Hierarchical cluster analysis as an approach for systematic grouping of diet constituents on basis of fatty acid, energy and cholesterol content:: application on consumable lamb products

The role of dietary fat in the etiology of chronic diseases is both a qualitative and a quantitative issue. The dietary fat intake is largely influenced by behavioral and social influences on food choice. Ongoing scientific research has led to dietary recommendations with main concerns being the percentage of saturated, essential fatty acids and cholesterol with respect to total energy intake. However, the compositional complexity of food choice constituting the diet is a critical concept complicating the interpretation of epidemiologic, clinical and laboratory evidence to define the role of dietary fat in the etiology of diseases. This study was conducted on the observation of the need to better systematically classify consumable food based on complex composition and lamb meat is randomly selected as a nonspecific subset for application of hierarchical cluster analysis method to obtain the dendogram using average linkage. Data on fat composition of consumable lamb prepared by different methods was obtained from USDA Nutrient Database for Standart Reference. Using agglomerative hierarchical cluster analysis lamb meat was grouped into two main clusters among which one divided into two families of which each was subdivided into two subfamilies based on fatty acids, cholesterol and energy composition. Present work may be considered as a leading study to systematically classify larger food sets. As high fat foods are rich in flavor and overall palatability, the outcome of this study may lead to behaviorally more acceptable but healthier dietary replacements. Besides future use of the results obtained may reveal the effect of complex compositional dietary influences on health and disease and may have superiority to studies questioning individual dietary items. Furthermore, hieararchial cluster analysis may be used to cluster food including other compositional data in food items like amino acids, vitamins, carbohydrates, as well. (C) 2000 Harcourt Publishers Ltd.