Recent 7-d pedometer data from a nationally representative sample of 6000 Canadian children and youth aged 5-19 yr indicate that 73%-91% do not accumulate sufficient daily steps (5). Likewise, the recent National Health and Nutrition Examination Survey data show that 58% of American children aged 6-11 yr and 92% of adolescents aged 12-19 yr are not meeting the recommended 60 min·d−1 of physical activity (33). Although these data suggest that many children and youth are inactive, they do not allow us to determine whether this was always the case or if physical activity has declined over time. Knowing how physically active children were when childhood obesity was rare may offer insight into obesity treatment and/or prevention. Unfortunately, longitudinal physical activity data on nationally representative samples of children and adolescents are lacking (17). However, data from questionnaire and time use studies may provide some insight into physical activity trends. For example, Canadian data show no change in leisure-time physical activity from 1981 to 1998 (7). These data are in line with more recent US data from the National Youth Risk Behavior Surveys that found no significant temporal trends in physical activity and physical inactivity between 1999 and 2005 (17).
In contrast to these trend data, evidence suggests that US children and youth walk and cycle less for transportation with active trips to school decreasing from 20.2% in 1977 to 12.5% in 2001 (29). Further, US time use data suggest that increased time spent in school, child care, studying, and reading have substantially decreased play and discretionary time from 1981 to 1997 (28). However, these conflicting data are based on proxy ecological evidence that does not take into account temporal changes in a certain domain (e.g., reductions in active commuting), which may be counterbalanced by opposing changes in another domain (e.g., increases in sports participation) (27). Moreover, it is generally felt that the major impact of technology on sedentary behavior was reasonably complete by the 1950s because of the phasing out of heavy manual labor (11). Although further reductions in physical activity were expected because of advances in computer and communication technology, it was believed they would be much more subtle. Until more robust measures of physical activity are integrated into public health surveillance systems, our ability to monitor trends accurately will continue to be severely limited (18).
In an effort to overcome the gap in historical trend data, Tremblay et al. (30) developed a unique model to assess the impact of modernity on activity levels in children. The authors reasoned that some cultures may preserve the inherently active lifestyle of earlier generations that preceded erosions to leisure time and occupational physical activity and the childhood obesity epidemic. Their data show that Old Order Mennonite (OOM) children are leaner, stronger, and more active than contemporary-living children from both rural and urban communities. Extending the model to look further back in time, Bassett et al. (2) found that Old Order Amish (OOA) children take more steps per day and have lower rates of overweight and obesity than contemporary-living children. Although much has been learned about the physical activity of OOA and OOM children, only the most basic variables have been explored to date (i.e., average moderate to vigorous physical activity [MVPA] per day and average steps per day). Although quantitative information is useful for informing evidence-based physical activity guidelines (i.e., how much physical activity one should do to be healthy), it offers no insights into how one might go about achieving the recommended levels of physical activity (4).
To fully understand the quantity and the quality of physical activity of these traditional groups, further work is needed to examine their activity patterns (i.e., specific intensities and timing of activity/inactivity) (1). Therefore, the purpose of this article was to profile the physical activity and inactivity behaviors of OOA, OOM, and contemporary-living children as a means of assessing the influence of lifestyle. Hypothesis 1 was that group differences in physical activity would be evident (i.e., Amish > Mennonite > contemporary-living children). Hypothesis 2 was that group differences in physical inactivity would be evident (i.e., contemporary-living > Mennonite > Amish children). Hypothesis 3 was that the timing (e.g., time of day and day of the week) of the physical activity of the Amish and Mennonite children would differ from that of the contemporary-living children.
The study used an ex post facto (comparative) design whereby the causes of group differences are assumed despite the lack of an experimental design. An overall sample of 474 children was drawn from four different groups: two living a traditional agrarian lifestyle (OOA and OOM) and two living a mainstream contemporary lifestyle (rural [RSK] and urban [USK]). The sampling procedures differed across the groups as follows: (i) OOA-sampled entire population of 81 children (i.e., 100% response rate); (ii) OOM-randomized by family from a prescreened list with a total eligible population of 300 with an achieved response rate of 40%; (iii) RSK-self-selected volunteers with a total population of 262 children and an achieved response rate of 59%; and (iv) USK-self-selected volunteers with a total population of 178 and an achieved response rate of 62%. As this study is based on secondary data analysis, each sample has been described in detail elsewhere (1,2,30,31). In brief, the lifestyle of OOA and OOM children emphasizes simplicity and traditional values rather than progress and technology. Ownership of automobiles is not permitted. Farm tractors, telephones, and bicycles are permitted, however, among the OOM. Farming and various labor trades are the preferred occupation. Children are educated in their own schools, and formal education takes place only through to eighth grade. Therefore, we reasoned that the lifestyles of OOA and OOM children would be representative of typical physical activity behaviors of 100 and 60 yr ago, respectively. The lifestyles of RSK children are typical of rural towns (population <5000), whereas the lifestyles of USK children are typical of urban centers (population >200,000) in Canada today.
As part of a larger battery of anthropometry and health-related fitness tests (peripheral to the present study), the 8- to 13-yr-old children had their height and body mass measured using a Health o meter® 402KL balance beam scale (Health o meter Inc., Bridgeview, IL). The attached height rod served as a stadiometer. The triceps skinfold was measured midway between the shoulder and the elbow on the right arm using Harpenden C136 skinfold calipers (British Indicators, West Sussex, England). All anthropometric measures were measured using the same procedures and equipment for all samples.
Each child's physical activity was objectively measured for seven consecutive days via accelerometry. All pertinent data collection and analytical procedures related to the accelerometry portion of the study are described in Table 1. The raw data were analyzed using custom software KineSoft version 2.0.95 (KineSoft, New Brunswick, Canada) to produce a series of standardized outcome variables similar to the procedures of Esliger et al. (8,9). The main variables of interest were average minutes of sedentary-, light-, moderate-, and vigorous-intensity physical activity per day. Because the timing of activity and inactivity was of interest, the intensity variables were analyzed with the following period groupings: weekday, weekend day, and hourly. Physical activity was measured using the same procedures and accelerometer model for all samples. In addition to parental consent, each child provided written assent to participate in the study. All procedures were approved by the Institutional Research Ethics Boards of the University of Saskatchewan and University of Tennessee.
One-way ANOVA were used to test for group differences in chronological age, month of data collection (i.e., seasonality), and accelerometer wear time. Subsequent analyses used MANCOVA models with chronological age and wear time as the covariate to determine group differences in anthropometric variables and physical activity and inactivity variables. Owing to the skewed distribution of the moderate and vigorous physical activity variables, these data were log transformed. All statistical tests were performed on the transformed data; however, the nontransformed means and SD are presented. Paired samples t-tests were used to determine within-group differences in the weekday versus weekend day physical activity variables. The influence of time of day on physical activity and inactivity was described visually via 24 h × 7 d area plots for each group. Where appropriate, models used Bonferroni adjustments for post hoc comparisons, and alpha was set at P < 0.05. All analyses were performed using SPSS for Windows version 15.0 (SPSS Inc., Chicago, IL).
The characteristics of the study sample are displayed by group in Table 2. The sample size lost to technical failure of the accelerometers was 13.9%, 0.0%, 8.5%, and 2.7% for the OOA, OOM, RSK, and USK groups, respectively. The traditional lifestyle groups were diligent in wearing their accelerometers and in recording on/off times on their logs. As a result, the average sample size lost to participant noncompliance (i.e., not wearing the accelerometer for at least 10 h·d−1 for at least four weekdays and one weekend day) was 0.0%, 0.0%, 12.6%, and 13.1% for the OOA, OOM, RSK, and USK groups, respectively. Impressively, 100% of the OOA and OOM children had 7 d of valid data. All groups exceeded the daily wear time requirement with OOA children having the greatest average amount of wear time (14.5 h·d−1) and RSK and USK children having the least (13.4 h·d−1). The extra wear time was a result of the 1- to 2-h earlier morning wake-up times of the traditional groups and their 30- to 70-min earlier bed times.
The descriptive characteristics of the participants are displayed by group in Table 3. The samples are sex balanced; however, because the OOM children were older than the other groups, all further analyses controlled for age because it is well established that physical activity declines with age (22). OOA children were shorter than RSK children, weighed less than OOM and RSK children, and had lower body mass index and triceps skinfolds than all other groups.
Influence of lifestyle on physical activity and inactivity.
Support for Hypothesis 1 was evident on weekdays where there were group differences in moderate-intensity physical activity (4-6.99 METs) between traditional lifestyle groups and contemporary lifestyle groups (OOA > OOM > USK > RSK; 90, 69, 58, 49 min per weekday, respectively; Fig. 1). On the weekend, the group differences in moderate-intensity physical activity persisted between, but not within, lifestyle groups (OOA = OOM > USK = RSK; 55, 54, 36, 40 min per weekend day, respectively; Fig. 2). There was relatively little vigorous physical activity (≥7 METs) accumulated on any day of the week. On weekdays, OOA children accumulated a greater amount of vigorous physical activity compared with all other groups (Fig. 1). Although the greater amount of vigorous physical activity in OOM children compared with RSK children was statistically significant, it is unlikely to have biological significance. With even fewer minutes of vigorous physical activity accumulated on the weekend by all groups, it was of little surprise that there were no differences between groups (Fig. 2). Likewise, there were no group differences in inactivity behaviors between lifestyle groups on weekdays (Fig. 1); however, on weekend days, the USK children accumulated more sedentary time than any other group, and the RSK children were more sedentary than the OOM children (Fig. 2).
Influence of day of the week and time of day on physical activity and inactivity.
All within-group differences between weekday and weekend day physical activity and inactivity were significant with the exception of the USK group for sedentary time and vigorous physical activity and the RSK group for light activity (Figs. 1 and 2).
Plotting and visualizing the hourly physical activity behaviors by intensity and group revealed some interesting temporal trends (Fig. 3). Focusing on school days (i.e., Monday to Friday from 0800 to 1500 h), it appears that the groups have similar patterns in physical activity and inactivity (i.e., they have the same activity and inactivity "hotspots"). Comparing the two peak sedentary hours of the school day (i.e., 0900-1000 and 1300-1400 h) showed that OOA and OOM children spent, on average, 50 of every 60 min during these two, 1-h blocks being sedentary compared with 40 of every 60 min for the contemporary children. Although being more sedentary during school hours, the OOA and the OOM children were more active overall as evidenced by the higher amounts of moderate physical activity. In fact, both traditional groups had a pronounced trimodal pattern of moderate physical activity, peaking during the morning commute (0800-0900 h), morning recess (1000-1100 h), and lunch (1200-1300 h) (Fig. 4). Notably, the hour during lunch break was, on average, the most active for all groups even when compared with free time on evenings and weekends. On average, OOA and OOM children spent 9 of every 60 min during these three, 1-h blocks being moderately active (compared with only 5 of every 60 min in the RSK and USK children). In fact, with the exception of the peak in moderate physical activity over the lunch break, it was difficult to discern clear peaks during the morning commute and/or the morning recess in the contemporary groups. The hourly contribution of vigorous physical activity was negligible across all groups over these three periods.
Compared with the in-school time, OOA and OOM children's out-of-school time (i.e., Monday to Sunday from 1500 to 2100 h) showed a marked decline in sedentary time and an increase in time spent in moderate physical activity. This favorable shift was less pronounced in the RSK and USK children (Fig. 3). Compared with in-school time, there was, on average, 44% less sedentary time out of school in the OOA and OOM children, amounting to 25 of every 60 min (27 min on weekend days) being spent sedentary during this 6-h block. However, in the RSK and USK children, sedentary time out of school was only 15% lower than the time in school, amounting to 31 of every 60 min (35 min on weekend days) being spent sedentary during this 6-h block. On weekend days, the RSK and the USK children's sedentary time was fairly consistent with roughly half of every hour spent sedentary. Although OOA and OOM children were more active on Saturday, they were more sedentary on Sunday. Collectively, these results support the hypothesis that the groups differ with respect to the day of the week and the time of the day they accumulated their physical activity and inactivity.
The purpose of this study was to determine whether differences in the lifestyles of children would be evident in their physical activity behavior profile (e.g., intensity and timing). In an attempt to provide a "window" to the past, we profiled the physical activity behaviors of OOA and Mennonite children known to differ in their use of modern conveniences such as labor-saving technologies and sedentary, multimedia-based leisure pursuits. Although cross sectional, these data suggest that contemporary/modern living is associated with lower levels of moderate physical activity compared with lifestyles representative of earlier generations (weekdays: OOA > OOM > USK > RSK; weekend days: OOA = OOM > USK = RSK). It is important to note that the physical activity levels of RSK and USK children (i.e., 51 and 58 min of MVPA per day) are comparable to other similarly analyzed data sets (e.g., National Health and Nutrition Examination Survey; 60 min of daily MVPA ). These comparative data imply that the differences between lifestyle groups are a result of OOA and OOM children being more active (i.e., 91 and 71 min of MVPA per day), not USK and RSK children being less active than typical children today. The fact that these traditional groups do not participate in organized sports or attend physical education classes suggests that the group differences are likely explained by differences in lifestyle-embedded physical activities such as farm and other manual chores, active commuting, and free play. This explanation seems logical when one considers the substantial contribution that lifestyle-embedded physical activities, also called NEAT (nonexercise activity thermogenesis) (19), make toward total daily physical activity energy expenditure (32).
Those living a traditional agrarian lifestyle, most notably OOA children, accumulated much physical activity and met or exceeded most national physical activity guidelines, including on weekdays, Health Canada's (12,13) recommendation of ≥90 min·d−1. Therefore, the physical activity and inactivity profiles of OOA and OOM children may be a useful behavioral model for contemporary children to emulate. Perhaps the most remarkable difference between lifestyle groups was the consistently greater amount of physical activity and less time spent sedentary exhibited by the OOA and OOM children compared with the RSK and USK children in their discretionary time. Although the context or mode of this "extra" physical activity is unknown, detailed information is available with respect to when/how this activity is accumulated. This "how-to" information is crucial for the development of effective physical activity messages and programs that are necessary to promote healthy behavior change (4). For example, data from OOA children highlight the fact that the physical activity gap that exists for contemporary children could be made up by encouraging them to accumulate small amounts of physical activity intermittently throughout the day (e.g., during the morning commute, recess, lunch, after school commute, and during the evening). These data, which are in agreement with those recently published by Riddoch et al. (25), show that making small changes in behaviors can add up to significant amounts of daily physical activity.
Living a lifestyle reminiscent of 100 yr ago, OOA children accumulated on average 55 min more MVPA per weekday and 18 min more per weekend day compared with contemporary children. The "extra" time spent being active in OOA children was significant and likely contributed to their low levels of overweight and obesity (2,31). Unfortunately, information about the mode and/or context of the physical activity is unknown; however, it is likely that agricultural chores contributed to the higher activity levels in the OOA and OOM children. On the basis of conservative estimates (i.e., using 4 kcal·kg−1·h−1), this translates to an activity energy expenditure deficit of 124 kcal per weekday and 51 kcal per weekend day for the contemporary children. All else being equal, if the energy imbalance was maintained over the long term, the contemporary children would be at risk of gaining approximately 5 kg of fat per year (assuming that 7700 kcal leads to an average of 1 kg weight gain as fat). This energy gap is similar to that found by Wang et al. (35), who suggested that consistent behavioral changes (i.e., reduced energy intake and/or increased energy expenditure) averaging 110-165 kcal·d−1 may be sufficient to counterbalance the energy gap in children. Unfortunately, maintaining energy balance in a modern world is difficult because the behavioral processes that evolved to ensure our survival are still intact. That is, people still eat when food is available and "rest" when physical activity is not required. The problem is food is nearly always available and physical activity is seldom required (3,24).
Although OOA and OOM children were more active than contemporary-living children regardless of the day, on weekdays both groups spent similar amounts of time being sedentary. There were, however, clear differences in how sedentary time was accumulated with group differences evident on weekend days. This is an important finding as there is mounting literature linking overall sedentary time and its pattern of accrual to health risks (10,14,15,23). For example, on Sundays, virtually all OOA and OOM children attend church. However, unlike the active commute to school, children travel to church with their families via horse and buggy and then sit through a service lasting two or more hours. These religious observances are easily discernable as sedentary bouts in Figure 3 when the activity profile on Saturday is compared with Sunday. No such clearly demarcated bout of sedentary time was found for the contemporary groups on Sunday. The notion that religious observance can have an impact on the accrual of physical activity and sedentary time has been studied before. Kahan (16) found that in a group of Jewish adolescents, those most likely to observe the Sabbath and attend a synagogue accrued less MVPA and less sedentary time compared with the less observant. Another contributing factor to the relatively much sedentary time during the school day relates to the fact that OOA and OOM children are educated in one-room school houses with limited space for moving around both during and between lessons. In contrast, contemporary children attend relatively large schools and often change classrooms between lessons and may have physical education classes scheduled throughout the school day. Cohen et al. (6) recently showed that the indoor square footage of a school can contribute 4% and 16% to the light and MVPA of adolescent girls, respectively.
A strength of this study was the use of objective measurements of physical activity (26). Another strength is the detailed analysis of the accelerometry data in terms of intensity and temporality. However, there are limitations to accelerometry, most notably, its inability to assess lifting and carrying activities, cycling, water-based activities, and the general lack of contextual information relating to activity mode and/or location/domain (21). For example, the fact that waist-mounted accelerometers do not measure cycling could have limited the true quantification of MVPA in the OOM, RSK, and USK children, all of whom ride bicycles; whereas the OOA children are disallowed. This explanation may partly explain the relatively high amounts of MVPA in the OOA children compared with the OOM children; however, one cannot forget the fact that OOM farms and homesteads are allowed some labor mechanization (e.g., tractors). Unfortunately, no information was collected on activity mode. Although attempts were made to control for the month of data collection, seasonal differences were not specifically measured. In addition, specific information was not collected regarding school start and end times, recess, or lunch. Another limitation was that selection bias could not be ruled out because of the nonrandom nature of the sample. Although virtually all OOA children were measured, less than half of the OOM children were sampled; however, it is unlikely that the OOM children are at risk of physical activity-related nonresponse bias because it was the most technologically conservative families that declined to participate. In contrast, the self-identifying process of contemporary school selection likely resulted in schools participating that were very supportive of physical activity. We are confident that, if anything, these limitations work to disprove our hypotheses and therefore strengthen our findings. Finally, it should be noted that this study included only children in technologically conservative Amish and Mennonite communities. Therefore, these results should not be generalized to other less conservative religiocultural communities where fewer families farm and modern technology is more prevalent because physical activity is likely to be lower in these groups.
Although cross sectional, these data suggest that contemporary/modern living is associated with lower levels of moderate- and vigorous-intensity physical activity compared with lifestyles representative of earlier generations. Analyzing the physical activity and inactivity patterns of traditional lifestyle groups such as the OOA and OOM can provide valuable insight into the quantity and quality of physical activity necessary to promote health. Future work in this area should address differences in how children fractionalize (i.e., accumulate in bouts) their physical activity and inactivity and the impact these behaviors have on health.
The authors express their appreciation to all the children who participated in this study and to David Luthy for his assistance and guidance with the Amish community and Levi Frey for his assistance with the Mennonite community. This research was supported by the Canadian Population Health Initiative of the Canadian Institute for Health Information and the Charlie and Mai Coffey endowment in Exercise Science at the University of Tennessee.
Conflict of interest: The authors declare no conflict(s) of interest. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Bassett DR. Physical activity of Canadian and American children: a focus on youth in Amish, Mennonite, and modern cultures. Appl Physiol Nutr Metab
2. Bassett DR Jr, Tremblay MS, Esliger DW, Copeland JL, Barnes JD, Huntington GE. Physical activity and body mass index of children in an Old Order Amish community. Med Sci Sports Exerc
3. Bellisari A. Evolutionary origins of obesity
. Obes Rev
4. Brawley LR, Latimer AE. Physical activity guides for Canadians: messaging strategies, realistic expectations for change, and evaluation. Can J Public Health
. 2007;98(2 suppl):S170-84.
5. Cameron C, Wolfe R, Craig CL. Physical Activity and Sport: Encouraging Children to be Active
. Ottawa (ON): Canadian Fitness and Lifestyle Research Institute; 2007. p. 14.
6. Cohen D, Scott M, Wang FZ, McKenzie TL, Porter D. School design and physical activity among middle school girls. J Phys Act Health
7. Eisenmann JC, Katzmarzyk PT, Tremblay MS. Leisure-time physical activity levels among Canadian adolescents, 1981-1998. J Phys Act Health
8. Esliger DW, Copeland JL, Barnes JD, Tremblay MS. Standardizing and optimizing the use accelerometer data for free-living physical activity monitoring. J Phys Act Health
9. Esliger DW, Tremblay MS. Physical activity and inactivity profiling: the next generation. Can J Public Health
. 2007;98(2 suppl):S195-207.
10. Hamilton MT, Hamilton DG, Zderic TW. Role of low energy expenditure
and sitting in obesity
, metabolic syndrome, type 2 diabetes, and cardiovascular disease. Diabetes
11. Haskell WL. Physical activity, sport, and health: toward the next century. Res Q Exerc Sport
. 1996;67(3 suppl):S37-47.
12. Health Canada and the Canadian Society for Exercise Physiology. Canada's Physical Activity Guide for Children. Cat. No. H39-611/2002-2E
. Ottawa (ON): Minister of Public Works and Government Services Canada; 2002.
13. Health Canada and the Canadian Society for Exercise Physiology. Canada's Physical Activity Guide for Youth. Cat. No. H39-611/2002-1E
. Ottawa (ON): Minister of Public Works and Government Services Canada; 2002.
14. Healy GN, Dunstan DW, Salmon J, et al. Breaks in sedentary
time: beneficial associations with metabolic risk. Diabetes Care
15. Healy GN, Wijndaele K, Dunstan DW, et al. Objectively measured sedentary
time, physical activity, and metabolic risk: the Australian Diabetes, Obesity
and Lifestyle Study (AusDiab). Diabetes Care
16. Kahan D. Relationships among religiosity, physical activity, and sedentary
behavior in Jewish adolescents. Pediatr Exerc Sci
17. Katzmarzyk PT, Baur LA, Blair SN, Lambert EV, Oppert JM, Riddoch C. Expert panel report from the International Conference on Physical Activity and Obesity
in Children, 24-27 June 2007, Toronto, Ontario: summary statement and recommendations. Appl Physiol Nutr Metab
18. Katzmarzyk PT, Tremblay MS. Limitations of Canada's physical activity data: implications for monitoring trends. Can J Public Health
. 2007;98(2 suppl):S185-94.
19. Levine JA, Eberhardt NL, Jensen MD. Role of nonexercise activity thermogenesis in resistance to fat gain in humans. Science
20. Mattocks C, Leary S, Ness A, et al. Intraindividual variation of objectively measured physical activity in children. Med Sci Sports Exerc
21. Montoye HJ, Kemper HCG, Saris WHM, Washburn RA. Measuring Physical Activity and Energy Expenditure
. Champaign (IL): Human Kinetics; 1996. p. 72-96.
22. Nader PR, Bradley RH, Houts RM, McRitchie SL, O'Brien M. Moderate-to-vigorous physical activity from ages 9 to 15 years. JAMA
23. Pate RR, O'Neill JR, Lobelo F. The evolving definition of "sedentary
". Exerc Sport Sci Rev
24. Peters JC, Wyatt HR, Donahoo WT, Hill JO. From instinct to intellect: the challenge of maintaining healthy weight in the modern world. Obes Rev
25. Riddoch CJ, Mattocks C, Deere K, et al. Objective measurement of levels and patterns of physical activity. Arch Dis Child
26. Shephard RJ. Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med
27. Stamatakis E, Ekelund U, Wareham NJ. Temporal trends in physical activity in England: the Health Survey for England 1991 to 2004. Prev Med
28. Sturm R. Childhood obesity
-what we can learn from existing data on societal trends, part 1. Prev Chronic Dis
29. Sturm R. Childhood obesity
-what we can learn from existing data on societal trends, part 2. Prev Chronic Dis
30. Tremblay MS, Barnes JD, Copeland JL, Esliger DW. Conquering childhood inactivity: is the answer in the past? Med Sci Sports Exerc
31. Tremblay MS, Esliger DW, Copeland JL, Barnes JD, Bassett DR. Moving forward by looking back: lessons learned from long-lost lifestyles. Appl Physiol Nutr Metab
32. Tremblay MS, Esliger DW, Tremblay A, Colley R. Incidental movement, lifestyle-embedded activity and sleep: new frontiers in physical activity assessment. Can J Public Health
. 2007;98(2 suppl):S208-17.
33. 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
34. Trost SG, Pate RR, Sallis JF, et al. Age and gender differences in objectively measured physical activity in youth. Med Sci Sports Exerc
35. Wang YC, Gortmaker SL, Sobol AM, Kuntz KM. Estimating the energy gap among US children: a counterfactual approach. Pediatrics
Keywords:©2010The American College of Sports Medicine
MODERNITY; TECHNOLOGY; SEDENTARY; ENERGY EXPENDITURE; OBESITY; ACCELEROMETRY