It has been estimated by various authorities that about one-third of cancers in Western high-income societies are attributable to factors relating to food, nutrition and physical activity. Identifying with confidence specific associations between dietary patterns, foods, body composition or individual nutrients is not simple because of the long latent period for cancer development, its complex pathogenesis and the challenge of characterising the multidimensional aspects of diet and activity over a lifetime. Reliable conclusions must therefore be drawn not only from randomised controlled trials but from a variety of methodological approaches, judged within a classic framework for inferring causality. Using a newly-developed method with a protocol for standardising the literature search and for analysis and display of the evidence, nine independent academic centres have conducted systematic reviews addressing the causal associations between food, nutrition and physical activity and risk of development of seventeen cancers, as well as of weight gain and obesity. A review has also examined the efficacy of such interventions in subjects with cancer. The reviews have been assessed by an independent Panel of twenty-one international experts who drew conclusions with grades of confidence in the causality of associations and made recommendations. Recommendations are given as public health goals as well as for individuals.