The Healthy Eating Index (HEI) is a tool developed by the United States Department of Agriculture and the National Cancer Institute to evaluate the extent to which diets are consistent with the Dietary Guidelines for Americans. Possible index points range from 0-100, with a higher score indicating greater consistency of the diet with the Dietary Guidelines for Americans. The HEI-2015 is the most recent version of the HEI and aligns with the 2015-2020 Dietary Guidelines for Americans. To learn more about the HEI including the components and scoring, click here to visit the NCI website.

HEI scores may be calculated using variables available in NDSR output files. To facilitate generating HEI-2015 scores we have created SAS programs that allow for calculating HEI-2015 scores in accord with two calculation approaches (per each intake day for each person and per person across all intake days). The programs for both calculation approaches calculate total HEI-2015 scores and scores for each of the 13 components of the index.

Note that two SAS programs are available for each approach. One program should be used for data collected using NDSR 2013 or any subsequent version of the program. The other program should be used for data collected using NDSR 2012 or an earlier version of the program. Please see the **special instructions** for calculating HEI-2015 scores for data collected using NDSR 2012 or an earlier version of the program.

A Guide is available to support you in using the SAS code for calculating HEI-2015 scores.

**Approach 1: SAS code for calculating HEI-2015 scores for each intake day for each person**

The SAS programs below calculate HEI-2015 scores for person-level NDSR data for each day of intake data, regardless of how many intake days are available for each person.

- SAS code for calculating HEI-2015 scores for each intake day for each person for data collected in NDSR 2013 or subsequent version
- SAS code for calculating HEI-2015 scores for each intake day for each person for data collected in NDSR 2012 or earlier version

**Approach 2: SAS code for calculating HEI-2015 scores per person when multiple days of intake data are available for each person**

The SAS programs below calculate one HEI-2015 score per individual across all days of recalls or records collected, even if some participants have varying number of intake days, including only one.

- SAS code for calculating HEI-2015 scores per person when multiple days of intake data are available for each person for data collected in NDSR 2013 or subsequent version
- SAS code for calculating HEI-2015 scores per person when multiple days of intake data are available for each person for data collected in NDSR 2012 or earlier version

**Special Instructions for Calculating HEI-2015 Scores for Data Collected using NDSR 2012 or an Earlier Version of the Program**

Before using one of the SAS programs provided for calculating HEI-2015 scores ** study data must be restored in NDSR 2012 or a subsequent version of the program if data were collected using an earlier version of the program. **Restoration is required so that all needed variables used for calculating HEI-2015 scores are available in the NDSR output files.

** If some of your data for a specific study was collected using NDSR 2012 or an earlier version of the program and some was collected using NDSR 2013 or a subsequent version, we recommend using the SAS programs provided for calculating HEI-2015 scores for data collected using NDSR 2012 or an earlier version of the program for all of your study data.** We recommend using the same SAS program to ensure HEI-2015 scores are calculated in an identical manner for all study participants and all dietary intake data collected from study participants. Rationale for this recommendation: The SAS programs for calculating HEI-2015 scores for data collected using older versus newer versions of NDSR rely on different variables in NDSR output files for measuring servings of whole grains and refined grains. Consequently, HEI-2015 scores calculated by each SAS program may differ as an artifact of the differences in whole grain and refined grain variables used in the calculations.