BACKGROUND: Active commuting is associated with various health benefits, but little is known about its causal relationship with body mass index (BMI). METHODS: We used cohort data from three consecutive annual waves of the British Household Panel Survey, a longitudinal study of nationally representative households, in 2004/2005 (n=15,791), 2005/2006 and 2006/2007. Participants selected for the analyses (n=4056) reported their usual main mode of travel to work at each time point. Self-reported height and weight were used to derive BMI at baseline and after 2 years. Multivariable linear regression analyses were used to assess associations between switching to and from active modes of travel (over 1 and 2 years) and change in BMI (over 2 years) and to assess dose-response relationships. RESULTS: After adjustment for socioeconomic and health-related covariates, the first analysis (n=3269) showed that switching from private motor transport to active travel or public transport (n=179) was associated with a significant reduction in BMI compared with continued private motor vehicle use (n=3090; -0.32 kg/m(2), 95% CI -0.60 to -0.05). Larger adjusted effect sizes were associated with switching to active travel (n=109; -0.45 kg/m(2), -0.78 to -0.11), particularly among those who switched within the first year and those with the longest journeys. The second analysis (n=787) showed that switching from active travel or public transport to private motor transport was associated with a significant increase in BMI (0.34 kg/m(2), 0.05 to 0.64). CONCLUSIONS: Interventions to enable commuters to switch from private motor transport to more active modes of travel could contribute to reducing population mean BMI.