The relationship of meat, fish, fat, fibre or calcium consumption to the risk of colon cancer was examined in a prospective study conducted by the Norwegian National Health Screening Service. Between 1977 and 1983, 50,535 Norwegian men and women aged 20-54 attended the health screening and completed a semi-quantitative questionnaire about food frequency. During a mean follow-up of 11.4 years, 143 cases of colon cancer were identified for analyses through a link with the Norwegian Cancer Registry. The relative risk of colon cancer was 3.5 (95% CI, 1.02-11.9) in women who consumed sausages as their main meal five or more times a month, compared with the risk in those who reported a consumption frequency of less than once a month. There was an increase in the relative risk with increasing frequency of consumption (P for linear trends = 0.03). Among men, the association was not statistically significant, but the trend was in the same direction as that of the women. The frequency of consuming meat meals in general, including meat stews, roasted meat, meat balls, fish or milk, was not associated with a risk of colon cancer. No trends in relative risks of colon cancer were found to be associated with intake of total energy intake or with energy-adjusted intake of total fat, saturated fat, monounsaturated fat, dietary fibre or calcium. In conclusions, this prospective study provided no evidence of association between intake of meat, fish, fat, energy, fibre or calcium and risk of colon cancer, although an increased risk with frequent consumption of sausages was suggested.