The USDA food coding scheme is a way of organizing the thousands of food codes used in NHANES and can be used as a guide to grouping foods for your particular research needs.
The first digit of each USDA food code represents one of nine major commodity groups listed below.
Milk and milk products
Meat, poultry, fish, and mixtures
Legumes, nuts, and seeds
Fats, oils, and salad dressings
Sugars, sweets, and beverages
The second, third, and sometimes fourth digits of a code identify increasingly more specific subgroups within the nine major groups. The remaining digits are used to identify specific foods within a numerical sequence. Here is an example for milk and milk products.
|Food Code Digits||Group/subgroup/description|
Milk and milk products
Milk and milk drinks
Flavored milk and milk drinks
Milk, chocolate, whole-milk-based
It is important to note that some individual food codes represent discrete food items while others represent mixed dishes consisting of multiple ingredients. If a food code represents a mixed dish, it is grouped according to its major ingredient. See Mixed Dishes, below, for more information on this topic.
During the collection and coding of dietary recall data, many individual food codes are linked together using “combination” codes. These codes allow investigators to account for individual foods that are consumed simultaneously, such as sugar in coffee or milk on cereal, or for food mixtures that are reported as discrete ingredients, such as a homemade sandwich reported separately as bread, cheese, lettuce and mayonnaise.
Combinations are defined using two separate variables. The first, the combination food number, flags foods as being eaten in combination. Each combination is given a unique combination food number, and these are listed in sequence (i.e., the first food combination reported by a participant is 1, the second is 2, and so on). The second variable, the combination food type, designates the type of combination, as shown below in the list.
Beverage with additions
Cereal with additions
Bread/baked products with additions
Ice cream/frozen yogurt with additions
Dried beans and vegetable with additions
Fruit with additions
Meat, poultry, fish
The following example shows the relationship of the combination food number and combination food type for three food items. Note that all the foods in a given combination are assigned the same combination food type code and combination food number.
|Combinations eaten||Food Items reported||Combination food number||Combination food type|
|Cereal with fruit and milk||Frosted flakes cereal||01||02|
|Coffee with sugar||Coffee||02||01|
|Ham and cheese sandwich||Ham||03||05|
Foods with a value of 00 for both combination number and combination type are either discrete food items that were not eaten in combination or mixed dishes coded with a single food code.
Mixed dishes include food items such as stews, soups, casseroles, sandwiches, pasta with meats and sauces, pizzas, and tortilla dishes (such as enchiladas and burritos). As mentioned above, mixed dishes can be coded either by an individual food code or several food codes—representing the ingredients of the mixture—linked by a combination food number and combination food type.
Individual food codes representing mixed dishes are included in many food groups and subgroups of the coding scheme. Generally, mixtures represented by individual food codes are placed in food groups based on the primary component or ingredient in the mixture. For example, a cheeseburger on a bun is assigned to the “meat, poultry, fish and mixtures” group because the hamburger is considered to be the main ingredient. Lasagna with meat is assigned to the “grain products” group because the noodles are considered to be the main ingredient.
Certain types of mixtures, such as sandwiches, salads, and soups, can be included in various food groups, depending on their main ingredient. Therefore, if you are interested in a particular type of mixture, it is important to note all the possible food groups that could contain codes related to that mixture. For example, different kinds of sandwiches can be found in various food groups and subgroups, as shown below.
Because of the level of detail in the descriptions for many FNDDS foods, it is unlikely that you will examine recall data by individual food codes. It is more likely that you will combine food codes representing similar foods into food groups and examine the data by food group.
If you plan to develop a food grouping scheme for your study, your analysis should take into consideration the widespread use of foods coded as multiple items and linked by a combination food number and type. In NHANES 2003-2004, about one-half of all food items were recorded as combination foods.
It may be useful to conduct a preliminary review of the data to ensure that your groups are going to be meaningful. Look at how often foods in your group(s) are reported and also the gram amounts consumed. Then decide whether to combine groups or to disaggregate to a greater level of detail. USDA's Food Surveys Research Group has developed a set of food groups for their reports and analyses. These food groups are currently based on the 8-digit food code. To learn more about these, see the “Task 3: How to Access the FSRG-Defined Food Groups".