Many studies have suggested that physical activity helps prevent coronary heart disease (CHD), but several others have shown no such association. Thus, evidence to support a beneficial association has been considered weak or questionable, primarily because physical activity is difficult to measure and assess (1).

An extensive review of studies on the possible association between physical activity and CHD focused on the quality of the measures and methods used. The results of that review indicate that physical activity does help prevent CHD.

A systematic review of the literature yielded 43 studies in English that provided relative risks or multiple regression coefficients of the association between physical activity and CHD. For each of these studies (36 cohort, three mortality, and four case-control studies), the reviewers used specific criteria to assess the quality of the physical activity measure, the CHD outcome measure, and the epidemiologic methods.

The seven criteria used in evaluating the physical activity measure were 1) clarity of the definition of physical activity, 2) reliability and validity of the measure, 3) assessment of individual physical activity rather than of group activity, 4) use of frequency, intensity, and duration of physical activity to characterize the behavior, 5) measurement of lifetime patterns of activity, 6) adherence to an activity pattern over time, and 7) systematic collection of the measure (usually via self-report surveys).

The four criteria used in assessing the CHD outcome measure were

1. specifically established diagnostic criteria, 2) objective diagnosis, 3) equal opportunity for diagnosis of CHD, and 4) systematic collection of CHD information.

The eight criteria used in evaluating the epidemiologic methods were 1) the temporal sequence of physical activity before CHD, 2) statistical control of other CHD risk factors, 3) representativeness of the sample, 4) whether subjects from the cohort studies who were lost to follow-up were located later or at least compared with the other subjects, 5) if random selection methods were used for placing subjects in active and inactive groups, 6) whether cases and controls were identified via predetermined selection criteria, 7) if they were equally subjected to exclusionary criteria, and 8) if neither subjects nor data abstractors were informed of the hypothesis being studied.

Primarily on the basis of these criteria, the authors considered 40% of the physical activity measures, 2% of the CHD outcome measures, and 30% of the epidemiologic methods to be unsatisfactory (Table 1).

These 43 studies reported 96 comparisons of the association between physical activity and CHD. The reviewers eliminated those comparisons that could not be interpreted (n=3), that focused only on angina (n=10) or women (n=15), that reported information on extra subpopulations or extra physical activity measures (n=5), and that reported multiple CHD outcomes for a given study (n=16). A total of 47 comparisons remained, and all of these were used to draw inferences about men.

Of these 47 comparisons, 32 (68%) showed a statistically significant inverse association between physical activity and CHD. Further, the reviewers' ability to detect such an association increased as the quality of the measures and methods improved (Table 2). For example, among the studies using unsatisfactory physical activity measures, the reviewers noted that 50% showed significant associations; among those using satisfactory measures, 76%; and among those using good physical activity measures, 88%. Similar trends were noted for CHD measures and epidemiologic methods.

The reviewers also examined the potential causal effect of physical activity on CHD by using six criteria: consistency of findings, strength of the association, appropriate temporal sequence, dose-response relationship, plausibility, and experimental evidence. A consistent statistically significant association between physical activity and CHD was found for more than two-thirds of the studies. The strength of the association between physical inactivity and CHD (median relative risk = 1.9 for the 47 comparisons) was of similar magnitude as that for several commonly accepted risk factors previously reported in the Coronary Pooling Project, which was based on five studies (2). In those studies, the median risk ratios were 2.1 for high systolic blood pressure (greater than150 millimeters of mercury (mm Hg) versus less than or equal to130 mm Hg), 2.4 for serum cholesterol (greater than268 milligrams per deciliter (mg/dl) versus less than or equal to218 mg/dl), and 2.5 for smoking (greater than or equal to1 pack of cigarettes/day versus no smoking). Most of the 43 studies reviewed showed that the activity assessment predated the CHD outcome, demonstrating an appropriate temporal sequence. More than two-thirds of the studies demonstrated a dose-response relationship, with lower levels of physical activity leading to more instances of CHD. There are plausible and coherent mechanisms whereby physical activity could exert a beneficial influence on CHD. Although no experimental evidence exists in the form of a randomized, controlled clinical trial, better studies (i.e., those in which the measures and methods used were judged to be good or satisfactory) were more likely to report a significant inverse association. On the basis of these criteria, the authors concluded that a causal inverse association exists between physical activity and CHD. Reported by: Behavioral Epidemiology and Evaluation Br, Div of Health Education, Center for Health Promotion and Education, CDC.

Editorial Note

Editorial Note:

Previous reviews have not provided sufficient evidence to show that the lack of physical activity is associated with coronary heart disease (CHD). In this study, by using criteria for causality, the authors systematically evaluated the most critical features of studies on the possible association between physical activity and CHD. Perhaps more importantly, the authors found that poor methods used in assessing physical activity had led to an erroneous inference (no causal association), illustrating the need for careful measures and methods in epidemiologic research.

Compared with measures of physical activity, fewer measures of CHD outcome were considered to be of poor quality, perhaps because CHD is easier to define and CHD-related morbidity and mortality are clearly measurable outcomes. Only recently have precise definitions of physical activity and exercise been offered (3). Unfortunately, physical activity is unlike the other CHD risk factors because standardized assessment methods do not exist (4). Methods for measuring physical activity are needed not only for determining its protective effect against CHD but also for determining its possible association with other diseases and health outcomes such as cancer, disability, and mental health.

When the results of this review were compared with the results of the Coronary Pooling Project, the strength of the association between lack of physical activity and CHD appears to be similar to that found for high serum cholesterol, high systolic blood pressure, and cigarette smoking (2). The relative risk ratios for these CHD risk factors appear to be similar to those for physical inactivity. Therefore, knowing the prevalence of each risk factor helps determine a U.S. population-based attributable risk (Figure 1). Using the measurements for each risk factor used by the Coronary Pooling Project (see above), the current nationwide prevalence estimates show that approximately 10% of persons have a systolic blood pressure greater than150 mm Hg (5), 10% have a serum cholesterol greater than268 mg/dl (6), and 18% smoke a pack or more of cigarettes per day (7). In the studies reviewed, the median risk ratio of 1.9 found for physical inactivity and CHD was based on study respondents considered to be least active versus those considered to be most active, and the contrast varied from study to study. Nevertheless, on the basis of study results, a nationwide prevalence estimate of minimal amounts of regular physical activity likely to protect against CHD can be determined. Approximately 59% of Americans do not perform physical activity regularly (three or more times per week for greater than or equal to20 minutes at a time) (8). Hence, the prevalence of persons at risk of CHD because of high serum cholesterol, high systolic blood pressure, or cigarette smoking is actually small compared with that of persons who do not perform regular physical activity. Since these four CHD risk factors are similar in strength, physical activity appears to be a far more important risk factor. This is so because the other three prevalence levels are comparatively low.

Current standards include a higher proportion of the population in the risk groups. For example, it is recommended that a systolic blood pressure greater than140 mm Hg and a diastolic blood pressure greater than90 mm Hg should be treated (5). At those levels, about 36% of Americans are at risk. A serum cholesterol greater than200 mg/dl is now considered the level at which treatment should begin (9). Approximately 25%-40% of Americans would be considered at risk at this level. As for cigarette smoking, approximately 30% of Americans are believed to be current smokers (7). In light of these standards, more Americans are at risk of CHD because of physical inactivity than because of the other three main risk factors viewed separately. Systematic reviews for these other risk factors as was done for physical inactivity and CHD-would be valuable. Besides being protective against CHD, increased levels of physical activity also have been protective against other chronic diseases (10-11). Eleven objectives of the 1990 Objectives for the Nation proposed by the Public Health Service pertain to physical fitness and exercise (12). Although these objectives promote regular and vigorous physical activity, less intensive-yet regular-physical activity is also beneficial (13). Since so many Americans are physically inactive, additional steps should be taken to promote a life-style that includes regularly scheduled physical activity (14).

References

Disclaimer   All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

Page converted: 08/05/98

HOME  |  ABOUT MMWR  |  MMWR SEARCH  |  DOWNLOADS  |  RSS |  CONTACT POLICY  |  DISCLAIMER  |  ACCESSIBILITY

Morbidity and Mortality Weekly Report Centers for Disease Control and Prevention 1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A Department of Health and Human Services

This page last reviewed 5/2/01