However, these findings are not necessarily generalisable to overall sedentary or sitting time because TV viewing is a complex exposure that seems to be a poor index of overall ST.17 In a recent US study comparing associations JAK1/2 inhibito between TV time and objectively measured ST, associations were of fair magnitude, but were not consistent across population subgroups.18 The results of the few studies that looked at overall (self-reported) sitting in relation to SEP are inconsistent. Higher social position was linked to higher overall sitting time among Australian women19 but education level was unrelated to sitting time among Portuguese adults.20 Objective measuring methodologies

such as accelerometers and inclinometers can give more comprehensive and complete estimates of total sedentary behaviour than partial self-reported indices such as TV viewing, or self-reported total sitting time, which may be more difficult to recall than TV viewing and therefore be subject to more measurement error. Besides, SEP characteristics that relate to occupational class and income will

naturally have an impact on work time sitting. For example, manual unskilled workers normally spend less time sitting during work than professionals in managerial office-based jobs.21 Similarly, higher incomes and the associated spending capacity might impact on the time spent sitting driving a car or commuting. To the best of our knowledge, no study has looked at the associations between SEP defined using education, occupational class, income and area deprivation indices, and SB estimated using self-reported sitting across different domains as well as objective methods. The aim of this study was to look at the associations between multiple SEP indicators and self-reported indices of sitting time and

SB as well as objectively-assessed total SB time. We used data from one of the largest European accelerometry general population studies, the 2008 Health Survey for England. Methods Study sample The Health Survey for England (HSE) is a repeated nationally representative study of individuals living in private households in England. We drew our sample from the 2008 HSE which had a special focus on physical activity and sedentary behaviour. The sample is drawn using multistage stratified probability sampling with postcode sectors as the primary sampling unit. More details of the sample Entinostat design are available elsewhere.8 The overall interview household response rate for the main sample of 15 102 adults was 64%, and for the accelerometer subsample of 4507 adults was 73%.8 In this analysis, we included adults aged 16 and above (age range 16–96 years) who had valid accelerometry and self-reported SB data. Participants provided written informed consent. An abridged methods section is presented here: the full methods section with more information can be found in online supplementary file S1 (Unabridged Methods).