Too Much Sitting: The Population Health Science of Sedentary Behavior : Exercise and Sport Sciences Reviews

Secondary Logo

Journal Logo

Perspectives for Progress

Too Much Sitting

The Population Health Science of Sedentary Behavior

Owen, Neville1,2; Healy, Geneviève N.1,2; Matthews, Charles E.3; Dunstan, David W.2

Author Information
Exercise and Sport Sciences Reviews 38(3):p 105-113, July 2010. | DOI: 10.1097/JES.0b013e3181e373a2
  • Free

Editor's Note: This Perspectives for Progress article is based on the corresponding author's President's Lecture from the American College of Sports Medicine's 56th Annual Meeting held in May 2009.


The physical, economic, and social environments in which modern humans sit or move within the contexts of their daily lives have been changing rapidly, and particularly so since the middle of the last century. These changes - in transportation, communications, workplace, and domestic entertainment technologies - have been associated with significantly reduced demands for physical activity. However, these reductions in the environmental demands for being physically active are associated with another class of health-related behaviors.

Sedentary behaviors (typically in the contexts of television (TV) viewing, computer and game console use, workplace sitting, and time spent in automobiles) have emerged as a new focus for research on physical activity and health (18,27,31-33). Put simply, the perspective that we propose is that too much sitting is distinct from too little exercise. Research findings on sedentary behavior and health have proliferated in the 10 yr after publication of our first Exercise and Sport Sciences Reviews article on this topic (32). As we will demonstrate, initial findings on the metabolic correlates of prolonged TV viewing time have since been confirmed by recent objective measurement studies, which also show that breaking up sedentary time can be beneficial. Furthermore, we describe recent studies from Canada, Australia, and the United States, which show prospective relationships of sedentary behaviors with premature mortality. Importantly, adults can meet public health guidelines on physical activity, but if they sit for prolonged periods, their metabolic health is compromised. This is a new and challenging area for exercise science, behavioral science, and population health research. However, many scientific questions remain to be answered before it can be concluded with a high degree of certainty that these adverse health consequences are uniquely caused by too much sitting, or if what has been observed so far can be accounted for by too little light, moderate, and/or vigorous activity.

The updated recommendation for adults on Physical Activity and Public Health from the American College of Sports Medicine and the American Heart Association (ACSM/AHA) clearly states that "the recommended amount of aerobic activity (whether of moderate or vigorous intensity) is in addition to routine activities of daily living, which are of light intensity, such as self-care, casual walking, or grocery shopping, or less than 10 min of duration such as walking to the parking lot or taking out the trash" ((20), p. 1426). Logically, doing such daily activities differently could involve reductions in sitting time, but sitting per se is not addressed specifically in the recommendations. In this context, the key question to be asked about the strength of the evidence on sedentary behavior and health that we present in this article is: Would one expect to see a statement on reducing sitting time included in future physical activity recommendations?

Sedentary Behavior

Sedentary behaviors (from the Latin sedere, "to sit") include sitting during commuting, in the workplace, the domestic environment, and during leisure time. Sedentary behaviors such as TV viewing, computer use, or sitting in an automobile typically are in the energy expenditure range of 1.0-1.5 METs (multiples of the basal metabolic rate) (1). Thus, sedentary behaviors are those that involve sitting and low levels of energy expenditure. In contrast, moderate- to vigorous-intensity physical activities such as bicycling, swimming, walking, or running may be done in a variety of body positions, but require an energy expenditure of 3-8 METs (1). In this perspective, light-intensity activity behaviors are those done while standing that require an expenditure of no more than 2.9 METs.

Addressing research on the health consequences of sedentary behavior requires some initial clarification of terminology. We refer to sedentary behaviors (different activities for different purposes in different contexts; see previous description). We also refer to sitting time, a generic descriptor covering what these sedentary behaviors primarily involve. As we demonstrate later, adults spend most of their waking hours either sitting or in light-intensity activity (predominantly standing with some gentle ambulation).

Time in sedentary behaviors is significant, if only because it displaces time spent in higher-intensity physical activity - contributing to a reduction in overall physical activity energy expenditure. For example, displacement of 2 h·d−1 of light-intensity activity (2.5 METs) by sedentary behaviors (1.5 METs) would be predicted to reduce physical activity energy expenditure by about 2 MET-h·d−1 or approximately the level of expenditure associated with walking for 30 min·d−1 (0.5 h * 3.5 METs = 1.75 MET-h).

Research on physical activity and health has concentrated largely on quantifying the amount of time spent in activities involving levels of energy expenditure of 3 METs or more, characterizing those with no participation at this level as "sedentary" (33). However, this definition neglects the substantial contribution that light-intensity activities make to overall daily energy expenditure (8) and also the potential health benefits of participating in these light-intensity activities, rather than sitting. Furthermore, although individuals can be both sedentary and physically inactive, there is also the potential for high sedentary time and physical activity to coexist (the Active Couch Potato phenomenon, which we will explain later). An example would be an office worker who jogs or bikes to and from work, but who then sits all day at a desk and spends several hours watching TV in the evening.

Common behaviors in which humans now spend so much time - TV viewing, computer use and electronic games, sitting in automobiles - involve prolonged periods of low-level metabolic energy expenditure. It is our contention that sedentary behavior is not simply the absence of moderate- to vigorous-intensity physical activities, but rather, is a unique set of behaviors with unique environmental determinants and a range of potentially unique health consequences (43). Our population health research perspective is on the distinct role of sedentary behavior, as it may influence obesity and other metabolic precursors of major chronic diseases (type 2 diabetes, cardiovascular disease, and breast and colon cancers).

Sedentary Behavior and Health: A Unique Underlying Biology?

Physiologically, distinct effects are observed between prolonged sedentary time and too little physical activity (17). There are broad consistencies between the patterns of findings from epidemiological studies on the cardiometabolic correlates of prolonged sitting that we will describe, and recent evidence on biological mechanisms ("inactivity physiology") identified in animal models. It seems likely that there is a unique physiology of sedentary time, within which biological processes that are distinct from traditionally understood exercise physiology are operating. The groundbreaking work of Hamilton and colleagues (3,16) provides a compelling body of evidence that the chronic unbroken periods of muscular unloading associated with prolonged sedentary time may have deleterious biological consequences. Physiologically, it has been suggested that the loss of local contractile stimulation induced through sitting leads to both the suppression of skeletal muscle lipoprotein lipase (LPL) activity (which is necessary for triglyceride uptake and high-density lipoprotein (HDL) cholesterol production) and reduced glucose uptake (3,16). A detailed account of findings and implications from the studies of Hamilton et al. (17,18) has been provided in recent reviews.

Findings by Hamilton et al. (17,18) suggest that standing, which involves isometric contraction of the antigravity (postural) muscles and only low levels of energy expenditure, elicits electromyographic and skeletal muscle LPL changes. However, in the past, this form of standing would be construed as a "sedentary behavior" because of the limited amount of bodily movement and energy expenditure entailed. This highlights the need for an evolution of the definitions used for sedentary behavior research. Within this perspective, standing would not be a sedentary activity, and our approach (subject to revision as further findings accumulate) is to equate "sedentary" with "sitting."


TV Viewing Time: The AusDiab Studies

The Australian Diabetes, Obesity and Lifestyle Study (AusDiab) was conducted initially in 1999/2000 on a common leisure time sedentary behavior - TV viewing time (TV time) - with biomarkers of cardiometabolic risk. AusDiab recruited a large population-based sample of some 11,000 adults from all Australian states and the Northern Territory. Some of our first AusDiab findings were that among adults without known diabetes, self-reported TV time was positively associated with undiagnosed abnormal glucose metabolism (12) and the metabolic syndrome (11). The strongest relationships were observed in the highest TV time category (4 h·d−1 or more). When TV time was considered as a continuous measure (10), a detrimental dose-response association was observed in women between TV time and 2-h plasma glucose and fasting insulin. Importantly, all of these associations persisted after adjustment for sustained moderate- to vigorous-intensity leisure time physical activity and waist circumference. Some of these cross-sectional relationships have been replicated recently in prospective analyses: increases in TV viewing during 5 yr predicted significant adverse changes in waist circumference for men and women and in diastolic blood pressure and a clustered cardiometabolic risk score for women. These associations were independent of baseline TV time, baseline physical activity and physical activity change, and other potential confounders (48).

Being Sedentary and Meeting Physical Activity Guidelines: The Active Couch Potato

We further examined relationships of TV time with continuous metabolic risk in men and women who reported at least 150 min·wk−1 of moderate- to vigorous-intensity physical activity - the generally accepted public health guidelines for health-enhancing physical activity (20). Among these healthy physically active adults, significant detrimental dose-response associations of TV time were observed with waist circumference, systolic blood pressure, and 2-h plasma glucose in both men and women, as well as fasting plasma glucose, triglycerides, and HDL cholesterol in women only (23). This observation - the Active Couch Potato phenomenon - is important. The particular metabolic consequences of time spent watching TV are adverse, even among those considered to be sufficiently physically active to reduce their chronic disease risk. This finding reinforces the potential importance of the deleterious health consequences of prolonged sitting time, which may be independent of the protective effect of regular moderate-intensity physical activity.

TV Viewing Time: Associations With Biomarkers for Men and for Women

One of the striking findings in the AusDiab TV time studies was that the associations with cardiometabolic biomarkers were stronger for women than for men (10-12,23). We subsequently examined the associations of both TV time and self-reported overall sitting time with these biomarkers in the 2004/2005 AusDiab sample (42). The TV time relationships for women were replicated, but for self-reported overall sitting time (which is inclusive of the TV time component), the associations were similar for men and women. So, the question remains as to whether there is a particular relationship of TV time with metabolic health for women. There are some testable hypotheses that can be put forward in this context: Are there dietary or TV time-related snacking differences between men and women? Are women (who have a lower average skeletal muscle mass and a higher average fat mass than men) metabolically more susceptible to the adverse influences of prolonged sitting after a typically large evening meal?

Although some of our most striking initial findings on the adverse health consequences of sedentary behavior have been for TV time, there should be caution in treating this common leisure time sedentary behavior as a marker for overall sedentary time. We have modest evidence (39) that for women, TV time is positively correlated with other leisure time sedentary behaviors and with being less likely to meet physical activity and health guidelines. However, these findings need to be replicated in other populations and with other measures. Furthermore, TV viewing is associated with other health-related behaviors (51), and those in the highest TV time categories are more likely to eat in front of the TV set (26). It is thus plausible that TV time will influence energy balance in two main ways. Most people sit to watch TV, and it has a lower energy cost than the alternative activities that it replaces. In addition, high levels of TV time are likely to increase energy intake because of prompts from frequent commercials about food and beverages, and unlike for many other activities, the hands are free to eat during TV time (51). It is thus a reasonable hypothesis that this latter factor may partially explain why higher levels of TV time are associated with higher waist circumferences and with adverse blood glucose and lipid profiles.

We must emphasize that TV time is one of a number of sedentary behaviors that characterize how adults go about their daily lives, and there is potential measurement error associated with using the self-report measures that are common to most TV time studies. However, based on our recent systematic review (6), we have some confidence that the TV time measures that we have used are reasonably reliable and valid.


Advances in the Objective Measurement of Sedentary Behavior

These Australian studies previously summarized have all relied on self-reported TV time or overall sitting time. However, advances in measurement technology now provide significantly enhanced scientific traction, which is helping to deal with the methodological limitation of measurement error related to the use of self-report items. Before summarizing findings from our objective measurement studies with AusDiab study participants, it is helpful to consider the new perspectives that emerge when accelerometer data on sedentary time and physical activity are examined. Accelerometers (as distinct from pedometers that count and display the number of steps taken) are small electronic devices worn on the hip that provide an objective record of the volume, intensity, and frequency of activity between and within days, which may be downloaded to computer databases and used to derive scientifically meaningful activity variables. Accelerometers have been used as part of the National Health and Nutrition Examination Survey (NHANES), gathering data from large population-based samples of adult residents of the United States. Findings reported to date suggest that compared with what has been assumed to be the case from self-report surveys, levels of participation in moderate- to vigorous-intensity physical activities are extremely low (44), and that some 60% or more of these adults' waking hours are spent sedentary (29).

Sedentary Behavior During Adults' Waking Hours

To illustrate the overall patterns of activity in adults' daily lives, Figure 1 shows a cluster heat map (50). This is a graphic representation from Genevieve Healy, showing accelerometer data for one individual during 1 wk, in the manner originally presented by Foulis et al. (15). The values taken by the accelerometer counts within each minute are represented as colors in the two-dimensional map. The dark blue shading shows accelerometer counts that are less than the currently used, but still debated, cutoff of 100 counts per min for sedentary time, and which are taken to be indicative predominantly of sitting (a caveat, however, is that some of the minutes shown as sedentary will include standing quite still). The pale blue through yellow colorings indicate light-intensity to moderate-intensity physical activities. The yellow through red colorings indicate moderate- to vigorous-intensity physical activity. From an energy expenditure perspective, the dark blue translates to very low levels of energy expenditure, with the red reflecting high energy expenditure levels. What is striking in Figure 1 is the extent to which this person spends his or her time either in light-intensity activities (pale blue to white) and being sedentary (dark blue). Although we would not contend that this is a totally precise and unambiguous representation of sitting time and light-, moderate-, and vigorous-intensity activities, it nevertheless is an informative perspective.

Being physically active, but also highly sedentary: 1 wk of accelerometer count data showing, on average, 31 min·d−1 moderate- to vigorous-intensity activity time (>1951 counts per minute) and 71% of waking hours sedentary (<100 counts per minute).

Figure 1 illustrates one of our key messages about the role of sedentary time in the physical activity and health equation: it is possible to achieve a level of activity consistent with the public health guidelines for health-related physical activity (30 min of moderate-intensity physical activity on most days of the week) but to spend most of waking hours involved in sedentary behaviors. In this case, we see that the accumulated moderate- to vigorous-intensity physical activity time is 31 min; however, this person spends 71% of his or her waking hours in sedentary time. Thus, it is possible for individuals to be physically active, yet highly sedentary - the Active Couch Potato phenomenon identified in the AusDiab TV time studies (24).

The main scientific caveat for this perspective is that these data show "activity," which we infer is reflective of "behavior." However, there are scientific devils in the details of these objective movement data. Debate remains about what are the most appropriate activity count cut points to identify sedentary and light-intensity activity time; also, different cut points may be appropriate for adults of different ages, race/ethnicity, and adiposity status.

Objectively Assessed Sedentary Time: Key Studies

As well as demonstrating remarkably low levels of physical activity and high levels of sedentary time within contemporary human environments (29,44), objective measures also have demonstrated the adverse impact of prolonged sedentary time on cardiometabolic biomarkers of risk. At least three studies in Europe and Australia have examined the associations of objectively measured sedentary time with continuous cardiometabolic biomarkers: the ProActive trial conducted in the United Kingdom (UK), the European RISC study, and the AusDiab study (2,13,14,23,25). For those in the UK ProActive trial (258 participants aged 30-50 yr with a family history of type 2 diabetes), sedentary time was detrimentally associated with insulin in the cross-sectional analysis (14) but was of borderline statistical significance (P = 0.07) in the 1-yr prospective analysis (13). Detrimental cross-sectional associations of sedentary time with insulin also were observed in participants of the European RISC study (801 healthy participants aged 30-60 yr), although the associations were attenuated after adjustment for total activity (2). In the AusDiab accelerometer study sample (169 participants aged 30-87 yr, general population), we observed detrimental associations of sedentary time with waist circumference, triglyceride levels, and 2-h plasma glucose (22,24). It is important to point out that the participants in all of these studies were primarily white adults of European descent (2,13,14,22,24). A key next step for this research is to examine whether the associations are consistent across different racial/ethnic groups, which is becoming feasible with the public availability of large multiethnic population-based data sets, particularly NHANES (29,44).

Objectively Assessed Sedentary Behavior: AusDiab Findings

We used accelerometers to assess sedentary time in a subsample of the AusDiab study participants. Sedentary time was defined as accelerometer counts less than 100 per minute (previously described) and was associated with a larger waist circumference and more adverse 2-h plasma glucose and triglyceride profiles, as well as a clustered metabolic risk score (22,24). The associations of sedentary time with these biomarkers (with the exception of triglyceride levels) remained significant after adjustment for time spent in moderate- to vigorous-intensity physical activities (22,24).

As logically would be expected, sedentary time and light-intensity activity time were highly negatively correlated (r = −0.96); more time spent in light-intensity activity is associated with less time spent sedentary. This suggests that it may be a feasible approach to promote light-intensity activities as a way of ameliorating the deleterious health consequences of sedentary time. Our evidence suggests that having a positive light-intensity activity/sedentary time balance (i.e., spending more time in light-intensity activity than sedentary time) is desirable because light-intensity activity has an inverse linear relationship with a number of cardiometabolic biomarkers (22,24).

Breaks in Sedentary Time: AusDiab Findings

One of the intriguing findings from our accelerometer measurement studies is that breaks in sedentary time (as distinct from the overall volume of time spent being sedentary) were shown to have beneficial associations with metabolic biomarkers (21). Sedentary time was considered to be interrupted if accelerometer counts rose up to or more than 100 counts per min (21). This can include behaviors that result in a transition from sitting to a standing position or from standing still to beginning to walk. Figure 2 is based on data from two of our AusDiab accelerometer study participants, showing a simple contrast between adults who have the same total volume of sedentary time, but who break up that time in contrasting patterns. The person whose data are shown in the right-hand panel of Figure 2 (the "Breaker") interrupts his or her sedentary time far more frequently than the person whose data are shown on the left panel (the "Prolonger").

Breaks in sedentary time: same amount of sedentary time, but different ways of accumulation. CPM, counts per minute. (Reprinted from Dunstan DW, Healy GM, Sugiyama T, Owen N. Too much sitting and metabolic risk - has modern technology caught up with us? US Endocrinol. 2009; 5(1):29-33. Copyright © 2009 Touch Briefings. Used with permission.)

Independent of total sedentary time, moderate- to vigorous-intensity activity time, and mean intensity of activity, we found that having a higher number of breaks in sedentary time was beneficially associated with waist circumference, body mass index, triglycerides, and 2-h plasma glucose (21). Figure 3 shows objectively measured waist circumference across quartiles of breaks in sedentary time. Those in the bottom quarter of the "breaks" distribution had, on average, a 6-cm larger waist circumference than did those in the top quarter of that distribution (21).

Associations of breaks in sedentary time with waist circumference (based on data from Healy et al. (21)).

These findings on breaks in sedentary time provide intriguing preliminary evidence on the likely metabolic health benefits of regular interruptions to sitting time, which we would argue are additional to the benefits that ought to accrue from reducing overall sedentary time. Interestingly, in a recent study (5), patterns of sedentary time accumulation (but not total sedentary time) were shown to differ among four groups of adults with various activity patterns (healthy group with active occupation, healthy group with sedentary occupation, group with chronic back pain, group with chronic fatigue syndrome). As we will go on to propose, although we believe that these are strongly indicative findings, there is the need to determine whether these associations can be confirmed in experimental manipulations of sitting time in the laboratory. Intervention studies where sedentary time is reduced or broken up in naturalistic settings such as the domestic environment or the workplace would also be needed.

Sedentary Behavior and Mortality

The significance of the evidence on the adverse cardiometabolic health consequences of prolonged sitting time is underscored by findings from a mortality follow-up of participants in the Canada Fitness Surveys. Canadians who reported spending most of their day sitting had significantly poorer long-term mortality outcomes than did those who reported that they spent less time sitting. These relationships with mortality were consistent across all levels of a self-report measure of overall sitting time. Participants estimated the broad fractions of their waking hours that were spent sitting. Importantly, the sitting time-mortality relationships were apparent even among those who were physically active and the relationships were stronger among those who were overweight or obese (25). In a follow-up of AusDiab study participants during 6.5 yr, high levels of TV time were significantly associated with increased all-cause and cardiovascular disease mortality rates (9). Each 1-h increment in TV time was found to be associated with an 11% and an 18% increased risk of all-cause and cardiovascular disease mortality rates, respectively. Furthermore, relative to those watching less TV (<2 h·d−1), there was a 46% increased risk of all-cause mortality and an 80% increased risk of cardiovascular disease mortality in those watching TV 4 h·d−1 or more. These increased risks were independent of traditional risk factors such as smoking, blood pressure, cholesterol level, and diet, as well as leisure time physical activity and waist circumference. A recent study from the United States (47) examined sedentary behaviors in relation to cardiovascular mortality outcomes based on 21 yr of follow-up of 7744 men. Those who reported spending more than 10 h·wk−1 sitting in automobiles (compared with <4 h·wk−1) and more than 23 h of combined television time and automobile time (compared with <11 h·wk−1) had an 82% and 64% greater risk of dying from cardiovascular disease, respectively. TV time alone was not a significant predictor (47).


Looking Back Through a Sedentary Behavior Lens

Emerging findings on sedentary behavior suggest a different perspective through which findings of earlier physical activity and health research studies may be reexamined (we thank William L. Haskell, Ph.D., FACSM for stimulating these observations). For example, physical activity epidemiology studies that have assessed physical activity comprehensively often have included measures of sitting time, which has been used mainly to derive overall daily energy expenditure estimates. We would predict (perhaps boldly) that if such studies were to be revisited, with further analyses being conducted using sitting time as a distinct exposure variable, that strong evidence would be found for deleterious effects on subsequent health outcomes, independent of those related to physical inactivity.

Other potentially fruitful areas in which the relevance of existing evidence could be reexamined are the NASA zero-gravity studies. Comparing findings of those studies (which relate to the metabolic consequences of extreme muscular unloading) with those of the recent findings from inactivity physiology (16,17) may yield further insights relating to the underlying biology of prolonged sedentary time.

Research on physical activity and health had its roots in early occupational epidemiological studies that assessed workers in jobs that primarily involve sitting as the comparison groups against which the protective benefits of physically active work were highlighted (4,17,18). In the perspective of the new evidence that we have highlighted, conducting further occupational epidemiological studies using new objective measurement capabilities and examining a range of cardiometabolic and inflammatory biomarkers as intermediate outcomes could yield valuable insights.

Sedentary Behavior Research Strategy

Our population health research program on sedentary behavior is guided by the behavioral epidemiology framework (34,36). Figure 4 shows six research phases. As we previously demonstrated, evidence within the first phase (examining the relationships of sedentary behavior to cardiometabolic biomarkers and health outcomes) has strengthened rapidly during the past 10 yr.

Behavioral epidemiology framework: phases of evidence for a population health science of sedentary behavior.

Prolonged periods of sitting in people's lives need to be measured precisely (phase 2). Their contextual determinants - that is, behavior settings (32,35) - need to be identified in domestic, workplace, transportation, and recreation contexts (phase 3). We have argued previously for a research focus on the distinct environmental determinants of sedentary behaviors, in contexts where they can be amenable to intervention (31,32,37,41). The feasibility and efficacy of such interventions need to be tested rigorously (phase 4). Importantly, public health policy responses need to be informed by evidence from all of these phases. Compared with the challenges for physical activity and public health (19), sedentary behavior may be less of a "moving target" in this context and may be shown to be a tractable public health objective (4).

The Population Health Science of Sedentary Behavior: Research Opportunities

Different sedentary behaviors and their interactions with physical activity need to be examined in a range of contexts. For example, we have demonstrated that leisure time Internet and computer use is related to overweight and obesity in Australian adults (45), and that habitual active transport reduces the impact of TV time on body mass index (40). Having identified these relationships, our program is now broadening the evidence base through research with other populations. New studies include work with the large population-based data set from the NHANES from the United States, examining potential racial and ethnic differences in the relationships of total sedentary time and breaks in sedentary time with cardiometabolic biomarkers. We have demonstrated significant associations of TV time with excess body weight among high school students in regional mainland China (52). In the context of the rapid economic development and increasing urbanization among the populations of many developing countries, documenting the health consequences of reductions in physical activity and increases in sedentary time will be crucial for informing preventive health measures (38).

Studies with high-risk groups also are required. For example, we examined accelerometer-derived physical activity, sedentary time, and obesity in breast cancer survivors, showing physical activity to be protective, but there was no deleterious relationship for sedentary time (28). Significant prospective relationships of TV time with weight gain during 3 yr were identified in a large population-based cohort of Australian colorectal cancer survivors (49). More such etiologic research is needed to examine potential relationships between too much sitting and the development of other diseases that have been linked to metabolic risk factors.

For the second phase of the behavioral epidemiology framework (measurement; Fig. 4), there is the need to identify the reliability and validity of self-report instruments (6). Population-based descriptive epidemiological studies using high-quality measures are needed. For example, we have shown that Australian adults with lower levels of educational attainment and those living in rural areas are more likely to be in the highest TV time categories (7). We also have demonstrated that for Australian women, being in the higher categories of TV time can be associated with a broader pattern of leisure time sedentary behavior and with being less likely to meet physical activity recommendations (39). Using American Cancer Society data from a large population-based study, we have identified clusters of adults in the 4 h or more category of TV time who are less educated, obese, and snacking while watching TV (26).

Studies have begun to identify the environmental correlates of sedentary behavior, and initial findings seem puzzling. Among urban Australians, lower levels of objectively assessed neighborhood walkability (poorly connected streets, low levels of residential density, and limited access to destinations) were found to be associated with higher TV time in women (41). However, a recent study in the city of Ghent, Belgium, showed higher levels of walkability to be associated with higher amounts of accelerometer-assessed sedentary time (46). These apparently contradictory outcomes require further research. Such findings have potential implications for the emerging area of research on built environment/obesity relationships, within which sedentary behavior is likely to have a significant role (30).

Research on sedentary behaviors also needs to be extended beyond the promising initial studies on TV time to understand the potential health consequences of other common sedentary behaviors. Evidence on the metabolic correlates of prolonged sitting in motor vehicles would be particularly informative in the light of recent evidence on relationships with premature mortality (47). The social and environmental attributes associated with high levels of time spent sitting in automobiles also need to be identified.

The highest priority for the sedentary behavior research agenda is to gather new evidence from prospective studies, human experimental work, and intervention trials. There is the particular need to build on the promising findings on relationships of sedentary time - overall sitting time, TV time, and time sitting in automobiles - with premature mortality (9,25,47). Controlled experimental studies with humans also should be particularly informative. For example, we are currently conducting a laboratory study experimentally manipulating different "sedentary break" conditions and examining associated changes in cardiometabolic biomarkers (focusing on levels of triglycerides, glucose, and insulin).

Field studies also are needed on the feasibility and acceptability of reducing and breaking up occupational, transit, and domestic sedentary time. For example, in a weight control intervention trial for adults with type 2 diabetes, we are testing the impact of a sedentary behavior reduction intervention module and examining behavioral and biomarker changes associated with reducing and breaking up sedentary time. There are multiple research opportunities that can be explored through integrating sedentary behavior change intervention into physical activity trials. When accelerometer data are gathered from such studies, sedentary time measures can be derived (21,22,24), and unique hypotheses may readily be tested. It is imperative that the field now moves to obtain such evidence through intervention trials, which will take the science beyond the inherent logical limitations of cross-sectional evidence.

Eleven Research Questions for a Science of Sedentary Behavior

  1. Can further prospective studies examining incident disease outcomes confirm the initial sedentary behavior/mortality findings?
  2. Can sedentary behavior/disease relationships be identified through reanalyses of established prospective epidemiological data sets by treating sitting time as a distinct exposure variable?
  3. What are the most valid and reliable self-report and objective measures of sitting time for epidemiological, genetic, behavioral, and population health studies?
  4. Are the TV time-biomarker relationships for women pointing to important biological and/or behavioral sex differences?
  5. What amounts and intensities of activity might be protective in the context of prolonged sitting time?
  6. What genetic variations might underlie predispositions to sit and greater susceptibility to the adverse metabolic correlates?
  7. What is the feasibility of reducing and/or breaking up prolonged sitting time for different groups (older, younger) in different settings (workplace, domestic, transit)?
  8. If intervention trials show significant changes in sitting time, are there improvements in the relevant biomarkers?
  9. What are the environmental determinants of prolonged sitting time in different contexts (neighborhood, workplace, at home)?
  10. What can be learned from the sitting time and sedentary time indices in built-environment/physical activity studies?
  11. Can evidence on behavioral, adiposity, and other biomarker changes be gathered from "natural experiments" (e.g., the introduction of height-adjustable workstations or new community transportation infrastructure)?


Practical and policy approaches to addressing too much sitting as a population health issue will involve innovations on multiple levels. For example, public information campaigns may emphasize reducing sitting time as well as increasing physical activity. There may be more widespread use of innovative technologies that can provide more opportunities to reduce sitting time (e.g., height-adjustable desks) or new regulations in workplaces to reduce or break up extended periods of job-related sitting. Active transport modes can be promoted not only as opportunities for walking, but also as alternatives to the prolonged periods that many people spend sitting in automobiles. Providing nonsitting alternatives at community entertainment venues or events also may be considered. If evidence on the deleterious health impact of too much sitting continues to accumulate as we predict, and if such innovations are implemented, there will be the need for systematic evaluations, particularly of approaches that have the potential for broader dissemination.

Anecdotally, the recent experience in Australia has been that initiatives in the final phase of the behavioral epidemiology framework ("using relevant evidence to inform public health guidelines and policy") have already begun. This is happening largely on the basis of the first-phase evidence presented in Figure 4 ("identifying relationships of sedentary behavior with health outcomes"). For example, the Australian National Preventative Health Task Force Report includes explicit recommendations to address prolonged sitting in the workplace in the context of reducing the burden of overweight and obesity, type 2 diabetes, and cardiovascular disease. The Western Australian state division of the Heart Foundation included reducing sitting time in a 2009 statewide mass media campaign for obesity prevention. In the state of Queensland, Health Promotion Queensland (a cross-departmental body) commissioned an evidence-based review in 2009 on health impacts and interventions to reduce workplace sitting, with a view to future practical initiatives. Thus, there are growing expectations in Australia that too much sitting is a real and substantial risk to health. However, it remains to be seen whether the science of sedentary behavior will deliver consistent new findings in all of the research areas that are needed to inform such innovations (Fig. 4).

Given the consistency of research findings reported thus far on sedentary behavior and health, we expect that in the near future there will be a stronger body of confirmatory evidence from prospective studies and intervention trials. Furthermore, we predict that the next iteration of the Physical Activity and Public Health recommendations of ACSM/AHA will include a statement on the health benefits of reducing and breaking up prolonged sitting time.


N. Owen is supported by a Queensland Health Core Research Infrastructure grant and by a National Health and Medical Research Council (NHMRC) Program grant funding (no. 301200; no. 569940). G.N. Healy is supported by an NHMRC (no. 569861)/National Heart Foundation of Australia (PH 08B 3905) Postdoctoral Fellowship. D.W. Dunstan is supported by a Victorian Health Promotion Foundation Public Health Research Fellowship.


1. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of physical activities: An update of activity codes and MET intensities. Med. Sci. Sports Exerc. 2000; 32(Suppl 9):S498-504.
2. Balkau B, Mhamdi L, Oppert JM, et al. Physical activity and insulin sensitivity: The RISC study. Diabetes 2008; 57(10):2613-8.
3. Bey L, Hamilton MT. Suppression of skeletal muscle lipoprotein lipase activity during physical inactivity: A molecular reason to maintain daily low-intensity activity. J. Physiol. 2003; 551(Pt 2):673-82.
4. Brown WJ, Bauman AE, Owen N. Stand up, sit down, keep moving: Turning circles in physical activity research? Br. J. Sports Med. 2009; 43(2):86-8.
5. Chastin SF, Granat MH. Methods for objective measure, quantification and analysis of sedentary behavior and inactivity. Gait Posture. 2010; 31(1):82-6.
6. Clark BK, Sugiyama T, Healy GN, et al. Validity and reliability of measures of television viewing time and other non-occupational sedentary behavior of adults: A review. Obes. Rev. 2009; 10(1):7-16.
7. Clark BK, Sugiyama T, Healy GN, et al. Socio-demographic correlates of prolonged television viewing time in Australian men and women: The AusDiab study. J. Phys. Act. Health (in press).
8. Donahoo WT, Levine JA, Melanson EL. Variability in energy expenditure and its components. Curr. Opin. Clin. Nutr. Metab. Care 2004; 7(6):599-605.
9. Dunstan DW, Barr ELM, Healy GN, et al. Television viewing time and mortality: The Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Circulation 2010; 121:384-91.
10. Dunstan DW, Salmon J, Healy GN, et al. Association of television viewing with fasting and 2-h postchallenge plasma glucose levels in adults without diagnosed diabetes. Diabetes Care 2007; 30(3):516-22.
11. Dunstan DW, Salmon J, Owen N, et al. Associations of TV viewing and physical activity with the metabolic syndrome in Australian adults. Diabetologia 2005; 48(11):2254-61.
12. Dunstan DW, Salmon J, Owen N, et al. Physical activity and television viewing in relation to risk of undiagnosed abnormal glucose metabolism in adults. Diabetes Care 2004; 27(11):2603-9.
13. Ekelund U, Brage S, Griffin SJ, Wareham NJ. Objectively measured moderate- and vigorous-intensity physical activity but not sedentary time predicts insulin resistance in high-risk individuals. Diabetes Care 2009; 32(6):1081-6.
14. Ekelund U, Griffin SJ, Wareham NJ. Physical activity and metabolic risk in individuals with a family history of type 2 diabetes. Diabetes Care 2007; 30(2):337-42.
15. Foulis SA, Larsen RG, Callahan DM, Kent-Braun JA. An accelerometer-based approach for measuring physical activity in young and older adults and its relevance to physical function measures. Med. Sci. Sports Exerc. 2008; 40(5):S62.
16. Hamilton MT, Hamilton DG, Zderic TW. Exercise physiology versus inactivity physiology: An essential concept for understanding lipoprotein lipase regulation. Exerc. Sport Sci. Rev. 2004; 32(4):161-6.
17. Hamilton MT, Hamilton DG, Zderic TW. Role of low energy expenditure and sitting in obesity, metabolic syndrome, type 2 diabetes, and cardiovascular disease. Diabetes 2007; 56(11):2655-67.
18. Hamilton MT, Healy GN, Dunstan DW, Zderic TW, Owen N. Too little exercise and too much sitting: Inactivity physiology and the need for new recommendations on sedentary behavior. Curr. Cardiovasc. Risk Rep. 2008; 2(4):292-8.
19. Haskell WL, Blair SN, Hill JO. Physical activity: Health outcomes and importance for public health policy. Prev. Med. 2009; 49(4):280-2.
20. 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(8):1423-34.
21. Healy GN, Dunstan DW, Salmon J, et al. Breaks in sedentary time: Beneficial associations with metabolic risk. Diabetes Care 2008; 31(4):661-6.
22. Healy GN, Dunstan DW, Salmon J, et al. Objectively measured light-intensity physical activity is independently associated with 2-h plasma glucose. Diabetes Care 2007; 30(6):1384-9.
23. Healy GN, Dunstan DW, Salmon J, et al. Television time and continuous metabolic risk in physically active adults. Med. Sci. Sports Exerc. 2008; 40(4):639-45.
24. 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 2008; 31(2):369-71.
25. Katzmarzyk PT, Church TS, Craig CL, Bouchard C. Sitting time and mortality from all causes, cardiovascular disease, and cancer. Med. Sci. Sports Exerc. 2009; 41(5):998-1005.
26. King AC, Goldberg JH, Salmon J, et al. Correlates of prolonged television viewing time in U.S. adults to inform program development. Am. J. Prev. Med. 2010; 38(1):17-26.
27. Levine JA, Schleusner SJ, Jensen MD. Energy expenditure of nonexercise activity. Am. J. Clin. Nutr. 2000; 72(6):1451-4.
28. Lynch BM, Dunstan DW, Healy GN, et al. Objectively measured physical activity and sedentary time of breast cancer survivors, and associations with adiposity: Findings from NHANES (2003-2006). Cancer Causes Control 2009; 21(2):283-8.
29. Matthews CE, Chen KY, Freedson PS, et al. Amount of time spent in sedentary behaviors in the United States, 2003-2004. Am. J. Epidemiol. 2008; 167(7):875-81.
30. Owen N. Effects of the build environment on obesity. In: Bouchard C, Katzmarczyk P, editors. Physical Activity and Obesity, Champaign, IL: Human Kinetics; 2010. p. 199-202.
31. Owen N, Bauman A, Brown W. Too much sitting: A novel and important predictor of chronic disease risk? Br. J. Sports Med. 2009; 43(2):81-3.
32. Owen N, Leslie E, Salmon J, Fotheringham MJ. Environmental determinants of physical activity and sedentary behavior. Exerc. Sport Sci. Rev. 2000; 28(4):153-8.
33. Pate RR, O'Neill JR, Lobelo F. The evolving definition of "sedentary." Exerc. Sport Sci. Rev. 2008; 36(4):173-8.
34. Sallis JF, Owen N. Physical Activity and Behavioral Medicine. Thousand Oaks, CA: Sage; 1999.
35. Sallis JF, Owen N, Fisher EB. Ecological models of health behavior. In: Glanz K, Rimer BK, and Viswanath K, editors. Health Behavior and Health Education: Theory, Research, and Practice, San Francisco, CA: Jossey-Bass; 2008. p. 465-82.
36. Sallis JF, Owen N, Fotheringham MJ. Behavioral epidemiology: A systematic framework to classify phases of research on health promotion and disease prevention. Ann. Behav. Med. 2000; 22(4):294-8.
37. Salmon J, Owen N, Crawford D, Bauman A, Sallis JF. Physical activity and sedentary behavior: A population-based study of barriers, enjoyment, and preference. Health Psychol. 2003; 22(2):178-88.
38. Spanier PA, Marshall SJ, Faulkner GE. Tackling the obesity pandemic: A call for sedentary behavior research. Can. J. Public Health. 2006; 97(3):255-7.
39. Sugiyama T, Healy GN, Dunstan DW, Salmon J, Owen N. Is television viewing time a marker of a broader pattern of sedentary behavior? Ann. Behav. Med. 2008; 35(2):245-50.
40. Sugiyama T, Merom D, Leslie E, Reeves M, Owen N. Habitual active transport moderates the association of TV viewing time with body mass index. J. Phys. Act Health. 2010; 7(1):11-6.
41. Sugiyama T, Salmon J, Dunstan DW, Bauman AE, Owen N. Neighborhood walkability and TV viewing time among Australian adults. Am. J. Prev. Med. 2007; 33(6):444-9.
42. Thorp AA, Healy GN, Owen N, et al. Deleterious associations of sitting time and television viewing time with cardio-metabolic risk biomarkers: AusDiab 2004-2005. Diabetes Care 2010; 33(2):327-34.
43. 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(Suppl 2):S208-217.
44. Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med. Sci. Sports Exerc. 2008; 40(1):181-8.
45. Vandelanotte C, Sugiyama T, Gardiner P, Owen N. Associations of leisure-time internet and computer use with overweight and obesity, physical activity and sedentary behaviors: Cross-sectional study. J. Med. Internet Res. 2009; 11(3):e28.
46. Van Dyck D, Cardon G, Deforche B, Owen N, Sallis JF, De Bourdeaudhuij I. Associations of neighborhood walkability with sedentary time in Belgian adults. Am. J. Prev. Med. (in press).
47. Warren TY, Barry V, Hooker SP, Sui X, Church T, Blair SN. Sedentary behaviors increase risk of cardiovascular disease mortality in men. Med. Sci. Sports Exerc. 2010; 42(5):879-85.
48. Wijndaele K, Healy GN, Dunstan DW, et al. Increased cardio-metabolic risk is associated with increased TV viewing time [published online ahead of print February 4, 2010]. Med. Sci. Sports Exerc. 2010.
49. Wijndaele K, Lynch BM, Owen N, et al. Television viewing time and weight gain in colorectal cancer survivors: A prospective population-based study. Cancer Causes Control. 2009; 20(8):1355-62.
50. Wilkinson L, Friendly M. The history of the cluster heat map. Am. Stat. 2009; 63(2):179-84.
51. Williams DM, Raynor HA, Ciccolo JT. A review of TV viewing and its association with health outcomes in adults. Am. J. Lifestyle Med. 2008; 2(3):250-9.
52. Xu F, Li J, Ware RS, Owen N. Associations of television viewing time with excess body weight among urban and rural high-school students in regional mainland China. Public Health Nutr. 2008; 11(9):891-6.

environmental and social change; TV time; breaks in sedentary time; accelerometer measurement; blood glucose; triglycerides; metabolic health

©2010 The American College of Sports Medicine