For much of the past 50 years, the study of physical activity epidemiology has focused on moderate-to-vigorous-intensity physical activity (MVPA) and its relationship with morbidity and premature mortality. However, there has been recent and increasing interest in sedentary behaviours as a distinct exposure.
Sedentary behaviours are pursuits carried out while sitting or reclining and that do not increase energy expenditure substantially above the resting level (e.g., 1.0–1.5 times the resting level). Independent of moderate–vigorous exercise undertaken, excessive time spent sitting has been associated with reduced levels of light-intensity activity and less energy expenditure (Owen et al, 2009
), as well as with poor metabolic health (Healy et al, 2008
), and possibly increased risk for certain cancers. These associations have been demonstrated not only for self-reported sitting time but also for objectively measured sedentary time (Healy et al, 2008
). Even among adults who meet the recommended 30
min or more of moderate–vigorous physical activity per day (e.g., by walking, playing sports, or exercising in a fitness centre), there seems to be deleterious metabolic consequences to sitting most of the rest of the day (Owen et al, 2010
). These consequences may stem from the tendency for sitting to displace light-intensity activities, such as time spent standing or in intermittent ambulation (Owen et al, 2010
). Displacement of 1
h of light-intensity activity by 1
h of sedentary activity results in an estimated loss of 1 MET-hday−1
) of physical activity energy expenditure, equivalent to about 15
min per day of walking (Owen et al, 2010
). Alternately, sitting time may have its own distinct insidious physiology. Prolonged periods of sitting result in reduced activity of skeletal muscle lipoprotein lipase (LPL)—an enzyme essential for triglyceride catabolism and HDL regulation—and inhibition of glucose uptake (Owen et al, 2010
To date, three prospective studies have examined sitting time in relation to endometrial cancer risk. Friberg et al
found that women who sat for 5 or more hours per day had increased risk of endometrial cancer relative to women who sat less than that; the RR (95% CI) for endometrial cancer was 1.80 (1.14–2.83) (Friberg et al, 2006
). Patel et al
compared women who sat 6+
those who sat for less than 3
h per day and reported increased endometrial cancer risk; RR=1.40 (95% CI=1.03–1.89) (Patel et al, 2008
). Gierach et al
examined women who sat for 7+
those who sat for less than 3
h per day and reported increased risk, with an RR of 1.56 and a 95% confidence interval of 1.22–1.99 (Gierach et al, 2009
). Whether sedentary behaviours contribute to risk independently from MVPA is not clear from these studies, as two of them (Patel et al, 2008
; Gierach et al, 2009
) did not adjust for level of MVPA. Indirect support for a link between sitting time and endometrial cancer risk also comes from studies of occupational physical activity, as they examine risk conferred by sedentary vs
non-sedentary occupations. The increased risk among women with sedentary jobs is very clear in these studies.
To further investigate the role of sedentary behaviours in endometrial cancer aetiology, we updated the analysis by Gierach et al
from the NIH-AARP Diet and Health Study by extending cancer follow-up through 31 December 2006 (Gierach et al, 2009
). Briefly, the study consisted of AARP members aged 50–71 years residing in 1 of 8 US states (CA, FL, PA, NJ, NC, LA, GA and MI). Participation in vigorous-intensity activities during the previous year was assessed on the basis of a baseline questionnaire and time spent sitting per day during the previous year was assessed on the basis of a second questionnaire. Our analysis included the 69
648 women studied by Gierach et al
who reported both their level of exercise and their time spent sitting per day, of whom 888 were diagnosed with endometrial cancer. All women included in the analysis had an intact uterus, no personal history of cancer, and no missing or extreme values of BMI.
In , we show that endometrial cancer risk is positively associated with time spent sitting (P
for trend <0.01), as previously reported by Gierach et al.
Adjustment for vigorous-intensity physical activity modestly attenuates RRs but sitting time still has a dose–response relationship with endometrial cancer risk (P
for trend <0.01). Among women who were active, that is, women who engaged in vigorous exercise three or more times per week, as well as among women who exercised less frequently, sitting time was associated with increased endometrial cancer risk (P
for trend <0.01). In a joint effects analysis, women who were inactive (as defined above) and who sat for 9+
h per day had twice the risk of endometrial cancer of active women who sat fewer than 3
h per day (RR=2.14; 95% CI=1.48, 3.10).
Table 1 Multivariate relative risks (RR) and 95% confidence intervals (CI) of endometrial cancer in relation to time spent sitting per day and joint categories of sitting and vigorous intensity physical activity among 69648 women in the NIH-AARP (more ...)
As a sensitivity analysis, we reran our analysis for all women with adjustment for BMI, as well as physical activity level (). Adjustment for BMI attenuated results substantially. Women who sat for 9+
those who sat for fewer than 3
h per day had an RR for endometrial cancer of 1.15 (95% CI=0.87–1.53) as compared with the previous RR of 1.45 (95% CI=1.10–1.92), although the trend remained statistically significant (P
for trend <0.01). Because sitting time is hypothesised to affect endometrial cancer risk at least partly through its effects on body weight, these results may be conservative.
Our data, and that of others (Friberg et al, 2006
; Patel et al, 2008
), suggest that sitting time contributes to endometrial cancer risk, independent from one's participation in MVPA. We earlier estimated that as many as 22% of endometrial cancers could be avoided if women exercised vigorously five or more times per week. However, if all women had both exercised at this level and
sat for 4 or fewer hours per day, then 34% of endometrial cancers could have been avoided. Thus, the incremental contribution of sedentary behaviours to the population level burden of endometrial cancer risk may be substantial.