Although the subjects observed in these examples both spent significant portions of the observed time engaging in sedentary (e.g., sitting) and light (e.g., slow walking) activities, the proportions were markedly different between the two. Light activity increases metabolic rate, and the energy cost of light activities accumulated throughout the day can contribute significantly to total daily energy expenditure. To illustrate this point, we estimated energy expenditure for the two subjects for the 13-h monitoring periods. Using the mean MET level for each intensity category (1.25 METs for sedentary, 2.2 METs for light, 4.5 METs for moderate, and 7.5 METs for vigorous activities), we multiplied METS by the number of hours engaged in each activity intensity. The subject in Figure 1, who failed to meet recommended levels of moderate-to-vigorous physical activity, performed an estimated 26.3 MET-hours of activity during the monitoring period. The subject in Figure 2, who engaged in an hour of structured exercise at a moderate-to-vigorous level, performed an estimated 23.6 MET-hours and expended less energy than the other subject (assuming equal body weights for the two persons). This example points to the importance of considering the full range of energy expenditure rates observed in the activity range below moderate intensity.
For the purpose of applying accelerometry to the study of sedentary behavior, it is important to carefully establish accelerometer count cut points that identify sedentary behavior and to recognize that these cut points may be different for adults and children. Accelerometers have been calibrated against energy expenditure, and regression models have been used to translate accelerometer counts into specific physical activity intensity ranges. Most studies addressing sedentary behavior have used an absolute count cutoff, typically less than 50 or 100 counts per minute. However, such an absolute cut point may not work well across diverse age groups. More calibration research is needed to determine the best definition of cut points for sedentary behavior with accelerometry, among both youth and adults (11).
Although accelerometers have created new opportunities for validly measuring the full range of energy expenditure, in the future, improved accelerometers or entirely new instruments may allow researchers to better quantify the energy expenditure of common daily exercise and nonexercise activities. Some of these instruments are currently in development and testing, and they are promising because they detect the energy expended in activities that typically are considered sedentary, such as sitting, lying down, changing postures, moving limbs, and fidgeting, as well as light, moderate, and vigorous activities (13,14).
STUDIES OF SEDENTARY BEHAVIOR
To test our hypothesis that researchers have inadequately defined or measured sedentary behavior, we searched the literature for studies that reported effects of sedentary behavior on health status. Most of the studies we identified that drew conclusions about the health effects of sedentary behavior (including some of our own) did not actually measure sedentary behavior. Although a complete summary of our literature review is beyond the scope of this article, several examples of these studies are presented below.
The Relationship of Physical Activity and Body Weight With All-Cause Mortality
The relationships of physical activity and obesity with all-cause mortality were studied in 9824 Puerto Rican men followed from 1962 to 1965 as part of the Puerto Rico Heart Health Program (1). Physical activity was self-reported, and the referent-sedentary group (lowest quartile of activity) included participants who reported engaging in little to no moderate or vigorous activities. All-cause mortality was significantly lower in the upper three activity quartiles, compared with the referent-sedentary group. Although sedentary activity was not measured, and the sedentary group included those who reported light activity, the authors concluded that "Puerto Rican men who are physically active experienced significant reductions in all-cause mortality compared with their sedentary counterparts."
Physical Inactivity and Overweight Among Los Angeles County Adults
The association between physical activity and health status was examined in 8353 Los Angeles County adults (12). Using a telephone survey, the researchers obtained self-reports of physical activity, body mass index, and mental health status variables. Individuals were defined as sedentary if they reported less than 10 min·wk−1 of continuous physical activity. In this report, being classified as sedentary was associated with an overall perception of poor health and with feelings of sadness and depression. In addition, obese adults were more likely to be classified as sedentary than overweight and normal weight adults. Although sedentary behavior was not measured in this study, the authors concluded that "mental and physical health status were prominent correlates of sedentariness."
The Association Between Television Viewing and Overweight Among Australian Adults
The authors studied the effect of physical activity on the association between television viewing and overweight in 3392 adults from New South Wales, Australia (8). Walking, moderate and vigorous leisure-time physical activities, television viewing habits, and body mass index were assessed using self-reports. Across low, moderate, and high physical activity categories, those who reported watching more than 4 h of television per day were twice as likely to be overweight as those who reported less than 1 h of television per day, irrespective of their physical activity level. Despite the fact that only one sedentary activity was measured and that an effect interaction was tested only for the total volume of leisure-time physical activity, the authors concluded that, "in order to reduce the prevalence of overweight and obesity or at least prevent weight gain, it may be not only important to increase participation in purposive physical activity, but also to reduce the time spent in sedentary behaviors."
Physical Activity and Determinants of Physical Activity in Obese and Nonobese Children
This study compared the physical activity levels of 54 obese and 133 nonobese children (9). Moderate, vigorous, and moderate-to-vigorous physical activity were measured with accelerometry, but the authors did not directly assess sedentary or light physical activity. Obese children engaged in less moderate, vigorous, and moderate-to-vigorous physical activity compared with nonobese children, but the authors concluded that, "Our findings are consistent with the hypothesis that physical inactivity is an important contributing factor in the maintenance of childhood obesity" (9(p826)).
These examples point to the tendency of investigators to draw conclusions about sedentary behavior without having measured it. However, we did find some examples of studies that did directly measure sedentary or light activity or that avoided drawing conclusions about sedentary behavior after having measured only one sedentary activity. Several of these studies are presented below.
Objectively Measured Light-Intensity Physical Activity Is Independently Associated With 2-h Plasma Glucose
The associations between time spent in sedentary, light, and moderate-to-vigorous activity with glucose metabolism were studied in 67 men and 106 women participating in the AusDiab study (4). Physical activity was assessed objectively during 7 d with an Actigraph accelerometer; and glucose metabolism, with an oral glucose tolerance test. In analyses that adjusted for potential confounders, including waist circumference, sedentary time was positively associated with, and both light and moderate-to-vigorous activity time were negatively associated with, 2-h postchallenge plasma glucose levels. The authors concluded that "light-intensity physical activity is beneficially associated with blood glucose and that sedentary time is unfavorably associated with blood glucose."
Ethnic Differences in Physical Activity and Inactivity Patterns and Overweight Status
The relationship between physical activity and overweight status was examined among 12,759 11- to 19-yr-olds participating in the National Longitudinal Study of Adolescent Health from 1995 to 1996 (3). Participants self-reported their low, moderate, and vigorous intensity activities over the past week. Light activity was defined as low-intensity activities ranging from 2-3 METs. Although both boys and girls reported the same amount of low-intensity activity per week, the change in low-intensity physical activity over 1 yr modestly affected overweight status in girls. The authors correctly identified low-intensity physical activity as such and concluded that "girls had 8% higher odds of overweight with 1-year change in low-intensity physical activity."
Television Watching and Other Sedentary Behaviors in Relation to Risk of Obesity and Type 2 Diabetes Mellitus in Women
Data from 50,277 women in the Nurse's Health Study cohort were used to investigate the longitudinal relationship between several sedentary and light-intensity activities and the risk of obesity and type 2 diabetes mellitus while adjusting for leisure-time exercise activities (5). Participants self-reported a range of activities, from sedentary and light intensity to vigorous. Each increment of 2 h·d−1 in television watching or sitting at work was associated with a 23% and 5% increase in obesity and a 14% and 7% increase in risk of diabetes, respectively. In contrast, standing or walking around at home (2 h·d−1) was associated with a 9% reduction in obesity and a 12% reduction in diabetes. The authors concluded that, "Independent of exercise levels, sedentary behaviors, especially television watching, were associated with significantly elevated risk of obesity and type 2 diabetes, whereas even light to moderate activity was associated with substantially lower risk."
These examples illustrate that, to date, few studies have measured sedentary behavior among adults and youth, and the health effects of very low activity levels are still unclear. In addition, the potential health benefits of accumulated light activity are unclear because, in most studies, light activity and no activity are combined in a sedentary or low-active category. Of the four summarized articles that measured sedentary or light activity, one study (4) used accelerometry to objectively measure light activity. More research is needed to measure the lower end of the activity continuum and to test its association with health outcomes.
Because of advancing methodologies and developing bodies of knowledge, we believe that several conventions should be applied widely and consistently in future studies of the relationships between physical activity, sedentary behavior, and health. Specifically:
- Whenever feasible and appropriate to the study aims, the full range of activity intensities should be observed, reported, and used analytically. This should include sedentary behavior and light physical activity as well as moderate and vigorous physical activity.
- While the full range of activity intensities (sedentary to vigorous) contribute to total energy expenditure, activity performed within a narrow intensity range (e.g., vigorous) may influence health in ways that are unique from other activity intensities. Accordingly, activity intensity categories should be treated as potential independent influences on health outcomes.
- Conclusions of studies should be phrased carefully so as to be consistent with the activity variables that were actually measured in the study. Conclusions regarding the influence of sedentary behavior should be drawn only if sedentary behavior was measured and used analytically. Likewise, study conclusions about the influence of physical activity should specify clearly the range of intensities included in the physical activity variable.
- If specific surrogate behaviors are used to study sedentary activity (e.g., television watching) or physical activity (e.g., walking), conclusions should be stated in terms that are limited to those behaviors.
- Operational definitions of activity constructs should be presented clearly by investigators. For example, accelerometry cut points for differentiation of sedentary, light, moderate, and vigorous activity should be reported. The same principle should be applied in studies using self-report instruments.
SUMMARY AND CONCLUSIONS
Although scientists have reported that being sedentary is associated with significant health risks, few studies to date have measured sedentary or low active behavior. In most cases, study participants who are reported to be sedentary or inactive are actually those who did not meet the study's criteria for moderate or higher levels of activity. This grouping of people at the lower end of the activity continuum may confound efforts to identify the health effects of sedentary and light-intensity activity behavior. To date, few self-report instruments have been developed to detect sedentary behavior and light-intensity physical activity, and even fewer have been validated. Recent advances in accelerometry have made possible the measurement of the full range of physical activity levels, from completely sedentary to extremely vigorous, with a single instrument. Thus, accelerometry is emerging as a valuable tool for exploring the independent associations of various activity levels with health outcomes. Future studies should measure sedentary and light activity to determine their independent and joint contributions to health outcomes.
The authors thank Gaye Groover Christmus, M.P.H., for assistance in the development of this article, and Kerry McIver, M.A., for assistance with the accelerometry examples.
1. Crespo, C.J., M.R. Palmieri, R.P. Perdomo, D.L. Mcgee, E. Smit, C.T. Sempos, I.M. Lee, and P.D. Sorlie. The relationship of physical activity and body weight with all-cause mortality: results from the Puerto Rico heart health program. Ann. Epidemiol
. 12:543-552, 2002.
2. Freedson, P.S., J. Sirard, and E. Debold. Calibration of the Computer Science and Applications, Inc. (CSA) accelerometer. Med. Sci. Sports Exerc
. 30:777-781, 1998.
3. Gordon-Larsen, P., L.S. Adair, and B.M. Popkin. Ethnic differences in physical activity and inactivity patterns and overweight status. Obes. Res
. 10:141-149, 2002.
4. Healy, G.N., D.W. Dunstan, J. Salmon, E. Cerin, J.E. Shaw, P.Z. Zimmet, and N. Owen. Objectively measured light-intensity physical activity is independently associated with 2-h plasma glucose. Diabetes Care
. 30:1384-1389, 2007.
5. Hu, F.B., T.Y. Li, G.A. Colditz, W.C. Willett, and J.E. Manson. Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women. J.A.M.A
. 289:1785-1791, 2003.
6. Lowry, R., H. Wechsler, D.A. Galuska, J.E. Fulton, and L. Kann. Television viewing and its associations with overweight, sedentary lifestyle, and insufficient consumption of fruits and vegetables among US high school students: differences by race, ethnicity, and gender. J. Sch. Health
. 72:413-421, 2002.
7. Paffenbarger, R.S., Jr., R.T. Hyde, A.L. Wing, and C.C. Hsieh. Physical activity, all-cause mortality, and longevity of college alumni. N. Engl. J. Med
. 314:605-613, 1986.
8. Salmon, J., A. Bauman, D. Crawford, A. Timperio, and N. Owen. The association between television viewing and overweight among Australian adults participating in varying levels of leisure-time physical activity. Int. J. Obes. Relat. Metab. Disord
. 24:600-606, 2000.
9. Trost, S.G., L.M. Kerr, D.S. Ward, and R.R. Pate. Physical activity and determinants of physical activity in obese and non-obese children. Int. J. Obes. Relat. Metab. Disord
. 25:822-829, 2001.
10. U.S. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General
. Atlanta, GA: 1996.
11. Ward, D.S., K.R. Evenson, A. Vaughn, A.B. Rodgers, and R.P. Troiano. Accelerometer use in physical activity: best practices and research recommendations. Med. Sci. Sports Exerc
. 37:S582-S588, 2005.
12. Yancey, A.K., C.M. Wold, W.J. McCarthy, M.D. Weber, B. Lee, P.A. Simon, and J.E. Fielding. Physical inactivity and overweight among Los Angeles County adults. Am. J. Prev. Med
. 27:146-152, 2004.
13. Zhang, K., F.X. Pi-Sunyer, and C.N. Boozer. Improving energy expenditure estimation for physical activity. Med. Sci. Sports Exerc
. 36:883-889, 2004.
14. Zhang, K., P. Werner, M. Sun, F.X. Pi-Sunyer, and C.N. Boozer. Measurement of human daily physical activity. Obes. Res
. 11:33-40, 2003.
Keywords:©2008 The American College of Sports Medicine
physical activity; health; accelerometry; adult; child