Article Text

Download PDFPDF
OP81 Does active commuting protect against obesity in mid-life? evidence from UK Biobank
  1. E Flint,
  2. S Cummins
  1. Social and Environmental Health Research, London School of Hygiene and Tropical Medicine, London, UK


Background Active commuting (AC) has been promoted as a way of addressing the population health effects of sedentary lifestyles in Western settings. This study examines the relationship between AC and obesity in mid-life – a key life-stage for CVD risk – using objectively measured anthropometric data from UK Biobank. Objectives are to test for associations between AC and markers of obesity, and to investigate confounding and interaction effects.

Methods Baseline (2006–2010) data from UK Biobank were used (n = 502,664, of which 264,341 commuted). Self-reported commuting mode was operationalised as a 7 category variable, ordered to reflect typical levels of physical exertion: car only; car and public transport; public transport only; car and mixed; public transport and cycling or walking; walking only; cycling only. Outcomes were objectively measured BMI and percentage bodyfat (continuous). Hypothesised confounders were identified. Nested, gender-stratified multivariate linear regression models were estimated for each outcome. Potential interactions with ethnicity and neighbourhood deprivation (Townsend score, quintiles) were investigated. Final complete case sample sizes were 68,484 males, 79,168 females for BMI analyses; 67,790 males, 78,351 females for percentage bodyfat analyses.

Results AC was significantly and independently predictive of lower BMI and percentage bodyfat for both men and women, with a graded ‘dose-response’ pattern apparent across the 7 ordered mode categories. Adjustment for confounders (socioeconomic, environmental, demographic, health, behavioural) did not attenuate the central relationships. A significant interaction was found between AC and deprivation quintiles but no interaction was found for ethnicity. In the fully adjusted model, compared to their car-driving counterparts, male public transport commuters had 0.6 kg/m2 lower BMI (95% CI –0.72 to –0.48). Males who mixed public transport with walking/cycling had 0.88 kg/m2 lower BMI (95% CI –1.02 to –0.73). Males who cycled had 1.5 kg/m2 lower BMI (95% CI –1.69 to –1.36). A similar pattern and magnitude of effects was observed for women, and for percentage bodyfat analyses across both genders. Sensitivity analyses did not affect the results.

Conclusion This study is the first to use UK Biobank data to address the topic of AC and obesity; finding robust, independent associations between AC and healthier body weight and composition. Mid-life is a crucial stage in the lifecourse for CVD risk, and this research supports of the promotion of AC as a population-level policy response. The cross-sectional design is a limitation, but future research will utilise the repeat-assessment subsample of UK Biobank to improve causal inference.

  • obesity
  • active travel
  • physical activity

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.