Healthy Eating Index 2010 calculation
The basic steps in calculating the Healthy Eating Index (HEI)-2010 component and total scores for individual food intake are described at National Cancer Institute's Applied Research website, where the SAS code is also archived. For this study the code for HEI-2010 for 24-hour recalls was used.
For a complete food equivalent data set, the MyPyramid Equivalents Database (MPED), version 2.0, the USDA's Center for Nutrition Policy and Promotion (CNPP) MPED Database for Whole Fruit and Fruit Juice for NHANES 2003-04, and the CNPP Addendum 2.0B were first combined. The CNPP Addendum to the MPED, version 2.0B was used since it contained food codes reported NHANES 2005-06, the same time period data were collected in HANDLS study. Then these data were merged with HANDLS baseline individual food intake data to calculate participants’ 1-day food-group intake. The number of equivalents per 100 grams of food was generated for the following food groups: Total Fruit, Total Vegetables, Dark Green Vegetables, Legumes, and Total Milk expressed in cup equivalents; Whole Grains, Meat, Poultry, and Fish, Eggs, Nut and Seeds, Soybean products, Fish and shellfish high in n-3 fatty acids, Fish and shellfish low in n-3, and Non-whole/refined Grains expressed in ounce equivalents; Discretionary solid fat expressed in grams; and Added sugars expressed in teaspoon equivalents. Foods containing fruit were assigned to either Whole Fruit or Fruit Juice in cup equivalents per 100 grams of foods. For foods which contained both whole fruit and fruit juice, Total Fruit equivalents were assigned to either Whole Fruit or Fruit Juice, based on the majority ingredient according to its description or recipe USDA’s FNDDS, Version 2.0. The data were reviewed by Registered Dietitian who assigned the appropriate food group for 53 items reported in the HANDLS study that did not match an 8-digit code in MPED and verified the assignment of missing total fruit items into either the whole fruit or fruit juice group.
In general, the initial step used to obtain the required variables to compute the components of the HEI-2010 as well as the total score, was to estimate each day’s intake using the grams consumed of each food code that was assigned with either the FNDDS (per gram of food: grams×nutrient.gram-1) or MPED database (per 100 g of food: (grams×100×(equivalent per 100g)). Daily intake was obtained by summing the grams per day of food groups and nutrients across food codes but within individuals for each day of intake. Special preliminary calculations were required for selected groups, such as empty calories and soy beverages. For instance, in the HEI-2010 calories from alcohol are considered to be empty calories, but only when alcohol is consumed beyond moderate amounts. Soy beverages were assigned to the Dairy component of the HEI-2010. However, in the MPED soy beverages were grouped with other soy products in the Meats and Beans group. Detailed descriptions of essential modifications to calculations can be found at HEI Tools for researchers.
Since the HEI-2010 is a multi-dimensional construct involving 12 densities (amounts of food per 1,000 calories and ratios of fatty acids), the use of a Monte Carlo simulation step was required for the calculation of standard errors. The SAS code for the generation of a Monte Carlo dataset was provided at the National Cancer Institute's Applied Research website.
Using the Monte Carlo data set, a SAS macro was run for the allocation of beans and peas. Legume intake was first allocated to Total Proteins group and once this standard was met, the additional intake was allocated to Total Vegetables, and Greens and Beans components. Then using the Monte Carlo data set with the beans and peas allocated, the HEI-2010 scoring macro was used to calculate densities for each HEI-2010 component and to apply the scoring algorithm. The total and component HEI-2010 scores were calculated for each recall day (day 1 and day 2) and then were averaged to obtain the mean HEI-2010 total and component scores for both days combined.