Food group categories used in dietary analysis can misrepresent the amount and type of fat present in foods

Nutrition Research Newsletter, Dec, 2006

Population-based dietary studies can be very difficult to compare because of differences in food classifications systems used and definitions of food groups. There have been considerable variations in the foods included as part of a defined food group in various research studies. In order to identify the true source of nutrients, particularly fat, consumed in published diets, it is critical to examine how the food groups and any subsets within a food group were defined. Food classification systems with small numbers of groups such as the 'core food groups' used for dietary intake advice are limited in their use for dietary analysis aimed at determining nutrient intake patterns in both small-scale dietary studies and population studies. The greater the division of foods into groups for dietary analysis, the greater the detail that can be obtained on dietary nutrient contribution from specific food commodities. However, expanding food groups into an excessively broad range of categories with narrow content defeats the intent of simplifying analysis to a manageable level. Obviously, a realistic balance is needed, along with an appreciation of the inherent limitations.

A comprehensive Australian food grouping system consisting of 19 categories was developed and used for the National Nutrition Survey (NNS) of 1995. However, even this system combined some dissimilar food items together into specific groups, a good example and the point of the present study being the 'meat, poultry and game products and dishes' group, which included processed meats, meat dishes, and fast foods or takeaway foods. A re-evaluation of the 1995 NNS data by CSIRO nutritionists used a 21 food group system that subdivided two NNS food groups (meat, poultry and game products and dishes; fish and seafood products and dishes) into four new groupings (red meat cuts, red meat products and dishes, white meat cuts, and white meat products and dishes, where 'white meat' included poultry, pork, fish, and seafood). Application of this expanded and subtly differing grouping system to the 1995 NNS dietary intake data gave a substantially differing interpretation of some nutrient intakes from specific food groupings, particularly the total fat and fat subgroup quantities from the meat-based groupings.

The quantity, type, and source of fats consumed in people's diets have become an important area of study because of their considerable health implications. Excessive fat intakes, particularly of saturated fatty acids, have been linked with increased risk of coronary heart disease, obesity, some cancers, and hypertension. The World Health Organization recommends that not more than 30% of energy intake be derived from fats, with 10% or less coming from saturated fats, which are found mainly in animal-based products, coconut and palm oils and vegetable shortening. The World Health Organization also suggests that not more than 300 mg/day cholesterol be consumed, as high-cholesterol levels appear to elevate the risk of cardiovascular diseases by increasing the risk of atherosclerosis. Cholesterol is found almost entirely in animal-based foods, mainly egg yolk, offal, meat, and dairy products.

In recent years, there has been a perception that 'red meat' is 'bad for you.' This arose because of advice from various public health agencies to eat less red meat and hence (it was implied) consume less fat in order to prevent CVD and cancer. Animal fats were, therefore, equated with saturated fats and thereby linked with atherosclerosis and heart disease. For this reason, people endeavouring to lead a healthy life may avoid eating meats in general. Yet lean meat is an important source of dietary protein, iron, zinc, vitamin B 12, thiamine, niacin, potassium and omega-3 fatty acids, so that omitting meat from the diet means missing out on a rich source of these valuable nutrients. This is well illustrated by the poor iron and zinc intake recorded in the 1995 NNS for women aged 19 to 29 years who were low meat consumers. More recently, analysis of the 1995 NNS food intake data using a fatty acid database of foods revealed red meat to be a critical source of the beneficial long-chain n-3 polyunsaturated fatty acids in the diet.

The present study aimed to determine the contribution of defined food groups (particularly the meat groups) to the total fat and fat type consumed by male omnivore subjects through analysis of their diets with three different food group classification systems in which meat, meat products, and meat dishes were either combined or separated. Dietary intake data were collected as semiquantitative FFQ and analyzed using the NUTTAB 95 database. The contribution of food groups to nutrient intakes was determined using three food classification systems.

When red meat was grouped with meat, poultry, and game products and dishes, 'meat' contributed 19.8% of total fat to a diet typical of an adult Australian man. When lean meat cuts were distinguished from meat products or dishes and fast foods, 'red meat cuts' provided only 4.3% of the total fat in this diet, whereas 'fast foods' provided 18.7%.