Cover Image
close this book Food Composition Data: A User's Perspective (1987)
close this folder The uses of food composition data
close this folder Need for a standardized nutrient data base in epidemiologic studies
View the document (introductory text)
View the document Introduction
View the document Limitations of diet-related epidemiologic studies
View the document Factors influencing diet-related epidemiologic studies, using diet and colon cancer studies as an illustration
View the document Some potential problems with incomplete and non-standardized nutrient data bases
View the document Summary
View the document References

Limitations of diet-related epidemiologic studies

Limitations of diet-related epidemiologic studies

Suitable and up-to-date food composition tables are practical tools for the identification of dietary problems and the planning of intervention programmes. Epidemiologic studies are largely dependent on food composition data bases because of the cost and impracticability of obtaining and assaying foods from the large number of free living subjects required for such studies. Therefore, food composition data bases should, whenever possible, give reliable representative data for indigenous foods reflecting the effect of growing conditions and treatment before consumption. They should include a wide variety of nutrients, making possible a comprehensive study of nutrient intake.

Advances in analytical chemical technology and the advent of high-speed computers have made feasible the processing of complex human diets. However, there is substantial criticism of diet-related epidemiologic research because the results of many studies have been weak, inconclusive, or equivocal, and at variance with animal models and in vitro evidence. Many problems with population-based diet studies relate to the following issues: (a) determining the strength of diet relationships to disease states which have multiple histologic and physiological characteristics; (b) identifying the significant dietary causal risk factors affecting the disease state; (c) having an incomplete or inappropriate nutrient data base to analyse data; (d) conducting studies with weak designs and limited technology; and (e) making inappropriate comparisons between study variables. Epidemiologic research related to diet and colon cancer can be used to illustrate how some of these problems can be influenced by food composition data, which in turn can influence the outcome of such studies. Colon cancer was selected as the example because it is a disease that has been strongly implicated with diet.