The results of this study suggest that middle school girls, regardless of weight status, spend the majority of their days in sedentary and light PA. The amount of time spent in moderate-to-vigorous activity is slightly higher among normal-weight than overweight girls. There also appears to be a distinct circadian rhythm to activity that differs across weekdays vs. weekends. The bursts of activity on weekday mornings and in the early afternoons do not occur on weekends, and without these bursts, there is less MVPA on weekends than weekdays. The monitor was worn for less time on the weekend than on the weekdays, and to standardize this, the percentage of time spent at these various intensities was calculated. The observation remains that less MVPA occurs on weekends than on weekdays.
A few energy expenditure calculations reveal the significance of the 26 vs. 18 minutes for MVPA on weekday vs. weekend days, respectively. In a study of adolescent girls (15
), the mean energy cost of a brisk walk was 19.1 kJ/min. The TAAG cutpoint for MVPA was based on the counts differentiating between a slow and brisk walk (12
). The energy expenditure of activity performed just above this MVPA threshold for 26 min/d for 5 weekdays would translate to 2483 kJ (593 kcal). Likewise, for 18 min/d for 2 days, the energy expenditure for activity would be 688 kJ (164 kcal). On a per day comparison, this would be 28 (119 kcal/d) vs. 20 (82 kcal/d) kJ/d. Girls with no decrease in PA over the weekends would expend an extra 18 kJ/wk (74 kcal/wk), and this would be considered of a modest clinical significance.
The distribution of time spent in the four levels of activity intensity (sedentary, light, moderate, and vigorous) suggests multiple opportunities for intervention. There is little time spent in MVPA (18 to 28 minutes per day). Because energy expenditure per minute is higher in moderate-to-vigorous activity than lower intensity activities, even small increases in the percentage of time spent in these activities would be useful toward increasing total daily energy expenditure. By contrast, the time spent in sedentary vs. light PA presents another opportunity to increase total PA. Reducing time spent in sedentary activity should be a goal to increase total activity. Also, the majority of energy expended in total PA is from light intensity (which has a lower energy expenditure per minute) rather than moderate or vigorous activity, simply because of the greater amount of time spent in light activity (40% to 45% of the day). Therefore, increasing light activity and decreasing sedentary activity may be as useful as increasing higher intensity activity time when attempting to increase total daily energy expenditure.
Patterns of activity in children and adolescents have typically been characterized using self-report methods or direct observation. Self-report studies have quantified PA by intensity level and by participation in sport teams and exercise programs such as in the National Health and Nutrition Examination Study (16
). The Youth Risk Behavior Survey used self-reports to quantify activity by intensity (moderate and vigorous activity) and included strength training, enrollment in physical education, and participation in sports programs (18
). These methods offer the advantage of describing types of activities and, in some cases, the social and physical environments in which they occur.
Objective measures, such as heart rate monitoring and accelerometry, can provide more precise estimates of the duration and the intensity of activities. Certain accelerometers, however, are required to be removed during water activities and are limited in their ability to detect certain activities such as biking. In TAAG, both accelerometers and self-reports were used to capture intensity and duration of activity and the context of activity. The types of activities were reported by using a self-report measure, the 3-day PA recall. Briefly, the top 10 activities from most to least reported included travel by walking, doing household chores, walking for exercise, running/jogging, dance, basketball, and playing with younger children (19
Several studies (20
) have examined the time spent in different intensities using objective measures of activity. In the present study, the proportion of time spent in sedentary and light activity constituted the majority of the day in these adolescent girls. This makes sense because sleeping constitutes a large part of the day, and the girls were sitting in class much of the school days. The relatively lower amount of time spent in MVPA was also documented. In a review of 26 studies using heart rate monitoring, Epstein et al. (20
) reported that youth of all ages attain ~30 min/d at an intensity of 50% or more of heart rate reserve. This would be comparable to our adolescent girls, where the MVPA during the week averaged 26 min/d. However, during the weekend, MVPA averaged only 18 min/d, which is lower than values previously reported (20
). A study (21
) of 9- and 15-year-old children and adolescents reported the time spent in MVPA (including at least 1 weekend day) measured by accelerometry was 73 ± 32 min/d in 15-year-old girls, suggesting girls in the present study were less active. However, moderate intensity was defined using a very different threshold of activity (MVPA defined as >1500 counts/min for 15-year olds, ~3 METs) and would account for most of this large difference between MVPA in adolescent girls in the United States vs. Europe. Also, this prior study did not provide separate weekday vs. weekend activity. Another investigation (22
) of middle school-aged boys and girls reported measures of all of the intensities using the TriTrac accelerometer. In girls, sedentary activity accounted for the majority of the day (74%), with light (18%), moderate (6.9%), and vigorous (0.5%) activity as the remainder. Clearly, sedentary activity and light activity are the dominant intensities of the day for many adolescent girls. The different accelerometer instruments and corresponding thresholds and much smaller sample size (n
= 17) could contribute to the different findings. Another study (23
) using the Actiwatch accelerometer reported the time spent in the various intensities as a percentage of the awake period of the day in 81 middle school boys and girls combined. The levels are more comparable to the present study with 51% sedentary, 43% light, 5% moderate, and 0.4% vigorous (23
). Thus, the U.S. studies give relatively comparable information on the amount of time spent in the various activities.
Comparisons across the studies for weekdays vs. weekends are somewhat difficult because the subject samples and measurement tools differ, as well as the expression of the activity data. Studies (24
) have reported differences in activity on weekdays and weekends using self-report in children. Activity in one study (25
) as measured by heart rate monitoring (expressed as percentage of the day active) reported slightly higher means on weekdays than on weekends at either moderate or vigorous intensity in young girls. In a longitudinal study (26
) of these same girls 8 to 10 years old, no differences were observed in weekdays vs. weekends. In the present study, differences in MVPA between weekdays and weekends were observed in 11- to 12-year-old girls. This has implications in terms of the age to target individuals for various intervention strategies that might focus on altering weekend behavior. The adolescent girls in the present study may obtain the higher intensity activity before, during, or after school, with lower intensity activities on the weekend. The differences noted may be due, in part, to greater television viewing time on weekends than weekdays. In the Eating and Activity Questionnaire Trial (Project EAST) (24
), middle school students self-reported 42 additional minutes of television viewing on weekend days compared with weekdays.
Comparing the activity by weight status revealed a lower MVPA on the weekdays and weekends in the at-risk and overweight girls than the normal-weight girls. These findings generally concur with self-report studies from the 1999 Youth Risk Behavior Survey, and from the National Health and Nutrition Examination Study in which adolescent girls and young adults who were overweight or at-risk of overweight were less likely to be physically active than normal-weight participants (16
). One study (27
) that measured activity by accelerometry reported that overweight adolescents were less active than non-overweight adolescents. The overweight adolescents spent 15.6 fewer minutes per day in MVPA, defined using a cut-off point of 3 METs (27
). The source of the activity differences between normal vs. at-risk plus overweight adolescents are most apparent in , which shows the differences distributed throughout the day. There is less MVPA among at-risk plus overweight girls compared with normal-weight girls in both cases. Early mornings, evenings, and school days are very similar between the two groups. Further, relative differences in the minutes spent in each intensity category were much smaller for sedentary activity (3% greater among at-risk plus overweight girls) and light-intensity activity (2% greater among normal weight) compared with 8% and 20% for moderate- and vigorous-intensity activities, respectively. This suggests that interventions designed to increase MVPA on weekday afternoons (after school) and/or weekends could make a modest impact on total PA among overweight girls.
In conclusion, adolescent girls are more active at moderate and vigorous intensity levels on weekdays than on the weekend, and girls who are at-risk of overweight and overweight have the lowest values of these levels of activity on weekdays and weekends. A general intervention approach for all girls could be to increase total daily activity in a healthful direction by reducing sedentary time, and this would have modest effects on improving energy balance.