Perhaps the greatest barriers to achieving major public health advances in the 21st century will result from pandemic paradigm paralysis or the widespread inability to envision alternative or new models of thinking. One potential example of this phenomenon could turn out to be the continued focus on moderate and vigorous physical activity as the dominant health-related aspect of human movement. The current model of physical activity and health is well supported by over 60 years of scientific inquiry, and the beneficial effects of moderate-to-vigorous physical activity have been more clearly defined in recent years (1 –4). However, if we are complacent with the existing paradigm—that increasing levels of moderate and vigorous levels of physical activity will result in the greatest improvements in public health—then we may not obtain the full return on investment with respect to improving quality of life and life expectancy through patterns of human movement. Emerging evidence for the role of sedentary behavior on health, which may be independent of physical activity per se, finds us at a crossroad with respect to prescribing optimal daily human movement patterns for health. Human movement represents a complex behavior that is influenced by personal motivation, health and mobility issues, genetic factors, and the social and physical environments in which people live. These factors undoubtedly exert an influence on the propensity to engage in sedentary behaviors as well as in physical activity. However, the biological, social, and environmental pathways leading to sedentary behavior versus physical activity may be different. Further, the health effects associated with sedentary behavior and physical activity may be the result of different biological mechanisms (5). Humans are designed for movement. Energy balance has been a central selective force throughout human evolutionary history, and humans have evolved to have high levels of energy expenditure, even more so than modern nonhuman primates (6). Obtaining dietary energy and nutrients from the environment traditionally required an expenditure of energy through human movement. Factors related to the expansion of the African grasslands between 2.5 and 1.5 million years ago and the emergence of Homo were major contributors to changes in both brain size and foraging behaviors (6,7). Early Homo (H. habilis and H. erectus) appeared at a time of rapid brain evolution with early Homo having an average brain size of 600–900 cc compared with earlier australopithecines with an average brain size of 400–500 cc (7). The larger brain size of Homo required higher quality diets, which necessitated larger foraging ranges, resulting in greater total energy expenditure. At the same time, the transition from a forest to savanna environment caused changes in resource distribution that would have also resulted in increases in foraging ranges and total energy expenditure (6). Much of human evolution has occurred as hunter-gatherers (3–4 million years), while recent advances in agriculture and technology have occurred over a short time frame (∼10,000 years). Eaton and Eaton (8) have estimated that Stone Age humans had an energy efficiency ratio of 2.25 (i.e., expending 1 kJ of energy to acquire 2.25 kJ of dietary energy) compared with an efficiency ratio of 3.66 for modern humans, which represents more than a 50% increase in efficiency. Modern humans in the Western world have relatively low levels of physical activity compared with contemporary hunter-gatherers. Hayes et al. (9) reported that the total energy expenditure/resting energy expenditure or Physical Activity Level (PAL) among subsistence-level human populations approximates 3.2, while among representative humans living in contemporary society, the PAL is ∼1.67. The impact of the transition from a semi-subsistent existence to a Western lifestyle on physical fitness levels are exemplified by work in an Inuit community (Igloolik, northern Canada) (10,11). Studies in the population from 1970 through 1990 demonstrated marked reductions in average aerobic fitness (ml · kg−1 · min−1) over time in all age-groups (10,11). Recent work among Old Order Amish living a traditional agricultural lifestyle indicates that this population engages in more daily movement than contemporary Americans. The average number of steps per day taken by Amish men and women were 18,425 steps per day and 14,196 steps per day, respectively (12). These values are considerably higher than recent estimates for contemporary U.S. adults (13,14) (Fig. 1). FIG. 1. Average steps per day among Old Order Amish men and women (12) compared with contemporary U.S. adults in the 2005–2006 U.S. NHANES (13) and the 2003 America on the Move Study (14). The weighted evidence indicates that humans evolved in environments that required higher levels of human movement than are required today. By becoming more efficient at extracting energy from the environment, there is now a lower level of expenditure required to subsist. Some studies have documented lower levels of physical activity among contemporary humans compared with those living in more primitive societies. A negative consequence to the observed improvements in energetic efficiency is the proliferation of health concerns that are related to low levels of physical activity and/or high levels of sedentary behavior. Physical activity and health. The modern field of physical activity epidemiology arguably began with the studies of Morris et al. (15) conducted in the early 1950s among employees of the London Transport Executive and Post Office employees. Their results demonstrated that physically active men (bus conductors and postmen) had lower mortality rates from heart disease than less active workers (bus drivers and telephone switchboard operators). These early studies provided evidence for a role of physical activity in averting premature mortality; however, it has also recently been hypothesized that some of the observed associations may be explained by differences in time spent sitting rather than being less physically active per se (i.e., bus drivers sit more than conductors) (5). The independent roles of sitting versus physical activity cannot be determined from these early studies. A great volume of evidence has accrued over the past 60 years on the relationship between physical activity and health. This culminated in the 1996 U.S. Surgeon General's report on Physical Activity and Health (3) and the 2008 Physical Activity Guidelines for Americans (16). Two classic studies are used here to illustrate the relationships between physical activity, cardiorespiratory fitness, and all-cause mortality. The first, the Harvard Alumni Study (Fig. 2 A) (17), was an analysis of physical activity and all-cause mortality over 16 years among ∼17,000 men that revealed an inverse dose-response relationship between physical activity and all-cause mortality rates. Greater physical activity was associated with a lower risk of death, and men expending >2000 kcal per week in physical activity had a 27% lower risk of mortality compared with men expending 12 h/week 0.96 (0.68–1.36) 0.94 Riding in car 10 h/week 1.37 (1.01–1.87) 0.01 EPIC-Norfolk Study (32) 13,197 men and women 9.5 years TV viewing All-cause, CVD, and cancer mortality‖ All-cause mortality per h/day 1.05 (1.01–1.09) CVD mortality per h/day 1.08 (1.01–1.16) Cancer mortality per h/day 1.04 (0.98–1.10) *Adjusted for age, geographic area, occupation, history of diabetes, smoking, alcohol intake, BMI, total energy intake, heavy physical work or strenuous exercise, walking or standing, and leisure-time sports or exercise; ‡adjusted for age, sex, smoking, alcohol consumption, leisure-time physical activity, and physical activity readiness; †adjusted for age, sex, smoking, education, total energy intake, alcohol intake, diet quality index, waist circumference, hypertension, cholesterol, HDL cholesterol, triglycerides, lipid-lowering medication use, glucose tolerance status, and exercise time; §adjusted for age, physical inactivity, current smoker, alcohol intake, BMI, family history of CVD, hypertension, diabetes, and hypercholesterolemia; and ‖adjusted for age, gender, education level, smoking status, alcohol consumption, history of diabetes, family history of CVD, family history of cancer, total physical activity energy expenditure, and medication use for hypertension or dyslipidemia (not in models for cancer mortality). NS, not significant. Although there is compelling evidence that sedentary behaviors such as sitting and TV viewing are related to premature mortality, a question that remains to be answered is whether these behaviors are independent of total physical activity levels per se. The studies presented in Table 1 provide evidence on this question using two strategies. First, all of the studies included physical activity in a final multivariate-adjusted regression model, and the results were largely unchanged from the models that did not include physical activity as a covariate (30 –34). Second, some studies stratified their analyses by physical activity level or included interaction terms in the statistical models. Interaction terms for sedentary behavior and physical activity in the AusDiab study, the Canada Fitness Survey, and the EPIC-Norfolk Study were not significant, and their inclusion did not significantly modify the observed relationships (30 –32). Stratifying analyses by physical activity level has led to different results. In the ACLS, there was a significant linear trend across categories of time spent riding in a car and CVD mortality in physically inactive men (P = 0.02) but not in physically active men (P = 0.13) (33). On the other hand, in the Canada Fitness Survey, there were significant positive associations between daily sitting time and all-cause mortality in both physically inactive (P 40 h per week of TV compared with women watching ≤1 h per week (36). The relationship between TV viewing and type 2 diabetes over 10 years was even stronger in men from the Health Professionals Follow-Up Study (HPFS). The multivariate-adjusted RR of developing type 2 diabetes was 3.02 (1.53–5.93) in men watching ≥40 h per week of TV compared with men watching ≤1 h per week, and these effects were largely independent of leisure-time physical activity (37). FIG. 4. Relationship between TV viewing and the development of obesity and type 2 diabetes over 6 years of follow-up in women 30–55 years of age from the Nurses' Health Study (36). Models are adjusted for age, smoking, alcohol use, hormone use, physical activity, total fat, cereal fiber, glycemic load, and total calories. Among Spanish university graduates followed prospectively for 40 months, those in the upper quartile of sedentary behavior had an RR of 1.48 (1.01–2.18) for developing hypertension compared with the lower quartile (38). However, in sub-analyses, the association with incident hypertension was evident only for driving and computer use and not for TV viewing. Among the participants in the Women's Health Initiative Observational Study (WHI-OS), the RR of incident CVD over 5.9 years of follow-up was 1.68 (1.07–2.64) among women sitting for ≥16 h per day compared with those sitting twofold increase) and continuous chronic inactivity (>threefold decrease) impact LPL mRNA. These different mechanisms suggest that the processes governing metabolism during common sedentary behaviors could be quite distinct from the effects observed in exercise studies. Further, a global gene-expression profiling study has identified 38 genes that are upregulated by just 12 h of physical inactivity (hind limb unloading) in rats, and 27 of these genes remained above control levels after returning to standing and ambulation of the hind limbs for 4 h, suggesting that some of the effects of sedentary behavior will persist long after the behavior is changed (63). Taken together, these results indicate that the gross metabolic disturbances observed with sedentary behavior result from metabolic alterations at the level of the muscle. Further research is required to elucidate the full spectrum of potential mechanisms in different organs and tissues that play a role in explaining the health effects associated with sedentary behavior. Conclusions. The current public health recommendations for moderate and vigorous physical activity are the result of more than 60 years of scientific inquiry that has produced evidence for a causal link between physical activity and health. This evidence comes from a spectrum of study designs including prospective observations, clinical intervention trials, and mechanistic studies in the laboratory. By comparison, the evidence for an independent effect of sedentary behavior on health is just now emerging. Given the rapid accumulation of this evidence over the last few years, it has been suggested that public health recommendations targeting sedentary behavior are needed (68). The evidence for an independent effect of sedentary behavior on health is both intriguing and convincing; however, several important questions remain. What are the dose-response relationships between sedentary behaviors and various health outcomes? Are health risks equivalent across all types of sedentary behaviors? Do reductions in sedentary behavior result in changes in health parameters or disease incidence? What types of interventions to reduce sedentary behavior are feasible from a public health standpoint? Given the ubiquitous nature of sedentary behaviors, what activities could feasibly be used to replace them? What are the distinct pathophysiological mechanisms linking sedentary behavior and health? These questions will provide a fertile area of research in the coming years. At present, the available evidence suggests that it is prudent to recommend that time spent in sedentary behaviors be minimized; however, optimal levels of sedentary behavior to recommend are not currently known. The emergence of the physical inactivity paradigm (5) has highlighted the potential role that all aspects of human movement can play in impacting health. Most current physical activity guidelines focus on achieving 30 min per day or 150 min per week of moderate-to-vigorous physical activity. This represents only 1.5% of a total week (10,080 min), or perhaps 3% of the time we spend awake. Recent data from NHANES 2003–2004 from objective physical activity monitoring (accelerometry) indicate that less than 5% of the population is obtaining the recommended level of physical activity (69). Thus, efforts must be redoubled in order to achieve demonstrable increases in physical activity levels. On the other hand, sedentary behaviors (<100 accelerometer counts per minute) account for ∼55% of an American's typical day (70). We must begin to explore novel approaches to reduce the widespread exposure to sedentary behaviors, as the potential health benefits to be gained could be substantial.
