In the linked systematic review and meta-analysis (doi:10.1136/bmj.d6617), Aune and colleagues assess the association between the intake of dietary fibre and whole grains and the risk of colorectal cancer.1

It was first hypothesised in 1988 that food items rich in dietary fibre may prevent colorectal cancer,2 but randomised trials studying dietary fibre have not supported the association.3 This is a classic situation within nutritional epidemiology; a food item is related to decreased incidence of disease and the biological effect is attributed to a single component, but when this component is tested in randomised trials the results are not what was expected. This ought to have taught researchers to study the dietary sources and not only one specific component, which is exactly what Aune and colleagues have done.1 Their systematic review and meta-analysis of prospective observational studies clearly shows that a high intake of fibre from cereals and high consumption of wholegrain foods is …