Physical activity and health: Dose-response issues

Over the past two decades, compelling scientific evidence has accumulated demonstrating that regular participation in physical activity provides important health benefits. As a result, promotion of physical activity has become recognized as a public health issue, and in recent years several prominent health organizations have issued pronouncements noting the disease prevention and health promotion benefits of physical activity (Fletcher et al., 1992; Killoran, 1994; Fate et al., 1995). In communicating the significance of regular physical activity to the public, it is important to indicate the types and amounts of physical activity that are needed for health benefits. This need raises several important questions and these include (a) how much and what types of physical activity are needed for each of the specific health benefits that have been associated with physical activity? (b) is there an optimal amount of physical activity that can be recommended? and (c) is there a minimal amount of activity that could be endorsed? A strategy that could be employed in answering these questions is to study the so-called dose-response relationship between physical activity and health (Blair, Wells, Weathers, & Paffenbarger, 1994; Haskell, 1994). This relationship describes the level of health benefits associated with various levels of physical activity. Because health is a multifactorial construct and because physical activity has been shown to affect numerous health parameters, a study of the global dose-response relationship for physical activity and health requires examining numerous specific dose-response gradients. The purpose of this paper is to examine our current knowledge of the dose-response relationship between physical activity and health. This examination will involve reviewing the dose-response relationships for several specific health outcomes that are known to be associated with habitual physical activity. These include (a) one physical fitness parameter (cardiorespiratory fitness); (b) three physiological factors associated with chronic disease risk (high-density lipoprotein cholesterol [HDL-C], resting blood pressure, and glucose tolerance); (c) morbidity from three chronic diseases (coronary heart disease, Type II diabetes, and hypertension); and (d) all-cause mortality. To the extent allowed by the existing scientific literature, the effects of two markers of physical activity ''dose'' will be considered for each health outcome. First, the effect of the amount of physical activity performed, as indicated by estimated weekly caloric expenditure (or a surrogate measure thereof),will be considered. Second, the effect of exercise intensity, independent of overall amount of activity, will be examined.