Associations between multiple indicators of objectively-measured and self-reported sedentary behaviour and...

To examine the associations between sedentary behaviour (SB) measured objectively and by self-report and cardiometabolic risk factors.

Cross-sectional analyses of adults ≥60 years who participated in the 2008 Health Survey for England. Main exposures were self-reported leisure-time SB consisting of TV/DVD viewing, non-TV leisure-time sitting, and accelerometry-measured SB. Outcomes included body mass index (BMI), waist circumference, cholesterol ratio (total/HDL), Hb1Ac and prevalent diabetes.

2765 participants (1256 men) had valid self-reported SB and outcomes/confounding variables data, of whom 649 (292 men) had accelerometer data. Total self-reported leisure-time SB showed multivariable-adjusted (including for moderate-to-vigorous physical activity) associations with BMI (beta for mean difference in BMI per 30 min/day extra SB: 0.088 kg/m(2), 95% CI 0.047 to 0.130); waist circumference (0.234, 0.129 to 0.339 cm); cholesterol ratio (0.018, 0.005 to 0.032) and diabetes (odds ratio per 30 min/day extra SB: 1.059, 1.030 to 1.089). Similar associations were observed for TV time while non-TV self-reported SB showed associations only with diabetes (1.057, 1.017 to 1.099). Accelerometry SB was associated with waist circumference only (0.633, 0.173 to 1.093).

In older adults SB is associated with cardiometabolic risk factors, but the associations are more consistent when is measured by self-report that includes TV viewing.