Patients with Crohn's disease show an increased consumption of refined sugar compared to controls. It was speculated that the frequent occurrence of Crohn's disease in developed countries might be related to a high intake of refined sugar. In contrast, others suggested margarine consumption to play a role in the etiology of Crohn's disease, the hypothesis being based on the seemingly simultaneous onset of margarine consumption and the first reports of granulomatous ileitis and the similarities in their geographic distribution. Both hypotheses, however, have not been subjected to a rigorous epidemiologic analysis. In the present study, the per capita consumption of refined sugar and margarine were correlated with the incidence and mortality of Crohn's disease from different countries. The time trends of mortality from Crohn's disease were compared to those of sugar and margarine intake. Two weak correlations were found between the geographic distribution of sugar and margarine consumption and mortality from Crohn's disease, the correlations resulting from the difference between two types of countries. Compared to Canada, the United States and most countries from Northern Europe on one side, Japan and most Mediterranean countries on the other side were characterized by low consumption of the two nutrients as well as low mortality from Crohn's disease. No consistent pattern was found, however, within each group of countries. No significant correlation was found between sugar or margarine consumption and the incidence of Crohn's disease. The time trends of Crohn's disease in different countries were not matched by similar time trends of either sugar or margarine consumption. The different epidemiologic behavior of the two nutrients and Crohn's disease indicates that variations in their dietary intake were not responsible for the geographic and temporal patterns of Crohn's disease. Other factors must be sought to explain the characteristic epidemiology of Crohn's disease.