Sedentary Behaviour and Health

Abstract

The term sedentary refers to a distinct class of activities which involve sitting or reclining and which do not cause an increase in energy expenditure above resting levels. Observational studies have reported positive associations between both sedentary time and the number of hours spent sitting per day, with risk for a number of health outcomes that are independent of moderate to vigorous physical activity (MVPA). The total time spent sitting can be amassed in different patterns (long and short bouts) and different types (watching TV, driving, working at a computer) that may have differential associations with health outcomes as well as different confounders that have yet to be properly explored. Further, limitations in current measures used to quantify sedentary behaviour and the possibility of residual confounding, mean that it is unclear whether the posture of sitting itself represents a risk to health or whether sitting is actually a proxy for low energy expenditure. This thesis aimed to examine; the associations between five separate sitting types with health risk, the prevalence of sitting behaviour in England, and the biological mechanisms which might underpin the observed negative health consequences of sitting.

Using data from the Whitehall II cohort study the first four studies of this thesis examined prospective associations between sitting at work, TV viewing, non-TV leisure time sitting, total leisure time sitting (TV and non-TV leisure sitting combined) and total sitting from work and leisure, with four health outcomes; mortality, cardiovascular disease, type II diabetes and obesity. No association between any of the sitting indicators with risk for mortality or incident cardiovascular disease was found. TV viewing and total sitting were associated with an increase in risk for type II diabetes following adjustment for sociodemographic covariates and MVPA, but were attenuated following further adjustment for body mass index. None of the five sitting indicators were associated with incident obesity but being obese prior to the measurement of sitting was associated with the number of reported hours of daily TV viewing. The final study of this thesis examined the acute effect of sustained versus interrupted sitting on glucose and insulin metabolism. Interrupting sitting with repeated short bouts of light intensity walking significantly improved insulin sensitivity while repeated short bouts of standing did not.

Sitting is a prevalent behaviour in English adults and varies by socio-demographic characteristics. Previously reported associations between sitting time and health risk may be confounded by light intensity physical activity and obesity. The absence of an effect of repeated standing bouts (a change in posture without a change in energy expenditure) suggests that promoting reductions in sitting without also promoting increases in movement are not likely to lead to improvements in metabolic health. New measures of sedentary behaviour are required that can be used in population studies, and can discriminate between the posture of sitting, standing and very low levels of physical activity of a light intensity. This would permit further studies that are needed to clarify the precise nature of the association between sitting and health.