Diet and cancer--an international study.

The notion that nutritional intake and cancer are interrelated is an old one and the strength of this relationship is dependent upon the proportion of underdeveloped and developed countries among the total number of countries investigated. The protein, fat and caloric intake is lower in underdeveloped versus developed countries. The positive association of intestine, rectum and breast cancers with food intake is biologically logical whereas the positive association of lymphatic, haemopoietic and myeloid leukaemias and skin cancer make no biological sense suggesting that strength of association by itself does not imply causality. The study points to the advantages and pitfalls of international data bases and broad correlation studies.