROC analysis were examined for classification accuracy (number of correct classifications/number of all classifications) by dividing the entire sample into subgroups stratified by sex and age (men: 20–29 y [n = 89], 30–39 y [n = 98], 40–49 y [n = 89], 50–59 y [n = 98], and 60–69 y [n = 86]; women: 20–29 y [n = 80], 30–39 y [n = 96], 40–49 y [n = 100], 50–59 y [n = 107], and 60–69 y [n = 97]). All analyses were conducted with SPSS 16.0 J for Windows (SPSS Japan Inc., Tokyo, Japan). Statistical significance was considered to be P < .05.
Results Subject characteristics are shown in Table 1. The mean age (standard deviation: SD) was 44.7 (13.9) years. The amount spent in MVPA included both minutes and PAEE were higher in men than in women, but no significant difference in step count per day existed between men and women. The percentage of subjects meeting the recommendation of 150 min/wk of MVPA was 62.6% in men and 54.8% in women; 18.5% of men and 15.0% of women met the recommendation of 23 MET-h/wk of MVPA. Linear regression analysis yielded the following equations for predicting the number of steps per day required to meet the 2 different MVPA recommendations: steps/d = 4702 + 280.8 × PAEE (R2 = .70, P < .001) and steps/d = 4356 + 22.4 × MVPA/wk (R2 = .70, P < .001). Using these equations, the recommendation of 23 MET-h/wk
Table 1 Physical Characteristics of Study Participants Variable
Total
Men
Women
n = 940
n = 460
n = 480
Age (y)
44.7 (13.9)
44.2 (14.0)
45.2 (13.7)
Height (cm)
163.8 (8.5)
170.2 (5.9)
157.6 (5.5)*
Body mass (kg)
59.3 (10.5)
66.3 (8.6)
52.5 (7.1)*
BMI (kg/m2)
22.0 (2.9)
22.9 (2.8)
21.1 (2.8)*
%Fat (%)
23.2 (6.8)
19.2 (5.2)
27.1 (5.7)*
78.8 (8.7)
81.4 (8.1)
76.3 (8.6)*
Step count (steps/d)
WC (cm)
8776 (3204)
8793 (3377)
8760 (3032)
MVPA (min/d)
28.2 (17.0)
29.7 (18.1)
26.8 (15.9)*
PAEE (MET-h/wk)
14.5 (9.5)
15.2 (10.0)
13.9 (8.9)*
150 min/wk of MVPA
551 (58.6)
288 (62.6)
263 (54.8)
23 MET-h/wk of MVPA
157 (16.7)
85 (18.5)
72 (15.0)
Number (%) of participants meeting physical activity recommendations
Note. Values are expressed as mean (SD) or number (%). Abbreviations: BMI, body mass index; %Fat, body fat percentage; WC, waist circumference; MVPA, moderate-to-vigorous physical activity; PAEE, physical activity-related energy expenditure spent in MVPA; MET, metabolic equivalents. * Significantly different from men, P < .05.
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Steps/Day for Meeting Physical Activity Guidelines 1369
of MVPA corresponded to 11,160 steps/d, and the recommendation of 150 min/wk of MVPA corresponded to 7716 steps/d (Table 2). ROC analysis revealed that 10,225 steps/d produced 0.92 sensitivity and 0.93 specificity for accumulating ≥ 23 MET-h/wk of MVPA (AUC = 0.93) and that 7857 steps/d produced 0.86 sensitivity and 0.90 specificity for accumulating ≥ 150 min/wk of MVPA (AUC = 0.91; Table 2). According to the ROC curve, the optimal steps/d thresholds for likelihood of accumulating ≥ 23 MET-h/wk of MVPA, or ≥ 150 min/wk of MVPA were 10,270 steps/d (sensitivity 0.92, specificity 0.83, AUC 0.93) and 7850 steps/d (sensitivity 0.85, specificity 0.83, AUC 0.92), respectively, for men and 10,660 steps/d (sensitivity 0.90, specificity 0.88, AUC 0.94) and 8065 steps/d (sensitivity 0.84, specificity 0.80, AUC 0.89), respectively, for women. The accuracy of daily step count thresholds derived from the 2 analytic approaches for the 2 different MVPA recommendations for subgroups stratified by sex and age are shown in Tables 3 and 4. The accuracy of steps/d thresholds to screen for accumulating ≥ 23 MET-h/wk of MVPA derived from linear regression analysis appears to be higher than the accuracy of steps/d thresholds derived from ROC analysis. However, the accuracy of steps/d thresholds to screen for accumulating ≥ 150 min/wk of MVPA derived from linear regression analysis was similar to the accuracy of steps/d thresholds derived from ROC analysis in all age groups.
Discussion This study investigated the relationship between steps/d and MVPA and the optimal number of steps/d needed to meet current public health recommendations for weekly physical activity. Using data from a large population sample of Japanese adults, we demonstrated that steps/d is positively associated with MVPA in terms of time (r = .83, P < .001) and energy expenditure (PAEE: r = .83, P < .001). Our findings from 2 different analytical models suggest that 10,000 to 11,000 and 7700 to 8000 steps/d represent the optimal thresholds for likelihood of accumulating ≥ 23 MET-h/wk of MVPA and ≥ 150 min/wk of MVPA, respectively. These thresholds are supported by good classification accuracy in all age subgroups. Participants in the current study appeared to take more steps/d than those in a Japanese National Health and Nutrition Survey9 conducted in 2007, using a spring-levered pedometer (AS-200, Yamasa Co. Ltd., Tokyo, Japan). The Japanese national survey reported that men and women took 7321 steps/d and 6267 steps/d, respectively.9 However, these findings do not necessarily mean that participants in the current study were more active compared with those in the 2007 survey, because differences in physical activity monitoring devices may explain some of this discrepancy. Studies have shown that accelerometer-based activity monitors are less affected by adi-
Table 2 Step Count per Day for Meeting 2 Different MVPA Recommendations Using 2 Analytic Approaches Step count per day for meeting MVPA recommendations Physical activity guidelines
n
Regression analysis
ROC curve analysis
23 MET-h/wk of MVPA All
940
11,160
10,225
Men
460
10,985
10,270
Women
480
11,358
10,660
All
940
7716
7857
Men
460
7523
7850
Women
480
7905
8065
150 min/wk of MVPA
Abbreviations: MVPA, moderate-to-vigorous physical activity; MET, metabolic equivalent; ROC, receiver operating characteristic.
Table 3 Accuracy of Step Count per Day Cut-Off Points Derived From Linear Regression Analysis for 2 MVPA Recommendations for Subgroups Stratified by Age Accuracy n
20–29 y
30–39 y
40–49 y
50–59 y
60–69 y
20–69 y
11,160 steps/d (23 MET-h/wk of MVPA) All
940
0.93
0.87
0.87
0.88
0.85
0.88
Men
460
0.91
0.86
0.83
0.87
0.85
0.86
Women
480
0.93
0.88
0.91
0.88
0.88
0.89
All
940
0.82
0.80
0.84
0.81
0.85
0.83
Men
460
0.87
0.82
0.91
0.80
0.87
0.85
Women
480
0.75
0.78
0.82
0.86
0.88
0.82
7716 steps/d (150 min/wk of MVPA)
Abbreviations: y, years; MVPA, moderate-to-vigorous physical activity; MET, metabolic equivalent; Accuracy = (Number of correct assessments)/(Number of all assessments).
1370 Cao et al
Table 4 Accuracy of Step Count per Day Cut-Off Points Derived From ROC Analysis for 2 MVPA Recommendations for Subgroups Stratified by Age Accuracy n
20–29 y
30–39 y
40–49 y
50–59 y
60–69 y
20–69 y
All
940
0.89
0.83
0.82
0.82
0.84
0.85
Men
460
0.87
0.86
0.79
0.87
0.84
0.84
Women
480
0.93
0.85
0.84
0.79
0.85
0.85
All
940
0.82
0.79
0.86
0.83
0.86
0.83
Men
460
0.88
0.80
0.90
0.81
0.86
0.85
Women
480
0.76
0.78
0.82
0.86
0.87
0.82
10,225 steps/d (23 MET-h/wk of MVPA)
7857 steps/d (150 min/wk of MVPA)
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Abbreviations: y, years; MVPA, moderate-to-vigorous physical activity; MET, metabolic equivalent; ROC, receiver operating characteristic; Accuracy = (Number of correct assessments)/(Number of all assessments).
posity and more sensitive to low force accelerations (such as slow stepping) than spring-levered pedometers.29,31 For example, Crouter et al29 found that the Yamax (Yamasa) pedometer undercounted steps in overweight, and obese individuals, whereas an accelerometerbased step counter (New Lifestyles NL-2000, Inc, Lees Summit, MO, USA) was more accurate. Le Masurier and Tudor-Locke30 found that the spring-levered pedometer detected fewer steps than the accelerometer at slow walking speed. Furthermore, Ayabe et al31 reported that the step counts determined by a spring-levered pedometer (EC-200, Yamasa Co. Ltd., Tokyo, Japan) were about 1400 steps/d less than those determined by an accelerometer-based activity monitor (Kenz Lifecorder) in Japanese adults. Recently, Inoue et al32 studied physical activity levels of a large population sample of adults (n = 790) by using the Kenz Lifecorder in 4 Japanese cities (Koganei, Tsukuba, Shizuoka, and Kagoshima). They reported that men and women took 8763 steps/d and 8242 steps/d, respectively, which is similar to our results. Therefore, the physical activity levels of participants in the current study are likely to be similar to those of the general Japanese urban population. Although previous physical activity guidelines in Japan have called for taking 10,000 steps/d (an amount equivalent to at least 20 to 30 min of moderate-intensity walking or other exercise most days of the week),33 more recent Japanese physical activity guidelines encourage at least 23 MET-h/wk of MVPA.8 The report from the Ministry of Health, Labor and Welfare of Japan33 indicates that 23 MET-h/wk of MVPA is approximately equivalent to 60 min of MVPA or 8000 to 10,000 steps/d. This recommended step count was based on an assumed daily level in Japanese adults of 2000 to 4000 steps/d of low intensity unconscious activity (< 3 METs), with each additional 60 min of MVPA adding 6000 steps. One previous study21 in a large sample of Japanese adults has attempted to determine daily step counts measured with a triaxial accelerometer (Actimarker EW4800; Panasonic Electric Works, Japan) that classify subjects as meeting the 23 MET-h/wk of MVPA. ROC methodology showed that 8500 to 10,000 steps/d, a number similar to current recommendations, was indicated as the optimal daily step count for achieving 23 MET-h/wk of MVPA.21 However, the activity monitor used in their study undercounted steps at slow walking speeds, and its production and sales have been suspended. In addition, subjects in their study had high levels of physical activity (subjects walked an average of 9600 steps/d, and approximately 48% of subjects achieved ≥ 23 MET-h/wk of MVPA) compared with the
general population (men in the general population take 7321 steps/d, and women take 6267 steps/d).9,21 In the current study, we found that current recommended step counts (8000 to 10,000 steps/d) in Japan are equivalent to only 12 to 19 MET-h/wk, falling short of the recommended 23 MET-h/wk of MVPA. We also found that the current recommended 8000 to 10,000 steps/d actually overestimate the proportion of our subjects who are meeting physical activity guideline (23 MET-h/wk of MVPA) up to 17% to 40%. This suggests that the current recommended step counts for Japanese adults may be largely overestimating the proportion of Japanese adults who are meeting physical activity guideline. Such an overestimation could be dangerous because it could lead individuals to underestimate their own health risks. Approximately 11,000 steps/d, corresponding to the “active” level of a graduated step index for healthy adults,34 are needed for Japanese adults to obtain at least 23 MET-h/wk of MVPA in this study. This threshold of 11,000 steps per day appears to have high classification accuracy for both men (83% to 91%) and women (83% to 89%) across all age groups. Recently, some international and national health guidelines have recommended that adults should accumulate at least 150 min/ wk of MVPA.3–7 A limited number of studies have investigated the number of steps per day needed to meet this guideline.17,20 To our knowledge, only 1 study has attempted to determine a direct step counts per day translation of 150 min/wk of accumulated MVPA. Tudor-Locke and colleagues20 measured steps and MVPA with the ActiGraph accelerometer in a large sample of US adults and analyzed the data with a linear regression model.20 They reported a strong linear relationship between step counts and MVPA and showed that 150 min/wk of MVPA translated to approximately 7000 steps/d (or 49,000 steps/wk).20 However, they did not evaluate the accuracy of their step count threshold in identifying accumulation of 150 min/wk of MVPA. Using linear regression analysis and ROC methodology, we found that 7700 to 8000 steps/d represented the optimal threshold for likelihood of accumulating ≥ 150 min/wk of MVPA in Japanese adults. To estimate the step count threshold’s performance, we also assessed classification accuracy by dividing the entire sample into subgroups stratified by sex and age. The results show a high correct classification rate for both men (79% to 89%) and women (74% to 86%) in all age groups, as well as higher classification accuracies across incremental age groups in women, but not in men. Slightly higher classification accuracy was observed in young and middleaged men and older women as compared with age-matched, opposite-
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Steps/Day for Meeting Physical Activity Guidelines 1371
sex groups. Our results provide evidence for the practical use of the step count threshold determined in the current study. In this population, Steps per day correlated strongly with time spent in MVPA (r = .830). Although few studies have investigated these associations, those existing have found similar results in healthy adult men35 (r = .84), children36 (r = .81), and cardiac rehabilitation patients37 (r = .85). A similar correlation was also seen between steps per day and PAEE spent in MVPA (r = .83). Considering PAEE was calculated by multiplying intensity of MVPA by time spent in MVPA, it is not surprising to find that there is an agreement relation between steps per day and the levels of MVPA expressed in terms of time or PAEE. This implicated that time spent in MVPA explained a large proportion of the variance in total PA expressed in term of steps per day, while intensity of MVPA did not account for an additional variance in total PA in this population. The amount of physical activity recommended by the Japanese physical activity guideline of 23 MET-h/wk of MVPA is more than twice the volume of activity in the 150 min/wk of MVPA recommendation. However, the magnitude of the difference was reduced after generating step-based translations of the 2 guidelines in this study. This discrepancy may be due in part to the fact that, unlike duration- and intensity-based physical activity guidelines that focus on MVPA, step-based translations take into account the total daily volume of ambulatory physical activity, including MVPA and light-intensity physical activity (< 3 METs). In the current study, participants engaged in about 2 hours per day of light-intensity physical activity, which is equivalent to 3500 steps. After removing 3500 steps/d from the step-based translations of the duration- and intensity-based physical activity guidelines, the magnitude of the difference for the 2-step count thresholds is similar to that for the 2 current MVPA recommendations. There are some limitations to our study. First, the current sample did not include older adults (aged 70 years and over) and included primarily individuals living in urban settings. Thus, our results may have limited generalizability. Second, our calculation of MVPA did not take into consideration MVPA bout length, even though current physical activity guidelines recommend that MVPA needs to be accumulated in bouts of at least 10 min. Despite these limitations, the current study has several strengths, including a large population sample of Japanese men and women, the use of a valid and reliable accelerometer-based activity monitor to concurrently quantify steps/d and MVPA, the use of both linear regression analysis and ROC methodology, and an assessment of classification accuracy of the step count threshold. In conclusion, our results suggest that 10,000 to 11,000 and 7700 to 8000 steps/d represented the optimal thresholds for accumulating ≥ 23 MET-h/wk of MVPA and ≥ 150 min/wk of MVPA, respectively, for Japanese adults. Acknowledgments This research was supported by a grant for Comprehensive Research on Cardiovascular and Life-Style Related Diseases from the Ministry of Health, Labour and Welfare, Japan (no.19160101) and by a grant-in-aid for the Global COE, Waseda University “Sport Sciences for the Promotion of Active Life” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan (no. A09338100).
References 1. Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence. CMAJ. 2006;174:801–809. PubMed doi:10.1503/ cmaj.051351
2. World Health Organization. Global status report on noncommunicable diseases 2010. Geneva: WHO Press; 2011. 3. Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc. 2007;39:1423–1434. PubMed doi:10.1249/ mss.0b013e3180616b27 4. World Health Organization. Global recommendations on physical activity for health. Geneva: WHO Press; 2010. 5. Tremblay MS, Warburton DE, Janssen I, et al. New Canadian physical activity guidelines. Appl Physiol Nutr Metab. 2011;36:36–46, 47–58. PubMed doi:10.1139/H11-009 6. Australian Government Department of Health and Ageing. Physical Activity Guidelines for Adults. [http://www.health.gov.au/internet/ main/publishing.nsf/Content/health-pubhlth-strateg-phys-actguidelines#guidelines_adults]. 7. Department of Health. UK Physical Activity Guidelines. [http://www. dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_127931]. 8. Ministry of Health Labour and Welfare of Japan. Exercise and Physical Activity Reference for Health Promotion 2006. [http://www.nih. go.jp/eiken/programs/pdf/epar2006.pdf.]. 9. Inoue S, Ohya Y, Tudor-Locke C, Tanaka S, Yoshiike N, Shimomitsu T. Time trends for step-determined physical activity among Japanese adults. Med Sci Sports Exerc. 2011;43:1913–1919. PubMed doi:10.1249/MSS.0b013e31821a5225 10. Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS. Physical activity of Canadian adults: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep. 2011;22:7–14. PubMed 11. Tudor-Locke C, Johnson WD, Katzmarzyk PT. Accelerometerdetermined steps per day in US adults. Med Sci Sports Exerc. 2009;41:1384–1391. PubMed doi:10.1249/MSS.0b013e318199885c 12. Petersen CB, Severin M, Hansen AW, Curtis T, Gronbaek M, Tolstrup JS. A population-based randomized controlled trial of the effect of combining a pedometer with an intervention toolkit on physical activity among individuals with low levels of physical activity or fitness. Prev Med. 2012;54:125–130. PubMed doi:10.1016/j. ypmed.2011.12.012 13. Freak-Poli R, Wolfe R, Backholer K, de Courten M, Peeters A. Impact of a pedometer-based workplace health program on cardiovascular and diabetes risk profile. Prev Med. 2011;53:162–171. PubMed doi:10.1016/j.ypmed.2011.06.005 14. Behrens TK, Hawkins SB, Dinger MK. Relationship between objectively measured steps and time spent in physical activity among freeliving college students. Meas Phys Educ Exerc Sci. 2005;9:67–77. doi:10.1207/s15327841mpee0902_1 15. Jordan AN, Jurca GM, Tudor-Locke C, Church TS, Blair SN. Pedometer indices for weekly physical activity recommendations in postmenopausal women. Med Sci Sports Exerc. 2005;37:1627–1632. PubMed doi:10.1249/01.mss.0000177455.58960.aa 16. Macfarlane DJ, Chan D, Chan KL, Ho EY, Lee CC. Using three objective criteria to examine pedometer guidelines for free-living individuals. Eur J Appl Physiol. 2008;104:435–444. PubMed doi:10.1007/ s00421-008-0789-4 17. Miller R, Brown W. Meeting physical activity guidelines and average daily steps in a working population. J Phys Act Health. 2004;1:218– 226. 18. Tudor-Locke C, Ainsworth BE, Thompson RW, Matthews CE. Comparison of pedometer and accelerometer measures of free-living physical activity. Med Sci Sports Exerc. 2002;34:2045–2051. PubMed doi:10.1097/00005768-200212000-00027
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
1372 Cao et al 19. Oshima Y, Hikihara Y, Ohkawara K, et al. Daily steps corresponding to the reference quantity of physical activity of Exercise and Physical Activity Reference for Health Promotion 2006 (EPAR2006) assessed by accelerometer. Jpn J Phys Fitness Sports Med. 2012;61:193–199 (in Japanese). doi:10.7600/jspfsm.61.193 20. Tudor-Locke C, Leonardi C, Johnson WD, Katzmarzyk PT, Church TS. Accelerometer steps/day translation of moderate-to-vigorous activity. Prev Med. 2011;53:31–33. PubMed doi:10.1016/j.ypmed.2011.01.014 21. Murakami H, Kawakami R, Ohmori Y, Miyatake N, Morita A, Miyachi M. Translating from 23METs-h/wk as physical activity reference value for Japanese to daily step counts. Jpn J Phys Fitness Sports Med. 2012;61:183–191 (in Japanese). doi:10.7600/jspfsm.61.183 22. Tudor-Locke C, Craig CL, Brown WJ, et al. How many steps/day are enough? For adults. Int J Behav Nutr Phys Act. 2011;8:79. PubMed doi:10.1186/1479-5868-8-79 23. Cao ZB, Miyatake N, Higuchi M, Miyachi M, Ishikawa-Takata K, Tabata I. Predicting VO2max with an objectively measured physical activity in Japanese women. Med Sci Sports Exerc. 2010;42:179–186. PubMed doi:10.1249/MSS.0b013e3181af238d 24. Schneider PL, Crouter SE, Bassett DR. Pedometer measures of free-living physical activity: comparison of 13 models. Med Sci Sports Exerc. 2004;36:331–335. PubMed doi:10.1249/01.MSS.0000113486.60548. E9 25. Arvidsson D, Fitch M, Hudes ML, Tudor-Locke C, Fleming SE. Accelerometer response to physical activity intensity in normal-weight versus overweight African American children. J Phys Act Health. 2011;8:682–692. PubMed 26. Kumahara H, Schutz Y, Ayabe M, et al. The use of uniaxial accelerometry for the assessment of physical-activity-related energy expenditure: a validation study against whole-body indirect calorimetry. Br J Nutr. 2004;91:235–243. PubMed doi:10.1079/BJN20031033 27. Wildman RP, Gu D, Reynolds K, Duan X, He J. Appropriate body mass index and waist circumference cutoffs for categorization of overweight and central adiposity among Chinese adults. Am J Clin Nutr. 2004;80:1129–1136. PubMed
28. Zhu S, Wang Z, Heshka S, Heo M, Faith MS, Heymsfield SB. Waist circumference and obesity-associated risk factors among whites in the third National Health and Nutrition Examination Survey: clinical action thresholds. Am J Clin Nutr. 2002;76:743–749. PubMed 29. Crouter SE, Schneider PL, Bassett DR, Jr. Spring-levered versus piezo-electric pedometer accuracy in overweight and obese adults. Med Sci Sports Exerc. 2005;37:1673–1679. PubMed doi:10.1249/01. mss.0000181677.36658.a8 30. Le Masurier GC, Tudor-Locke C. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med Sci Sports Exerc. 2003;35:867–871. PubMed doi:10.1249/01. MSS.0000064996.63632.10 31. Ayabe M, Ishii K, Takayama K, Aoki J, Tanaka H. Comparison of interdevice measurement difference of pedometers in younger and older adults. Br J Sports Med. 2010;44:95–99. PubMed doi:10.1136/ bjsm.2007.045179 32. Inoue S, Ohya Y, Odagiri Y, et al. Sociodemographic determinants of pedometer-determined physical activity among Japanese adults. Am J Prev Med. 2011;40:566–571. PubMed doi:10.1016/j. amepre.2010.12.023 33. Ministry of Health Labour and Welfare of Japan. Exercise and physical activity reference for health promotion: physical activity. Exercise, and Physical Fitness; 1993. (in Japanese) 34. Tudor-Locke C, Hatano Y, Pangrazi RP, Kang M. Revisiting “how many steps are enough?”. Med Sci Sports Exerc. 2008;40:S537–S543. PubMed doi:10.1249/MSS.0b013e31817c7133 35. Cao ZB, Miyatake N, Higuchi M, Miyachi M, Tabata I. Predicting VO2max with an objectively measured physical activity in Japanese men. Eur J Appl Physiol. 2010;109:465–472. PubMed doi:10.1007/ s00421-010-1376-z 36. Colley RC, Janssen I, Tremblay MS. Daily step target to measure adherence to physical activity guidelines in children. Med Sci Sports Exerc. 2012;44:977–982. PubMed doi:10.1249/MSS.0b013e31823f23b1 37. Ayabe M, Brubaker PH, Dobrosielski D, et al. Target step count for the secondary prevention of cardiovascular disease. Circ J. 2008;72:299– 303. PubMed doi:10.1253/circj.72.299
Official Journal of ISPAH www.JPAH-Journal.com ORIGINAL RESEARCH
Steps Per Day Required for Meeting Physical Activity Guidelines in Japanese Adults Zhen-Bo Cao, Taewoong Oh, Nobuyuki Miyatake, Kazuyo Tsushita, Mitsuru Higuchi, and Izumi Tabata Background: This study aimed to determine the optimal number of steps per day needed to meet the current physical activity guidelines in a large population sample of Japanese adults. Methods: An accelerometer-based activity monitor (Kenz Lifecorder) was used to simultaneously measure moderate-to-vigorous intensity physical activity (MVPA) and step counts in 940 Japanese adults (480 women) aged 20 to 69 years. The step count per day equivalents to 2 different physical activity recommendations (23 MET-h/wk and 150 min/wk of MVPA) were derived using linear regression analysis and receiver operating characteristic (ROC) methodology. Results: Linear regression analysis showed that daily step counts correlated with weekly PAEE (r = .83) and daily minutes of MVPA (r = .83). Linear regression analysis also showed that 23 MET-h/wk of MVPA is equivalent to 11,160 steps/d, and 150 min/wk of MVPA is equivalent to 7716 steps/d. ROC analysis yielded similar findings: 10,225 steps/d are required to accumulate ≥ 23 MET-h/wk of MVPA and 7857 steps/d are needed to meet the recommendation of ≥ 150 min/wk of MVPA. Conclusions: The findings suggest that 10,000 to 11,000 and 7700 to 8000 steps/d represent the optimal thresholds for accumulating ≥ 23 MET-h/wk of MVPA and ≥ 150 min/wk of MVPA, respectively, for Japanese adults. Keywords: accelerometer, energy expenditure, pedometer Physical inactivity is a well-documented risk factor for coronary heart disease and is associated with increased risk for other chronic diseases, including obesity, Type II diabetes mellitus, hypertension, colon cancer, depression, and osteoporosis.1 Accordingly, promoting physical activity has become a public health priority worldwide,2 and numerous physical activity guidelines have been published by government agencies, professional organizations, and associations to provide information and guidance on the types and amounts of physical activity sufficient to offer substantial health benefits.3–8 Usually, physical activity guidelines are expressed in terms of time or energy expenditure (metabolic equivalent, MET). A number of international and national health authorities have produced similar guidelines recommending that adults accumulate at least 150 min of moderate-to-vigorous intensity physical activity (MVPA) per week.3–7 In 2006, the Ministry of Health, Labor and Welfare of Japan revised its previous physical activity guidelines and recommended that Japanese adults should engage in a minimum of 23 MET-h per week of MVPA,8 which is more than twice the volume of activity in the recommendation of 150 min/wk of MVPA.3 Recently, steps per day, a simple, useful indicator of physical activity volume, widely accepted by researchers, practitioners, and the general public, has been used in large, nationally representative surveillance studies in Japan,9 Canada,10 and the USA11 and in physical activity intervention research.12,13 If current physical activity guidelines can be translated into this indicator, steps per day would
be a useful additional measurement for surveillance of adherence to these recommendations. Some studies have tried to determine how many steps per day equate to current MVPA guidelines in freeliving samples.14–22 However, most of the studies conducted to date did not use 2 classic analytic approaches, linear regression analysis and receiver operating characteristic (ROC) methodology, nor did they assess classification accuracy of the step count thresholds. Furthermore, many of these prior studies14–19 have used relatively small and/or selected samples such as college students, postmenopausal women, or workplace employees. Only 2 studies used large population samples.20,21 To the best of our knowledge, no study has used both linear regression analysis and ROC methodology to determine the step-based translations of 23 MET-h/wk of MVPA and 150 min/wk of MVPA in Japanese adults. The purposes of this study, therefore, were to examine the relationship between accelerometer-assessed step count per day and MVPA and to determine the optimal number of steps per day needed to meet current public health recommendations for weekly physical activity (23 MET-h/wk of MVPA and 150 min/wk of MVPA) by using linear regression analysis and ROC methodology in a large population sample of Japanese adults.
Cao ([email protected]) is with School of Kinesiology, Shanghai University of Sport, Shanghai, China. Higuchi is with the Faculty of Sport Sciences, Waseda University, Tokorozawa, Japan. Oh is with the Faculty of Human Health Science, Matsumoto University, Matsumoto, Japan. Miyatake is with the Dept of Hygiene, Faculty of Medicine, Kagawa University, Miki, Japan. Tsushita is with the Comprehensive Health Science Center, Aichi Health Promotion Foundation, Higashiura, Japan. Tabata is with the Faculty of Sport and Health Science, Ritsumeikan University, Kusatsu, Japan.
This cross-sectional study included 940 Japanese adults (480 women) aged 20 to 69 years who participated in the Exercise and Physical Activity Reference and Guide for Health Promotion Study (EPARGHP), which was designed to investigate Japanese levels of physical activity and cardiorespiratory fitness and the association of these levels with other risk factors for lifestyle-related diseases. Data were collected from June 2007 to November 2009 in 4 Japanese cities (Tokyo, Chita-gun, Okayama, and Matsumoto). Study
Methods Subjects
1367
1368 Cao et al
procedures were approved by the Ethical Committee of the National Institutes of Health and Nutrition. Written informed consent was obtained for all participants.
Anthropometrics Body mass was measured to the nearest 0.1 kg and height was measured to the nearest 0.1 cm. Body mass index (BMI) was calculated according to the formula [body mass (kg)/height squared (m2)]. Waist circumference (WC) was measured at the umbilical level with a nonelastic tape to the nearest 0.1 cm.23
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Physical Activity The Kenz Lifecorder PLUS (LC; Suzuken Co. Ltd., Nagoya, Japan), an accelerometer-based activity monitor, was used to record participants’ step count data along with time and physical activity-related energy expenditure (PAEE) spent in MVPA for 7 consecutive days. All participants were asked to engage in their normal daily pattern of physical activity. The average steps per day (steps/d), accumulated MVPA per day (min/d), and PAEE per week (MET-h/wk) were calculated using data from at least 10 h per day for 7 days (including 2 weekend days), without consideration of MVPA bout length. Previous investigations24–26 have demonstrated that the LC provides a valid assessment of step count, as well as physical activity intensity and PAEE. Additional details of the methodology and equipment have been published elsewhere.23,26
Statistical Analyses Gender differences were evaluated with a Student’s t test (normal distribution) or Mann-Whitney U test (nonnormal distribution). Linear regression analysis was used to estimate relationships between step count per day and MVPA and PAEE. The optimal number of steps/d required to meet the different physical activity recommendations was determined using 2 analytic approaches:
linear regression analysis and ROC curves. In the ROC analysis, we considered meeting or not meeting the different MVPA recommendations (23 MET-h/wk or 150 min/wk) as an outcome variable and step count per day as a testing variable. The distance between each point on the ROC curve and the upper left corner was calculated as the square root of [(1 – sensitivity)2 + (1 – specificity)2], and the point with the shortest distance was judged to correspond to the optimal cutoff value.27,28 Furthermore, we calculated the area under the curve (AUC) and its 95% confidence interval (CI). The optimal step counts thresholds derived from linear regression analysis or ROC analysis were examined for classification accuracy (number of correct classifications/number of all classifications) by dividing the entire sample into subgroups stratified by sex and age (men: 20–29 y [n = 89], 30–39 y [n = 98], 40–49 y [n = 89], 50–59 y [n = 98], and 60–69 y [n = 86]; women: 20–29 y [n = 80], 30–39 y [n = 96], 40–49 y [n = 100], 50–59 y [n = 107], and 60–69 y [n = 97]). All analyses were conducted with SPSS 16.0 J for Windows (SPSS Japan Inc., Tokyo, Japan). Statistical significance was considered to be P < .05.
Results Subject characteristics are shown in Table 1. The mean age (standard deviation: SD) was 44.7 (13.9) years. The amount spent in MVPA included both minutes and PAEE were higher in men than in women, but no significant difference in step count per day existed between men and women. The percentage of subjects meeting the recommendation of 150 min/wk of MVPA was 62.6% in men and 54.8% in women; 18.5% of men and 15.0% of women met the recommendation of 23 MET-h/wk of MVPA. Linear regression analysis yielded the following equations for predicting the number of steps per day required to meet the 2 different MVPA recommendations: steps/d = 4702 + 280.8 × PAEE (R2 = .70, P < .001) and steps/d = 4356 + 22.4 × MVPA/wk (R2 = .70, P < .001). Using these equations, the recommendation of 23 MET-h/wk
Table 1 Physical Characteristics of Study Participants Variable
Total
Men
Women
n = 940
n = 460
n = 480
Age (y)
44.7 (13.9)
44.2 (14.0)
45.2 (13.7)
Height (cm)
163.8 (8.5)
170.2 (5.9)
157.6 (5.5)*
Body mass (kg)
59.3 (10.5)
66.3 (8.6)
52.5 (7.1)*
BMI (kg/m2)
22.0 (2.9)
22.9 (2.8)
21.1 (2.8)*
%Fat (%)
23.2 (6.8)
19.2 (5.2)
27.1 (5.7)*
78.8 (8.7)
81.4 (8.1)
76.3 (8.6)*
Step count (steps/d)
WC (cm)
8776 (3204)
8793 (3377)
8760 (3032)
MVPA (min/d)
28.2 (17.0)
29.7 (18.1)
26.8 (15.9)*
PAEE (MET-h/wk)
14.5 (9.5)
15.2 (10.0)
13.9 (8.9)*
150 min/wk of MVPA
551 (58.6)
288 (62.6)
263 (54.8)
23 MET-h/wk of MVPA
157 (16.7)
85 (18.5)
72 (15.0)
Number (%) of participants meeting physical activity recommendations
Note. Values are expressed as mean (SD) or number (%). Abbreviations: BMI, body mass index; %Fat, body fat percentage; WC, waist circumference; MVPA, moderate-to-vigorous physical activity; PAEE, physical activity-related energy expenditure spent in MVPA; MET, metabolic equivalents. * Significantly different from men, P < .05.
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Steps/Day for Meeting Physical Activity Guidelines 1369
of MVPA corresponded to 11,160 steps/d, and the recommendation of 150 min/wk of MVPA corresponded to 7716 steps/d (Table 2). ROC analysis revealed that 10,225 steps/d produced 0.92 sensitivity and 0.93 specificity for accumulating ≥ 23 MET-h/wk of MVPA (AUC = 0.93) and that 7857 steps/d produced 0.86 sensitivity and 0.90 specificity for accumulating ≥ 150 min/wk of MVPA (AUC = 0.91; Table 2). According to the ROC curve, the optimal steps/d thresholds for likelihood of accumulating ≥ 23 MET-h/wk of MVPA, or ≥ 150 min/wk of MVPA were 10,270 steps/d (sensitivity 0.92, specificity 0.83, AUC 0.93) and 7850 steps/d (sensitivity 0.85, specificity 0.83, AUC 0.92), respectively, for men and 10,660 steps/d (sensitivity 0.90, specificity 0.88, AUC 0.94) and 8065 steps/d (sensitivity 0.84, specificity 0.80, AUC 0.89), respectively, for women. The accuracy of daily step count thresholds derived from the 2 analytic approaches for the 2 different MVPA recommendations for subgroups stratified by sex and age are shown in Tables 3 and 4. The accuracy of steps/d thresholds to screen for accumulating ≥ 23 MET-h/wk of MVPA derived from linear regression analysis appears to be higher than the accuracy of steps/d thresholds derived from ROC analysis. However, the accuracy of steps/d thresholds to screen for accumulating ≥ 150 min/wk of MVPA derived from linear regression analysis was similar to the accuracy of steps/d thresholds derived from ROC analysis in all age groups.
Discussion This study investigated the relationship between steps/d and MVPA and the optimal number of steps/d needed to meet current public health recommendations for weekly physical activity. Using data from a large population sample of Japanese adults, we demonstrated that steps/d is positively associated with MVPA in terms of time (r = .83, P < .001) and energy expenditure (PAEE: r = .83, P < .001). Our findings from 2 different analytical models suggest that 10,000 to 11,000 and 7700 to 8000 steps/d represent the optimal thresholds for likelihood of accumulating ≥ 23 MET-h/wk of MVPA and ≥ 150 min/wk of MVPA, respectively. These thresholds are supported by good classification accuracy in all age subgroups. Participants in the current study appeared to take more steps/d than those in a Japanese National Health and Nutrition Survey9 conducted in 2007, using a spring-levered pedometer (AS-200, Yamasa Co. Ltd., Tokyo, Japan). The Japanese national survey reported that men and women took 7321 steps/d and 6267 steps/d, respectively.9 However, these findings do not necessarily mean that participants in the current study were more active compared with those in the 2007 survey, because differences in physical activity monitoring devices may explain some of this discrepancy. Studies have shown that accelerometer-based activity monitors are less affected by adi-
Table 2 Step Count per Day for Meeting 2 Different MVPA Recommendations Using 2 Analytic Approaches Step count per day for meeting MVPA recommendations Physical activity guidelines
n
Regression analysis
ROC curve analysis
23 MET-h/wk of MVPA All
940
11,160
10,225
Men
460
10,985
10,270
Women
480
11,358
10,660
All
940
7716
7857
Men
460
7523
7850
Women
480
7905
8065
150 min/wk of MVPA
Abbreviations: MVPA, moderate-to-vigorous physical activity; MET, metabolic equivalent; ROC, receiver operating characteristic.
Table 3 Accuracy of Step Count per Day Cut-Off Points Derived From Linear Regression Analysis for 2 MVPA Recommendations for Subgroups Stratified by Age Accuracy n
20–29 y
30–39 y
40–49 y
50–59 y
60–69 y
20–69 y
11,160 steps/d (23 MET-h/wk of MVPA) All
940
0.93
0.87
0.87
0.88
0.85
0.88
Men
460
0.91
0.86
0.83
0.87
0.85
0.86
Women
480
0.93
0.88
0.91
0.88
0.88
0.89
All
940
0.82
0.80
0.84
0.81
0.85
0.83
Men
460
0.87
0.82
0.91
0.80
0.87
0.85
Women
480
0.75
0.78
0.82
0.86
0.88
0.82
7716 steps/d (150 min/wk of MVPA)
Abbreviations: y, years; MVPA, moderate-to-vigorous physical activity; MET, metabolic equivalent; Accuracy = (Number of correct assessments)/(Number of all assessments).
1370 Cao et al
Table 4 Accuracy of Step Count per Day Cut-Off Points Derived From ROC Analysis for 2 MVPA Recommendations for Subgroups Stratified by Age Accuracy n
20–29 y
30–39 y
40–49 y
50–59 y
60–69 y
20–69 y
All
940
0.89
0.83
0.82
0.82
0.84
0.85
Men
460
0.87
0.86
0.79
0.87
0.84
0.84
Women
480
0.93
0.85
0.84
0.79
0.85
0.85
All
940
0.82
0.79
0.86
0.83
0.86
0.83
Men
460
0.88
0.80
0.90
0.81
0.86
0.85
Women
480
0.76
0.78
0.82
0.86
0.87
0.82
10,225 steps/d (23 MET-h/wk of MVPA)
7857 steps/d (150 min/wk of MVPA)
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Abbreviations: y, years; MVPA, moderate-to-vigorous physical activity; MET, metabolic equivalent; ROC, receiver operating characteristic; Accuracy = (Number of correct assessments)/(Number of all assessments).
posity and more sensitive to low force accelerations (such as slow stepping) than spring-levered pedometers.29,31 For example, Crouter et al29 found that the Yamax (Yamasa) pedometer undercounted steps in overweight, and obese individuals, whereas an accelerometerbased step counter (New Lifestyles NL-2000, Inc, Lees Summit, MO, USA) was more accurate. Le Masurier and Tudor-Locke30 found that the spring-levered pedometer detected fewer steps than the accelerometer at slow walking speed. Furthermore, Ayabe et al31 reported that the step counts determined by a spring-levered pedometer (EC-200, Yamasa Co. Ltd., Tokyo, Japan) were about 1400 steps/d less than those determined by an accelerometer-based activity monitor (Kenz Lifecorder) in Japanese adults. Recently, Inoue et al32 studied physical activity levels of a large population sample of adults (n = 790) by using the Kenz Lifecorder in 4 Japanese cities (Koganei, Tsukuba, Shizuoka, and Kagoshima). They reported that men and women took 8763 steps/d and 8242 steps/d, respectively, which is similar to our results. Therefore, the physical activity levels of participants in the current study are likely to be similar to those of the general Japanese urban population. Although previous physical activity guidelines in Japan have called for taking 10,000 steps/d (an amount equivalent to at least 20 to 30 min of moderate-intensity walking or other exercise most days of the week),33 more recent Japanese physical activity guidelines encourage at least 23 MET-h/wk of MVPA.8 The report from the Ministry of Health, Labor and Welfare of Japan33 indicates that 23 MET-h/wk of MVPA is approximately equivalent to 60 min of MVPA or 8000 to 10,000 steps/d. This recommended step count was based on an assumed daily level in Japanese adults of 2000 to 4000 steps/d of low intensity unconscious activity (< 3 METs), with each additional 60 min of MVPA adding 6000 steps. One previous study21 in a large sample of Japanese adults has attempted to determine daily step counts measured with a triaxial accelerometer (Actimarker EW4800; Panasonic Electric Works, Japan) that classify subjects as meeting the 23 MET-h/wk of MVPA. ROC methodology showed that 8500 to 10,000 steps/d, a number similar to current recommendations, was indicated as the optimal daily step count for achieving 23 MET-h/wk of MVPA.21 However, the activity monitor used in their study undercounted steps at slow walking speeds, and its production and sales have been suspended. In addition, subjects in their study had high levels of physical activity (subjects walked an average of 9600 steps/d, and approximately 48% of subjects achieved ≥ 23 MET-h/wk of MVPA) compared with the
general population (men in the general population take 7321 steps/d, and women take 6267 steps/d).9,21 In the current study, we found that current recommended step counts (8000 to 10,000 steps/d) in Japan are equivalent to only 12 to 19 MET-h/wk, falling short of the recommended 23 MET-h/wk of MVPA. We also found that the current recommended 8000 to 10,000 steps/d actually overestimate the proportion of our subjects who are meeting physical activity guideline (23 MET-h/wk of MVPA) up to 17% to 40%. This suggests that the current recommended step counts for Japanese adults may be largely overestimating the proportion of Japanese adults who are meeting physical activity guideline. Such an overestimation could be dangerous because it could lead individuals to underestimate their own health risks. Approximately 11,000 steps/d, corresponding to the “active” level of a graduated step index for healthy adults,34 are needed for Japanese adults to obtain at least 23 MET-h/wk of MVPA in this study. This threshold of 11,000 steps per day appears to have high classification accuracy for both men (83% to 91%) and women (83% to 89%) across all age groups. Recently, some international and national health guidelines have recommended that adults should accumulate at least 150 min/ wk of MVPA.3–7 A limited number of studies have investigated the number of steps per day needed to meet this guideline.17,20 To our knowledge, only 1 study has attempted to determine a direct step counts per day translation of 150 min/wk of accumulated MVPA. Tudor-Locke and colleagues20 measured steps and MVPA with the ActiGraph accelerometer in a large sample of US adults and analyzed the data with a linear regression model.20 They reported a strong linear relationship between step counts and MVPA and showed that 150 min/wk of MVPA translated to approximately 7000 steps/d (or 49,000 steps/wk).20 However, they did not evaluate the accuracy of their step count threshold in identifying accumulation of 150 min/wk of MVPA. Using linear regression analysis and ROC methodology, we found that 7700 to 8000 steps/d represented the optimal threshold for likelihood of accumulating ≥ 150 min/wk of MVPA in Japanese adults. To estimate the step count threshold’s performance, we also assessed classification accuracy by dividing the entire sample into subgroups stratified by sex and age. The results show a high correct classification rate for both men (79% to 89%) and women (74% to 86%) in all age groups, as well as higher classification accuracies across incremental age groups in women, but not in men. Slightly higher classification accuracy was observed in young and middleaged men and older women as compared with age-matched, opposite-
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
Steps/Day for Meeting Physical Activity Guidelines 1371
sex groups. Our results provide evidence for the practical use of the step count threshold determined in the current study. In this population, Steps per day correlated strongly with time spent in MVPA (r = .830). Although few studies have investigated these associations, those existing have found similar results in healthy adult men35 (r = .84), children36 (r = .81), and cardiac rehabilitation patients37 (r = .85). A similar correlation was also seen between steps per day and PAEE spent in MVPA (r = .83). Considering PAEE was calculated by multiplying intensity of MVPA by time spent in MVPA, it is not surprising to find that there is an agreement relation between steps per day and the levels of MVPA expressed in terms of time or PAEE. This implicated that time spent in MVPA explained a large proportion of the variance in total PA expressed in term of steps per day, while intensity of MVPA did not account for an additional variance in total PA in this population. The amount of physical activity recommended by the Japanese physical activity guideline of 23 MET-h/wk of MVPA is more than twice the volume of activity in the 150 min/wk of MVPA recommendation. However, the magnitude of the difference was reduced after generating step-based translations of the 2 guidelines in this study. This discrepancy may be due in part to the fact that, unlike duration- and intensity-based physical activity guidelines that focus on MVPA, step-based translations take into account the total daily volume of ambulatory physical activity, including MVPA and light-intensity physical activity (< 3 METs). In the current study, participants engaged in about 2 hours per day of light-intensity physical activity, which is equivalent to 3500 steps. After removing 3500 steps/d from the step-based translations of the duration- and intensity-based physical activity guidelines, the magnitude of the difference for the 2-step count thresholds is similar to that for the 2 current MVPA recommendations. There are some limitations to our study. First, the current sample did not include older adults (aged 70 years and over) and included primarily individuals living in urban settings. Thus, our results may have limited generalizability. Second, our calculation of MVPA did not take into consideration MVPA bout length, even though current physical activity guidelines recommend that MVPA needs to be accumulated in bouts of at least 10 min. Despite these limitations, the current study has several strengths, including a large population sample of Japanese men and women, the use of a valid and reliable accelerometer-based activity monitor to concurrently quantify steps/d and MVPA, the use of both linear regression analysis and ROC methodology, and an assessment of classification accuracy of the step count threshold. In conclusion, our results suggest that 10,000 to 11,000 and 7700 to 8000 steps/d represented the optimal thresholds for accumulating ≥ 23 MET-h/wk of MVPA and ≥ 150 min/wk of MVPA, respectively, for Japanese adults. Acknowledgments This research was supported by a grant for Comprehensive Research on Cardiovascular and Life-Style Related Diseases from the Ministry of Health, Labour and Welfare, Japan (no.19160101) and by a grant-in-aid for the Global COE, Waseda University “Sport Sciences for the Promotion of Active Life” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan (no. A09338100).
References 1. Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence. CMAJ. 2006;174:801–809. PubMed doi:10.1503/ cmaj.051351
2. World Health Organization. Global status report on noncommunicable diseases 2010. Geneva: WHO Press; 2011. 3. Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc. 2007;39:1423–1434. PubMed doi:10.1249/ mss.0b013e3180616b27 4. World Health Organization. Global recommendations on physical activity for health. Geneva: WHO Press; 2010. 5. Tremblay MS, Warburton DE, Janssen I, et al. New Canadian physical activity guidelines. Appl Physiol Nutr Metab. 2011;36:36–46, 47–58. PubMed doi:10.1139/H11-009 6. Australian Government Department of Health and Ageing. Physical Activity Guidelines for Adults. [http://www.health.gov.au/internet/ main/publishing.nsf/Content/health-pubhlth-strateg-phys-actguidelines#guidelines_adults]. 7. Department of Health. UK Physical Activity Guidelines. [http://www. dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_127931]. 8. Ministry of Health Labour and Welfare of Japan. Exercise and Physical Activity Reference for Health Promotion 2006. [http://www.nih. go.jp/eiken/programs/pdf/epar2006.pdf.]. 9. Inoue S, Ohya Y, Tudor-Locke C, Tanaka S, Yoshiike N, Shimomitsu T. Time trends for step-determined physical activity among Japanese adults. Med Sci Sports Exerc. 2011;43:1913–1919. PubMed doi:10.1249/MSS.0b013e31821a5225 10. Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS. Physical activity of Canadian adults: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep. 2011;22:7–14. PubMed 11. Tudor-Locke C, Johnson WD, Katzmarzyk PT. Accelerometerdetermined steps per day in US adults. Med Sci Sports Exerc. 2009;41:1384–1391. PubMed doi:10.1249/MSS.0b013e318199885c 12. Petersen CB, Severin M, Hansen AW, Curtis T, Gronbaek M, Tolstrup JS. A population-based randomized controlled trial of the effect of combining a pedometer with an intervention toolkit on physical activity among individuals with low levels of physical activity or fitness. Prev Med. 2012;54:125–130. PubMed doi:10.1016/j. ypmed.2011.12.012 13. Freak-Poli R, Wolfe R, Backholer K, de Courten M, Peeters A. Impact of a pedometer-based workplace health program on cardiovascular and diabetes risk profile. Prev Med. 2011;53:162–171. PubMed doi:10.1016/j.ypmed.2011.06.005 14. Behrens TK, Hawkins SB, Dinger MK. Relationship between objectively measured steps and time spent in physical activity among freeliving college students. Meas Phys Educ Exerc Sci. 2005;9:67–77. doi:10.1207/s15327841mpee0902_1 15. Jordan AN, Jurca GM, Tudor-Locke C, Church TS, Blair SN. Pedometer indices for weekly physical activity recommendations in postmenopausal women. Med Sci Sports Exerc. 2005;37:1627–1632. PubMed doi:10.1249/01.mss.0000177455.58960.aa 16. Macfarlane DJ, Chan D, Chan KL, Ho EY, Lee CC. Using three objective criteria to examine pedometer guidelines for free-living individuals. Eur J Appl Physiol. 2008;104:435–444. PubMed doi:10.1007/ s00421-008-0789-4 17. Miller R, Brown W. Meeting physical activity guidelines and average daily steps in a working population. J Phys Act Health. 2004;1:218– 226. 18. Tudor-Locke C, Ainsworth BE, Thompson RW, Matthews CE. Comparison of pedometer and accelerometer measures of free-living physical activity. Med Sci Sports Exerc. 2002;34:2045–2051. PubMed doi:10.1097/00005768-200212000-00027
Downloaded by Alderman Library on 09/22/16, Volume 11, Article Number 7
1372 Cao et al 19. Oshima Y, Hikihara Y, Ohkawara K, et al. Daily steps corresponding to the reference quantity of physical activity of Exercise and Physical Activity Reference for Health Promotion 2006 (EPAR2006) assessed by accelerometer. Jpn J Phys Fitness Sports Med. 2012;61:193–199 (in Japanese). doi:10.7600/jspfsm.61.193 20. Tudor-Locke C, Leonardi C, Johnson WD, Katzmarzyk PT, Church TS. Accelerometer steps/day translation of moderate-to-vigorous activity. Prev Med. 2011;53:31–33. PubMed doi:10.1016/j.ypmed.2011.01.014 21. Murakami H, Kawakami R, Ohmori Y, Miyatake N, Morita A, Miyachi M. Translating from 23METs-h/wk as physical activity reference value for Japanese to daily step counts. Jpn J Phys Fitness Sports Med. 2012;61:183–191 (in Japanese). doi:10.7600/jspfsm.61.183 22. Tudor-Locke C, Craig CL, Brown WJ, et al. How many steps/day are enough? For adults. Int J Behav Nutr Phys Act. 2011;8:79. PubMed doi:10.1186/1479-5868-8-79 23. Cao ZB, Miyatake N, Higuchi M, Miyachi M, Ishikawa-Takata K, Tabata I. Predicting VO2max with an objectively measured physical activity in Japanese women. Med Sci Sports Exerc. 2010;42:179–186. PubMed doi:10.1249/MSS.0b013e3181af238d 24. Schneider PL, Crouter SE, Bassett DR. Pedometer measures of free-living physical activity: comparison of 13 models. Med Sci Sports Exerc. 2004;36:331–335. PubMed doi:10.1249/01.MSS.0000113486.60548. E9 25. Arvidsson D, Fitch M, Hudes ML, Tudor-Locke C, Fleming SE. Accelerometer response to physical activity intensity in normal-weight versus overweight African American children. J Phys Act Health. 2011;8:682–692. PubMed 26. Kumahara H, Schutz Y, Ayabe M, et al. The use of uniaxial accelerometry for the assessment of physical-activity-related energy expenditure: a validation study against whole-body indirect calorimetry. Br J Nutr. 2004;91:235–243. PubMed doi:10.1079/BJN20031033 27. Wildman RP, Gu D, Reynolds K, Duan X, He J. Appropriate body mass index and waist circumference cutoffs for categorization of overweight and central adiposity among Chinese adults. Am J Clin Nutr. 2004;80:1129–1136. PubMed
28. Zhu S, Wang Z, Heshka S, Heo M, Faith MS, Heymsfield SB. Waist circumference and obesity-associated risk factors among whites in the third National Health and Nutrition Examination Survey: clinical action thresholds. Am J Clin Nutr. 2002;76:743–749. PubMed 29. Crouter SE, Schneider PL, Bassett DR, Jr. Spring-levered versus piezo-electric pedometer accuracy in overweight and obese adults. Med Sci Sports Exerc. 2005;37:1673–1679. PubMed doi:10.1249/01. mss.0000181677.36658.a8 30. Le Masurier GC, Tudor-Locke C. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med Sci Sports Exerc. 2003;35:867–871. PubMed doi:10.1249/01. MSS.0000064996.63632.10 31. Ayabe M, Ishii K, Takayama K, Aoki J, Tanaka H. Comparison of interdevice measurement difference of pedometers in younger and older adults. Br J Sports Med. 2010;44:95–99. PubMed doi:10.1136/ bjsm.2007.045179 32. Inoue S, Ohya Y, Odagiri Y, et al. Sociodemographic determinants of pedometer-determined physical activity among Japanese adults. Am J Prev Med. 2011;40:566–571. PubMed doi:10.1016/j. amepre.2010.12.023 33. Ministry of Health Labour and Welfare of Japan. Exercise and physical activity reference for health promotion: physical activity. Exercise, and Physical Fitness; 1993. (in Japanese) 34. Tudor-Locke C, Hatano Y, Pangrazi RP, Kang M. Revisiting “how many steps are enough?”. Med Sci Sports Exerc. 2008;40:S537–S543. PubMed doi:10.1249/MSS.0b013e31817c7133 35. Cao ZB, Miyatake N, Higuchi M, Miyachi M, Tabata I. Predicting VO2max with an objectively measured physical activity in Japanese men. Eur J Appl Physiol. 2010;109:465–472. PubMed doi:10.1007/ s00421-010-1376-z 36. Colley RC, Janssen I, Tremblay MS. Daily step target to measure adherence to physical activity guidelines in children. Med Sci Sports Exerc. 2012;44:977–982. PubMed doi:10.1249/MSS.0b013e31823f23b1 37. Ayabe M, Brubaker PH, Dobrosielski D, et al. Target step count for the secondary prevention of cardiovascular disease. Circ J. 2008;72:299– 303. PubMed doi:10.1253/circj.72.299
