Aims/hypothesis We examined whether small amounts of low-intensity physical activity were associated with reduced risk of developing type 2 diabetes in a national sample of people aged 50 years and over. Methods The sample comprised 7,466 individuals (55.9% women) free from self-reported doctor-diagnosed diabetes and was prospectively followed for a mean of 45.3 months. Baseline self-reported physical activity was categorised as physical inactivity, low-and vigorous/moderate-intensity physical activity at least once a week. Cox proportional hazard regression was used to model the association between baseline physical activity and incident type 2 diabetes. Results Vigorous/moderate-intensity physical activity at least once a week was associated with reduced risk of type 2 diabetes (HR 0.64, 95% CI 0.43-0.95, p=0.026) but lowintensity physical activity at least once a week was not (HR 0.87, 95% CI 0.58-1.30, p=0.497) after adjustment for all covariates. However, age-stratified analysis showed that low-intensity physical activity at least once a week was associated with reduced risk of type 2 diabetes for those aged 70 years and over (HR 0.53, 95% CI 0.28-1.02, p= 0.059), but not for those aged 50 to 59 years (HR 1.09, 95% CI 0.52-2.29, p=0.828) or those aged 60 to 69 years (HR 1.15, 95% CI 0.55-2.41, p=0.715) after adjustment for all covariates. Conclusions/interpretation Compared with physical inactivity, any type of physical activity was associated with reduced risk of type 2 diabetes in adults aged 70 years and over, while in adults aged 50 to 69 years, physical activity needed to be vigorous/moderate in intensity to be associated with reduced risk of type 2 diabetes.