Until the beginning of the 20th century, physical activity was an inevitable part of people’s lives; almost all daily endeavours, including hunting and gathering food, required physical exertion. During the 20th century, the balance shifted from active to more sedentary lifestyles, with increasingly prevalent adverse health consequences. Now, at the beginning of the 21st century, most forms of transport and work are automated (and often involve long periods of sitting), and the “non-working” day includes mostly sedentary leisure time pursuits and light domestic tasks.

The reduction in daily activity is clearly illustrated by comparing data from two studies. The first, conducted in 2002, was a study of physical activity in the 15 EU nations. It reported an average energy expenditure of 24 MET.hours/week, which equates to less than 1 h of moderate-intensity activity each day.1 The second was a study in an Old Order Amish community in Canada, which reported an average energy expenditure of 253 MET.hours/week (or almost 10 h/day of activity) in 2004. In this community, the use of motorised transportation, electricity and modern conveniences is banned, and most people work on farms, tilling the soil with horses, or grow vegetables in family gardens.2

The lifestyle differences of the participants in these two studies illustrate the drastic reduction in total daily energy expenditure that is now occurring in both developed and developing countries, and which is in turn prompting changes to the ways we think about the behavioural epidemiology of physical activity and sedentary time.

FROM “OCCUPATIONAL SITTING” TO “EXERCISE”

The first epidemiological studies that connected lack of physical activity with adverse health outcomes were by Jeremy Morris who published a series of papers in England in the 1950s. This seminal research identified higher rates of cardiovascular events in occupationally sedentary bus drivers and mail sorters than in more-active bus conductors and postal delivery workers.3 These, and subsequent epidemiological studies, identified the independent relationship between low levels of energy expenditure and adverse health outcomes.4^–6 The focus in these early epidemiological studies was on the health consequences of inactivity, especially time spent sitting in the occupational setting.

From the 1960s, the focus of much of the epidemiological work in this field changed to activity, particularly in the form of structured leisure time exercise, and the vast range of health benefits associated with it.7 Most research studies in this period recommended aerobic fitness training through vigorous exercise at least three times a week for more than 20 min.8 Although interventions achieving this level of vigorous exercise produced physiological benefits, early estimates across different surveys suggested that only about 10% of adults in Western developed countries complied with the “3×20” recommendation.9 This prompted another change in thinking and research, which began to challenge the public health utility of the “physical exercise” approach.

FROM (VIGOROUS) EXERCISE TO MODERATE PHYSICAL ACTIVITY

By the 1990s, epidemiological evidence showed that moderate-intensity activity could, through its effect on population-attributable risk of inactivity, have a greater impact on population health than vigorous-intensity activity.10 Increasing evidence from large-scale epidemiological studies (mostly of men) led to the public health recommendation that adults should accumulate 30 min of at least moderate-intensity physical activity on most days of the week.11 The evidence, summarised in the 1996 US Surgeon General’s report on physical activity and health,12 showed a clear dose–response relationship between activity and health outcomes, across increments of physical activity. Maximal benefit was observed in those who were moderately active on most days, with small additional increments in benefit accruing to those who reported doing vigorous exercise. This evidence has strengthened and remained consistent over time.13 Furthermore, a 2007 review which focused exclusively on studies of women found strong evidence for the health benefits of moderate-intensity activity for mid-age and older women.14 Importantly, this review found that it is never too late to start physical activity for improved health outcomes in mid and later life.

The 1996 US Surgeon General’s report12 and consistent subsequent epidemiological evidence led public health policy makers around the world to adopt the “moderate physical activity” message. The concept of accumulating physical activity, with short bouts of activity in diverse settings across the day—for example, in transport, occupations, gardening, housework and active play with children—gained increasing prominence through what has become known as the “active living” or “healthy lifestyle” approach.15 Although the epidemiological evidence in support of this accumulation hypothesis was limited,16 it was proposed that energy expenditure in bouts of as little as 10 min duration could summate across the day to provide health benefits.12

UPDATING AND CLARIFYING THE EVIDENCE

Accumulating evidence relating to the health benefits of physical activity resulted in an update of the US recommendations on physical activity and health in 2007.17 The recommended frequency of moderate-intensity activity was clarified as “on at least 5 days per week” with increased duration (to 60–90 min daily) proposed for obesity prevention and control. Strength training was recommended for the prevention of falls. Revised US guidelines, published in October 2008, emphasised the dose–response relationship between physical activity and health outcomes, with further benefits (above those resulting from the previously stated 150 min/week) for even higher levels of activity (more than 300 min of moderate-vigorous activity per week). Additional guidelines included muscle strengthening activities (resistance training on at least 2 days/week) for adults, and a recommendation of an hour/day of physical activity for children and adolescents.18

SEDENTARY BEHAVIOURS

Although some of the early research in this field was on the hazards of sitting (relative to being active) at work,3 5 and despite the fact that most working adults now spend many more hours sitting (not only at work, but also in transport and in leisure time), the revised guidelines did not say anything about sedentary behaviour. In the last 5 years, we have, however, started to see a revival of interest in the health effects of sitting, both at work and in leisure time, as well as in the benefits of active transport, such as cycling to work.19

Several cross-sectional studies have shown relationships between self-reported sitting time and body mass index and other health outcomes.20^–22 There are also early indications that self-reported sitting time (as a marker of sedentary behaviour) may be a predictor of weight gain in Australian women, even after adjustment for energy intake and leisure time activity.23 24 At the population level, there is also indirect evidence that contemporary patterns of transport, as well as occupational, domestic and leisure activities, impact negatively on daily total energy expenditure.25^–27

CHANGING PARADIGMS IN (IN)ACTIVITY RESEARCH

These studies signal a paradigm shift in thinking about the balance of activity and inactivity in all aspects of daily life, including transport, occupational and domestic work, and leisure, and its impact on population health outcomes. This is important, because, as we move towards the second decade of the 21st century, it is unlikely that, in the context of otherwise sedentary lifestyles, leisure time activity will be sufficient to prevent increasing population levels of overweight, obesity and chronic disease.28 29 Indeed, it may be that we have to decrease sitting and increase activity in transport and at work to restore the energy balance that resulted in much more stable body weights at the beginning of the last century than are evident today.

As the world oil price threatens to rise, we may be at the onset of a natural experiment that will see reduced sitting and increased activity for transport in the major centres where people live and work. We should be ready to evaluate any changes that occur by investing more effort now in the development of accurate population measures of sitting time and daily energy expenditure, and ensuring that these are included in our ongoing intervention and cohort studies, so that the health effects of changes in the balance of activity and sitting time can be evaluated in due course.

TURNING FULL CIRCLE?

Research into the epidemiology of physical activity and health has come full circle, from its early roots in occupational sitting, through aerobic fitness training, then moderate-intensity physical activity, to a contemporary perspective on the balance of inactivity and activity in different domains (transport, occupation, domestic and leisure) of everyday life. Understanding this balance between activity and sedentariness is fundamental to halting population weight gain and its adverse metabolic and health consequences. As Morris and Paffenbarger so eloquently showed us in their early work in this field, we must now start to evaluate the optimal patterns of activity and inactivity for the prevention or delayed onset of obesity and chronic disease in adult populations.