Chapter 4 Introduction

The Effects of Physical Activity on Health and Disease

This chapter examines the relationship of physical activity and cardiorespiratory fitness to a variety of health problems. The primary focus is on diseases and conditions for which sufficient data exist to evaluate an association with physical activity, the strength of such relationships, and their potential biologic mechanisms. Because most of the research to date has addressed the health effects of endurance-type physical activity (involving repetitive use of large muscle groups, such as in walking and bicycling), this chapter focuses on that type of activity. Unless otherwise specified, the term physical activity should be understood to refer to endurance-type physical activity. Less well studied are the health effects of resistance-type physical activity (i.e., that which develops muscular strength); when this type of physical activity is discussed, it is specified as such. Much of the research summarized is based on studies having only white men as participants; it remains to be clarified whether the relationships described here are the same for women, racial and ethnic minority groups, and people with disabilities.

Physical activity is difficult to measure directly. Three types of physical activity measures have been used in observational studies over the last 40 years. Most studies have relied on self-reported level of physical activity, as recalled by people prompted by a questionnaire or interview. A more objectively measured characteristic is cardiorespiratory fitness (also referred to as cardiorespiratory endurance) which is measured by aerobic power (see Chapter 2 for more information on measurement issues). Some studies have relied on occupation to classify people according to how likely they were to be physically active at work.

Epidemiologic studies of physical activity and health have compared the activity levels of people who have or develop diseases and those who do not. Cohort studies follow populations forward in time to observe how physical activity habits affect disease occurrence or death. In case-control studies, groups of persons who have disease and separate groups of people who do not have disease are asked to recall their previous physical activity. Cross-sectional studies assess the association between physical activity and disease at the same point in time. Clinical trials, on the other hand, attempt to alter physical activity patterns and then assess whether disease occurrence is modified as a result.

Results from epidemiologic studies can be used to estimate the relative magnitude or strength of an association between physical activity and a health outcome. Two such measures used in this chapter are risk ratio (RR) and odds ratio (OR). For these measures, an estimate of 1.0 indicates no association, when the risk of disease is equivalent in the two groups being compared. RR or OR estimates greater than 1.0 indicate an increase in risk; those less than 1.0 indicate a decreased risk. Confidence intervals (CI) reported with estimates of association indicate the precision of the estimate, as well as its statistical significance. When the CI range includes 1.0, the effect is considered likely to have occurred by chance; therefore the estimate of association is not considered statistically significantly different from the null value of 1.0.