- G David Batty (), senior research fellow in epidemiology,
- I-Min Lee, associate professor of medicine
- Department of Social Medicine, Institute of Public Health, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen N, Denmark
- Division of Preventive Medicine, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, 900 Commonwealth Avenue East, Boston, MA 02215, USA
Fifty years of research confirms inverse relationship
Fifty years ago the first empirical investigation of what was subsequently termed the exercise hypothesis—physical activity reduces the occurrence of coronary heart disease—was undertaken by Morris et al.1 Using data from two cohorts of British workers, they reported lower rates of coronary heart disease in bus conductors than in less occupationally active bus drivers, and in postmen relative to deskbound telephonists and other office based employees. Although this research was pioneering, it was not without its shortcomings. Early statistical methods were limited in their capacity to explore the issue of confounding—for example, it was possible that higher levels of overweight, high blood pressure, stress, or pre-existing ischaemia in the less active groups, rather than their sedentary behaviour, placed them at increased risk of coronary heart disease. Further, the study focused exclusively on work activity. Morris et al, and subsequently Paffenbarger et al, went on to address these issues, showing physical activity in leisure time to be cardioprotective, an effect that held after a range of covariates were controlled for.2 3
The work of these researchers prompted a series of other investigations, including the study of the association between cardiorespiratory fitness—a physiological outcome of physical activity and therefore an objective proxy for it—and cardiovascular disease, which showed that higher levels, which were none the less eminently attainable by non-athletes, conferred protection against coronary heart disease.4 These studies, and those of physical activity, represent a range of methodological rigour and have, with few exceptions, shown an inverse association between activity and coronary heart disease, which is testimony to the robustness of the relation. Until recently this research has focused on men. In the past two decades, however, the same degree of consistency has been observed in women—a dose-response effect is again evident, with the highest rates of coronary heart disease seen in people who are inactive or have low cardiorespiratory fitness. This level of risk may be of slightly lower magnitude than that reported in men.
We still do not fully understand the pathways underlying the protective effect of physical exertion against coronary heart disease. In addition to well established improvements in blood pressure, lipid profile, insulin sensitivity, and body weight, activity may improve endothelial function and coronary blood flow and may be associated with beneficial changes in haemostatic and inflammatory variables.5 6 These warrant further examination. Given the public's reluctance to be physically active, future research also needs to identify the minimum dose (the total volume of activity, as well as the intensity, duration, and frequency), type (aerobic, strength training), and mode (walking, swimming) of activity associated with reduced coronary heart disease risk that will be most palatable to the sedentary populations. For the prevention of a range of chronic diseases, including coronary heart disease, the current recommendation is 30 minutes or more of moderately intense aerobic activity such as brisk walking or cycling on five or more occasions per week.7 Worryingly, however, the prevalence of activity in most Western societies does not match this: in England, for example, only about one quarter of men and women currently achieve this level,8 and these figures are lower still in elderly people.
For the sedentary majority, commonly cited barriers to participation include inconvenience (for example, lack of easily accessible facilities) and danger (in the United States injuries and fatalities among pedestrians and cyclists far exceed those for car occupants). Therefore, efforts to promote activity that focus on personal behaviour change—often delivered via face to face contact in a small group—may not be sufficiently effective. An alternative approach is one that also recognises the importance of physical and social environments as crucial determinants of physical activity both of a utilitarian nature and for leisure. Utilitarian activity levels seem to be higher in residentially dense neighbourhoods in mixed use (business and residential) that offer a greater concentration of street connections than neighbourhoods that are more sprawling in design where, presumably by necessity, travel by motorised transportation predominates.9 Perhaps because of these differing degrees of physical activity, people residing in more compact neighbourhoods are also less likely to report obesity and hypertension.10 Moreover, in Japan the life expectancy of people with more easy access to facilities conducive to physical activity in leisure time, such as parks and tree lined streets, exceeds that of people without such resources, even after the obvious and important socioeconomic differences between such areas are taken into account.11
Although the governments of many countries, including the United Kingdom, have the opportunities to implement such changes in urban form in order to make environments more conducive to physical activity, what they may lack is the political resolve. This may be short sighted. Given the high prevalence of sedentary behaviour and its association with a range of chronic diseases—which include not only coronary heart disease, but also stroke, type 2 diabetes, certain cancers, and osteoporotic fractures12—modification of physical activity may, as Morris indicated four decades after the publication of his seminal work, represent today's best buy in public health.13
Funding GDB is supported by a fellowship from the University of Copenhagen.
Conflict of interest None declared.