We addressed three questions: 1) Is there a dose-response relation between physical activity and health? 2) Is there a dose-response relation between cardiorespiratory fitness and health? 3) If both activity and fitness have a dose-response relation to health, is it possible to determine which exposure is more important? We identified articles by PubMed search (restricted from 1/1/90 to 8/25/00) using keywords related to physical activity, physical fitness, and health. An author scanned titles and abstracts of 9831 identified articles. We included for thorough review articles that included three or more categories of activity or fitness and a health outcome and excluded articles on clinical trials, review papers, comments, letters, case reports, and nonhuman studies. We used an evidence-based approach to evaluate the quality of the published data. We summarized results from 67 articles meeting final selection criteria. There is good consensus across studies with most showing an inverse dose-response gradient across both activity and fitness categories for morbidity from coronary heart disease (CHD), stroke, cardiovascular disease (CVD), or cancer; and for CVD, cancer, or all-cause mortality. All studies reviewed were prospective observational investigations; thus, conclusions are based on Evidence Category C. 1) There is a consistent gradient across activity groups indicating greater longevity and reduced risk of CHD, CVD, stroke, and colon cancer in more active individuals. 2) Studies are compelling in the consistency and steepness of the gradient across fitness groups. Most show a curvilinear gradient, with a steep slope at low levels of fitness and an asymptote in the upper part of the fitness distribution. 3) It is not possible to conclude whether activity or fitness is more important for health. Future studies should define more precisely the shape of the dose-response gradient across activity or fitness groups, evaluate the role of musculoskeletal fitness, and investigate additional health outcomes.
