The inverse associations between physical activity and overall mortality, as well as the risk of cardiovascular disease, diabetes, osteoporosis and obesity have been well documented. A protective role of physical activity regarding cancer incidence and mortality has also been identified from relevant cohort and case-control studies which have been included in the constantly updated World Cancer Research Fund Report  (Table 1).
Table 1.Physical activity and cancer (Source: Modified from WCRF, 2007)
|Cancer site||Evidence for a protective effect of physical activity|
Definitions, types and measures of physical activity
Physical activity is defined as “any movement using skeletal muscles”  and is conventionally classified into four types:
- occupational (at work)
- household (at home)
- transport (e.g. travelling to and from work)
- recreational (during leisure time)
Activities are also characterized by their intensity, that is, the degree to which they increase energy expenditure above basal metabolism. Intensity is usually measured either by the amount by which an activity increases the heart and breathing rates, or by Metabolic Equivalents (METs) which express the intensity of an activity relative to a person’s resting metabolic rate . Based on its intensity, physical activity is usually classified as: vigorous, moderate, light, or sedentary. METs can be used to measure the combined contribution of multiple types of activities (overall physical activity level) over specific time periods (e.g. days or weeks) by multiplying the MET value of each activity by the corresponding duration of the activity(in minutes, hours etc) and then summing up the resulting MET-time products (e.g. MET-hours). In epidemiological studies, data on physical activity are usually collected through questionnaires designed to capture the average level of physical activity during shorter or longer time periods. Information included in these questionnaires varies with respect to the range of and details about the activities covered.
Association of physical activity with specific cancer sites
Colorectal cancer: An inverse association between physical activity and colon cancer risk has been consistently reported in epidemiologic studies; for rectal cancer, however, the evidence is still inconclusive. In meta-analyses, highest compared to lowest levels of physical activity have been associated with a reduction in colon cancer risk by 20–25% , the risk reductions being similar for occupational or recreational physical activity. Biological mechanisms that may underlie this association include the reduction in insulin resistance and the effect of physical activity on endogenous steroid hormone metabolism. Physical activity before as well as after colorectal cancer diagnosis has also been associated with improved survival among colon cancer patients.
Postmenopausal breast cancer: Higher rather than lower levels of total, occupational and/or recreational lifetime physical activity have been consistently associated with decreased postmenopausal breast cancer risk, the risk reduction ranging in cohort studies from 3% to more than 20%. Potential mechanisms that may underlie this association include the beneficial effect of physical activity on body fatness and the metabolism of endogenous steroid hormone, as well as a possible beneficial effect on the immune system. Evidence suggests that the association between physical activity and the risk of postmenopausal breast cancer may be modified by body mass index, or estrogen receptor status. Pre- and post-diagnosis physical activity has also been reported to reduce all-cause and breast cancer-specific mortality, as well as disease recurrence among patients .
Endometrial cancer: The evidence regarding regular physical activity and endometrial cancer risk is based predominantly on case-control studies. A meta-analysis of prospective cohort studies up until 2009 associates an estimated 20% reduction in endometrial cancer risk with high levels of recreational and occupational physical activity. These associations may be partly explained by the mechanisms indicated above with respect to the risk of postmenopausal breast cancer.
Other cancer sites : There are reports from prospective and case-control studies suggesting that overall, occupational and recreational physical activity may protect against premenopausal breast cancer, but the overall evidence is inconclusive. The evidence on physical activity in relation to the risk of lung cancer is also (if not more) inconclusive and no plausible mechanisms have been invoked. Moreover, one cannot ignore the potential residual confounding due to smoking, a factor that has been taken into account in only some of the relevant studies. For pancreatic cancer, a meta-analysis of 28 studies showed reduced risk of the disease associated with total physical activity. Lack of heterogeneity among the studies makes the interpretation of these findings difficult. Finally, the evidence on the association of physical activity with prostate cancer risk is still inconclusive.
There is converging evidence in support of an inverse association of physical activity with colon, endometrial and postmenopausal breast cancer. For colon cancer the association is characterized as “convincing”, whereas for endometrial and postmenopausal breast cancer the associations are considered “probable”. The evidence on therole of physical activity on premenopausal breast cancer risk, as well as the risk of lung and pancreatic cancer, is inconclusive and the respective associations are considered as “possible”. Protection against risk of cancer holds for any type of physical activity (i.e., occupational, recreational or total) and the respective associations are monotonic, within the range of physical activity examined across studies: “the more physically active people are, the better” . To better understand and quantify these associations, studies with detailed lifetime histories of physical activity across various population subgroups are needed. Given that physical activity is a modifiable risk factor, relevant public health interventions may be of major importance for the primary prevention of cancer .
Christina Bamia, Assistant Professor Medical School, University of Athens, Dept of Hygiene, Epidemiology and Medical Statistics