Physical (in)activity over 20y in adulthood: Associations with adult lipid levels in the 1958 British birth cohort

Objectives: To investigate associations between physical (in)activity at different life-stages and lipids in mid-adulthood, examining the role of potential confounding and mediating factors, such as adiposity. Methods: Data from the 1958 British birth cohort (n= 7824) were examined. Using linear regression, we analysed prospectively reported frequency of activity and TV-viewing (23, 33, 42 and 45. y) in relation to total, LDL-, HDL-cholesterol and triglycerides, at 45. y. Results: Activity at different ages was associated with HDL-cholesterol and triglycerides at 45. y: e.g. in men, a 1 day/week greater activity frequency at 42. y was associated with 0.006. mmol/L higher HDL-cholesterol and 1.4% lower triglycerides. Most associations attenuated, but were not entirely explained by adjustment for covariates (life-styles and socio-economic factors): e.g. among men, the estimated 2.0% lower triglycerides per 1 day/week greater frequency at 33. y reduced to 1.8% after adjustment. Among women, though not men, activity at both 23 and 45. y contributed cumulatively to HDL-cholesterol. For sedentary behaviour, associations were found for sitting at work: a 1. h/day greater sitting among men was associated with a 0.012. mmol/L lower HDL-cholesterol after adjustment for covariates. Associations were seen for TV-viewing: e.g. in men, a 0.04. mmol/L lower HDL-cholesterol and 5.9% higher triglycerides per hour/day greater TV-viewing at 45. y, attenuated, respectively, to 0.03. mmol/L and 4.6% after adjustment for covariates. Associations attenuated further after adjustment for current BMI. Associations for total and LDL-cholesterol were less consistent. Conclusion: Activity and sedentary behaviour at different adult ages were associated with HDL-cholesterol and triglycerides in mid-adulthood. Associations were partly mediated by other life-style factors and by BMI.