Background Higher status occupations are associated with higher levels of occupational sedentary time, yet lower levels of obesity compared to lower social class groups. We hypothesised that leisure-time physical activity and sedentary time explain this paradox: when not at work individuals in higher status occupations spend more time physically active and less time sedentary, offsetting their sedentary jobs. This study examines if the positive association between occupational sedentary time and obesity persists after adjustment for leisure-time physical activity and leisure-time sedentary time.

Methods The sample comprised of 4282 economically active participants from the 2008 Health Survey for England (mean age 43.4 ± 13.2, 2160 women). Respondents reported sedentary time at work (sitting/standing), TV viewing, leisure-time non-TV sitting and moderate to vigorous physical activity (MVPA). Logistic regression models were run for obesity (BMI> = 30) with occupational sedentary time, TV time, non-TV leisure-time sitting, and leisure-time MVPA as main exposures. Covariates included age, sex, indicators of socioeconomic position (occupational social class, household income quartile, highest educational qualification and area deprivation quintile) smoking status, frequency of alcohol consumption, and frequency of fruit and vegetable consumption. The complex sample design was taken into account and non-response weights applied.

Results Occupational sedentary time was associated with obesity after adjusting for TV time and leisure-time MVPA (OR 1.08 per hour [95% CI 1.04, 1.12]). Leisure-time non-TV sitting was not associated with obesity. TV time was positively associated with obesity (1.2 [1.14, 1.27]), and leisure-time MVPA was negatively associated with obesity (0.8 [0.7, 0.9]).

Conclusion The associations between occupational sedentary time and obesity were independent of leisure-time sedentary behaviour and MVPA. Sedentary occupations are associated with obesity regardless of occupational social class, TV viewing and leisure-time MVPA. Lower levels of obesity among higher social class groups, despite higher levels of occupational sedentary time, cannot be explained by leisure-time sedentary and physical activity behaviours.