Policy Research for Front of Package Nutrition Labeling: Developing and Testing a Summary System Algorithm

Publication Date

Apr 30, 2011

Prepared for

Kathleen Koehler
Department of Health and Human Services
Office of Assistant Secretary for Planning and Evaluation
Office of Science and Data Policy
200 Independence Avenue SW
Washington, DC 20201

Prepared by

Joanne E. Arsenault, RTI International
Victor Fulgoni, Nutrition Impact, LCC
James C. Hersey, RTI International
Mary K. Muth, RTI International
RTI International
3040 Cornwallis Road
Research Triangle Park, NC 27709
RTI Project Number 0212050.009
RTI International is a trade name of Research Triangle Institute.

Disclaimer
This report was prepared by RTI International, under contract to the Assistant Secretary for Planning and Evaluation. The findings and conclusions of this report are those of the author(s) and do not necessarily represent the views of ASPE or HHS.

Executive Summary

Front-of-package (FOP) and shelf nutrition labeling of food products can potentially improve diets and health of the U.S. population if they are easy to use and understand and accurately reflect the nutritional quality of foods. The Department of Health and Human Services (DHHS) and the Food and Drug Administration (FDA) are interested in ensuring that nutrition criteria that manufacturers use on FOP or shelf labels are based on evidence. The Assistant Secretary for Planning and Evaluation (ASPE) contracted with RTI International to examine the nutrient criteria behind existing summary FOP or shelf systems and develop and test an algorithm for a summary FOP system.

A summary FOP or shelf system rates the overall nutritional quality of a food with an overall numeric score, a multiple category rating, or a single icon to indicate whether a food meets specified nutrient criteria. Summary systems are one of two general types of FOP systems; the other type is a nutrient-specific system that displays a few nutrients and their amounts per serving and sometimes includes the percentage of a daily recommended value. In some instances, both a summary and nutrient-specific system may be applied to a food item (e.g., a summary check mark and the number of calories per serving or a summary score on a shelf tag and a nutrient-specific label on the front of the food package). On the one hand, nutrient-specific systems can help consumers identify key nutrients in a food but do not provide an overall assessment of the product. Therefore, the consumer may focus on a single nutrient (e.g., calories) and not get a sense of the overall nutritional quality of the food. On the other hand, summary systems can provide an overall assessment of the nutritional quality of a food but do not tell the consumer why the food received the score or rating (e.g., high calcium and low saturated fat content of low-fat milk). Few studies have directly compared how consumers understand the two types of systems, but it appears consumers can identify healthier food choices more easily with a nutrient-specific system with traffic light colors of red, yellow, and green than a summary check mark type system. This report considers only summary systems.

RTI developed and tested a nutrient density-based algorithm that included positive scores for nutrients that should be encouraged and negative scores for nutrients that should be limited in the diet. We scored a set of foods using the algorithm and compared the average scores of food groupings. As expected, nutrient-dense foods (i.e., foods with substantial amounts of vitamins and minerals and few calories) scored high, and foods that are low in nutrient density (i.e., that supply calories but relatively small amounts of micronutrients) scored low. Fruits, vegetables, and legumes and nuts had the highest group scores, and the lowest group scores were seen with fats and oils, sweets, and beverages (e.g., coffee, tea, soft drinks, and fruit drinks). A series of modifications were made to the algorithm, adding and removing various nutrients and food components, and effects on food scores and the ability of the algorithm to predict overall dietary quality were assessed. As a measure of overall dietary quality, we used the Healthy Eating Index (HEI) developed by the Center for Nutrition Policy and Promotion, U.S. Department of Agriculture (USDA) and the National Cancer Institute, U.S. Department of Health and Human Services. HEI scores are calculated by assigning points for adherence to food group-based guidelines in MyPyramid and are based primarily on recommended volume amounts per 1,000 kcal. Scores for each food-group or component of the HEI are capped and the highest HEI score is 100. Although the HEI scores total diet, it is inherently different from a food-scoring algorithm; nevertheless, it provides a mechanism to evaluate how well the individual food scores of an individual's diet relate to overall diet quality. In this report, we demonstrate that a final algorithm incorporated weighting factors to nutrients that were derived from statistical analyses of nutrient intakes of the U.S. population resulted in higher prediction of dietary quality than seen with existing nutrient density algorithms that have been tested similarly. Our final algorithm explained two-thirds of the variation in HEI scores, compared with one-third to one-half with other nutrient density algorithms. Our algorithm included nutrients or food components with positive weighting factors for protein, unsaturated fat, fiber, calcium, and vitamin C and with negative weighting factors for saturated fat, sodium, and added sugars. The use of nutrient values per 100 kcal was slightly better at predicting overall dietary quality than using nutrients per reference serving sizes (reference amounts customarily consumed, RACC). Among the top-scoring foods were raw and leafy green vegetables on a per 100 kcal basis and avocado, almonds, oranges, and strawberries on a per RACC basis. The algorithm worked well in predicting dietary quality across various population groups, such as age, ethnicity, socioeconomic status, and weight status.

Nutrients or components included in the algorithm:
Positive: Protein Unsaturated fat Fiber Calcium Vitamin C	Negative: Saturated fat Sodium Added sugar

This project examined various summary FOP systems, the selection of nutrients and nutrient criteria for FOP systems, issues in developing and testing a summary system, and the subsequent development and testing of a system. The results from this project reveal relatively small differences in predicting overall dietary quality with inclusion or exclusion of various nutrients; therefore, using a "best model" approach to determine which algorithm explains the greatest variation in dietary quality is one approach to determining the selection of nutrients. Consumers selecting foods in the supermarket may not think in terms of how a food fits into their overall diet; therefore, it remains to be determined if this type of system will assist consumers in the supermarket.

The ability of consumers to understand and select healthier food choices using summary systems needs further testing before further recommendations can be made. This project began to explore a new approach to identify nutrients and weighting factors to include in a scoring system; however, additional nutrients or alternative weighting of nutrients could be explored further. A summary system is not necessarily intuitive or transparent regarding nutrients that influence the score or rating; therefore, it is important that it be accepted by a respected regulatory agency such as FDA.

1. Introduction

Food manufacturers have recently been adding summary nutrition information on the fronts of packages in addition to the back or side panel Nutrition Facts Panel (NFP). In addition, some supermarkets have added summary information or symbols on shelf labels. The Department of Health and Human Services (HHS), Office of the Assistant Secretary for Planning and Evaluation (ASPE) and the Food and Drug Administration (FDA) are interested in ensuring that nutrition criteria that manufacturers use to make broad front-of-package (FOP) or shelf-label claims concerning the nutritional quality of a food are based on evidence and are not misleading to consumers. It is recognized that multiple FOP labeling schemes can cause confusion for consumers and that a uniform system based on current science may be optimal.

Figure 1-1. Examples of Summary FOP or Shelf Nutrition Labels

Example of a numerical scoring shelf-label system (NuVal; reprinted with permission from NuVal, LLC, U.S.)	Example of a threshold multiple category rating shelf-label system (Guiding Stars; reprinted with permission from the Guiding Stars Licensing Company, U.S.)
Example of a threshold system FOP icon system (Australian/New Zealand National Heart Foundation Tick; reprinted with permission from the National Heart Foundation of Australia)

The science of ranking foods based on their nutrient content is referred to as "nutrient profiling." A nutrient profile is a ranking, either on a continuous scale (e.g., 1 to 100) or categorical (such as low, medium, and high). Nutrient profiling is the basis for summary FOP and shelf-label systems. Some systems use nutrient criteria or thresholds to determine whether a food receives an icon signifying a healthy or nutritious food. Most systems are based on similar dietary recommendations, such as the 2005 Dietary Guidelines for Americans (2005 DGA) (U.S. Department of Health and Human Services & U.S. Department of Agriculture, 2005), which encourage the consumption of nutrient-dense foods to meet recommended nutrient intakes. The 2005 DGA note particular nutrients of importance that are consumed in less than optimal amounts by the population, such as fiber; vitamins A, C, and E; and calcium, magnesium, and potassium, and nutrients to limit, such as saturated fat, added sugars, and sodium. FDA criteria for nutrition labeling regulations and claims are often used in summary FOP systems, particularly threshold-based systems. For example, a "healthy" food as defined by FDA has to contain at least 10% of the daily value of one or more of these six nutrients: protein, fiber, vitamin A, vitamin C, calcium, or iron per reference amount. In addition, it has to be low in total fat (<3 g), saturated fat (<1 g), cholesterol (<60 mg), and sodium (<480 mg) per reference amount. Many other summary FOP systems use these FDA threshold criteria.

Some key factors must be considered in developing an FOP labeling scheme:

selection of the nutrients to include in the algorithm
weighting of nutrients
assignment of an overall rating score, a category rating, or a symbol (e.g., check mark)
validation of the ability of the system to rank foods and predict diet quality

FOP labeling systems use algorithms containing selected nutrients to calculate an overall score of nutritional quality or to determine if a food meets specified criteria. Nutrient selection involves the choice of "positive" or "negative" nutrients:

Positive nutrients are those associated with health effects and are to be encouraged.
Negative nutrients are related to obesity and chronic disease when consumed in excess and should be limited.

Nutrient profiling systems can use either positive, negative, or both kinds of nutrients. Some systems use additional nonnutrient components such as whole grains or glycemic load (Katz et al., 2009).

Nutrient density is a basis for some nutrient profiling systems and typically refers to the amount of nutrient per kcal (or some multiple of kcal, for example, 100 kcal). Nutrient criteria for calculating a nutrient density index have been examined using various combinations of nutrients to encourage and nutrients to limit. One nutrient density system found that a combination of nine nutrients to encourage and three nutrients to limit best predicted an overall measure of dietary quality (Fulgoni, Keast, & Drewnowski, 2009).

Weighting factors can be applied to nutrients in a scoring system, where some nutrients are assigned higher weights than others. Weighting can be based on bioavailability, biological quality, or presence in the food supply. For example, iron in meat is more bioavailable than iron from plant foods. Animal protein is higher quality (defined by amino acid content) than plant protein. Some nutrients are only present in selected foods, such as vitamin B₁₂, which is only present in animal foods. Few systems actually use weighting factors because the optimal method to weigh nutrients within algorithms is uncertain; however, this remains a topic of interest for nutrient profiling research.

Numerous FOP labeling schemes assign a rating, symbol, or icon only if a food meets certain nutrient criteria. For example, the American Heart Association assigns a check mark to foods meeting criteria for total fat, saturated fat, trans fat, cholesterol, sodium, and fiber. Guiding Stars assigns one, two, or three stars to foods meeting different criteria for vitamins A and C, calcium, iron, zinc, fiber, whole grains, saturated fat, trans fat, cholesterol, added sugars, and sodium.

There is no gold standard to measure the performance of nutrient-based FOP labeling systems in ranking foods by nutritional quality. Often the developers of these systems have asked nutrition experts to rank the foods based on their subjective expert opinion. Different FOP labeling systems can be compared by selecting specific foods and comparing the ranking of the food using the various systems. Current opinion among nutrient profile researchers for validating FOP labeling systems is to compare the total rankings of all foods consumed by individuals in a dietary survey against an overall measure of dietary quality using the same dietary intake survey data. The best-performing system can be assessed by the degree of variance explained by the system in regression analyses. We used this methodology to test our algorithms in this project.

Numerous issues must be considered when evaluating FOP labeling systems with the ultimate goal of identifying or developing a single system for the United States. The best nutrient-based system will only be effective if the consumer understands and uses the system to make healthier choices. Critical evaluation of both the consumer and nutritional aspects of FOP labeling systems is necessary to inform the development of effective policy.

1.1 Overview of the Project Approach

The overall project consisted of two parts, one was to conduct an environmental scan of consumer, producer, and retailer responses to FOP systems, and the second part was to examine nutrient criteria of summary systems and develop and test algorithms to rank the nutritional quality of foods. The results of the environmental scan are reported separately, and this report focuses on the examination of nutrient criteria of summary systems and development and testing of algorithms.

For this project, RTI reviewed summary FOP systems in the United States and abroad in terms of nutrient criteria, aspects of development of an algorithm to rate the nutritional quality of foods, and testing and validation methods. This task involved reviewing the literature regarding nutrient criteria, development, and testing of summary FOP or shelf systems.

Based on the review of literature, RTI developed three options of algorithms that could be developed or modified to test various concepts identified to be important in a nutrient profiling system. The options were presented to the ASPE/FDA team and decisions were made to develop one of the algorithms and test various modifications of the algorithm. After reviewing results of the testing of the algorithm and modifications in their ability to rank the nutritional quality of foods and predict overall dietary quality, a final algorithm was decided upon for final analyses that included creating categorical rankings from a continuous score and predicting overall dietary quality among select subpopulations.

1.2 Report Organization

This report reviews nutrient criteria and validation of FOP systems and describes the development and testing of FOP algorithms.

Section 2 presents a description of summary FOP or shelf systems based on available information in the scientific literature and on Internet Web pages associated with the systems.
Section 3 describes the options for algorithm development and testing proposed by RTI and agreed upon by ASPE/FDA.
Section 4 describes the methodology, results, and discussion on the development and testing of the algorithm.
Section 5 provides conclusions, limitations, and knowledge gaps.
Appendices include additional details on existing summary FOP systems and the methodology and results of algorithm development and testing.

2. Description of Summary FOP Systems

2.1 Introduction

The purpose of this section is to

examine the nutrient criteria of selected summary systems, including FOP, shelf, and nutrient profiling systems;
assess the ability of summary systems to rank foods according to contribution to diet quality; and
identify criteria and considerations for developing summary systems.

The information about the specific systems included in this review reflects the information that was available and attainable at the time of this review. Thus, some details on the systems are not known. For example, a methodology paper on Guiding Stars is in press and may provide more transparency for the nutrient criteria. The review focuses on 13 specific systems that were identified based on literature searches, particularly on those systems that have been validated and systems mentioned in the scope of work for the task order. The systems include FOP or shelf-labeling systems (e.g., Guiding Stars, NuVal) or other nutrient profiling systems for deciding which foods can be advertised to children (e.g., Ofcom model) or to improve a manufacturer's product line (e.g., Unilever's Nutrition Enhancement Program) (Table 2-1).

Table 2-1. List of FOP, Shelf, or Nutrient Profiling Systems Reviewed
Name of System (Abbreviation)	Country	Purpose	Application of Nutrient Criteria	Type of Nutrient Rating
Guiding Stars (GS)	United States	Shelf	Food categories (4)	Threshold
Heart Check Mark (AHA)	United States	FOP	Food categories (6)	Threshold
Smart Choices Program (SC)	United States	FOP	Food categories (19)	Threshold
Nutrient Rich Foods Index (NRFI)	United States	Nutrient profiling	Across the board	Numerical score
NuVal (NV)	United States	Shelf	Across the board, with food category adjustments	Numerical score
Sensible Solution (SS)	United States	FOP	Food categories (16)	Threshold
UK Ofcom Nutrient Profiling Model (Ofc)	United Kingdom	Nutrient profiling	Across the board	Threshold
Choices Programme (CP)	Global	FOP	Food categories (17)	Threshold
Keyhole symbol (K)	Sweden	FOP	Food categories (25)	Threshold
Pick the Tick (Tick)	New Zealand, Australia	FOP	Food categories (55+)	Threshold
Heart Check (HC)	Canada	FOP	Food categories (84)	Threshold
Nutrition Enhancement Program (NEP)	Global (Unilever)	Nutrient profiling	Food categories (8)	Threshold
Netherlands Tripartite Classification System (NTri)	Netherlands	Nutrient profiling	Food categories (14)	Threshold

The Heart Check Mark of the American Heart Association (AHA) was the first FOP system in the United States, launching in 1995. The intention of the system is to provide guidance to consumers for cardiovascular disease (CVD) risk reduction. The system has been periodically updated based on new scientific and regulatory labeling information that has become available. Food manufacturers followed with their own systems. For example, Kraft Foods introduced Sensible Solution for its product line in 2005. Guiding Stars was introduced as a shelf-labeling system in 2006 by Hannaford Supermarkets. The NuVal system was introduced as a shelf-labeling system in 2007 by an independent scientific panel supported by Griffin Hospital in Connecticut. The Nutrient Rich Food Index was developed by researchers based on the concept of nutrient density promoted by the Nutrient Rich Foods Coalition. Smart Choices was developed in 2008 by a coalition of scientists, food manufacturers, and retailers, but the program was suspended after complaints about the system appearing on sweetened breakfast cereals and a letter from FDA regarding concerns about the criteria used in the system.

Outside of the United States, the Keyhole system was developed in Sweden in 1989 and was expanded to Denmark and Norway. The Pick the Tick program was developed in 1991 by the Australia and New Zealand Heart Foundation. The Heart Check program was launched in 1999 by Canada's Heart and Stroke Foundation. The Choices Programme was developed based on criteria set by an independent international scientific committee and the program was implemented by the Choices International Foundation. The program was first introduced in the Netherlands in 2006. The UK Ofcom Nutrient Profiling Model is used to define healthy and unhealthy food products for television advertising to children. Two other nutrient profiling systems are the Nutrition Enhancement Program, developed by Unilever to evaluate and improve the nutritional quality of their products, and the Netherlands' Tripartite Classification System.

2.2 Across-the-Board Versus Food Category - Specific Summary Systems

Systems rank or rate foods in nutritional quality by using nutrient criteria that are either

across the board, the same for all foods, or
food category-specific, different for food categories (Table 2-1).

Most systems are category specific and are meant to compare the nutritional quality of a food to other foods in the same category. The number of food categories used in category-specific criteria varies; some systems use only a few categories and others use over 80 categories (e.g., Heart Check). The category-specific systems may reflect dietary guidelines, such as the recommendation to consume low-fat or nonfat milk instead of full-fat products.

An across-the-board approach allows for the comparison of foods across categories, although it also has the advantage of allowing within-food category comparisons. Another advantage of across-the-board systems is that they are simple because they have only one set of nutrient criteria.

Food category-specific systems incorporate intrinsic properties of foods that may be specific to a food group. For example, vitamin B₁₂ is only in animal-based foods, so the nutrient criteria for milk may include vitamin B₁₂, but B₁₂would not be included in the criteria for fruits. The consensus of participants in an International Life Sciences Institute (ILSI) Europe workshop on nutrient profiling was that a category-specific approach would be preferred, because this approach addresses the intrinsic differences in foods (Tetens, Oberdorfer, Madsen, & de Vries, 2007). A disadvantage of the approach is that it can be complex, because it may be difficult to categorize some foods, and different nutrient criteria thresholds must be developed for each category.

To assess whether a nutrient profiling system should be generic or category specific, Scarborough et al. (2010) compared the diets of participants in a British dietary survey with average scores using the algorithm from the UK Ofcom Nutrient Profiling Model, referred to as a WXYfm score. It was determined that to improve their diets similar to the level of the healthy-diet consumers, consumers of unhealthy diets would have to both consume healthier versions of foods within food categories and consume greater amounts of healthy foods. Individuals with a healthy diet consumed healthier versions of foods in the following categories: meat, dairy (except cheeses), bread (with minimal fiber), cereal products (except breakfast cereals), ready meals/soups/sandwiches, and breakfast cereals. These results suggest that a category-specific system may be preferable for nutrient profiling models designed to promote healthy diets. The authors suggest that these results could be used as a basis for deciding which food categories should be included in a system, that is, only those foods for which evidence exists that individuals with a healthy diet consume healthier versions of foods within those categories than those with an unhealthy diet. The results of this study also suggest that a limited number of food categories are probably necessary. As the authors state, "it would seem unlikely that people who achieve a healthy diet do so by consuming healthier versions of ice cream than those who consume an unhealthy diet" (p. 6).

Food category-specific systems created by a single manufacturer may be of limited use in general because they are likely designed to fit their product line and may not allow for comparison to other brands in that category. Other brand products in a category may have another profiling system using different criteria or may not have a profiling system at all.

2.3 Type of Nutrient Ratings of Foods

Ratings of foods can be presented as

an overall summary score or
a threshold (see Table 2-1).

A threshold rating system uses cutoff values to determine whether a food meets specified nutrient criteria. Often there is one set of criteria used to determine whether the system receives the rating, and the rating is often displayed as an icon such as a check mark. Some systems use several threshold levels for a graded rating system, such as the Guiding Stars, which rates foods with none, one, two, or three stars.

Systems that use an overall numerical score, such as the Nutrient Rich Food Index and NuVal systems, rate all foods on a continuum. For example, the algorithm for the NuVal system, which is called the Overall Nutrient Quality Index, scores foods from less than 1 to a large finite number but then converts the scores to 1 to 100 for consumer use. Threshold systems use nutrient criteria to determine if a food receives a rating, which is often a pictorial icon such as a check mark. An example of a threshold system is the Ofcom Nutrient Profiling Model, which uses ratings of healthy and less healthy. Icons are popular on FOP and shelf labels because they are graphic and "eye catching." The food receives an icon if it meets specified nutrient "thresholds." Icons usually represent a dichotomous condition, either the food product meets the criteria or it does not. Some systems use multiple ratings with the icon (e.g., Guiding Stars provides either no star or one to three stars based on different nutrient thresholds for each star).

Threshold systems generally categorize foods as good or bad, whereas numerical scoring systems do not. Some nutritional professionals believe there are no bad foods and that it is overall dietary intake that is important; therefore, they criticize threshold schemes that label foods as "bad." Another criticism of the use of thresholds to determine a rating or whether a food qualifies for an icon is that the cutoff values are chosen somewhat subjectively because there is no absolute scientific basis for them.

2.4 Nutrient Criteria for Summary Systems

The nutrient criteria used to develop summary systems are based on

dietary recommendations set by governmental or other authoritative organizations,
scientific and epidemiological evidence regarding associations with nutrients and chronic disease risk, and
nutrient intake shortfalls among the general population or subgroups.

Some practical limitations preclude the inclusion of some nutrients in scoring systems, such as the nutrient information not being readily available on the NFP or in computerized nutrient databases.

Many of the summary FOP systems in the United States focus on nutrients of concern as identified in the 2005 DGA (U.S. Department of Health and Human Services & U.S. Department of Agriculture, 2005). The 2005 DGA note particular nutrients of concern that are consumed in less than optimal amounts by the population, such as fiber; vitamins A, C, and E; and calcium, magnesium, and potassium. The 2005 DGA report notes additional nutrients of concern for specific subpopulation groups: iron, folic acid, vitamin B₁₂, and vitamin D. The 2005 DGA also recommend Americans limit their intake of total fat, saturated fat, trans fat, cholesterol, sodium, and sugar based on scientific evidence linking excess intake of these nutrients with a disease. The DGA were recently updated, and the 2010 DGA specify nutrients of concern consumed in low amounts by the U.S. population, such as fiber, vitamin D, calcium, and potassium; and nutrients of concern for specific subpopulations include iron, folic acid, and vitamin B₁₂ (U.S. Department of Agriculture & U.S. Department of Health and Human Services, 2010). The 2010 DGA report that Americans consume excessive energy (particularly from solid fats and added sugars), saturated fats as a percentage of total energy, cholesterol (men only), sodium, and refined grains.

FDA criteria for nutrition labeling regulations and claims may also be a basis for nutrient criteria in summary FOP systems. For FDA to consider a food healthy, the food has to contain at least 10% of the daily value of one or more of these six nutrients: protein, fiber, vitamin A, vitamin C, calcium, or iron per reference amount. Foods are disqualified from making a health claim if they contain ≥13 g of fat, ≥4 g of saturated fat, ≥60 mg of cholesterol, or ≥480 mg of sodium per reference amount (Code of Federal Regulations, 2001 amended).

FOP labeling systems use algorithms containing selected nutrients to calculate an overall score of nutritional quality. Nutrient selection involves choosing positive or negative nutrients:

Positive nutrients are those associated with health effects and are to be encouraged.
Negative nutrients are related to obesity and chronic disease when consumed in excess and should be limited.

Nutrient profiling systems use either positive, negative, or both type of nutrients (Table 2-2). Some systems use food groups (e.g., the Ofcom model includes criteria for fruit, vegetable, and nut content) or additional nutrition-related concepts (e.g., NuVal includes glycemic load and protein quality). For some systems, foods are only awarded an icon if they meet the threshold criteria for both positive and negative nutrients. Other systems score foods according to the amount of positive and negative nutrients (e.g., NRFI, NuVal), therefore allowing positives to potentially compensate for negatives. A philosophy of systems that require the food meets thresholds for both positive and negative nutrients (e.g., Smart Choices, NEP) is that positive nutrients should not compensate for negative nutrients (Lupton et al., 2010; Nijman et al., 2007).

Table 2-2. FOP and Nutrient Profiling Systems and their Nutrients or Nutrition-Related Components
Name of System	Negative Nutrients	Positive Nutrients
Guiding Stars (GS)	Saturated fat Trans fat Cholesterol Added sugars Added sodium	Vitamins A, C Calcium Iron Zinc Fiber Whole grains
Heart Check Mark (AHA)	Total fat Saturated fat Trans fat Cholesterol Sodium	Vitamins A, C Calcium Iron Protein Fiber Whole grains Beta-glucan (whole oat soluble fiber)
Smart Choices Program (SC)	Total fat Saturated fat Trans fat Cholesterol Added sugars Sodium	Vitamins A, C, E Calcium Potassium Magnesium Fiber Whole grains Fruits Vegetables Nonfat/low-fat dairy
Nutrient Rich Foods Index (NRFI)	Saturated fat Added sugars Sodium	Protein Fiber Vitamins A, C, E Calcium Iron Magnesium Potassium
NuVal (NV)	Energy density Saturated fat Trans fat Cholesterol Total sugar Added sugar Sodium Glycemic load	Protein quality Fiber Vitamins A, C, D, E, B₆, B₁₂ Folate Calcium Iron Zinc Magnesium Potassium Fat quality Omega-3 fatty acids Total bioflavonoids Total carotenoids
Sensible Solution (SS)	Energy Total fat Saturated fat Trans fat Added sugars Sodium	Protein Vitamins A, C, E Calcium Iron Magnesium Potassium Fiber Whole grain
UK Ofcom Nutrient Profiling Model (Ofc)	Energy Saturated fat Total sugars Sodium	Protein Fiber "Fruit, vegetables, and nuts"
Choices Programme (CP)	Energy Saturated fat Trans fat Added sugars Sodium	Fiber
Keyhole symbol (K)	Total fat Saturated fat Total sugars Refined sugars Sodium	Fiber
Pick the Tick (Tick)	Energy density Total fat Saturated fat Trans fat Partially hydrogenated fat Sodium	Protein Calcium Fiber Whole grains Vegetables % of key ingredients (e.g., fish)
Heart Check (HC)	Total fat Saturated fat Trans fat Sodium	Protein Vitamins A, C Folate Calcium Iron Fiber
Nutrition Enhancement Program (NEP)	Saturated fat Trans fat Total sugars Added sugars Sodium	(None)
Netherlands Tripartite Classification System (NTri)	Saturated fat (include trans)	Vitamin C Fiber

Of the six U.S.-based systems reviewed, five stated that the 2005 DGA was the basis for developing nutrient criteria. The Heart Check Mark uses criteria in line with the FDA health claim regulations for coronary heart disease (CHD). As demonstrated in Table 2-3, many of the nutrients identified in the 2005 DGA as of concern for the population (as discussed previously) are commonly used by FOP systems in the United States and outside the United States. Outside the United States, criteria for systems are based on country-specific dietary recommendations or World Health Organization (WHO) guidelines for nutrition and chronic disease (WHO, 2003).

Table 2-3. Nutrients or Nutrition-Related Components and the Number of FOP Systems Including Them as Criteria
Nutrient or Component	Number	Systems
GS, Guiding Stars; AHA, Heart Check Mark; SC, Smart Choices Program; NRFI, Nutrient Rich Foods Index; NV, NuVal; SS, Sensible Solution; Ofc, UK Ofcom Nutrient Profiling Model; CP, Choices Programme; SK, Swedish Keyhole Symbol; Tick, Pick the Tick, HC, Heart Check; NEP, Nutrition Enhancement Program; NTri, Netherlands Tripartitite Classification System
*Negative*
Energy	5	Ofc, CP, NV, SS, Tick
Total fat	6	AHA, SS, SC, K, Tick, HC
Saturated fat	13	GS, AHA, Ofc, CP, NRFI, NV, SS, SC, K, Tick, HC, NEP, NTri
Trans fat	10	GS, AHA, CP, NV, SS, SC, Tick, HC, NEP, NTri
Partially hydrogenated fat	1	Tick
Cholesterol	4	GS, AHA, NV, SC
Total sugars	4	Ofc, NV, K, NEP
Added sugars	6	GS, CP, NRFI, SS, SC, NEP
Refined sugars	1	K
Sodium	11	AHA, Ofc, CP, NRFI, NV, SS, SC, K, Tick, NEP, HC
Added sodium	1	GS
Glycemic load	1	NV
*Positive*
Protein	6	AHA, Ofc, NRFI, SS, Tick, HC
Protein quality	1	NV
Vitamin A	7	GS, AHA, NRFI, NV, SS, SC, HC
Vitamin C	8	GS, AHA, NRFI, NV, SS, SC, HC, NTri
Vitamin D	1	NV
Vitamin E	4	NRFI, NV, SS, SC
Vitamin B6	1	NV
Folate	2	NV, HC
Vitamin B₁₂	1	NV
Calcium	8	GS, AHA, NRFI, NV, SS, SC, Tick, HC
Iron	6	GS, AHA, NRFI, NV, SS, HC
Zinc	2	GS, NV
Magnesium	4	NRFI, NV, SS, SC
Potassium	4	NRFI, NV, SS, SC
Fiber	12	GS, AHA, Ofc, CP, NRFI, NV, SS, SC, K, Tick, HC, NTri
Whole grains	5	GS, AHA, SS, SC, Tick
Beta-glucan	1	AHA
Fat quality	1	NV
Omega-3 fatty acids	2	GS, NV
DHA/EFA	1	GS
Total bioflavonoids	1	NV
Total carotenoids	1	NV
Nonfat/low-fat dairy	1	SC
Fruits	1	SC
Vegetables	2	SC, Tick
Fruit, vegetables, and nuts	1	Ofc
% of key ingredients (e.g., fish)	1	Tick

2.4.1 Negative Nutrients and Components

Most summary systems include negative nutrients, and the evidence for effects on health is more clearly known than for negative than for positive nutrients.

The 2010 DGA recommends that saturated fat intake be less than 10% of recommended energy intake as an interim goal, with an ultimate goal of <7%. The 2005 DGA had a goal of less than 10% of energy from saturated fat, which is 20 g based on a 2,000 kcal/day reference diet. Saturated fat is associated with increased low-density lipoprotein (LDL), and high serum LDL concentrations are associated with increased risk of CVD (Mensick, Zock, Kester, & Katan, 2003). Intake of trans fat should be "minimal" or "as low as possible" as recommended by the 2010 and 2005 DGA, respectively. Trans fat adversely lowers high density lipoprotein (HDL) cholesterol (Mensick et al., 2003) and has similar effects on LDL and CVD as saturated fats (Mozaffarian, Katan, Ascherio, Stampfer, & Willett, 2006). The 2010 and 2005 DGA recommend limiting cholesterol to 300 mg/d. Cholesterol is associated with increased risk of CVD, but changes in dietary cholesterol have only modest effects on plasma cholesterol concentrations (Connor & Connor, 2002). FDA has evaluated the scientific evidence for saturated fat and cholesterol and risk of CHD and issued health claim regulation regarding saturated fat and cholesterol and risk of CHD (Code of Federal Regulations, 1993b). Total fat includes both "bad" fats (saturated and trans fatty acids) and "good" fats, (polyunsaturated and monounsaturated fatty acids). The 2005 DGA recommends intake of total fat to be between 20% and 35% of calories for adults, but the 2010 DGA do not recommend a specific limit to total fat intake. Although the evidence for the effect of total fat on disease risk is not supported by science, a high intake of total fat generally increases saturated fat and energy intakes. The percentage of energy intake from fat is not associated with obesity (Willett, 2002). The 2010 DGA recommend replacing saturated fats with monounsaturated and polyunsaturated fatty acids in an effort to reduce intake of saturated fats to less than 10% of calories.

Sugar has long been known to cause dental caries, but recent data have implicated sugar, primarily from high intake of sweetened beverages, in obesity (Malik, Schulze, & Hu, 2006). Associations between sugar intake and obesity do not imply causality and are confounded by the fact that sugar is a source of energy. The 2010 DGA does not set a specific guideline for sugar intake; however, it states that Americans consume too many calories from solid fats and added sugars (SoFAS), and intake should be reduced to no more than 5 to 15% of total calories from SoFAS. The Institute of Medicine recommendation for added sugar intake is no more than 25% of total energy, or 125 g for a 2,000 kcal/d diet (Food and Nutrition Board. Institute of Medicine, 2002). The WHO recommends limiting daily intake of "free" sugar to no more than 10% of total energy (WHO, 2003).

Excess sodium intake is associated with elevated blood pressure (Obarzanek et al., 2003), and salt restriction lowers blood pressure (Law, Frost, & Wald, 1991). FDA reviewed the evidence and issued health claim regulation for sodium and hypertension (Code of Federal Regulations, 1993a). Elevated blood pressure is a risk factor for CHD and stroke. The 2005 and 2010 DGA recommend a sodium intake of 2,300 mg/d, but an ultimate goal of 1,500 mg/d is recommended by the 2010 DGA.

2.4.2 Positive Nutrients and Components

Most systems included positive nutrients, that is, the nutrients that should be encouraged because of health benefits and shortfalls in the average diet of the population. However, the evidence for effects on health is less clearly known for positive than for negative nutrients.

One exception is fiber, which has a role in body weight regulation (Howarth, Saltzman, & Roberts, 2001), helps prevent type 2 diabetes (Salmerón et al., 1997), reduces total and LDL cholesterol (Queenan et al., 2007), and is associated with reduced risk of CHD (Truswell, 2002). FDA has evaluated the scientific evidence for fiber, particularly soluble fiber, and risk of CHD and issued health claim regulation for soluble fiber and risk of CHD (Code of Federal Regulations, 2008 amended-b). In the United States, the recommended intake of fiber is 14 g per 1,000 kcal consumed (Food and Nutrition Board. Institute of Medicine, 2002).

Evidence for single vitamins or minerals in chronic disease risk is less well established. Calcium and vitamin D have a critical role in bone development and prevention of osteoporosis. FDA has issued food labeling regulation to allow a health claim for calcium, vitamin D, and osteoporosis (Code of Federal Regulations, 2008 amended-a). Folate prevents neural tube defects, and FDA has issued regulations for health claims (Code of Federal Regulations, 2000 amended) and fortification (Code of Federal Regulations, 1996). Contradictory studies have confused the scientific community and the public, for example, with regard to vitamin E in heart disease and beta-carotene in cancer prevention. It is not possible to include in this report a full review of the literature on the role of micronutrients and disease risk.

A large number of potential positive nutrients could be included in a summary system, and the content of these nutrients (e.g., vitamins and minerals) varies greatly in foods. Some foods are good sources of particular vitamins or minerals but are low in other nutrients. Most threshold systems that include criteria for positive nutrients do not require that the food be a good source of all of the positive nutrients specified, but they do require that one or two of the nutrients are provided in a specified amount (e.g., ≥10% of the daily recommended value). In comparison to icon or rating systems, summary score systems apply criteria for positive nutrients to all foods rather than only foods that meet thresholds. Therefore, summary systems can test the impact of including specific positive nutrients to determine the overall nutritional quality of foods. Summary scoring systems can also test how the ratings of the foods related to overall diet quality. Most systems cap positive nutrients at a certain level or include only the nutrients intrinsic to the food (not added via fortification). The reasons for capping positive nutrients include to prevent the score from being overly influenced by a large quantity of a single nutrient and to prevent over-fortification by food manufacturers to meet a system's criteria.

Many systems base their inclusion of positive nutrients on those that the population consumes in less than recommended amounts, as reported in the DGA report (U.S. Department of Health and Human Services & U.S. Department of Agriculture, 2005). Although these nutrients are essential for health and their consumption should be promoted, some may question whether an FOP system is the appropriate method to encourage consumption.

2.4.3 Practical Considerations Regarding Nutrients and Components

Some systems have used practical criteria to decide which nutrients to include. As noted previously, nutrients may be excluded because the nutrient information is not available on the NFP or in the nutrient database used to score foods. For example, trans fat was not included in the NRFI system because it was missing from the USDA nutrient database at the time of development. Publically available nutrient databases do not contain all foods available on the market. The NuVal system used in some supermarkets has gone to great lengths to create and continually update a database of branded food items and has scored more than 40,000 products (Katz, Njike, Rhee, Reingold, & Ayoob, 2010). For some systems, nutrients were excluded because the information does not appear on the NFP, making it difficult to assess the nutrient value for a specific product. There are various types of fiber and these are not separated out on the NFP. Another reason nutrients have been excluded is the lack of an analytical method to obtain the nutrient value of a food. For example, no analytical method exists to differentiate added sugars from the sugars intrinsic to the food. The NuVal system applies glycemic load as a proxy for carbohydrate quality in grain foods and foods with added sugar (Katz, 2007). However, values for glycemic load do not exist in food composition databases and are not readily available for all foods.

The consensus of the ILSI Europe workshop on nutrient profiling was that systems should include either disqualifying (negative) nutrients only or a combination of both disqualifying and qualifying (positive) nutrients (Tetens et al., 2007). There was no consensus on the approach for compensation of disqualifying nutrients with qualifying nutrients, although it was generally believed that qualifying nutrients should not compensate for disqualifying nutrients. The report focused on disqualifying nutrients and suggested that total fat, saturated fat, trans fat, sodium, and sugar should be considered. However, consensus could not be reached on sugar and total fat, and it was suggested that energy should be taken into account instead. The workshop consensus was that nutrient criteria should be based on actual nutrient recommendations rather than dietary guidelines. The report states that "[n]utrient recommendations are scientifically based and independent of dietary habits, variability in consumption, and availability of foods" (Tetens et al., 2007, p. 10).

2.4.4 Comparison of Specific Nutrient Criteria and Ranking of Foods in Summary Systems

Appendix A contains detailed information on the specific nutrient criteria or algorithms for the 13 summary systems that were reviewed to the extent the information was available from the published literature and the systems' Web sites. A comprehensive analysis would involve rating a large number of foods with each system and comparing the rankings across systems, both across foods and within categories. A comprehensive evaluation of food scores among various systems was beyond the scope of this project; however, some limited comparisons were made, and anomalies among systems are noted. Anomalies among the different systems are more likely to occur with foods that have a mixture of healthy and unhealthy attributes. For example, breakfast cereals may be high in fiber but also high in sugar; peanut butter is high in fat and calories, but the majority of the fat is monounsaturated.

Example of Breakfast Cereals

There are numerous breakfast cereals with varying ingredient and nutrient content on the market. The Smart Choices system received criticism for awarding its icon to sweetened cereals (Neuman, 2009). Smart Choices allows up to 12 g of added sugar per labeled serving (~30 g). However, Guiding Stars does not award stars to any food products with added sugar. Sensible Solution has a generous allotment of added sugars (25% of kcal) for cereals, and the Keyhole has no sugar limit for cereals, but criteria for fiber would preclude some sweetened cereals from obtaining ratings. The Choices Programme has stricter criteria for added sugars in breakfast cereal (<28 g/100 g or ~8 g per 30 g serving), precluding some sweetened cereals from obtaining its rating. The Choices Programme specifies further restrictions in the next 3 years (to 24 g/100 g) and 6 years (20 g/100 g). The staged restrictions are designed to allow manufacturers to gradually reformulate their cereals to decrease the sugar content.

Example of Peanut Butter

Peanut butter is difficult to characterize in terms of the overall nutritional quality because it is very high in fat, which makes it high in calories, and although some of the fat is saturated, most of it is monounsaturated. Based on nutrient values for generic peanut butter in the USDA food composition table, the total and saturated fat would disqualify peanut butter from the AHA Heart Check. Peanut butter would not qualify for the Choices Programme icon because it exceeds the saturated fat criterion of up to 13% of total calories (peanut butter is approximately 16%). However, peanut butter does qualify for the Smart Choices icon because the criterion for saturated fat is 28% of total fat (approximately 21% of the fat calories in peanut butter are from saturated fat). Peanut butter would classify as less healthy using the Ofcom model. With Guiding Stars, peanut butters vary from one to three stars presumably because of the saturated fat and added sugars in some brands.

2.4.5 Unit Basis of Nutrient Criteria for Summary Systems - per 100 Kcal, per 100 g, or per Serving

Nutrient criteria need to be based on a unit of measure. Most systems are calculated based on per 100 kcal, per 100 g, or per serving. U.S. food labeling regulations specify reference amounts customarily consumed (RACC), providing standard serving sizes that can be used for FOP systems. In the European Union, labeling information is provided per 100 g, and many of the FOP systems in Europe use this as their basis. Each method has advantages and disadvantages as described below.

The basis of per 100 kcal allows for comparisons with nutrient recommendations and guidelines, typically expressed per 2,000 kcal. However, foods with low energy content and high nutrient content will be scored disproportionately high (Drewnowski, Maillot, & Darmon, 2009). Energy density of foods is defined as energy or calories per unit of weight. Foods with high water content are often low in energy density because water contributes weight but no calories. Green leafy vegetables have low energy density because of their high water content and are typically consumed in much smaller portions than 100 kcal, but their high nutrient content can cause extremely high scores on a per 100 kcal basis. One cup of raw spinach provides about 7 calories, and a 100 kcal portion of spinach would be about 14 cups, much more than anyone would consume; therefore, a high food score on a per 100 kcal basis would be unrealistic. For example, the NRFI score for spinach is 694.8 per 100 kcal and 135.8 per RACC (see online supporting material in Fulgoni et al., 2009). Some low-fat spreads such as light mayonnaise may exceed sodium criteria on a per 100 kcal basis but would not on a typical portion size basis (personal communication, Lisa Sutherland).

Scoring foods on a per 100 g basis, which is consistent with nutrient labeling in European countries, allows for comparisons of foods in the same category that would be served in similar amounts. However, it is not very useful for across-the-board approaches because different foods are consumed in very different amounts than 100 g (Drewnowski et al., 2009). For example, typical servings of soup or beverages may be approximately 250 g, while a typical serving of oil or butter may be less than 10 g. The water content of foods greatly influences the nutrient content per unit by weight. Sweetened beverages would benefit from a system based on 100 g that includes a limit on sugar because the sugar content of 100 g may meet the criteria. A version of the Food Standard Agency's Ofcom model makes an allowance for this by scoring beverages on a 200 g basis (Scarborough, Boxer, Rayner, & Stockley, 2007).

Both per 100 kcal and per 100 g may be difficult for consumers to understand. Europeans are used to seeing label information on a per 100 g basis. Some people understand calories, but many do not know how 100 kcal relates to a typical portion of food. Calculation on a per-serving basis makes sense because it generally represents how people eat. However, this assumes consumption patterns are similar to standardized or reference portions.

To examine the performance of four versions of a nutrient-rich scoring system calculated on a per 100 g, per 100 kcal, and per-serving basis, Drewnowski et al. (2009) calculated scores using a food list of 378 foods from a food frequency questionnaire. A version of the algorithm that used only negative nutrients expressed on a per 100 g basis penalized energy-dense foods such as butter, oils, and cheeses because of the high amount of negative nutrients in these foods and no balance of positive nutrients. The algorithm expressed on per 100 kcal produced better scores than the per 100 g algorithm for some energy-dense foods that have an energy content greater than 100 kcal per 100 g (e.g., chips and chocolate bars). Algorithms that used positive nutrients only performed similarly when calculated on 100 kcal and on RACC. A version of the NRFI that contained a combination of negative and positive nutrients performed similarly when expressed per RACC and per 100 kcal in models where food scores from the index were regressed on an overall index of dietary quality, the Healthy Eating Index (HEI) (Fulgoni et al., 2009). However, models per 100 kcal explained slightly more of the variation in HEI scores.

2.5 Validation of Summary Systems to Rank Foods in Nutritional Quality

The optimal method of assessing validity of a summary system to profile the nutritional quality of foods would be to test for criterion validity, that is, to compare the new method with a gold standard. However, no gold standard for profiling the nutritional quality of foods exists; thus, researchers must opt for other methods of validation. Construct validity measures how well the measure relates to its theoretical concept, that is, how well the summary system relates to other measures of the nutritional quality of foods.

Approaches to construct validity of summary systems have focused on nutritional ratings of the foods themselves or on overall dietary quality. In construct validity of foods themselves, foods are rated using the summary system, and the results are compared with results from other measures of nutrition quality of foods, such as expert ratings or authoritative recommendations. In construct validity focused on overall diet quality, foods are rated using the summary system and aggregate results for diets are compared with other measures of overall diet quality, such as the HEI.

Convergent and discriminant validity are two types of construct validity that have been used in testing FOP systems:

Convergent validity tests that constructs that are expected to be related are, in fact, related.
Discriminant validity tests that constructs that should not be related are not related.

The application of convergent or discriminant validity tests to a summary system will differ, depending on whether the summary system is a scoring or a threshold system and whether the validation is focused on foods themselves or on overall diet quality.

Comparing food rankings by the nutrient profiling system with the food rankings by expert opinion is relatively easy and inexpensive to conduct, but the relative importance of this method to validate the accuracy of a nutrient profiling system was deemed to be low-medium (Townsend, 2010). The subjective nature of expert opinion is a disadvantage of using this method to validate systems. Rankings by professionals could be biased by the nutrient information provided to them, as well as by the food descriptions (Scarborough, Rayner, Stockley, & Black, 2007).

In one study, to assess the characteristics of expert food ratings, 850 nutrition professionals from the British Dietetic Association and the (British) Nutrition Society were asked to rank 120 foods in nutritional quality (Scarborough, Rayner et al., 2007). The experts were given the nutrient values for 10 nutrients for each food. The average rankings and standard deviations for each food were calculated and grouped into food categories based on the UK food guide "The Balance of Good Health" (BGH). The results were that the average expert rankings of foods were in accordance with the guidance in the food guide; that is, the highest average ranks were attained by foods in the fruit and vegetable group, and the lowest average ranks were attained by foods in the foods high in fat, foods high in sugar group. The composite foods showed the highest variance in ranks by nutrition professionals reflecting difficulty in categorizing these foods. Most of the variation in scores was explained by providing the nutritionists with nutrient values for fat, total sugars, sodium, and nonstarch polysaccharides. The nutritionists were influenced by food descriptions; for example, "wholemeal fruit crumble" was ranked slightly higher than "apple, stewed with sugar" despite the fact that the crumble contained more sugar, fat, and saturated fat per 100 g than the stewed fruit.

The scores of foods by these experts as described previously (Scarborough, Rayner et al., 2007) were also compared with the categorization of foods as "healthy" and "less healthy" by the Ofcom/WXYfm model, a threshold system (Scarborough, Boxer et al., 2007). Researchers found a strong relationship between quintiles of food scores by the experts and categorization of foods by the Ofcom/WXYfm model (χ² = 64.8). They also compared the rankings of foods by the experts and the model and found that the ranking of 11 out of 120 foods differed by 40 or more positions in rank.

Expert rankings of foods were used during the development of the Overall Nutrition Quality Index (ONQI), an algorithm that is the basis for the NuVal system, a scoring system. Members of the scientific expert panel were asked to rank approximately 1,000 foods and correlation analyses were used to compare the expert ranks to rankings produced by the ONQI algorithm (Katz et al., 2009). Any apparent anomalies were examined and the ONQI algorithm was adjusted as needed. The final version of the ONQI algorithm was highly correlated with the pooled expert panel ranking of 21 diverse foods (Spearman rank correlation coefficient 0.92, p < 0.001).

Arambepola et al. (2007) tested for convergent validity of the Ofcom/WXYfm model, a threshold system. The researchers used the model to categorize foods consumed by adults in the British National Diet and Nutrition Survey as healthy or less healthy and compared the categorized foods to food group recommendations in the BGH, the authoritative UK food guide. The model classified as healthy 97% of fruit and vegetables and 72% of bread, other cereals, and potatoes as classified by the BGH. In addition, 95% of fatty and sugary foods as classified by the BGH were classified as less healthy by the model. The к-value for the level of agreement between the model and BGH in categorizing these foods was 0.69, considered good agreement.

In the UK, indicator foods derived from a healthy eating index were used in another validation of the WXY model. First, indicator foods were identified using national dietary surveys by categorizing the population into quintiles of the healthy eating index and identifying foods that were eaten in statistically different amounts by individuals in the first and fifth quintiles of healthy eating index scores (Volatier et al., 2007). Then, the validation tested the ability of the Ofcom/WXY model, a threshold system, to correctly classify the indicator foods (Quinio et al., 2007). The WXY model identified 73.7% of indicator foods as healthy that were classified as healthy by the reference index.

Some FOP or nutrient profiling systems have tested convergent validity to examine the relation between the healthiness of diets measured by algorithm scores of the foods consumed and the healthiness of diets measured by an overall diet quality score (Fulgoni et al., 2009; Katz et al., 2010).

Fulgoni et al. (2009) assessed convergent validity of a nutrient density index, a scoring system, by calculating the mean nutrient density scores of foods consumed by participants in the National Health and Nutrition Examination Survey (NHANES) 1999-2002. The calculated scores were regressed against HEI scores, an indicator of diet quality, in the NHANES sample. For a nutrient density index that included nine positive nutrients and three negative nutrients to avoid, the regression model explained 45.3% of the variation in HEI scores in the NHANES sample (Fulgoni et al., 2009). This validation also examined scores of foods themselves within food categories. Whole grain products scored higher than nonwhole grain foods, fruits with less added sugar scored higher than those with more added sugar, and 100% fruit products scored higher than soft drink choices. This agreement between food scores and dietary recommendations indicated that an across-the-board index can be useful for ranking foods within food categories.

Katz et al. (2010) also validated a scoring system, the ONQI algorithm, by calculating mean scores of foods consumed in a national survey, using dietary intake data from NHANES 2003-2006. The calculated ONQI scores were regressed against HEI scores in models adjusted for age, race, and gender. The beta coefficients for ONQI scores to predict the HEI scores were significantly different from zero (p < 0.001), and the model explained 29% of the variation in HEI scores in the NHANES sample (Katz et al., 2010).

Relatively few researchers have used discriminant validity with respect to FOP systems or nutrient profiling. The Ofcom/WXY model was used to define healthy and less healthy foods consumed by adults in the National Diet and Nutrition Survey (Arambepola et al., 2007). An overall score of dietary quality was calculated using the Diet Quality Index (DQI). The validation compared the energy intake from "less healthy" foods, as defined by the Ofcom/WXY, among two groups of adults in the survey - those with the least healthy and most healthy diets according to the DQI. As predicted, the group with the least healthy diet by DQI had higher energy intake (by about a factor of two) from less healthy foods as defined by the Ofcom/WXY model, compared with those with the healthiest diets according to the DQI.

For the ONQI algorithm, mean scores of foods in a 7-day meal plan from the Dietary Approaches to Stop Hypertension (DASH) diet, representing a "healthy diet," were compared with the mean scores of foods in the typical American diet using data from NHANES 2003-2006 (Katz et al., 2009; Katz et al., 2010). The aggregate ONQI score for the 7-day DASH meal plan at the 2,300 mg sodium level was 46 (95% CI, 40-53) and for the typical American diet in the NHANES cohort was 26.5 (95% CI, 26.2-26.7). As expected, the aggregate score for the healthy diet (DASH) was significantly higher than the score for the typical diet (NHANES) (p < 0.05).

The ONQI algorithm was also validated with chronic disease outcomes, including CVD, cancer, and diabetes, by scoring diets from over 100,000 participants in Harvard's Nurses' Health Study and the Health Professionals Follow-Up Study (Chiuve, Sampson, & Willett, 2011). This validation approach could be considered a type of criterion validity as defined by Townsend (2010). Dietary data were collected from food frequency questionnaires that were administered to subjects at baseline. Each food was scored by the ONQI algorithm, and the average ONQI score for the diet consumed by each participant was computed. The ONQI score was inversely associated with risk of total chronic disease, CVD, diabetes, and all-cause mortality, but not cancer, in both cohorts. The multivariate relative risk of chronic disease, comparing the highest to lowest quintile of ONQI scores, was 0.91 (95% CI: 0.87-0.95) in women and 0.88 (95% CI: 0.83-0.93) in men. A limitation of the study is that the diet information is for only one point in time.

2.6 Strengths and Limitations of Summary Systems

2.6.1 Transparency

A limitation of summary systems is that the nutrient criteria are not transparent to the consumer at the point of purchase. The nutrient criteria may or may not be available to the consumer in printed information, the Internet, or scientific literature. Most systems reviewed were transparent with respect to providing nutrient criteria from published literature or Web sites, with the exception of Guiding Stars, NuVal, and Pick the Tick, but were not transparent to the consumer at the point of purchase.

2.6.2 Food Categories

A strength of summary systems that rate all or most foods (e.g., NuVal, NRFI) is that they allow the consumer to use the system to compare products across foods, while also allowing for comparisons within food categories. However, they are limited by the fact that they do not necessarily inform the consumer what the actual difference means in terms of magnitude of a difference to affect the diet, and these systems do not tell the consumer what nutrient(s) account for the differences. Although across-food category comparisons make sense for some foods, for example, if a consumer is choosing a dessert and comparing fruit to a fruit crisp, it would not make practical sense to compare fruit with chicken.

2.6.3 Nutrients

Another limitation of summary score systems is that positive nutrients could outweigh or compensate for negative nutrients. Many summary systems place a cap on the amount of positive nutrients that factor into a food score to prevent excessive influence of positive nutrients. A strength of summary systems is that they do account for positive attributes to foods, as opposed to nutrient-specific systems that usually only include negative nutrients. A limitation of summary systems is that they may not be as useful as nutrient-specific systems for consumers interested in specific nutrients; however, that information is on the NFP. On the other hand, although the summary score or rating is based on the nutrient content of foods, summary systems may be beneficial for consumers who do not think about food in terms of specific nutrients.

2.6.4 Discrimination between Foods

Threshold systems do allow for discrimination between foods with different nutrient content. Multiple thresholds can be used; for example, Guiding Stars awards one, two, or three stars based on ranges of nutrient criteria. However, the range between threshold values may be too large for fine distinctions. For example, if a rating allows for fat to be between 5 and 20 g, and two foods are being compared that contain 6 and 18 g of fat each, the system will rate the two foods equally. Systems that provide an overall score to foods are sensitive to small differences between foods, although the reason for the difference in scores is not apparent to the consumer.

2.6.5 Simplicity

A strength of summary systems, both scoring and threshold-type systems, is that they are simple and provide potential to communicate the overall nutritional value of a food to people with varying levels of education and literacy.

2.7 Criteria and Considerations for Developing Summary Systems

This section describes the following considerations for developing summary FOP nutrition labeling systems:

purpose and specific goals of the system
nutrients to include in the system
unit basis for measuring nutrients
type of scoring system
weighting of nutrients

2.7.1 Purpose and Specific Goals of Summary System

The purpose of the summary system relates to the health outcome that is to be addressed with the system. The purpose of this project was to develop options for summary FOP algorithms that address chronic disease and obesity and to provide accurate and useful information to meet the Dietary Guidelines. We do not consider nutrient-specific FOP systems for this project. The DGA recommend avoiding excess intake of components that are associated with chronic disease and obesity and encourage components that are associated with health and that the population consumes in lower than optimal amounts.

The purpose of the algorithm affects the choice of nutrients or food components. For example, obesity is caused by excess energy intake. Although excess intakes of fat and added sugars result in increased energy intakes, the inclusion of these components in an algorithm is debated. The scientific evidence for chronic disease risk is strongest for "negative" nutrients: saturated fat, trans fat, and sodium. Foods associated with these nutrients are foods that are highly processed and have a high energy density (kcal per gram), except for a small number of energy-dense foods with unsaturated fats (nuts, oils). Nutrients to encourage whose consumption is low according to the 2010 DGA include fiber, vitamin D, calcium, and potassium. Shortfall nutrients from the 2005 DGA also included magnesium and vitamins A, C, and E, and for specific population groups nutrients of concern - vitamin B₁₂, folic acid, and iron. Food-based guidelines are usually based on meeting nutrient requirements. Foods that should be encouraged include fruits, vegetables, whole grains, low-fat milk and milk products, and oils (to replace solid fats, not to increase calories). An algorithm could incorporate the amounts of certain food groups or components themselves instead of nutrients.

Potential goals of summary FOP systems are the following:

To correspond to the NFP: If the goal is for the FOP label to correspond to the NFP, the system could be limited to the nutrients listed on the NFP or potentially others of interest. Some systems that restricted nutrients or food components to those listed on the NFP have excluded added sugars because they are not listed on the NFP. Some systems use the nutrient values on the NFP to score foods or set criteria based on the label serving sizes.
To warn consumers of "bad" attributes of foods: An FOP system designed to warn consumers of "bad" attributes of foods would best be served with negative nutrients. Considering summary systems, a threshold-type system (check mark or stars) might be better than an overall scoring system because the icon signifies whether the food meets specified levels of all of the negative nutrients in a simple format.
To represent the overall nutritional profile of a food: If the goal is to represent the overall nutritional profile of a food, then a combination of negative and positive nutrients is probably most appropriate. In terms of a summary system, an overall score or a threshold system works. An overall score communicates differences between products in a more refined manner than a threshold system; that is, overall numerical scores of 30 and 50 signify a difference between two products, whereas with a threshold system, those same two foods may both meet the criteria for a check mark, and the consumer would not know if one is slightly more nutritious than the other. However, thresholds with transparent criteria may inform the consumer that the product meets specified levels of certain nutrients, whereas the overall score does not communicate to the consumer what nutrients contribute to a difference in the scores.
To help consumers choose foods within or across food categories: An FOP system can be designed to help consumers choose foods within categories (analogous to comparing foods in the same supermarket aisle) or choose foods across categories (analogous to comparing foods across the supermarket). An overall numerical score is designed for both types of comparisons. In contrast, a threshold system (e.g., check mark or star) is not as refined for either type of comparison. A wide range of nutrient levels may be allowed within the rating for a check mark; therefore, this type of system would have less discriminating power. For example, within the category of fruits and vegetables, all would get the check mark, but an orange has more nutrients than a banana (although a multiple rating system like one to three stars would be more helpful), or across food categories, an orange and a low-fat cookie may meet the thresholds. Threshold systems usually set different criteria for different food categories, which allow for within-category comparisons but do not allow for across-category comparisons, although a graded threshold system could serve this purpose (e.g., one to three stars).

2.7.2 Nutrients to Include in the System

The first decision regarding nutrients is whether to include negative only, positive only, or both negative and positive nutrients. The purpose and specific goals of the system may influence this decision, as discussed previously. The decision of which nutrients to include should be based on science and dietary recommendations. The scientific evidence for nutrients was described in Sections 2.4.1 and 2.4.2. Another approach to be considered is including food components or groupings, such as the percentage of whole grains or fruits and vegetables in the food.

Some decisions are needed with regard to the representation of the nutrient. For example, saturated fat may be represented in an absolute amount (grams), as a percentage of energy, or as a percentage of fat. Saturated fat as a percentage of total fat could be useful as an indicator of fat quality, in particular, for foods high in unsaturated fats such as nuts. Trans fat has most often been represented in absolute amounts, but some systems have used percentage of energy. Some systems have proposed combining saturated and trans fat. An alternative could be to use the amount of solid fats, a component in the HEI.

FOP systems may use specific foods or food groupings instead of or in addition to nutrients. For example, a system may include nuts as an indicator of a "good" unsaturated fat. One system that we reviewed, the FSA Ofcom model, uses the percentage of nuts, fruits, and vegetables as a positive component in a score (Scarborough, Boxer et al., 2007). This particular system did not include the positive nutrients that are in these foods as criteria (unsaturated fat or vitamins) but did include other positive nutrients such as fiber and protein.

Collinearity of nutrients and foods could be a consideration in the decision. Many nutrients are correlated, and inclusion of both correlated nutrients may not add predictability of the algorithm score to ranking of an overall index of dietary quality. For example, total fat and saturated fat are correlated, so including both in an algorithm may not improve the ability of the score to explain diet quality. Likewise, vitamin D is found in calcium-rich foods; thus, including both may not be necessary. This suggests use of a statistical approach to select nutrients. The scientific approach to selecting nutrients is critical; however, the science is less clear on which "positive" nutrients are most important to include in a nutritional profile of foods compared with which "negative" nutrients to include.

2.7.3 Unit Basis for Measurement of Nutrients

Nutrient criteria are based on either a weight (per 100 g or serving) or energy (per 100 kcal). Some issues exist with each type of unit basis. Since food is consumed in servings, not on a 100 g or kcal basis, the serving size basis seems logical. At least one system found that a per 100 kcal basis performed slightly better in predicting overall diet quality than a serving size basis; however, anomalies do occur with very low-energy dense foods consumed in smaller portions than 100 kcal, such as green leafy vegetables.

2.7.4 Type of Scoring System

Systems that produce a numeric score using negative and positive nutrients are calculated either by sums (subtract the negative subscore from the positive subscore or vice versa) or as a ratio of the two subscores. Ratios can be greatly affected by small differences and, when tested against sums, did not correlate as well with an overall index of diet quality (Fulgoni et al., 2009); thus, ratios are not recommended.

Whether to score foods across the board or by food categories is also a consideration. As discussed in Section 2.2, across-the-board scoring systems can also be used to rank foods within food groups, but category-specific criteria incorporate intrinsic differences in foods so that certain nutritional qualities of specific foods are accounted for (e.g., vitamin B₁₂ only in animal foods).

2.7.5 Weighting of Nutrients

Most existing algorithms have applied the same weight to all nutrients in the scoring. Few systems have incorporated weighting, probably because the evidence base for these decisions has not been strong. However, the NuVal system applies proprietary weighting to nutrients based on the prevalence of health conditions associated with specific nutrients, the severity of the condition, and the relative strength of the association (Katz, 2007). Prevalence measures were derived from NHANES. The severity and relative strength determinations were made based on epidemiological evidence and interpretation of that evidence by consensus of the expert panel that designed the algorithm. A simpler approach could be used, such as applying a greater weight to the negative nutrients than the positive nutrients. This approach was used in the design of the HEI, whereby calories from solid fats, added sugars, and alcohol were weighted twice as heavily as other components, and saturated fat was also included as a separate component (Guenther, Reedy, Krebs-Smith, Reeve, & Basiotis, 2007). The determination of weighting factors using other methods is an important topic for future research.

3. Consideration of Options for Developing and Testing Algorithms

In the past, most FOP systems and algorithms have been developed by an expert panel that comes to consensus over time on the nutrients to include, the nutrient criteria, and the ability of the system to rank foods. Based on our review of existing systems and assessment of their strengths and limitations, we identified three options that could be developed or modified, and we selected the options for testing in consultation with ASPE and FDA. Examples of the application of each option to a food item are provided in Appendix B.

3.1 Proposed Options

3.1.1 Option 1 - Overall Score Based on Nutrient Density

Nutrient density is typically defined as the nutrient amount per kcal (or some multiple of kcal, for example, 100 kcal). A nutrient density score is a summation of nutrient values per kcal amount in relation to the daily recommended intake value of each nutrient. Some have suggested expressing nutrient density-based algorithms on a weight basis (i.e., per 100 gram or per serving size) as an alternative to per kcal amount (Drewnowski, Maillot, & Darmon, 2009). Various nutrients can be included that characterize the overall nutritional quality of a food. An advantage of this type of system over a threshold-type system is that it does not assign arbitrary thresholds to determine a score.

Using an overall score of nutrient density, potential modifications to test included the following:

unit basis (e.g., per 100 kcal or per RACC serving)
nutrients (e.g., include a fat-quality indicator such as saturated fat as a percentage of total fat, with or without added sugars)
weighting (e.g., assigning more weight to negative nutrients as positive nutrients)
bioavailability (e.g., protein quality score, adjustment for nonheme iron that is less well absorbed)
overall score or food category-specific (e.g., categorize final score into two [healthy or not] or three [low, medium, high] categories using one set of cutoffs or food category-specific cutoffs)

3.1.2 Option 2 - Overall Score Using Thresholds to Assign Points

Another option for a scoring system is based on thresholds (e.g., FSA XYW model/Ofcom). This type of system uses a cutoff or threshold nutrient level to assign points for negative and positive attributes, subtracts the positives from the negatives to get a continuous score, and then categorizes or dichotomizes the score. For the XYW model/Ofcom, there are a series of 11 thresholds (0 to 10 points) for the negative nutrients that include energy, saturated fat, total sugar, and sodium, and there are 6 thresholds for positive nutrients/components that include the percentage of fruits, vegetables and nuts, fiber, and protein. This algorithm is unique in that it contains food groups in addition to nutrients as a positive component. However, the selection of thresholds to assign points is not necessarily clearly justified.

Using an overall score employing thresholds to assign points, potential modifications to test include the following:

unit basis (e.g., per 100 kcal or per RACC serving instead of per 100 g)
nutrients (e.g., include a fat-quality indicator such as saturated fat as a percentage of total fat; replace total sugar with added sugar in the model; remove energy from the model; or include vitamins or minerals such as vitamin D, iron, potassium, or calcium)
food groups (e.g., with or without percentage of fruits, vegetables, and nuts; include low-fat dairy)
weighting (e.g., assigning more weight to negative nutrients as positive nutrients)
conversion of overall score to categorical rankings (e.g., categorize final score into two [healthy or not] or three [low, medium, high] categories using one set of cutoffs or food category-specific cutoffs)

3.1.3 Option 3 - Threshold System with Food Categories

Any threshold system could be modified using different nutrients, thresholds, or food categories. Many threshold systems use the FDA criteria for a "healthy" claim and these criteria could be a starting point for developing a new threshold system. Most threshold systems assign food category-specific criteria, which adds complexity to the system's design and there is no solid evidence base for these criteria.

Using a threshold system with categories, potential concepts to test include the following:

nutrients (e.g., modifications to the FDA "healthy" claim nutrients - exclude total fat and cholesterol; include a fat-quality indicator; with or without added sugars; include additional nutrients of interest)
positive vs. negative nutrients (e.g., include both or only negative nutrients)
weighting (e.g., assigning more weight to negative nutrients than positive nutrients)
use of food category-specific nutrient criteria instead of across-the-board criteria (one set of criteria)

3.2 Decision on Algorithm to Develop and Test

The ASPE/FDA team selected Option 1, the nutrient density-based scoring system to test for this project. The modifications they were most interested in testing were

unit basis,
nutrients such as fat quality and added sugar, and
categorization of scores.

RTI devised a detailed plan describing the algorithm, modifications, and methods of testing. The baseline algorithm for a nutrient density-based score (NDS) is depicted in Figure 3-1.

Figure 3-1. Algorithm for a Nutrient Density-Based Score (NDS) on a per 100 Kcal Basis

NDS1KCAL^a =
(Protein g per kcal/50 g + Fiber g per kcal/25 g + Vitamin E IU per kcal/30 IU + Vitamin D IU per kcal/400 IU + Calcium mg per kcal/1,000 mg + Iron mg per kcal/18 mg + Potassium mg per kcal/3,500 mg + Unsaturated fat g per kcal/44 g) * 100/8
− (Saturated fat g per kcal/20 g + Added sugars g per kcal/50 g + Sodium mg per kcal/2,400 mg) * 100/3

^a NDS1KCAL is an abbreviation for the baseline algorithm per 100 kcal. Positive nutrients were capped at 100% of the recommended intake.

We selected the nutrients to be included in the baseline algorithm for these reasons:

positive nutrients of importance that were identified as consumed in less than optimal amounts and are of concern for the entire U.S. population as specified in the 2010 DGA (vitamin D, calcium, potassium, and fiber);
some positive nutrients that were previously identified as low in the American diet in the 2005 DGA report (iron and vitamin E);
some negative nutrients that are overconsumed as defined by both 2005 and 2010 DGA (saturated fat and sodium);
added sugars because of their contribution to excess energy intakes;
protein because of its dietary importance; and
unsaturated fat because of the recognition of its importance and recommendation in the 2010 DGA to replace saturated fatty acids with polyunsaturated and monounsaturated fatty acids.

We identified additional positive nutrients that were identified as nutrients of concern in the American diet in the 2005 DGA to be tested in modified versions of the algorithm (as described further in Step 5 below).

The unit basis for the baseline algorithm is per 100 kcal. The basis for the denominators for most of the nutrients is the daily value (DV) or daily reference value (DRV) (Code of Federal Regulations, 2010a). For unsaturated fat, the denominator was derived from the total fat DRV of ~30% of total kcal minus the saturated fat DRV of ~10% total kcal; therefore, 20% of total energy from unsaturated fat or 44 g was used as the basis for unsaturated fat. There is no daily value for added sugars; we used as a basis the recommendation 10% of total energy (WHO, 2003). Positive nutrients were capped at 100% of DV to avoid undue influence by one nutrient and to discourage fortification to yield high scores. The total contribution from the positive and negative nutrients is equalized by dividing the total positive portion of the score by the number of positive nutrients (eight) and the total negative portion of the score by the number of negative nutrients (three).

We followed a step-wise approach to testing modifications to the base algorithm. Details of the methodology used to test the algorithm are presented in Section 4.

Step 1: Test the baseline algorithm. Baseline algorithm scores were calculated for a set of foods based on nutrient values per 100 kcal (500 foods most commonly consumed from the NHANES survey). The rankings of foods were examined across and within food groupings. Major food groupings will be selected based on the USDA food coding scheme (for example, dairy, meat/poultry/fish, legumes, grains, fruits, vegetables, fats/oils, and sweets). We looked for discriminations between foods to determine if expected rankings are evident. Finally, we determined the amount of variation explained by the algorithm in accounting for HEI scores. Details of the methods are presented in Section 4.

Step 2: Test the unit basis. The unit basis for the amount of nutrient components was calculated "per RACC" serving, and Step 1 was repeated.

Step 3: Test a "fat quality" component. We removed the unsaturated fat component of the algorithm and repeated Step 1 to determine if inclusion of a "good" fat improved discrimination of food rankings and variation explained in HEI scores.

Step 4: Test an added sugar component. We removed the added sugar component of the algorithm and repeated Step 1 to determine if it improved discrimination of food rankings and variation explained in HEI scores.

Step 5: Test other positive nutrients. We added other positive nutrients individually into the algorithm. We proposed nutrients that have been previously identified as being important nutrients to encourage in the diet of Americans: magnesium, vitamin A, vitamin C, folate, and vitamin B₁₂.

Step 6: Test the concept of whole grains. We added the percentage of grains that are whole grains to the positive nutrients in the base algorithm. The data for this component were derived from the MyPyramid database by dividing the ounce-equivalents of whole grains by the ounce-equivalents of total grains. The recommendation that at least 50% of grains should be consumed as whole grains was used as the "daily value."

Step 7: Decide on final algorithm based on modification results. We determined which nutrient components to keep in a final algorithm. We planned to base this on the results of testing in the previous steps; however, we developed a new statistical approach to determine the final algorithm. The details of the new statistical approach and the resulting final algorithm are described in detail in Section 4.4. We tested the unit basis again to verify the final algorithm.

Step 8: Convert continuous scores to categorical rankings. Using the final algorithm, we examined tertile and quintile rankings and categorization of foods across and within food categories. Such a categorization could be considered, for example, for an FOP system with an overall "low, medium, high" rating scheme.

Step 9: Conduct additional validations with subpopulations. We conducted additional validations of the algorithm and HEI scores among different subgroups of the population (e.g., age, ethnicity, or health conditions).

Results of testing are described in Section 4.

4. Results and Discussion of Algorithm Testing

This section presents the methodology, results, and discussion of the testing of the nutrient density scoring algorithm described in Section 3.2.

4.1 Methods for Developing Nutrient Density-Based Scores

4.1.1 Calculation of NDS Scores for Food Items

Scores were calculated for all food items reportedly consumed by participants in NHANES 2005-2008. The first dietary interview day was used. Individuals were excluded if they were under 2 years of age, pregnant, or lactating or had unreliable records. Dietary intake records from 16,587 individuals were included.

Nutrient values corresponding to foods consumed in NHANES were from the USDA Food and Nutrient Database for Dietary Studies (FNDDS) Version 3.0 (for NHANES 2005-2006) and Version 4.0 (for NHANES 2007-2008) (U.S. Department of Agriculture, 2008; U.S. Department of Agriculture, 2010b). Adjustments to the available nutrient values were made. Vitamin D values were not available in FNDDS 3.0 and thus were calculated using data from the USDA Standard Reference, Release 22 database (U.S. Department of Agriculture, 2009) and FNDDS 4.0. The amount of the food components added sugar and whole grains were obtained from the MyPyramid Equivalents Database, 2.0 for USDA Survey Foods, 2003-2004 (Bowman, Friday, & Moshfegh, 2008), and values for new foods in 2005-2008 were determined by Nutrition Impact, LCC, primarily by matching to similar foods. Added sugar is provided in teaspoon equivalent units and was converted to grams using a factor of 4.2 grams per teaspoon.

Standard serving sizes or RACCs were assigned to all foods using FDA guidance in the CFR (Code of Federal Regulations, 2010b). Reference amounts in units other than grams (cup, teaspoon, piece, etc.) were converted to grams using conversions in FNDDS 4.0 or earlier versions. Some judgments had to be made for foods, such as frozen meals or mixed dishes that are specified in the regulation as 1 cup if measurable with a cup or 140 g if not measurable with a cup. The data file provided to ASPE includes RACC serving sizes.

The baseline algorithm (Figure 3-1) calculates a score based on the nutrient value of foods per 100 kcal divided by the daily recommended value of the nutrient, summing these values for all of the positive nutrients and dividing by the number of positive nutrients (eight), and subtracting the sum of the negative nutrients divided by the number of negative nutrients (three). The equation for the baseline algorithm as depicted in Figure 3-1 is [(Protein g per kcal/50 g + Fiber g per kcal/25 g + Vitamin E IU per kcal/30 IU + Vitamin D IU per kcal/400 IU + Calcium mg per kcal/1,000 mg + Iron mg per kcal/18 mg + Potassium mg per kcal/3,500 mg + Unsaturated fat g per kcal/44 g) * 100/8] − [(Saturated fat g per kcal/20 g + Added sugars g per kcal/50 g + Sodium mg per kcal/2,400 mg) * 100/3]. The rationale for nutrient selection was discussed in Section 3.2.

4.1.2 Modifications to the Baseline Algorithm

We tested a series of modifications to the baseline algorithm that included (1) calculating the score using nutrient values on a per RACC basis instead of per 100 kcal basis, (2) removing unsaturated fat from the algorithm, (3) removing added sugars, (4) adding individual positive nutrients (magnesium, vitamin A, vitamin C, folate, and vitamin B₁₂), and (5) adding the percentage of total grains that are whole grains. After examining the results of the modifications, we tested an additional modified algorithm with vitamin C and whole grains. As the algorithms were modified to include or exclude nutrients/components, the divisor was changed to adjust to the new numbers of positive or negative nutrients. We calculated all modified algorithms on a per 100 kcal and RACC basis. The modified algorithms or NDSs are presented in Figure 4-1.

Figure 4-1. Baseline and Modified Nutrient Density-Based Algorithms Tested

NDS1	Baseline with unit basis test	(Protein g per unit^a/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44) * 100/8 − (Saturated fat g per unit/20 + Added sugars g per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS2	Remove unsaturated fat	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500) * 100/7 − (Saturated fat g per unit/20 + Added sugars g per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS3	Remove added sugars	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44) * 100/8 − (Saturated fat g per unit/20 + Sodium mg per unit/2,400) * 100/2
NDS4M	Add magnesium	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44 + Magnesium mg per unit/400) * 100/9 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS4A	Add vitamin A	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44 + Vitamin A RAE mcg per unit/1,500) * 100/9 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS4C	Add vitamin C	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44 + Vitamin C mg per unit/60) * 100/9 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS4F	Add folate	(Protein g per unit /50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3500 + Unsaturated fat g per unit/44 + Folate mcg per unit/400) * 100/9 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS4B₁₂	Add vitamin B₁₂	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3500 + Unsaturated fat g per unit/44 + Vitamin B₁₂ mcg per unit/6) * 100/9 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2,400) * 100/3
NDS5	Add whole grains	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44 + [Whole grains g per unit/Total grains g per unit/0.5]) * 100/9 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2400) * 100/3
NDS6	Add vitamin C and whole grains	(Protein g per unit/50 + Fiber g per unit/25 + Vitamin E IU per unit/30 + Vitamin D IU per unit/400 + Calcium mg per unit/1,000 + Iron mg per unit/18 + Potassium mg per unit/3,500 + Unsaturated fat g per unit/44 + Vitamin C mg per unit/60 + [Whole grains g per unit/Total grains g per unit/0.5]) * 100/10 − (Saturated fat g per unit/20 + Added sugars mg per unit/50 + Sodium mg per unit/2,400) * 100/3

^a All algorithm modifications were tested with both unit bases - per 100 kcal and per RACC. Positive nutrients were capped at 100% of recommended intake.

4.1.3 Scoring and Ranking of Foods

To examine the scores and rankings of foods across and within food groupings, we initially proposed to select 300 of the most commonly consumed foods from the most recent NHANES survey. After discussion with the ASPE/FDA team, we increased the number of foods to 500 to ensure more foods from all major food categories were represented. In addition, ASPE/FDA requested several other foods, such that the total number of foods was 570.¹ For each algorithm, we calculated an overall mean score across the 570 foods and mean scores for each of nine major USDA food groupings: (1) dairy; (2) meat, poultry, and fish (MPF); (3) eggs; (4) legume and nuts; (5) grain products; (6) fruits; (7) vegetables; (8) fats, oils, and dressings; and (9) sweets and beverages. Some groups were broken down further into subgroupings to allow for more detailed comparisons of similar foods within a grouping. For example, the grain products group includes a variety of types of foods such as breads, quickbreads, cakes, cereals, salty snacks, and mixed foods, all having a grain as a major component. The scores of these foods were provided to ASPE in separate Microsoft Excel workbooks for each algorithm. In this report, summary figures (box plots) provide the mean scores and distribution of scores for food groups for selected algorithms. Box plots for additional algorithm modifications are found in Appendix F.

We examined the rankings and distributions of scores for foods and food groups, as discussed in Section 4.2, but we did not conduct formal validity tests focused on the food scores themselves.

4.1.4 Regression Models

To determine the ability of algorithm scores to predict overall dietary quality, we calculated a composite algorithm score for each of the 16,587 individuals based on foods consumed in NHANES 2005-2008 based on 1-day intakes for each individual. To calculate the composite algorithm score for each individual, we summed the algorithm scores for each food consumed on the survey day and divided by the number of 100 kcal units or RACC servings in the daily intake. A detailed example of the calculated composite algorithm score for an individual is provided in Appendix C. HEI scores, which measure diet quality based on the 2005 DGA, were calculated using methods described by Guenther et al. (2007). Details of the HEI calculation are presented in Appendix D. Composite algorithm scores were then regressed against HEI scores. The beta-coefficient for the algorithm score and R² were generated using SUDAAN (Version 10) procedure Proc Regress, which accounts for the complex sampling design of NHANES by including information supplied in the NHANES data set: day 1 dietary weights, strata, and primary sampling units (PSU). Covariates included age (as continuous variable), gender, and ethnicity (as five categories: Mexican American, other Hispanic, non-Hispanic white, non-Hispanic black, and other race). R² values of the models (e.g., the coefficient of determination that measures the proportion of variation explained by the model) were examined. Because of the large sample size in NHANES 2005-2008, we expect that small differences in R² between two models could be statistically different. It was not feasible to statistically test the difference in R² between the various models; however, based on previous experience of Nutrition Impact, LCC with testing different algorithms for NRFI, we consider a 5% unit difference in R² to be meaningful.

4.2 Algorithm Performance in Ranking the Nutritional Quality of Foods

The rankings of foods using the algorithm are presented for each of the modifications as follows:

baseline algorithm with unit basis modification (NDS1): Section 4.2.1
modification removing unsaturated fat (NDS2): Section 4.2.2
modification removing added sugars (NDS3): Section 4.2.3
modification with vitamins (only vitamin C presented, NDS4C): Section 4.2.4
modification with whole grains (NDS5): Section 4.2.5
modification with vitamin C and whole grains (NDS6): Section 4.2.6

4.2.1 Ranking of Foods and Food Groups for the Baseline Algorithm, Comparing the Unit Basis (NDS1)

The mean algorithm scores and distributions of scores of major food groupings using the baseline algorithm are presented in Figures 4-2 and 4-3 on a per 100 kcal basis (NDS1KCAL) and a per RACC basis (NDS1ACC), respectively. Positive scores mean that positive nutrients outweigh negative nutrients included in the algorithm. Likewise, negative scores mean that negative nutrients outweigh positive nutrients. The range of scores is much larger for NDS1KCAL than for NDS1RACC. The mean score of all 570 foods on a per RACC basis was lower than the mean score per 100 kcal.

Figure 4-2. Box Plot of Nutrient Density Scores per 100 Kcal for Baseline Algorithm (NDS1KCAL)

Box plot shows the mean and distribution of nutrient density scores for all 570 foods and for major food groups. The mean is shown by the diamond; the left side of the boxes represents the 25th percentile of scores, and the right side of the boxes represents the 75th percentile. The minimum and maximum scores are represented at the ends of the horizontal lines.

Figure 4-3. Box Plot of Nutrient Density Scores per RACC for Baseline Algorithm (NDS1RACC)

Fruits had the highest mean score among the food groups on a per 100 kcal basis and legumes had the highest mean score on a per RACC basis. Vegetables and eggs had the third and fourth highest mean scores. The mean score of grains was fifth for NDS1KCAL and eighth for NDS1RACC. The grain group contained cookies, cakes, and pastries, which have RACC servings greater than 100 kcal, and the high fat and sugar content resulted in lower scores for these foods. Fats, oils, and dressings scored higher on a per RACC basis because their portion sizes are smaller than 100 kcal portions. The sweets and beverages group had the lowest mean score on a per RACC basis.

Some of the major food groupings contain a diverse variety of foods; therefore, it is useful to examine mean scores of subgroupings for these foods (Tables E-1 and E-2 in Appendix E) and scores of selected foods (Table E-3). In the dairy group, rankings of individual foods were similar between both unit bases (Table E-3). As expected, the milks had decreasing scores with increasing fat content. Chocolate milk, even the low-fat version, scored lower than regular whole milk. Among the yogurts, the nonfat versions scored the highest. The fruit variety of whole-milk yogurt had the lowest ranking for both unit bases but had a more extreme lower score per RACC because the large portion contributes more sugar and fat to the score.

Among legumes and nuts, there were higher scores for nuts on a per RACC basis because the larger RACC serving size, 1 ounce, is greater than 100 kcal, resulting in higher positive nutrient values, particularly for unsaturated fat.

For many of the grain subgroups, similar rankings of foods on a per RACC and per 100 kcal were seen for some subgroups, particularly when the RACC serving size was close to 100 kcal (e.g., breads, cereals, cookies, crackers, and salty snacks). On the other hand, high-calorie and high-fat cakes and mixed foods such as pizza resulted in more extreme lower values on a per RACC basis. For example, a RACC serving of chocolate cake provides 468 kcal, 30% of the daily saturated fat value, and 116% of the daily added sugar value.

The vegetable group had an overall negative mean score. Some foods in this group were condiments that contained a large amount of sodium (e.g., salsa, catsup, pickles). Among the highest scoring vegetables were leafy greens. The NDS1KCAL scores were very high because of the large serving size of 100 kcal portions (e.g., 435 g raw spinach). Pickles had the lowest ranking per 100 kcal because of the large serving (833 g) per 100 kcal and high sodium content. Potato chips are included in the vegetable group and, interestingly, scored higher than baked potatoes regardless of the unit basis. Potato chips have higher unsaturated fat and vitamin E. However, in the nutrient database used for NHANES, baked potatoes have higher sodium than potato chips because the baked potatoes have salt added by default. If the sodium value for baked potatoes without salt were used (5 mg/100 kcal), then the NDS1KCAL score would be 3.11 rather than −1.16 (with salt), which is higher than the score for potato chips (1.93).

Among the fats, oils, and salad dressings group, low-calorie salad dressings had very low scores on a per 100 kcal basis because of high sodium content. For example, the lowest scoring salad dressing was fat-free, reduced-calorie Italian dressing (−33.41), providing 100% of the daily sodium value on a per 100 kcal basis. On a RACC basis, the score was −4.71, which is 14% of the daily sodium value. In contrast, scores for regular Italian dressing were similar with the different unit bases (−8.98 and −7.84 on a per 100 kcal and RACC basis, respectively).

Among beverages, coffee and tea scored very high on a per 100 kcal basis because of their low calorie content, resulting in large portions on a per 100 kcal basis and high potassium values driving the high scores. Sugar-sweetened beverages were similar between both unit bases because a RACC of 240 g is equivalent to ~100 kcal. Sugar-free, calorie-free beverages were assigned a 1 kcal/100 g value to obtain a score on a per 100 kcal basis, which resulted in relatively high nutrient values on a per 100 kcal basis.

4.2.2 Ranking of Foods and Food Groups with Modified Algorithm, Removing Unsaturated Fat (NDS2)

The modification removing unsaturated fat from the algorithm results in changes in scores regardless of whether the food contains unsaturated fat because the total number of positive nutrients in this algorithm is 7 and the sum of the positive nutrient percentages was divided by 7 instead of 8 in the baseline algorithm. In general, if a food did not contain unsaturated fat, then the score increased because the positive nutrients were now divided by 7 rather than 8. Conversely, in general, if a food contained unsaturated fat, then the score decreased when it was removed from the algorithm. The mean and distribution of algorithm scores by major food groupings using the modified algorithm scoring foods on a per 100 kcal basis (NDS2KCAL) and a per RACC basis (NDS2RACC) are presented in Figures F-1 and F-2.

The overall mean scores of the 570 foods with the modified algorithms were similar to the overall mean scores for foods using the baseline algorithm. The rankings of the mean scores from the major USDA food groupings changed slightly. Fruits ranked first for both algorithms, although the mean score for fruits increased with the modified algorithm. The scores for most fruits increased because they do not contain unsaturated fat; however, the score for avocado decreased with the modified algorithm because avocados are high in unsaturated fat (i.e., from 4.42 to 3.02 on a per 100 kcal basis and from 9.89 to 6.76 on a per RACC basis). Vegetables and eggs reversed within the second and third rankings on a per kcal basis with the modification, and scores for eggs decreased because they contain some unsaturated fat. As in NDS1, potato chips still had higher scores than baked potatoes, but the gap was smaller because the score for potato chips decreased and the score for baked potatoes increased with the removal of unsaturated fat from the algorithm. Dairy scores generally increased because they contain relatively small amounts of unsaturated fat and high amounts of other positive nutrients. A few foods in the dairy group that contained more unsaturated fats (e.g., cream substitutes) had lower scores with the modified algorithms (NDS2KCAL and NDS2RACC). The mean score for legumes/nuts decreased slightly. Although the scores for all of the nut foods decreased, some of the bean scores increased. The top-scoring food in the legume/nut group was almonds; its score decreased from 7.33 to 6.11 on a per 100 kcal basis and from 12.65 to 10.54 on a per RACC basis. The largest decrease in mean food group scores with the modified algorithm was for the fats, oils, and dressings group because of the high unsaturated fat content of many of the foods in this group. For example, the top-ranking food, olive oil, decreased from 1.14 to −1.39, and margarine decreased from −1.78 to −4.06 on a per 100 kcal basis. The fats and oils remained the group with the lowest mean score on a per 100 kcal basis but decreased from the fifth to the eighth ranking on a per RACC basis.

4.2.3 Ranking of Foods and Food Groups with Modified Algorithm, Removing Added Sugars (NDS3)

The baseline algorithm was modified to remove the added sugars component from the negative nutrients for a total of two negative nutrients instead of three in the baseline algorithm. The average score of the 570 foods with the modified algorithm was slightly lower than the mean score of foods using the baseline algorithm (Figures F-3 and F-4). In general, foods that contained the other two negative nutrients, saturated fat and sodium, had lower scores when added sugar was removed from the algorithm.

The scores for most fruits and fruit juices did not change; however, canned fruits containing added sugar had higher scores when added sugar was removed from the algorithm. Scores of foods in the sweets/beverages group increased substantially. The mean score of grains increased slightly on a per 100 kcal basis and decreased slightly on a per RACC basis. Among the grain products, scores for cakes, cookies, and cereals increased, and scores for crackers and mixed foods decreased. The decrease in scores for mixed foods was greater on a per RACC basis than on a per 100 kcal basis. For example, macaroni and cheese decreased from −10.10 to −24.07 because the high amounts of saturated fat and sodium on a per RACC basis were divided by two instead of three in the modified algorithm; on a per 100 kcal basis, the score decreased from −2.06 to −4.90. Among dairy foods, scores for yogurts and ice cream generally increased, while scores for cheeses decreased. Low-fat chocolate milk scored higher than white whole or 2% milk, whereas in the original algorithm all chocolate milks scored lower than all of the white milk varieties. Scores for fats, eggs, and meats decreased because these foods do not contain sugar, and the other negative nutrients were more heavily weighted because of dividing by two rather than three negative nutrients.

4.2.4 Ranking of Foods and Food Groups with Modified Algorithm, Adding Vitamin C (NDS4)

The mean scores of the 570 foods with vitamin C added to the algorithm were higher than the respective baseline algorithms (Figures F-5 and F-6).

The scores for fruits increased substantially with the addition of vitamin C to the algorithm; the average score increased from 3.06 to 8.77 on a per 100 kcal basis and from 2.47 to 7.49 on a per RACC basis. Fruits with the highest vitamin C content were capped at 100% of daily value to avoid extreme scores. One of these foods was raw oranges; its score increased from 5.86 to 16.31 on a per 100 kcal basis and from 3.85 to 14.53 on a per RACC basis. The highest score using both the original and modified algorithm was orange juice with calcium added; its score increased from 10.15 to 20.12 on a per 100 kcal basis when vitamin C was added to the algorithm. Scores for vegetables increased moderately. The mean scores of the other food groups were similar to the original scores.

4.2.5 Ranking of Foods and Food Groups with Modified Algorithm, Adding Whole Grains (NDS5)

In this algorithm, whole grains are expressed as the percentage of total grains compared with the recommendation to consume 50% of grains as whole grains. A second method of expressing whole grains was tested based on the recommendation of three servings per day of whole grains. In this section we present a summary of the scores for the former method of expressing whole grains as the percentage of total grains (Figures F-7 and F-8).

Except for increases in scores for the grain products, the mean scores and rankings of the food groupings remained similar to the original algorithm. Grain products increased from the fourth to the second highest ranking group on a per 100 kcal basis and from the eighth to the fifth ranking group on a per RACC basis. Scores for some breads increased substantially; for example, multigrain bread increased from −0.59 to 10.02 and wheat bread increased from −0.29 to 6.81 on a per 100 kcal basis. White bread decreased slightly, from −1.79 to −2.15. Granola bars increased from −4.49 to 6.32 on a per 100 kcal basis. In the original algorithm, fruit-based cereal bars had slightly higher scores than granola bars, but this was reversed with the modification because of a higher whole grain content. The score for Cheerios increased from 8.23 to 18.03 on a per 100 kcal basis. It should be noted that whole grains contribute other positive nutrients that are included in the algorithm, such as fiber and vitamin E.

4.2.6 Ranking of Foods and Food Groups with Modified Algorithm, Adding Vitamin C and Whole Grains (NDS6)

The mean score of the 570 foods with vitamin C and whole grains added on a per 100 kcal basis was the highest mean score of all modifications (−2.13). The mean and distribution of algorithm scores by major food groupings using the modified algorithm scoring foods on a per 100 kcal basis (NDS6KCAL) and a per RACC basis (NDS6RACC) are presented in Figures F-9 and F-10.

Fruits, vegetables, and legumes were the food groups with the highest ranking mean scores with both unit bases, and the rankings of mean scores of all food groups were the same for both unit bases.

4.2.7 Synopsis of Algorithm Performance in Ranking the Nutritional Quality of Foods

Based on examination of food scores using the various algorithms, it is difficult to assess the differences of the various algorithms to determine which nutrients should be included in an algorithm; however, the process does provide insights into the effects of various modifications. Examination of food scores using the various algorithms demonstrates that the algorithm discriminates between more and less healthy versions of foods; for example, low-fat dairy products score higher than high-fat dairy products, and fruits score higher than sweetened beverages. Food scores are sensitive to the nutrients in the algorithm; for example, scores for nuts, which are high in unsaturated fat, decreased when unsaturated fat as a positive nutrient was removed from the algorithm. Some anomalies became evident when comparing the scores based on per 100 kcal or RACC basis; for example, green leafy vegetables score very high on a per 100 kcal basis because of their low energy content but 100 kcal is a very large portion size for these foods.

As noted previously, we did not conduct formal validity tests focused on the food scores themselves. We did conduct formal validity tests of the ability of the algorithms to predict overall diet quality by scoring foods in diets and comparing with the HEI, a measure of diet quality. Section 4.3 describes the results of regression models predicting HEI scores using the various algorithms to score foods in diets.

4.3 Algorithm Performance in Predicting Overall Dietary Quality

4.3.1 Results of Regression Models Predicting HEI Scores

The performance of the algorithms in predicting overall dietary quality was assessed by calculating algorithm scores for foods consumed by the 15,576 individuals in NHANES 2005-2008. The results for the regression of the composite algorithm scores for the NHANES participant diets on the HEI diet quality scores for the participant diets are presented in Table 4-1. The models included covariates for age, gender, and ethnicity. The details of the methodology are described at the end of Section 4.1.

Table 4-1. Weighted Mean Scores of Foods Consumed by 16,587 Individuals in NHANES 2005-2008 and Linear Regression Models on the Healthy Eating Index
Algorithm	Modification	Weighted Mean Scores (SE)	Beta Coefficient (SE)	p-value	R²
Notes: A model with no nutrients (only with covariates age, gender, and ethnicity) per 100 kcal has an R² of 4.17%. A per 100 kcal model with only negative nutrients (saturated fat, added sugars, and sodium) has an R² of 35.19%, and a model with only the positive nutrients (protein, fiber, vitamin E, vitamin D, calcium, iron, potassium, and unsaturated fat) has an R² of 32.66%. - Means no data.
Per 100 kcal	-	-	-	-	-
NDS1KCAL	-	−2.79 (0.06)	4.12 (0.08)	<0.0001	46.19%
NDS2KCAL	− Unsaturated fat	−2.92 (0.06)	3.96 (0.07)	<0.0001	46.49%
NDS3KCAL	− Added sugar	−2.55 (0.03)	4.63 (0.11)	<0.0001	36.23%
NDS4AKCAL	+ Vitamin A	−3.01 (0.06)	4.14 (0.08)	<0.0001	46.16%
NDS4B12KCAL	+ Vitamin B₁₂	−2.78 (0.05)	3.99 (0.07)	<0.0001	44.42%
NDS4CKCAL	+ Vitamin C	−2.58 (0.06)	3.95 (0.07)	<0.0001	49.58%
NDS4FKCAL	+ Folate	−2.97 (0.05)	4.13 (0.08)	<0.0001	45.64%
NDS4MKCAL	+ Magnesium	−2.84 (0.06)	4.10 (0.08)	<0.0001	46.41%
NDS5KCAL	+ Whole grains	-	-	-	-
-	% of total/50%	−2.83 (0.06)	4.00 (0.07)	<0.0001	48.59%
-	# servings	−3.11 (0.06)	4.20 (0.08)	<0.0001	47.36%
NDS6KCAL	+ Vitamin C and whole grains (% of total/50%)	−2.64 (0.07)	3.89 (0.06)	<0.0001	51.57%
Per RACC	-	-	-	-	-
NDS1RACC	-	−2.92 (0.06)	3.34 (0.08)	<0.0001	38.99%
NDS2RACC	− Unsaturated fat	−3.06 (0.06)	3.19 (0.07)	<0.0001	38.99%
NDS3RACC	− Added sugar	−2.58 (0.03)	3.91 (0.11)	<0.0001	32.48%
NDS4ARACC	+ Vitamin A	−3.13 (0.06)	3.28 (0.08)	<0.0001	38.40%
NDS4B12RACC	+ Vitamin B₁₂	−2.91 (0.06)	3.26 (0.08)	<0.0001	37.75%
NDS4CRACC	+ Vitamin C	−2.74 (0.07)	3.31 (0.08)	<0.0001	42.21%
NDS4FRACC	+ Folate	−3.09 (0.06)	3.31 (0.08)	<0.0001	38.42%
NDS4MRACC	+ Magnesium	−2.99 (0.06)	3.31 (0.08)	<0.0001	39.11%
NDS5RACC	+ Whole grains	-	-	-	-
-	% of total/50%	−2.98 (0.07)	3.26 (0.07)	<0.0001	40.78%
-	# servings	−3.24 (0.06)	3.30 (0.08)	<0.0001	39.11%
NDS6RACC	+ Vitamin C and whole grains (% of total/50%)	−2.82 (0.07)	3.25 (0.07)	<0.0001	43.58%

In all of the linear regression models, the weighted mean scores were significantly associated with HEI scores (p < 0.0001). The weighted mean scores of foods consumed by individuals were slightly higher on a per 100 kcal than on a per RACC basis. For the baseline algorithm (NDS1KCAL), the model with scores based on a per 100 kcal basis explained 46.2% of the variation in HEI scores compared with 39.0% explained by the per RACC scores (NDS1RACC). For each modification to the baseline algorithm, the variation in HEI scores was better explained by the algorithm calculated on a per 100 kcal basis than a per RACC basis. The model with the highest R² was the modification with vitamin C and whole grains on a per 100 kcal basis (NDS6), explaining 51.6% of the variance in HEI scores. The algorithm with vitamin C had the second highest R² (49.6%).

A better explanation of HEI scores when the food scores were calculated on a per 100 kcal basis rather than a per RACC basis could possibly be that the range of scores on a per 100 kcal basis is more extreme. The fact that the HEI is based on nutrient standards on a per 1,000 kcal basis could also contribute to a better prediction by an algorithm scored on a per 100 kcal basis.

4.3.2 Plots of Predicted HEI Scores

Figure 4-4 illustrates the prediction of HEI scores by the covariates only (age, gender, and race), which explained 4.17% of the variance in HEI scores. To illustrate good prediction of the algorithm score at the high and low ends of actual HEI scores, Figure 4-5 displays HEI scores predicted by the modified algorithm on a per 100 kcal basis with vitamin C added.

Figure 4-4. Plot of Predicted HEI Scores with Covariates Only (Age, Gender, Race)

The figure shows a scatterplot for the predicted HEI scores (on the vertical axis) from a regression model with only the covariates, age, gender, and race as independent variables. The actual HEI scores are on the horizontal axis. The diagonal line shows theoretical perfect prediction of the HEI. The model showed that the covariates alone predicted only 4.17% of variation in the HEI. Data are from 16,587 participants in NHANES 2005-2008.

Figure 4-5. Plot of Predicted HEI Scores with Modified Algorithm on per 100 Kcal with Vitamin C Added

The figure shows a scatterplot for the predicted HEI scores (on the vertical axis) from a regression model for a modified algorithm on a per 100 kcal basis with vitamin C added (NDS4CKCAL). The actual HEI scores are on the horizontal axis. The diagonal line shows theoretical perfect prediction of the HEI. The model showed that the algorithm accounted for 45.59% of variation in the HEI. Agreement was reasonably good at high and low HEI values, as shown by points lying near the diagonal line. Data are from 16,587 participants in NHANES 2005-2008.

4.4 A New Statistical Approach to Developing a Nutrient Density-Based Algorithm

There are some limitations to the approach taken in deciding which nutrients to include in an algorithm and which algorithm to choose as the final algorithm. We did not test every possible combination of nutrients proposed. As previously mentioned, the selection of negative nutrients to include in an algorithm is more clearly based on scientific evidence, but the selection of positive nutrients is somewhat arbitrary. We chose to test various micronutrients that were known to be consumed in less than adequate amounts in the U.S. population and should be encouraged according to the DGA. With 17 different nutrients or food components, there are 2¹⁷ (131,072) possible nutrient combinations. It is difficult to assess this number of combinations, but insights can be gained from examining correlations between nutrients.

4.4.1 Correlation Analyses to Inform Development of a New Approach

We conducted correlation analyses using SAS Proc Corr with day 1 dietary weights with variables representing the nutrient intakes as a percentage of recommended intake levels per 100 kcal consumed for the 17 nutrients/components and the HEI scores of individuals in NHANES. The inclusion in the model of nutrients that are highly correlated with each other may not provide more information to the prediction of dietary quality than if only one of the nutrients was included. The correlation coefficients are provided in Table 4-2.

Table 4-2. Correlation Matrix of Nutrient or Component Intakes as a Percentage of Recommended Intake Levels Per 100 Kcal and HEI Score^a
Nutrient/ Component	HEI	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsatu- rated Fat	Magne- sium	Vitamin A	Vitamin C	Folic Acid	Vitamin B12	Satu- rated Fat	Sodium	Added Sugar	Whole Grains (% of Whole)
^a Nutrient intakes and HEI scores are from 1-day dietary intakes for 16,587 participants in NHANES 2005-2008. - Means no data.
HEI	1.00	0.33	0.60	0.35	0.22	0.21	0.32	0.50	0.04	0.52	0.29	0.42	0.16	0.14	−0.40	−0.02	−0.39	0.35
Protein	-	1.00	0.16	0.09	0.29	0.27	0.26	0.42	0.03	0.36	0.23	0.09	0.01	0.39	0.07	0.35	−0.45	0.09
Fiber	-	-	1.00	0.32	0.05	0.16	0.37	0.53	−0.07	0.60	0.31	0.41	0.14	0.03	−0.26	0.10	−0.32	0.36
Vitamin E	-	-	-	1.00	0.04	0.08	0.22	0.25	0.28	0.37	0.28	0.29	0.08	0.11	−0.11	0.06	−0.22	0.09
Vitamin D	-	-	-	-	1.00	0.62	0.27	0.33	−0.16	0.25	0.44	0.13	0.26	0.61	0.05	−0.04	−0.13	0.17
Calcium	-	-	-	-	-	1.00	0.25	0.36	−0.21	0.43	0.45	0.21	0.24	0.41	0.17	0.06	−0.22	0.23
Iron	-	-	-	-	-	-	1.00	0.26	−0.16	0.35	0.42	0.21	0.70	0.50	−0.20	0.16	−0.18	0.34
Potassium	-	-	-	-	-	-	-	1.00	−0.13	0.71	0.38	0.47	−0.02	0.25	−0.14	0.15	−0.39	0.22
Unsaturated fat	-	-	-	-	-	-	-	-	1.00	−0.10	−0.11	−0.17	−0.23	−0.13	0.38	0.15	−0.29	−0.09
Magnesium	-	-	-	-	-	-	-	-	-	1.00	0.37	0.33	0.07	0.19	−0.23	0.11	−0.39	0.34
Vitamin A	-	-	-	-	-	-	-	-	-	-	1.00	0.30	0.27	0.38	0.02	0.07	−0.16	0.25
Vitamin C	-	-	-	-	-	-	-	-	-	-	-	1.00	0.08	0.08	−0.24	0.01	−0.16	0.17
Folic acid	-	-	-	-	-	-	-	-	-	-	-	-	1.00	0.40	−0.16	0.06	−0.07	0.25
Vitamin B₁₂	-	-	-	-	-	-	-	-	-	-	-	-	-	1.00	0.09	0.06	−0.13	0.14
Saturated fat	-	-	-	-	-	-	-	-	-	-	-	-	-	-	1.00	0.08	−0.17	−0.10
Sodium	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	1.00	−0.32	0.00
Added sugar	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	1.00	−0.13
Whole grains	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	1.00

These results show that calcium and vitamin D are highly correlated (r = 0.62). Dairy foods are primary sources of both calcium and vitamin D to the daily diet, and some fish are high in both of these nutrients. Folic acid is highly correlated with iron (r = 0.70), so the algorithm that included folic acid may not have improved the prediction of HEI over the baseline algorithm because of the collinearity with iron, which is likely due to fortification of grains with both folate and iron. Magnesium is strongly correlated with potassium (r = 0.71), so the addition of magnesium to the baseline algorithm that contained potassium may not have added to the explanation of dietary quality. HEI is strongly correlated with fiber (r = 0.60), magnesium (r = 0.52), potassium (r = 0.50), vitamin C (r = 0.42), saturated fat (r = −0.40), added sugar (r = −0.39), vitamin E (r = 0.35), and whole grains (r = 0.35).

4.4.2 Maximum R-Square Approach for Calculating Nutrient Density Scores

After running the analysis of the proposed algorithms for RTI, Nutrition Impact, LLC developed a new approach to decide which nutrients to include in a nutrient density index. The method involves examining the various nutrient combinations in models with individual nutrient intake values as independent variables rather than as a composite score. Nutrient intake values were capped at 100% of the daily value. HEI was the dependent variable, and other covariates were age, gender, and ethnicity as defined previously. The MAXimum R² (MAXR) option in some SAS procedures allows for examining every possible combination of variables of interest. However, a MAXR option is not available in the SurveyReg procedure necessary to analyze NHANES data, so a unique macro was developed to assess every possible combination of the 17 nutrients or components included in our testing. This method identifies the best one-variable model producing the highest R², the best two-variable model, etc. The MAXR approach is different from stepwise regression in that it evaluates the possible switching of the order of variables entered into the model, which can affect the model results. These analyses resulted in evaluation of 131,072 regression models for nutrients or components expressed as per 100 kcal and the same number of models with the nutrients or components expressed on a per RACC basis. Adjusted R² values were used to compare the various models because the number of variables in the model affects the R².

The MAXR analyses were conducted with some modifications. We conducted the analyses with three sets of nutrients/components: (1) with whole grains (17 nutrients/components), (2) without whole grains (16 nutrients/components), and (3) with total sugars in place of added sugars (17 nutrients/components). We proceeded with further testing of algorithms without whole grains and with added sugars after consultation with the ASPE/FDA team. Tables 4-3 and 4-4 present the beta coefficients and p-values, respectively, from regressions yielding the best models containing 1 through 16 nutrients or food components as a percentage of recommended intake levels per 100 kcal. The third columns of Tables 4-3 and 4-4 show the adjusted R² for the models. Tables 4-5 and 4-6 present the regression results for models for nutrients or components as a percentage of recommended intake levels per RACC. (Results of the analyses with whole grains and with total sugars are presented in Appendix G.)

Table 4-3. Beta Coefficients from Regression Models Using the Maximum R² Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per 100 Kcal^a
Number of Variables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsatu- rated Fat	Magne- sium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Satu- rated Fat	Sodium	Added Sugar
^a Nutrient intakes and HEI scores are from 1-day dietary intakes for 16,587 participants in NHANES 2005-2008. - Means no data.
0	0.0417	0.0413	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1	0.3657	0.3655	-	5.3217	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2	0.4346	0.4343	-	4.6677	-	-	-	-	-	-	-	-	-	-	-	−1.8039	-	-
3	0.5212	0.5209	-	3.5910	-	-	-	-	-	-	-	-	-	-	-	−2.3942	-	−0.8846
4	0.5526	0.5524	-	3.5525	-	1.2382	-	-	-	-	-	-	-	-	-	−2.4271	-	−0.8191
5	0.5819	0.5816	1.4179	3.5979	-	-	-	-	-	-	-	-	-	-	-	−2.3481	−1.3151	−0.7920
6	0.6081	0.6078	1.5858	3.7412	-	-	-	-	-	1.7797	-	-	-	-	-	−2.7342	−1.4101	−0.6454
7	0.6330	0.6328	1.3746	3.5073	-	-	1.1420	-	-	2.4082	-	-	-	-	-	−3.0957	−1.3673	−0.5568
8	0.6476	0.6473	1.3963	3.1259	-	-	1.0006	-	-	2.5145	-	-	0.3651	-	-	−2.9478	−1.3350	−0.5164
9	0.6519	0.6516	1.2879	3.1872	-	0.5950	0.6705	-	-	2.4946	-	-	0.3655	-	-	−2.8976	−1.2675	−0.5204
10	0.6536	0.6533	1.3078	2.8759	-	-	0.8162	-	0.8945	2.6864	-	-	0.3159	0.4951	-	−2.8498	−1.3595	−0.4647
11	0.6556	0.6552	1.2476	2.9657	-	0.4258	0.6125	-	0.7275	2.6423	-	-	0.3255	0.4127	-	−2.8308	−1.3072	−0.4768
12	0.6565	0.6561	1.2723	3.0880	-	0.3824	0.7285	-	0.9552	2.6759	−0.5357	-	0.3102	0.4037	-	−2.9089	−1.3266	−0.4890
13	0.6569	0.6566	1.3278	3.0675	-	0.4769	0.7201	-	0.9772	2.6714	−0.5278	-	0.3115	0.4800	−0.1722	−2.8825	−1.3307	−0.4807
14	0.6574	0.6570	1.3088	3.1120	-	0.3569	0.6839	−0.3275	0.9391	2.6825	−0.4921	0.2453	0.3034	0.5766	-	−2.9330	−1.3146	−0.4879
15	0.6577	0.6572	1.3442	3.0821	-	0.4281	0.6805	−0.2609	0.9569	2.6762	−0.5003	0.2509	0.3033	0.5957	−0.1392	−2.9116	−1.3212	−0.4833
16	0.6579	0.6574	1.3505	3.0722	0.2420	0.4419	0.6843	−0.2546	0.9741	2.5964	−0.5626	0.2185	0.2926	0.5862	−0.1544	−2.8918	−1.3188	−0.4838

Table 4-4. P-Values from Regression Models Using the Maximum R² Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per 100 Kcal^a
Number of Variables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsatu- rated Fat	Magne- sium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Satu- rated Fat	Sodium	Added Sugar
^a Nutrient intakes and HEI scores are from 1-day dietary intakes for 16,587 participants in NHANES 2005-2008. - Means no data.
0	0.0417	0.0413	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1	0.3657	0.3655	-	0.0000	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2	0.4346	0.4343	-	0.0000	-	-	-	-	-	-	-	-	-	-	-	0.0000	-	-
3	0.5212	0.5209	-	0.0000	-	-	-	-	-	-	-	-	-	-	-	0.0000	-	0.0000
4	0.5526	0.5524	-	0.0000	-	0.0000	-	-	-	-	-	-	-	-	-	0.0000	-	0.0000
5	0.5819	0.5816	0.0000	0.0000	-	-	-	-	-	-	-	-	-	-	-	0.0000	0.0000	0.0000
6	0.6081	0.6078	0.0000	0.0000	-	-	-	-	-	0.0000	-	-	-	-	-	0.0000	0.0000	0.0000
7	0.6330	0.6328	0.0000	0.0000	-	-	0.0000	-	-	0.0000	-	-	-	-	-	0.0000	0.0000	0.0000
8	0.6476	0.6473	0.0000	0.0000	-	-	0.0000	-	-	0.0000	-	-	0.0000	-	-	0.0000	0.0000	0.0000
9	0.6519	0.6516	0.0000	0.0000	-	0.0000	0.0000	-	-	0.0000	-	-	0.0000	-	-	0.0000	0.0000	0.0000
10	0.6536	0.6533	0.0000	0.0000	-	-	0.0000	-	0.0002	0.0000	-	-	0.0000	0.0000	-	0.0000	0.0000	0.0000
11	0.6556	0.6552	0.0000	0.0000	-	0.0000	0.0000	-	0.0007	0.0000	-	-	0.0000	0.0000	-	0.0000	0.0000	0.0000
12	0.6565	0.6561	0.0000	0.0000	-	0.0000	0.0000	-	0.0000	0.0000	0.0187	-	0.0000	0.0000	-	0.0000	0.0000	0.0000
13	0.6569	0.6566	0.0000	0.0000	-	0.0000	0.0000	-	0.0000	0.0000	0.0188	-	0.0000	0.0000	0.0023	0.0000	0.0000	0.0000
14	0.6574	0.6570	0.0000	0.0000	-	0.0000	0.0000	0.0006	0.0000	0.0000	0.0305	0.0372	0.0000	0.0000	-	0.0000	0.0000	0.0000
15	0.6577	0.6572	0.0000	0.0000	-	0.0000	0.0000	0.0127	0.0000	0.0000	0.0258	0.0356	0.0000	0.0000	0.0255	0.0000	0.0000	0.0000
16	0.6579	0.6574	0.0000	0.0000	0.3175	0.0000	0.0000	0.0156	0.0000	0.0000	0.0139	0.0845	0.0000	0.0000	0.0150	0.0000	0.0000	0.0000

Table 4-5. Beta Coefficients from Regression Models Using the Maximum R² Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per RACC^a
Number of Variables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsatu- rated Fat	Magne- sium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Satu- rated Fat	Sodium	Added Sugar
^a Nutrient intakes and HEI scores are from 1-day dietary intakes for 16,587 participants in NHANES 2005-2008. - Means no data.
0	0.0417	0.0413	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1	0.2074	0.2071	-	3.9647	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2	0.4036	0.4033	-	4.7469	-	-	-	-	-	-	-	-	-	-	-	−2.0714	-	-
3	0.4733	0.4730	-	4.5968	-	-	-	-	-	-	-	-	-	-	-	−1.7867	-	−0.6530
4	0.5218	0.5216	-	3.1268	-	-	-	-	3.1408	-	-	-	-	-	-	−2.1353	-	−0.5858
5	0.5416	0.5413	-	2.8474	-	-	-	-	3.1095	1.3402	-	-	-	-	-	−2.7629	-	−0.6115
6	0.5655	0.5652	-	3.1201	-	-	-	-	3.3032	1.7655	-	-	-	-	-	−2.5132	−1.0466	−0.6171
7	0.5883	0.5880	1.4790	3.6687	-	-	-	-	-	1.6051	-	-	0.4734	-	-	−2.5592	−1.5007	−0.6410
8	0.6042	0.6039	1.1864	3.4481	-	-	0.9462	-	-	1.9831	-	-	0.3892	-	-	−2.9859	−1.4494	−0.6371
9	0.6066	0.6063	1.0272	3.1973	-	-	0.8442	-	0.9134	1.9644	-	-	0.3236	-	-	−2.9811	−1.4161	−0.6241
10	0.6080	0.6076	1.0243	3.1658	-	-	0.7168	-	0.8498	1.9767	-	0.4500	0.3133	-	-	−2.9757	−1.4145	−0.6260
11	0.6084	0.6080	1.0624	3.2439	-	-	0.7307	−0.2021	0.8251	1.9845	-	0.5244	0.3162	-	-	−2.9937	−1.3871	−0.6177
12	0.6093	0.6088	1.0874	3.2450	-	-	0.6794	−0.4727	0.9240	2.0045	-	0.5220	0.3139	0.3325	-	−2.9897	−1.4063	−0.6167
13	0.6095	0.6091	1.0799	3.3072	-	0.1796	0.5987	−0.4655	0.8039	1.9986	-	0.4698	0.3248	0.3088	-	−2.9727	−1.3880	−0.6172
14	0.6097	0.6092	1.0814	3.2919	0.2105	0.1926	0.5910	−0.4675	0.7920	1.9330	-	0.4339	0.3196	0.3009	-	−2.9493	−1.3830	−0.6155
15	0.6099	0.6094	1.1092	3.3693	0.2896	0.1878	0.6276	−0.4352	0.9035	1.9384	−0.3446	0.4369	0.3132	0.2870	-	−2.9663	−1.4041	−0.6181
16	0.6099	0.6094	1.1021	3.3763	0.2832	0.1729	0.6297	−0.4505	0.8976	1.9441	−0.3368	0.4352	0.3138	0.2855	0.0291	−2.9710	−1.4014	−0.6182

Table 4-6. P-Values from Regression Models Using the Maximum R² Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per RACC^a
Number of Variables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsatu- rated Fat	Magne- sium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Satu- rated Fat	Sodium	Added Sugar
^a Nutrient intakes and HEI scores are from 1-day dietary intakes for 16,587 participants in NHANES 2005-2008. - Means no data.
0	0.0417	0.0413	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
1	0.2074	0.2071	-	0.0000	-	-	-	-	-	-	-	-	-	-	-	-	-	-
2	0.4036	0.4033	-	0.0000	-	-	-	-	-	-	-	-	-	-	-	0.0000	-	-
3	0.4733	0.4730	-	0.0000	-	-	-	-	-	-	-	-	-	-	-	0.0000	-	0.0000
4	0.5218	0.5216	-	0.0000	-	-	-	-	0.0000	-	-	-	-	-	-	0.0000	-	0.0000
5	0.5416	0.5413	-	0.0000	-	-	-	-	0.0000	0.0000	-	-	-	-	-	0.0000	-	0.0000
6	0.5655	0.5652	-	0.0000	-	-	-	-	0.0000	0.0000	-	-	-	-	-	0.0000	0.0000	0.0000
7	0.5883	0.5880	0.0000	0.0000	-	-	-	-	-	0.0000	-	-	0.0000	-	-	0.0000	0.0000	0.0000
8	0.6042	0.6039	0.0000	0.0000	-	-	0.0000	-	-	0.0000	-	-	0.0000	-	-	0.0000	0.0000	0.0000
9	0.6066	0.6063	0.0000	0.0000	-	-	0.0000	-	0.0000	0.0000	-	-	0.0000	-	-	0.0000	0.0000	0.0000
10	0.6080	0.6076	0.0000	0.0000	-	-	0.0000	-	0.0001	0.0000	-	0.0209	0.0000	-	-	0.0000	0.0000	0.0000
11	0.6084	0.6080	0.0000	0.0000	-	-	0.0000	0.0375	0.0001	0.0000	-	0.0048	0.0000	-	-	0.0000	0.0000	0.0000
12	0.6093	0.6088	0.0000	0.0000	-	-	0.0000	0.0013	0.0000	0.0000	-	0.0043	0.0000	0.0017	-	0.0000	0.0000	0.0000
13	0.6095	0.6091	0.0000	0.0000	-	0.0422	0.0000	0.0010	0.0003	0.0000	-	0.0111	0.0000	0.0021	-	0.0000	0.0000	0.0000
14	0.6097	0.6092	0.0000	0.0000	0.4645	0.0211	0.0000	0.0012	0.0003	0.0000	-	0.0115	0.0000	0.0040	-	0.0000	0.0000	0.0000
15	0.6099	0.6094	0.0000	0.0000	0.2161	0.0177	0.0000	0.0002	0.0000	0.0000	0.4285	0.0095	0.0000	0.0023	-	0.0000	0.0000	0.0000
16	0.6099	0.6094	0.0000	0.0000	0.2222	0.0296	0.0000	0.0002	0.0000	0.0000	0.4399	0.0099	0.0000	0.0027	0.6419	0.0000	0.0000	0.0000

The best one-nutrient variable model is fiber, explaining 36.6% of the variance in HEI scores on a per 100 kcal basis and 20.7% on a per RACC basis. The highest R² two-nutrient variable model included fiber and saturated fat, and the highest R² three-nutrient variable model included fiber, saturated fat, and added sugars. There was a consistent increase in adjusted R² up until about eight nutrient/component variables, explaining approximately 65% of the variation in HEI on a per 100 kcal basis and 60% on a per RACC basis, which is more than a 40% improvement from our original baseline algorithm presented in Table 4-1. These eight nutrients/components were protein, fiber, calcium, unsaturated fat, vitamin C, saturated fat, sodium, and added sugars, for both per 100 kcal and per RACC basis. Supplementary analyses were conducted with the addition of whole grains and the replacement of added sugars with total sugars. The results of these supplementary analyses are reported in Appendix G.

The results of these analyses (Tables 4-3 through 4-6) could be used to help identify nutrients to include in a nutrient density algorithm. The beta coefficients signify a relative importance of nutrients in predicting the HEI and could potentially be considered as weighting factors. However, these data should be interpreted with several factors in mind.

First, the calculation method of the HEI score will influence the strength of the associations with nutrients. Second, foods are complex and contain a combination of positive and negative nutrients. For example, the beta coefficients in Table 4-3 for magnesium and vitamin B₁₂ are negative, which may be due to their association with high-meat diets that are high in saturated fat and thus are associated with lower HEI scores. Saturated fat is weighed heavily in the negative direction in the HEI scoring, because it is included both as a nutrient and in the "calories from solid fat" component. Fortified breakfast cereals are also a primary source of vitamin B₁₂, and the associated added sugar may also be driving the negative coefficient for vitamin B₁₂.

Nevertheless, these nutrients do not seem to be contributing more information to the overall explanation of HEI, because the R² of the models with magnesium and vitamin B₁₂ (13- to 16-nutrient/component models) is not much higher than the eight- to nine-nutrient/component models.

The MAXR approach does allow further insight into the nutrients selected in the proposed RTI algorithms described in Sections 3 and 4. As shown in Table 4-3, vitamin E was only retained in the 16-nutrient model and iron only remained in models with 14 to 16 nutrients. Interestingly, vitamin D remained in seven of the higher variable number models despite a high correlation with calcium, which was also retained in all models with seven or more nutrients, and vitamin D was retained in the four-nutrient model and calcium was not. Unsaturated fat seems to be important; it was included in all models with six or more nutrients. Of the positive nutrients tested individually in the RTI algorithm (vitamin A, vitamin C, vitamin B₁₂, folic acid, and magnesium), vitamin C was the only one included in the eight- and nine-nutrient models. Magnesium, vitamin A, and vitamin B₁₂ were retained in higher variable number models but did not add to the explanation of variance in HEI scores.

The results of the MAXR analyses show the best regression models for 1 through 16 (Tables 4-3 through 4-6) and 1 through 17 (Tables G-1 through G-4, G-7, and G-8) nutrients or components. Additional analyses were performed to examine the distributions of R² of models and the properties of "next-best" models (i.e., those that had the next lowest R² from the best R² model). These analyses were conducted with the 1 through 17 nutrients or food components models that included whole grains. These results are presented in Appendix G. In brief, there was a wide distribution of R² among the 24,310 possible eight nutrient or food component models, with models on a per 100 kcal basis having a minimum R² of 0.21 and interquartile range of 0.44 to 0.53. We also examined the top 10 eight-nutrient or food component models to examine differences in R², nutrients, and beta coefficients (Tables G-5 and G-6). The R² for the top models were high and close to the maximum R² values. All of the top 10 models included fiber, unsaturated fat, saturated fat, sodium, and added sugar. All but one of the models included protein. Whole grains were not retained in the top two models but were present in 4 or 5 of the top 10 models for both unit bases. Vitamin D replaced calcium in the second highest models for both unit bases. Some of the top 10 models had potassium, one had vitamin A, one had vitamin B₁₂, but none included vitamin E, iron, magnesium, or folic acid. The fact that the R² values for the top 10 models were extremely close suggests that there are a number of possible eight-nutrient or food component algorithms that would be similar in predicting dietary quality based on the HEI.

After discussions with the ASPE/FDA team, it was decided to use information from the MAXR analyses to develop final models and conduct further analyses. RTI recommended the eight-nutrient/component model, because the higher-term models did not significantly improve R². The ASPE/FDA team requested models scored using both unit bases, per 100 kcal and per RACC.

4.4.3 Final Algorithm Development and Testing Using the MAXR Approach

The beta coefficients from the eight-nutrient/component models per 100 kcal (Table 4-3) and per RACC (Table 4-5) were used to create algorithms to score foods. The algorithms are depicted in Figure 4-6.

Figure 4-6. Final Algorithms Using the MAXR Approach

NDSKCAL	[(1.40 * Protein g per 100 kcal/50 g) + (3.13 * Fiber g per 100 kcal/25 g) + (1.00 * Calcium mg per 100 kcal/1,000 mg + (2.51 * Unsaturated fat g per 100 kcal/44 g) + (0.37 * Vitamin C mg per 100 kcal/60 mg) − (2.95 * Saturated fat g per 100 kcal/20 g) - (0.52 * Added sugars g per 100 kcal/50 g) - (1.34 * Sodium mg per 100 kcal/2,400 mg) ] * 100
NDSRACC	[(1.19 * Protein g per RACC/50 g) + (3.45 * Fiber g per RACC/25 g) + (0.95 * Calcium mg per RACC/1,000 mg + (1.98 * Unsaturated fat g per RACC/44 g) + (0.39 * Vitamin C mg per RACC/60 mg) − (2.99 * Saturated fat g per RACC/20 g) - (0.64 * Added sugars g per RACC/50 g) − (1.45 * Sodium mg per RACC/2,400 mg) ] * 100

Positive nutrients were capped at 100% of recommended intake.

Methods for Subgroup Analyses for Final Algorithms

These algorithms are inherently different from the previous algorithms tested in Section 4.3 because different weights are applied to each nutrient and the values for each nutrient are summed rather than taking an average of the positive and negative nutrients. The scores were calculated for all foods in the NHANES database, and scores for the 570 foods were examined by food groupings as described previously. The food scores were regressed on HEI scores for all individuals and for the following subpopulations:

age groups (children 2 to 18 years, adults 19+ years, and adults 50+ years)
ethnic groups (non-Hispanic white, non-Hispanic black, and Mexican American)
income status (poverty income ratio ≤1.85, and >1.85)
weight status (for adults, obese: body mass index [BMI] ≥ 30, overweight: BMI ≥ 25 and < 30, normal: BMI < 25; for children, obese: BMI-for-age z-score ≥ 95th percentile, overweight: BMI-for-age z-score ≥ 85th and < 95th percentile, normal: BMI-for-age z-score < 85th percentile)
lipid status (normal LDL: <130 mg/dl, elevated LDL: ≥130 mg/dl)

Regression models were conducted as previously described in Section 4.1, using SUDAAN Version 10 to account for the complex sampling design of NHANES. Covariates included age (as a continuous variable), gender, and ethnicity. Models for subpopulations by age included covariates for gender and ethnicity, and models for ethnic groups included covariates for age and gender.

Methods for Categorical Ratings for Final Algorithms

The continuous nutrient density-based scores were converted into 3- and 5-point scales or ratings using tertiles of scores for the 3-point rating and quintiles for the 5-point rating. To create a normal distribution of scores, the scores were standardized within each major USDA food grouping using the following equation:

standardized score within food group = (NDS food score - median NDS score
within food group)/interquartile range of NDS scores within food group

and across all food groups using this equation:

standardized score across food groups = (NDS food score - median NDS score
of all foods)/interquartile range of NDS scores of all foods

Next, separate tertile cutoffs were identified for the within-food group and across all foods' standardized scores. Each standardized score was assigned a rating of 1 to 3 based on its corresponding tertile. Next, separate quintile cutoffs were identified for the within-food group and across all foods standardized scores, and each standardized score was assigned a rating of 1 to 5 based on its corresponding quintile.

The categorical scores are described and compared for each of the two final algorithms shown in Figure 4-6.

Performance of Final Algorithms in Ranking the Nutritional Quality of Foods and Food Groups

The mean and distribution of algorithm scores for all foods and major food groupings are presented in Figures 4-7 and 4-8 for NDSKCAL and NDSRACC, respectively. The absolute algorithm scores cannot be compared with previous algorithms tested in Section 4.3 because in these final algorithms different weights were applied to each nutrient (Figure 4-6), and the values for each nutrient were summed rather than taking an average of the positive and negative nutrients.

Figure 4-7. Box Plot of Nutrient Density Scores per 100 Kcal for Final Algorithm (NDSKCAL)

Figure 4-8. Box Plot of Nutrient Density Scores per RACC for Final Algorithm (NDSRACC)

The overall mean score of the 570 foods was slightly higher, and the range of scores was wider, on a per 100 kcal unit basis than on a per RACC basis. The highest ranking mean score for a major food group was fruit on a per 100 kcal basis and legumes on a per RACC basis. The dairy mean score ranked eighth on both unit bases. Fats and oils had the lowest mean score ranking on a per 100 kcal basis, while the sweets and beverages mean score ranked lowest on a per RACC basis.

Mean scores of subgroupings of foods and scores of selected foods are presented in Appendix H. The top 10 highest scoring foods were all vegetables on a per 100 kcal basis, mostly raw and leafy green vegetables. Among the top 10 highest ranking foods on a per RACC basis were avocado, Fiber One bar, calcium-fortified orange juice, almonds, pinto beans, raw oranges, and strawberries.

Among the dairy products, nonfat yogurts and milk scored highest, and ice cream and cheeses scored lowest. Among legumes and nuts, the scores per RACC were higher for nuts because of the larger portion size of a RACC (1 ounce) compared with a 100 kcal portion. Among the highest scoring foods in the grain group were fortified bars and cereals, and popcorn. The range of scores for grains was much larger on a per RACC basis than on a per 100 kcal basis. Cakes and pastries had much lower scores on a per RACC basis than on a per 100 kcal basis because of the larger portion sizes and, therefore, higher fat and sugar content. Avocados ranked much higher than other fruits on a per RACC basis, with a score of 175 compared with a score of 98 for the next highest scoring fruit (calcium-fortified orange juice). The mean score of vegetables was much higher on a per 100 kcal basis than on a per RACC basis, because of the exaggerated scores of low-calorie leafy green vegetables.

Performance of Final Algorithms in Predicting HEI Scores for All Individuals and Subpopulations

The performance of the final algorithms in predicting overall dietary quality was assessed by calculating composite algorithm scores for 1-day food intakes in the NHANES sample and regressing composite algorithm scores on HEI scores for all individuals and for subpopulations. Results of regression models are presented in Table 4-7. Overall, the models explained approximately two-thirds of the variance in HEI scores, compared with the best of the previous models (Section 4.3.1) that explained approximately one-half of the variance in the HEI. The model with the final algorithm scored on a per 100 kcal basis had a higher R² (64.76%) than the per RACC algorithm (60.42%). We would expect similar R²s from these regressions of HEI scores on food scores using the final algorithms as the R² from the eight-nutrient model in Tables 4-3 and -5, because they include the same nutrients and beta-coefficients. In addition, we can examine how well the algorithm works for various subpopulations.

Table 4-7. Weighted Mean Scores of Foods Consumed by Individuals in NHANES 2005-2008 and Linear Regression Models on Healthy Eating Index Using Final Algorithms^a
Per 100 kcal (NDSKCAL)	n	Weighted Mean Scores (SE)	Beta Coefficient (SE)	p-value	R²
- Means no data. ^a Regression models were adjusted for complex sampling of NHANES and included covariates for age, gender, and ethnicity. ^b Regression models included covariates for gender and ethnicity. ^c Regression models included covariates for age and gender.
Overall	16,587	7.40 (0.29)	1.00 (0.02)	<0.0001	64.76%
Age groups ^b	-	-	-	-	-
Children (2-18 y)	6,706	5.08 (0.23)	1.12 (0.02)	<0.0001	64.48%
Adults (19+ y)	9,881	8.15 (0.35)	0.96 (0.02)	<0.0001	65.54%
Older adults (50+ y)	4,792	10.58 (0.33)	0.96 (0.02)	<0.0001	66.81%
Ethnicity ^c	-	-	-	-	-
Non-Hispanic white	6,700	6.99 (0.37)	1.00 (0.02)	<0.0001	66.07%
Non-Hispanic black	4,110	6.31 (0.33)	0.99 (0.02)	<0.0001	60.05%
Mexican Americans	3,781	9.89 (0.30)	0.99 (0.03)	<0.0001	60.11%
Poverty income ratio	-	-	-	-	-
≤1.85	7,353	6.83 (0.49)	1.00 (0.01)	<0.0001	64.63%
>1.85	8,175	7.53 (0.29)	1.00 (0.02)	<0.0001	64.87%
Child BMI-for-age status	-	-	-	-	-
Obese (≥95th pct)	1,262	4.87 (0.50)	1.02 (0.06)	<0.0001	65.00%
Overweight (≥85th and <95th pct)	996	4.76 (0.63)	1.11 (0.04)	<0.0001	69.17%
Normal weight (<85th pct)	4,448	5.20 (0.26)	1.15 (0.02)	<0.0001	63.44%

(continued)

Table 4-7. Weighted Mean Scores of Foods Consumed by Individuals in NHANES 2005-2008 and Linear Regression Models on Healthy Eating Index Using Final Algorithms^a (continued)
Per RACC (NDSRACC)	n	Weighted Mean Scores (SE)	Beta Coefficient (SE)	p-value	R²
- Means no data. ^a Regression models were adjusted for complex sampling of NHANES and included covariates for age, gender, and ethnicity. ^b Regression models included covariates for gender and ethnicity. ^c Regression models included covariates for age and gender.
Overall	16,587	1.16 (0.29)	1.00 (0.01)	<0.0001	60.42%
Adult BMI status	-	-	-	-	-
Obese (≥30)	3,457	7.58 (0.35)	0.97 (0.03)	<0.0001	64.19%
Overweight (≥25 and <30)	3,316	8.72 (0.41)	0.95 (0.03)	<0.0001	65.05%
Normal weight (<25)	2,965	8.03 (0.46)	0.98 (0.03)	<0.0001	67.70%
LDL status	-	-	-	-	-
Normal (<130 mg/dl)	4,132	7.13 (0.34)	0.95 (0.04)	<0.0001	64.10%
Elevated (≥130 mg/dl)	1,450	7.38 (0.60)	0.95 (0.03)	<0.0001	65.10%
Age groups ^b	-	-	-	-	-
Children (2-18 y)	6,706	−0.67 (0.28)	0.91 (0.02)	<0.0001	62.73%
Adults (19+ y)	9,881	1.76 (0.32)	1.02 (0.02)	<0.0001	60.22%
Older adults (50+ y)	4,792	4.01 (0.27)	1.11 (0.02)	<0.0001	63.03%
Ethnicity ^c	-	-	-	-	-
Non-Hispanic white	6,700	0.66 (0.35)	1.04 (0.02)	<0.0001	61.43%
Non-Hispanic black	4,110	−0.04 (0.36)	0.92 (0.03)	<0.0001	57.82%
Mexican Americans	3,781	4.26 (0.30)	0.91 (0.02)	<0.0001	57.08%
Poverty income ratio	-	-	-	-	-
≤1.85	7,353	0.77 (0.52)	0.95 (0.03)	<0.0001	59.78%
>1.85	8,175	1.20 (0.25)	1.03 (0.02)	<0.0001	60.77%
Child BMI-for-age status	-	-	-	-	-
Obese (≥95th pct)	1,262	−0.99 (0.60)	0.87 (0.04)	<0.0001	62.62%
Overweight (≥85th and <95th pct)	996	−1.37 (0.68)	0.92 (0.06)	<0.0001	66.11%
Normal weight (<85th pct)	4,448	−0.44 (0.33)	0.91 (0.02)	<0.0001	62.02%
Adult BMI status	-	-	-	-	-
Obese (≥30)	3,457	1.02 (0.29)	1.08 (0.03)	<0.0001	61.01%
Overweight (≥25 and <30)	3,316	2.27 (0.37)	1.04 (0.03)	<0.0001	61.10%
Normal weight (<25)	2,965	1.77 (0.40)	1.00 (0.03)	<0.0001	60.58%
LDL status	-	-	-	-	-
Normal (<130 mg/dl)	4,132	0.78 (0.33)	1.01 (0.03)	<0.0001	60.76%
Elevated (≥130 mg/dl)	1,450	0.90 (0.54)	1.00 (0.03)	<0.0001	60.71%

As shown in Table 4-7, the algorithms explained more of the variance in HEI scores for some subpopulations than others. The R² of the model with older adults was higher than models with all adults or children only. Older adults also had more nutrient-dense diets; the weighted mean score was higher than the other age groups. Mexican Americans had a higher weighted mean score than the other ethnic groups but that model had a lower R² than the model with non-Hispanic whites. A more nutrient-dense diet as measured by weighted mean score does not equate to a higher R² because the R² is driven by better agreement between algorithm scores and HEI scores across all levels of the HEI. This point is illustrated in Appendix I, which contains the mean scores by deciles of HEI. Although Mexican Americans had higher HEI scores and higher mean algorithm scores than whites, Mexican Americans had lower algorithm scores per 100 kcal than whites in both the lowest and highest deciles of the HEI. Models with overweight children had higher R²s than models with normal or obese children. There was little difference in R²s in models by poverty status or LDL status.

Results of Categorical Ratings Based on Final Algorithms

Category ratings within or across food groups for the 4,059 foods in NHANES 2007-2008 were assigned to foods according to tertiles or quintiles of standardized scores of final algorithms, NDSKCAL and NDSRACC. The ranges of the standardized scores corresponding to each categorical rating are provided in Tables 4-8 (for the three-category rating scheme) and 4-9 (for the five-category rating scheme). By definition, the median value of the standardized scores is zero, and each category (tertile or quintile) has approximately the same number of food items.

Table 4-8. Range of Standardized Nutrient Density Scores for Three-Category Rating Scheme for 4,097 Foods in NHANES 2007-2008
Standardized Scores	Tertile 1	Tertile 2	Tertile 3
Category ratings within and across food groups were assigned to foods according to tertiles of standardized scores on final algorithms, NDSKCAL or NDSRACC.
Category Rating	1	2	3
Per 100 kcal (NDSKCAL)
Within food groups	−37.55 to −0.34	−0.35 to 0.27	0.28 to 15.52
Across food groups	−30.02 to −0.21	−0.22 to 0.35	0.35 to 7.94
Per RACC (NDSRACC)
Within food groups	−65.63 to −0.33	−0.34 to 0.10	0.11 to 11.32
Across food groups	−17.92 to −0.17	−0.18 to 0.35	0.36 to 7.84

Table 4-9. Range of Standardized Nutrient Density Scores for Five-Category Rating Scheme for 4,097 Foods in NHANES 2007-2008
Standardized Scores	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5
Category ratings within and across food groups were assigned to foods according to tertiles of standardized scores on final algorithms, NDSKCAL or NDSRACC.
Category Rating	1	2	3	4	5
Per 100 kcal (NDSKCAL)
Within food groups	−37.55 to −0.68	−0.69 to −0.16	−0.17 to 0.08	0.09 to 0.62	0.63 to 15.52
Across food groups	−30.02 to −0.62	−0.63 to −0.05	−0.06 to 0.19	0.20 to 0.78	0.79 to 7.94
Per RACC (NDSRACC)
Within food groups	−65.63 to −0.81	−0.82 to −0.14	−0.15 to 0.05	0.06 to 0.34	0.35 to 11.32
Across food groups	−17.92 to −0.54	−0.55 to −0.02	−0.03 to 0.12	0.13 to 0.78	0.79 to 7.84

The category ratings of a selection of 100 foods based on per 100 kcal standardized NDSs are presented in Table 4-10 for three- and five-category ratings within and across food groups. These 100 foods were selected to illustrate a variety of foods, with some examples of foods that rated differently across or within food groups. Foods in Table 4-10 are grouped by major food groupings and sorted by descending NDSs, and the three- and five-category ratings appear from highest (3 or 5) to lowest (1). Categories shown in Table 4-10 could be considered, for example, for an FOP system with an overall "low, medium, high" rating scheme (three-category rating) or for an FOP system with a 5-point rating scheme (five-category rating). The results for selected foods in Table 4-10 illustrate some characteristics of category rating systems based on the final NDS algorithm, NDSKCAL.

Table 4-10. Categorical Ratings of 100 Foods Based on the per 100 Kcal Standardized Nutrient Density-Based Scores, for Three- and Five-Category Ratings Within and Across Food Groups
Food Code	Food Description	Nutrient Density Score	Three- Category Ratings Within Food Groups	Three- Category Ratings Across Foods	Five- Category Ratings Within Food Groups	Five- Category Ratings Across Foods
Foods are sorted by major food groups and descending nutrient density-based scores. The three- and five-category ratings appear from highest (3 or 5) to lowest (1).
	DAIRY FOODS
11433500	Yogurt, fruit variety, nonfat milk, sweetened with low-calorie sweetener	67.71	3	3	5	5
11411300	Yogurt, plain, nonfat milk	54.99	3	3	5	5
11113000	Milk, cow's, fluid, skim or nonfat, 0.5% or less butterfat	54.73	3	3	5	5
11112210	Milk, cow's, fluid, 1% fat	28.39	3	3	5	4
11411200	Yogurt, plain, low-fat milk	27.80	3	3	5	4
11511400	Milk, chocolate, low-fat milk-based	19.07	3	2	4	4
14204010	Cheese, cottage, low-fat (1-2% fat)	14.51	3	2	4	3
11432500	Yogurt, fruit variety, low-fat milk, sweetened with low-calorie sweetener	14.31	3	2	4	3
11433000	Yogurt, fruit variety, nonfat milk	11.67	3	2	4	3
11112110	Milk, cow's, fluid, 2% fat	7.62	3	2	4	3
11511200	Milk, chocolate, reduced fat milk-based, 2% (formerly "low-fat")	7.00	3	2	4	3
12210200	Cream substitute, liquid	1.55	2	2	4	2
11432000	Yogurt, fruit variety, low-fat milk	1.44	2	2	4	2
11511100	Milk, chocolate, whole milk-based	−3.10	2	1	4	2
14107030	Cheese, Mozzarella, part skim	−4.49	2	1	3	2
11111000	Milk, cow's, fluid, whole	−6.63	2	1	3	2
11411100	Yogurt, plain, whole milk	−9.41	2	1	3	2
13160410	Fat-free ice cream, chocolate	−10.31	2	1	3	2
11431000	Yogurt, fruit variety, whole milk	−11.93	2	1	3	2
13130310	Light ice cream, chocolate (formerly ice milk)	−18.71	2	1	2	1
13110110	Ice cream, regular, chocolate	−32.38	1	1	2	1
14104010	Cheese, natural, Cheddar or American type	−35.98	1	1	1	1
13110130	Ice cream, rich, chocolate	−39.47	1	1	1	1
	MEATS, POULTRY, AND FISH
24122120	Chicken, breast, roasted, broiled, or baked, skin not eaten	37.87	3	3	5	4
26137120	Salmon, baked or broiled	34.10	3	3	5	4
21401120	Beef, roast, roasted, lean only eaten	21.31	3	2	4	4
27111410	Chili con carne with beans	20.02	3	2	4	4
28141250	Chicken with rice-vegetable mixture (diet frozen meal)	19.59	3	2	4	4
22101120	Pork chop, broiled or baked, lean only eaten	18.07	3	2	4	4
21501000	Ground beef, less than 80% lean, cooked (formerly regular)	−0.94	2	1	2	2
25210110	Frankfurter, wiener, or hot dog, NFS	−26.73	1	1	1	1
	EGGS
31103000	Egg, whole, boiled	5.04	1	2	2	2
	LEGUMES AND NUTS
42100100	Almonds, NFS	74.53	3	3	5	5
41104020	Pinto, calico, or red Mexican beans, dry, cooked, fat not added in cooking	66.28	3	3	5	5
41208030	Pork and beans	44.07	2	3	3	4
42110000	Mixed nuts, NFS	43.05	2	3	2	4
42202150	Peanut butter, reduced fat	31.91	1	3	2	4
42202000	Peanut butter	31.32	1	3	2	4
	GRAIN PRODUCTS
53540300	Fiber One Chewy Bar	80.61	3	3	5	5
53544450	PowerBar (fortified high energy bar)	50.30	3	3	5	5
57123000	Cheerios	49.47	3	3	5	5
51601020	Bread, multigrain	41.34	3	3	5	4
56203010	Oatmeal, cooked, regular, fat not added in cooking	35.46	3	3	5	4
54403000	Popcorn, popped in oil, unbuttered	25.90	3	3	5	4
51301010	Bread, wheat or cracked wheat	21.94	3	3	4	4
53542210	Granola bar, nonfat	21.27	3	2	4	4
53540400	Kellogg's Nutri-Grain Cereal Bar	20.27	3	2	4	4
58301110	Vegetable lasagna (frozen meal)	16.89	3	2	4	4
55201000	Waffle, plain	16.44	3	2	4	4
51101000	Bread, white	9.49	2	2	3	3
53234000	Cookie, peanut butter	7.15	2	2	2	3
57135000	Corn flakes, Kellogg's	3.87	1	2	2	2
58130011	Lasagna with meat	2.90	1	2	2	2
58304200	Ravioli, cheese-filled, with tomato sauce (diet frozen meal)	1.79	1	2	2	2
54101010	Cracker, animal	1.15	1	2	2	2
53204010	Cookie, brownie, without icing	0.78	1	2	2	2
58106225	Pizza, cheese, regular crust	0.07	1	2	2	2
54408010	Pretzels, hard	−0.41	1	2	2	2
58145110	Macaroni or noodles with cheese	−1.17	1	1	2	2
53121260	Cake, yellow, with icing, made from home recipe or purchased ready-to-eat	−6.86	1	1	1	2
	FRUIT
61119010	Orange, raw	115.15	3	3	5	5
61210250	Orange juice, with calcium added, canned, bottled or in a carton	91.76	3	3	5	5
63105010	Avocado, raw	82.10	3	3	5	5
63101000	Apple, raw	65.53	3	3	4	5
63143010	Plum, raw	58.07	2	3	4	5
61210620	Orange juice, frozen (reconstituted with water)	48.81	2	3	3	5
63107010	Banana, raw	45.35	2	3	2	5
63149010	Watermelon, raw	42.05	1	3	2	4
63101120	Applesauce, stewed apples, unsweetened	36.01	1	3	2	4
63123000	Grapes, raw, NS as to type	29.49	1	3	1	4
63135150	Peach, cooked or canned, drained solids	22.76	1	3	1	4
64104010	Apple juice	20.97	1	2	1	4
62125100	Raisins	20.19	1	2	1	4
	VEGETABLES
72125100	Spinach, raw	215.74	3	3	5	5
72201100	Broccoli, raw	163.50	3	3	5	5
71201010	White potato, chips	33.46	2	3	3	4
74303000	Tomato and vegetable juice, mostly tomato	26.00	2	3	3	4
71401030	White potato, French fries, from frozen, deep fried	18.85	2	2	3	4
71101000	White potato, baked, peel not eaten	17.85	2	2	3	4
75510000	Olives, NFS	17.23	2	2	3	4
	FATS, OILS, AND DRESSINGS
82104000	Olive oil	30.83	3	3	5	4
83107000	Mayonnaise, regular	20.85	3	2	4	4
81102000	Margarine, NFS	12.83	2	2	4	3
83104000	French dressing	6.50	2	2	3	3
83202020	French dressing, reduced calorie	−22.12	1	1	1	1
81101000	Butter, stick, salted	−90.35	1	1	1	1
83205500	Italian dressing, reduced calorie, fat-free	−115.13	1	1	1	1
	SWEETS AND BEVERAGES
92101000	Coffee, made from ground, regular	48.61	3	3	5	5
92550030	Fruit juice drink, low calorie, with high vitamin C	47.98	3	3	5	5
92306000	Tea, herbal	14.89	2	2	4	3
92410320	Soft drink, cola-type, sugar-free	8.07	2	2	3	3
91201010	Sugar substitute, aspartame-based, dry powder	1.66	2	2	3	2
92541010	Fruit flavored drink, made from powdered mix	−3.01	2	1	2	2
91401000	Jelly, all flavors	−13.25	1	1	2	2
91405500	Jelly, reduced sugar, all flavors	−14.83	1	1	2	2
91745020	Hard candy	−17.08	1	1	2	1
92410310	Soft drink, cola-type	−24.77	1	1	2	1
91101010	Sugar, white, granulated or lump	−26.86	1	1	1	1
92560100	Gatorade Thirst Quencher sports drink	−28.56	1	1	1	1
91705010	Milk chocolate candy, plain	−35.79	1	1	1	1

Among dairy foods, nonfat milk and yogurts were rated highest both across all foods and within the group of dairy foods. Low-fat milk and yogurt were also rated highest across all foods and within dairy foods with the three-category scheme, but using the five-category scheme they were rated highest (5) within the group and next highest (4) across foods. Whole milk was rated intermediate (2) within the group and lowest (1) across foods with the three-category scheme, and intermediate (3) within dairy foods and next lowest (2) across all foods using five categories.
Among the meats, poultry and fish group, baked chicken and salmon ranked highest within and across foods using three categories and highest (5) within the group and next highest (4) across foods using five categories.
Among legumes and nuts, almonds scored in the highest category both within the group and across foods using either three or five categories. Peanut butter had a score of lowest (1) within the legume and nut group despite scoring in the highest category (3) across all foods using three categories.
Grain products contain many varied foods. Foods that scored high both across foods and within the food group include fortified energy bars, Cheerios, oatmeal, and whole wheat bread. Some of the lowest scoring grains within the grain group with the three-category scheme scored intermediate (2) across foods (e.g., animal crackers, cheese pizza, and pretzels); with the five-category scheme, these foods scored next lowest (2) both within the group and across foods.
Fruits that scored highest across foods using three categories but lowest within the fruit group included applesauce, grapes, canned peaches, and watermelon. Grapes and peaches also scored lowest (1) within the fruit group using five categories and scored next highest (4) across foods.

Categorical ratings of the same 100 selected foods based on the final algorithm per RACC, NDSRACC, are presented in Table 4-11. There were some differences in categorical scores compared with the per 100 kcal rankings (Table 4-10), attributable to the differences in algorithm scores with the different unit bases; however, the differences were small, usually one category difference. Standardization of scores normalizes the distribution of scores. The normalization probably contributes to more similar categorical ratings per 100 kcal and per RACC than if the absolute scores were categorized without standardization and compared between the two unit bases.

Table 4-11. Categorical Ratings of 100 Foods Based on the per RACC Standardized Nutrient Density-Based Scores for Three- and Five-Category Ratings Within and Across Food Groups
Food Code	Food Description	Nutrient Density Score	Three- Category Ratings Within Food Groups	Three- Category Ratings Across Foods	Five- Category Ratings Within Food Groups	Five- Category Ratings Across Foods
Foods are sorted by major food groups and descending nutrient density-based scores. The three- and five-category ratings appear from highest (3 or 5) to lowest (1).
	DAIRY FOODS
11433500	Yogurt, fruit variety, nonfat milk, sweetened with low-calorie sweetener	64.61	3	3	5	5
11411300	Yogurt, plain, nonfat milk	60.71	3	3	5	5
11113000	Milk, cow's, fluid, skim or nonfat, 0.5% or less butterfat	40.05	3	3	5	5
11411200	Yogurt, plain, low-fat milk	30.01	3	3	5	5
11432500	Yogurt, fruit variety, low-fat milk, sweetened with low-calorie sweetener	25.40	3	3	5	4
11112210	Milk, cow's, fluid, 1% fat	22.39	3	3	5	4
11511400	Milk, chocolate, low-fat milk-based	22.02	3	3	5	4
11433000	Yogurt, fruit variety, nonfat milk	11.94	3	2	5	4
11511200	Milk, chocolate, reduced fat milk-based, 2% (formerly "low-fat")	5.31	3	2	5	4
14204010	Cheese, cottage, low-fat (1-2% fat)	2.82	3	2	4	3
11112110	Milk, cow's, fluid, 2% fat	1.47	3	2	4	3
12210200	Cream substitute, liquid	−1.65	3	2	4	2
14107030	Cheese, Mozzarella, part skim	−11.75	2	1	3	2
11432000	Yogurt, fruit variety, low-fat milk	−11.93	2	1	3	2
11511100	Milk, chocolate, whole milk-based	−16.62	2	1	3	2
11111000	Milk, cow's, fluid, whole	−18.83	2	1	2	2
13160410	Fat-free ice cream, chocolate	−19.43	2	1	2	1
11411100	Yogurt, plain, whole milk	−21.61	2	1	2	1
13130310	Light ice cream, chocolate (formerly ice milk)	−29.77	1	1	2	1
11431000	Yogurt, fruit variety, whole milk	−49.36	1	1	1	1
14104010	Cheese, natural, Cheddar or American type	−53.58	1	1	1	1
13110110	Ice cream, regular, chocolate	−54.48	1	1	1	1
13110130	Ice cream, rich, chocolate	−84.67	1	1	1	1
	MEATS, POULTRY, AND FISH
24122120	Chicken, breast, roasted, broiled, or baked, skin not eaten	38.00	3	3	5	5
27111410	Chili con carne with beans	37.80	3	3	5	5
26137120	Salmon, baked or broiled	33.16	3	3	5	5
28141250	Chicken with rice-vegetable mixture (diet frozen meal)	28.67	3	3	5	4
21401120	Beef, roast, roasted, lean only eaten	20.17	3	3	5	4
22101120	Pork chop, broiled or baked, lean only eaten	13.19	3	2	4	4
21501000	Ground beef, less than 80% lean, cooked (formerly regular)	−23.43	1	1	2	1
25210110	Frankfurter, wiener, or hot dog, NFS	−62.82	1	1	1	1
	EGGS
31103000	Egg, whole, boiled	−3.13	3	2	5	2
	LEGUMES AND NUTS
42100100	Almonds, NFS	114.43	3	3	5	5
41104020	Pinto, calico, or red Mexican beans, dry, cooked, fat not added in cooking	70.54	3	3	4	5
42110000	Mixed nuts, NFS	63.32	2	3	3	5
41208030	Pork and beans	52.82	2	3	2	5
42202150	Peanut butter, reduced fat	45.48	1	3	2	5
42202000	Peanut butter	41.54	1	3	2	5
	GRAIN PRODUCTS
53540300	Fiber One Chewy Bar	118.62	3	3	5	5
53544450	PowerBar (fortified high energy bar)	71.03	3	3	5	5
57123000	Cheerios	53.93	3	3	5	5
51601020	Bread, multigrain	52.70	3	3	5	5
56203010	Oatmeal, cooked, regular, fat not added in cooking	49.86	3	3	5	5
58301110	Vegetable lasagna (frozen meal)	33.95	3	3	5	5
54403000	Popcorn, popped in oil, unbuttered	33.70	3	3	5	5
55201000	Waffle, plain	31.56	3	3	5	5
53542210	Granola bar, nonfat	28.14	3	3	5	4
51301010	Bread, wheat or cracked wheat	25.78	3	3	5	4
53540400	Kellogg's Nutri-Grain Cereal Bar	24.33	3	3	5	4
51101000	Bread, white	8.82	2	2	3	4
57135000	Corn flakes, Kellogg's	2.76	2	2	2	3
53234000	Cookie, peanut butter	0.17	2	2	2	3
54408010	Pretzels, hard	−3.38	1	2	2	2
54101010	Cracker, animal	−4.12	1	2	2	2
58130011	Lasagna with meat	−4.89	1	2	2	2
53204010	Cookie, brownie, without icing	−7.18	1	1	2	2
58304200	Ravioli, cheese-filled, with tomato sauce (diet frozen meal)	−8.99	1	1	2	2
58106225	Pizza, cheese, regular crust	−17.70	1	1	2	2
58145110	Macaroni or noodles with cheese	−31.79	1	1	1	1
53121260	Cake, yellow, with icing, made from home recipe or purchased ready-to-eat	−38.51	1	1	1	1
	FRUIT
63105010	Avocado, raw	174.89	3	3	5	5
61210250	Orange juice, with calcium added, canned, bottled or in a carton	98.42	3	3	5	5
61119010	Orange, raw	93.81	3	3	5	5
63107010	Banana, raw	60.68	3	3	4	5
63101000	Apple, raw	51.91	2	3	3	5
61210620	Orange juice, frozen (reconstituted with water)	51.43	2	3	3	5
63143010	Plum, raw	39.59	1	3	2	5
63149010	Watermelon, raw	36.38	1	3	2	5
63123000	Grapes, raw, NS as to type	29.99	1	3	2	5
62125100	Raisins	25.39	1	3	1	4
64104010	Apple juice	24.20	1	3	1	4
63135150	Peach, cooked or canned, drained solids	23.85	1	3	1	4
63101120	Applesauce, stewed apples, unsweetened	23.02	1	3	1	4
	VEGETABLES
72201100	Broccoli, raw	77.00	3	3	5	5
72125100	Spinach, raw	50.92	3	3	5	5
71201010	White potato, chips	42.89	3	3	5	5
74303000	Tomato and vegetable juice, mostly tomato	32.55	3	3	4	5
71401030	White potato, French fries, from frozen, deep fried	32.16	3	3	4	5
71101000	White potato, baked, peel not eaten	18.20	2	3	2	4
75510000	Olives, NFS	1.04	1	2	2	3
	FATS, OILS, AND DRESSINGS
82104000	Olive oil	22.85	3	3	5	4
83107000	Mayonnaise, regular	10.45	3	2	5	4
81102000	Margarine, NFS	1.05	3	2	4	3
83104000	French dressing	−4.38	2	2	3	2
83205500	Italian dressing, reduced calorie, fat-free	−18.07	1	1	2	2
83202020	French dressing, reduced calorie	−20.25	1	1	2	1
81101000	Butter, stick, salted	−98.00	1	1	1	1
	SWEETS AND BEVERAGES
92550030	Fruit juice drink, low calorie, with high vitamin C	43.95	3	3	5	5
92101000	Coffee, made from ground, regular	0.95	2	2	4	3
92306000	Tea, herbal	0.30	2	2	3	3
92410320	Soft drink, cola-type, sugar-free	0.15	2	2	3	3
91201010	Sugar substitute, aspartame-based, dry powder	0.05	2	2	3	3
91101010	Sugar, white, granulated or lump	−5.12	2	2	2	2
91405500	Jelly, reduced sugar, all flavors	−6.46	1	1	2	2
92541010	Fruit flavored drink, made from powdered mix	−6.95	1	1	2	2
91401000	Jelly, all flavors	−8.58	1	1	2	2
91745020	Hard candy	−12.39	1	1	2	2
92560100	Gatorade Thirst Quencher sports drink	−21.33	1	1	1	1
92410310	Soft drink, cola-type	−27.31	1	1	1	1
91705010	Milk chocolate candy, plain	−86.49	1	1	1	1

5. Conclusions

5.1 Summary of Findings

Although the HEI scores total diet, it is inherently different from a food-scoring algorithm; nevertheless, it provides a mechanism to evaluate how well the individual food scores of an individual's diet relate to overall diet quality. In this report, we demonstrate that a final algorithm with weighting factors for nutrients that were derived from statistical analyses of nutrient intakes of the U.S. population resulted in higher prediction of dietary quality than seen with existing nutrient algorithms that have been tested similarly. Our final algorithm explained two-thirds of the variation in HEI scores, compared with one-third to one-half with other nutrient density algorithms. Our algorithm included nutrients or food components with positive weighting factors for protein, unsaturated fat, fiber, calcium, and vitamin C and with negative weighting factors for saturated fat, sodium, and added sugars. The use of nutrient values per 100 kcal was slightly better at predicting overall dietary quality than using nutrients per reference serving sizes (RACC). Among the top-scoring foods were raw and leafy green vegetables on a per 100 kcal basis and avocado, almonds, oranges, and strawberries on a per RACC basis. The algorithm worked well in predicting dietary quality across various population groups, such as age, ethnicity, socioeconomic status, and weight status.

Numerous criteria and considerations must be accounted for when developing a nutrient scoring system for foods. The selection of nutrients or food components is probably the most critical component of the process. We began by selecting nutrients that were deemed important and were limited in the diets of Americans. However, nutrients coexist in foods, and the addition of nutrients to a scoring system does not necessarily help improve prediction of overall dietary quality. The unit basis for the nutrient data is important because it determines the amount of the nutrient that is included in the calculation of the score. We found that when examining scores of foods, the algorithms based on RACC servings seemed to reduce extreme values for some foods such as low-calorie vegetables. A scoring system based on RACC servings is intuitively appealing because it accounts for serving size differences among various types of foods. However, the algorithms based on RACC servings were slightly less predictive of overall dietary quality. More extreme scores using algorithms based on 100 kcal servings, particularly for fruits and vegetables, may have driven the higher R² values of models predicting HEI. The fact that HEI is based on nutrient standards on a per 1,000 kcal basis could also contribute to a better prediction by an algorithm scored on a per 100 kcal basis.

Our original baseline algorithm and modifications that weighed positive and negative nutrients equally performed reasonably well in predicting HEI scores. Approximately 50% of the variance in HEI score was explained by the respective algorithms. Weighting of nutrients has been used in very few existing nutrient scoring systems because there is a lack of concrete scientific evidence to support specific weighting factors to apply to nutrients. Our new approach, developed by Nutrition Impact, LCC, used weighting factors obtained from beta coefficients of nutrient intake variables in regression models predicting HEI scores. The final nutrient density algorithm resulted in greater prediction of dietary quality assessed by HEI score than our baseline algorithm and its modifications. The final algorithms predicted approximately two-thirds of the variance in HEI scores (R² of 65% for the 100 kcal- and 60% for RACC-based algorithm) compared with one-half of the variance explained by the modified algorithm with vitamin C and whole grains (R² of 52% for the 100 kcal- and 44% for RACC-based algorithm). Previous published validation studies of nutrient density indexes reported R² values of 45% with the NRFI (Fulgoni et al., 2009) and 29% with the ONQI associated with the NuVal shelf-labeling system (Katz et al., 2010). Additional regression models with our final algorithm using various subpopulations demonstrated R² values comparable to those for the overall population. Further analyses of the top 10 eight-nutrient or food component models showed that the R² values were extremely close to each other, suggesting that a number of potential algorithms include eight nutrients or food components that would be roughly equivalent in predicting dietary quality based on the HEI.

Summary systems can be simplified for the consumer by categorizing a score into different levels. For example, categories used by others have included three levels, such as traffic light colors or text signifying "low, medium, or high." We assessed categories that used both three- and five-point categorization of scores using the final algorithm. This resulted in reasonable rankings of foods based on three- or five-point ratings. The three-category system performed as well as the five-category system in distinguishing between common foods (e.g., whole grain vs. white bread, nonfat vs. whole-milk yogurt). The use of such three- or five-category rankings of foods may be more helpful to consumers than a continuous score, although this needs to be tested with consumers. Categorical rankings of foods using this algorithm could also be compared with other existing ranking systems.

5.2 Limitations

Some of the limitations of our design are described in this section. The scoring of foods using the algorithms depends on the nutrient database used. The FNDDS has some limitations. For example, trans fat is not in FNDDS, and it would have been time prohibitive to add this information to each food in the database. Another limitation of the FNDDS is that the database uses a default choice of added salt, which resulted in high sodium content for some foods (e.g., baked potato) that was reflected in lower algorithm scores (the dietary recall in NHANES asks respondents if they added salt to foods at the time of eating, and the dietary intake data reflect the respondent's answer). The use of the major food grouping that is used in FNDDS to examine scores by food groups was limiting. The classification of some food items into major groupings was awkward, such as pickles and salsa in the vegetable group. Because of the high sodium content, various condiments had very low scores and skewed the overall vegetable score. Some food groups had a smaller number of foods and could be skewed by outliers. In theory, we could have examined scores of all 4,000+ foods in FNDDS, but it would have been difficult to assess the scores of that number of foods given the time constraints.

The selection of nutrients would need to be reassessed in the future as new evidence emerges for nutrients of importance and new federal dietary guidance is issued. The daily values used in the algorithm would also need to be updated when daily values change based on new federal dietary recommended intake guidance. The nutrient values per RACC would also need to be updated if standard serving sizes change in the future.

The two methods used in the project to test the algorithms for construct validity were to examine food scores for reasonableness and to validate composite algorithm scores for dietary intake against a measure of overall dietary quality. Examination of food scores is an important step but is a descriptive process rather than a formal quantitative assessment. It is difficult to compare food scores among all the algorithm modifications and identify which set of food scores is the best representative of nutritional quality. Validation against a measure of diet quality is a recommended approach to evaluate a nutrient scoring system, but it depends on the validity of the index used to measure dietary quality. The HEI is based on federal dietary recommendations found in MyPyramid and the 2005 Dietary Guidelines and has been validated (Guenther, Reedy, Krebs-Smith, & Reeve, 2008). The relationship between HEI and a nutrient scoring system also depends on the particular components in each score and how much these components influence the overall score. The HEI validation study demonstrated that certain components of the HEI have more influence on the score, that is, calories from SOFAAS (solid fats, alcohol and added sugars) and fruits (Guenther et al., 2008). Our algorithms demonstrated high food scores for fruits and showed better agreement with HEI scores when added sugar was present in the algorithm. With the new 2010 Dietary Guidelines, the HEI may be updated and it will be important to reevaluate the algorithm.

The focus of our formal validation was the ability of the algorithm to predict dietary quality, an important property of a summary scoring system. Nevertheless, if a summary scoring system is used in FOP labeling, the score will be shown on each food, and the ability of the algorithm to validly represent the nutritional value of each food would also be important. We qualitatively assessed the scores of foods and food groups, but we did not have an available method for formal validation of individual food scores.

The final algorithms were derived from dietary intake data from NHANES 2005-2008. Specifically, the weighting factors are the beta coefficients from regressions models of nutrient intakes against HEI scores. The reliability of the algorithm with other dietary intake data sets is unknown and will be important to test with future NHANES data when it becomes available.

Our results apply only to summary FOP systems; we did not consider nutrient-specific systems that are the other general type of FOP system. On the one hand, nutrient-specific systems can help consumers identify key nutrients in a food, but they do not provide an overall assessment of the product. On the other hand, summary systems can provide an overall assessment of the nutritional quality of a food, but they do not tell the consumer why the food received the score or rating (e.g., high calcium and low saturated fat content of low-fat milk).

5.3 Identification of Knowledge Gaps and Next Steps

Further analyses of our final algorithms could include the following next steps:

assessing additional nutrients using the new approach to identify nutrients for a nutrient density algorithm,
assessing alternative weighting of nutrients,
creating a database of additional brand-specific foods and a calculator to determine scores using the selected system,
comparing scores or rankings for foods with rankings of foods using other existing summary systems,
conducting initial consumer testing of the selected FOP summary system,
conducting an experimental study to measure changes in consumer behavior in response to the system, and
studying consumer response and behavior for an identified summary system algorithm in comparison with consumer response and behavior for nutrient-specific systems.

Several important issues regarding determination of an index to measure nutritional quality of foods have been addressed in this report. Some general questions still remain regarding the development and testing of nutrient density-based FOP systems.

Which positive nutrients are most important in lowering risk of chronic disease? The evidence for single "positive" nutrients in the prevention of chronic disease is not as well established as for "negative" nutrients such as saturated fat. Longitudinal studies of diet and disease relationships are needed to help identify and quantify (for weighting) nutrients that best predict lower disease risk.

What is the best method to validate a nutrient scoring system? We discussed previously the limitations of validation methods. There is no gold standard to evaluate how well a system ranks foods and predicts chronic disease risk. Longitudinal studies will be needed to test the effects of using the system with consumers on dietary quality and health outcomes.

How well will consumers understand and use a nutrient density-based system? Addressing this issue was not within the scope of this project. Knowledge about the ability for consumers and subpopulations who are at particular risk of obesity and chronic disease to understand and use a nutrient density system is extremely important.

How do consumer understanding and use of a nutrient density system compare with other summary systems and with nutrient-specific systems? Previous research suggests that consumers can identify healthier food choices more easily with a nutrient-specific system with traffic light colors of red, yellow, and green than a summary check mark type system (Hersey, Wohlgenant, Kosa, Arsenault, & Muth, 2011). Assessments of consumer understanding of nutrient-specific systems compared to nutrient density scoring systems or graded rating systems are limited.

References

Arambepola, C., Scarborough, P., & Rayner, M. (2007). Validating a nutrient profile model. Public Health Nutrition, 11(4), 371-378.

Bowman, S. A., Friday, J. E., & Moshfegh, A. (2008). MyPyramid Equivalents Database, 2.0 for USDA Survey Foods 2003-2004. Beltsville, MD: U.S. Department of Agriculture, Agricultural Research Service, Food Surveys Research Group, http://www.ars.usda.gov/ba/bhnrc/fsrg.

Chiuve, S. E., Sampson, L., & Willett, W. C. (2011). The association between a nutritional quality index and risk of chronic disease. American Journal of Preventive Medicine, 40(5), 505-513.

Code of Federal Regulations. (1993a). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart E - Specific Requirements for Health Claims. § 101.74 Health claims: Sodium and hyptertension.

Code of Federal Regulations. (1993b). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart E - Specific Requirements for Health Claims. § 101.75 Health claims: Dietary saturated fat and cholesterol and risk of coronary heart disease.

Code of Federal Regulations. (1996). Title 21: Food and Drugs, Parts 136, 137, and 139 - Food Standards: Amendment of Standards of Identity for Enriched Grain Products to Require Addition of Folic Acid.

Code of Federal Regulations. (2000 amended). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart E - Specific Requirements for Health Claims. § 101.79 Health claims: Folate and neural tube defects.

Code of Federal Regulations. (2001 amended). Title 21, Food and drugs. Part 101-Food Labeling. Subpart A-General Provisions, Sec. 101.14 Health claims: General requirements.

Code of Federal Regulations. (2008 amended-a). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart E - Specific Requirements for Health Claims. § 101.72 Health claims: Calcium, vitamin D, and osteoporosis.

Code of Federal Regulations. (2008 amended-b). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart E - Specific Requirements for Health Claims. § 101.81 Health claims: Soluble fiber from certain foods and risk of coronary heart disease (CHD).

Code of Federal Regulations. (2010a). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart A - General Provisions. § 101.9 Nutrition labeling of food.

Code of Federal Regulations. (2010b). Title 21: Food and Drugs, Part 101 - Food Labeling, Subpart A - General Provisions. § 101.12 Reference amounts customarily consumed per eating occasion.

Connor, W. E., & Connor, S. L. (2002). Dietary cholesterol and coronary heart disease. Current Atherosclerosis Reports, 4, 425-432.

Drewnowski, A., Maillot, M., & Darmon, N. (2009). Should nutrient profiles be based on 100 g, 100 kcal or serving size? European Journal of Clinical Nutrition, 63, 898-904.

Food and Nutrition Board. Institute of Medicine. (2002). Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, DC: National Academy Press.

Fulgoni, V. L., III, Keast, D. R., & Drewnowski, A. (2009). Development and validation of the nutrient-rich foods index: a tool to measure nutritional quality of foods. Journal of Nutrition, 139, 1549-1554.

Guenther, P. M., Reedy, J., Krebs-Smith, S. M., & Reeve, B. B. (2008). Evaluation of the Healthy Eating Index-2005. Journal of American Dietetic Association, 108, 1854-1864.

Guenther, P. M., Reedy, J., Krebs-Smith, S. M., Reeve, B. B., & Basiotis, P. P. (2007). Development and Evaluation of the Healthy Eating Index-2005. Technical Report (Vol. ). Center for Nutrition Policy and Promotion, U.S. Department of Agriculture. Available at http://www.cnpp.usda.gov/HealthyEatingIndex.htm.

Hersey, J. C., Wohlegenant, K. C., Kosa, K. M., Arsenault, J. E., & Muth, M. K. (2011). Policy Research for Front of Package Nutrition Labeling: Environmental Scan and Literature Review. Research Triangle Park: RTI International.

Howarth, N. C., Saltzman, E., & Roberts, S. B. (2001). Dietary fiber and weight regulation. Nutrition Reviews, 59(5), 129-139.

Katz, D. L. (2007). ONQI Version 1 Abbreviated Reference Manual. Derby, CT: Griffin Hospital.

Katz, D. L., Njike, V. Y., Faridi, Z., Rhee, L. Q., Reeves, R. S., Jenkins, D. J. A., & Ayoob, K. T. (2009). The stratification of foods on the basis of overall nutritional quality: the Overall Nutritional Quality Index. American Journal of Health Promotion, 24(2), 133-143.

Katz, D. L., Njike, V. Y., Rhee, L. Q., Reingold, A., & Ayoob, K. T. (2010). Performance characteristics of NuVal and the Overall Nutritional Quality Index (ONQI). American Journal of Clinical Nutrition, 91(Suppl), 1102S-1108S.

Law, M. R., Frost, C. D., & Wald, N. J. (1991). By how much does dietary salt reduction lower blood pressure? III. Analysis of data from trials of salt reduction. British Medical Journal, 302, 819-824.

Lupton, J. R., Balentine, D. A., Black, R. M., Hildwine, R., Ivens, B. J., Kennedy, E. T., . . . Story, M. (2010). The Smart Choices front-of-package nutrition labeling program: rationale and development of the nutrition criteria. American Journal of Clinical Nutrition, 91(Suppl), 1078S-1089S.

Malik, V. S., Schulze, M. B., & Hu, F. B. (2006). Intake of sugar-sweetened beverages and weight gain: a systematic review. American Journal of Clinical Nutrition, 84, 274-288.

Mensick, R. P., Zock, P. L., Kester, A. D. M., & Katan, M. B. (2003). Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. American Journal of Clinical Nutrition, 77, 1146-1155.

Mozaffarian, D., Katan, M. B., Ascherio, A., Stampfer, M. J., & Willett, W. C. (2006). Trans fatty acids and cardiovascular disease. New England Journal of Medicine, 354, 1601-1613.

Neuman, W. (2009). For your health, froot loops. The New York Times. Retrieved from http://www.nytimes.com/2009/09/05/business/05smart.html?_r=3.

Nijman, C. A. J., Zijp, I. M., Sierksma, A., Roodenburg, A. J. C., Leenen, R., van den Kerkhoff, C., . . . Meijer, G. W. (2007). A method to improve the nutritional quality of foods and beverages based on dietary recommendations. European Journal of Clinical Nutrition, 61, 461-471.

Obarzanek, E., Proschan, M. A., Vollmer, W. M., Moore, T. J., Sacks, F. M., Appel, L. J., . . . Cutler, J. A. (2003). Individual blood pressure responses to changes in salt intake. Results from the DASH-sodium trial. Hypertension, 42, 459-467.

Queenan, K. M., Stewart, M. L., Smith, K. N., Thomas, W., Fulcher, G., & Slavin, J. L. (2007). Concentrated oat beta-glucan, a fermentable fiber, lowers serum cholesterol in hypercholesterolemic adults in a randomized controlled trial. Nutrition Journal, 6, 6. doi: 10.1186/1475-2891-6-6

Quinio, C., Biltoft-Jensen, A., De Henauw, S., Gibney, M. J., Huybrechts, I., McCarthy, S. N., . . . Volatier, J. L. (2007). Comparison of different nutrient profiling schemes to a new reference method using dietary surveys. European Journal of Nutrition, 46(Suppl 2), 37-46.

Salmerón, J., Ascherio, A., Rimm, E. B., Colditz, G. A., Spiegelman, D., Jenkins, D. J., . . . Willett, W. C. (1997). Dietary fiber, glycemic load, and risk of NIDDM in men. Diabetes Care, 20(4), 545-550.

Scarborough, P., Arambepola, C., Kaur, A., Bhatnagar, P., & Rayner, M. (2010). Should nutrient profile models be "category specific" or "across-the-board"? A comparison of the two systems using diets of British adults. European Journal of Clinical Nutrition, 64(6), 553-560. doi: 10.1038/ejcn.2010.31

Scarborough, P., Boxer, A., Rayner, M., & Stockley, L. (2007). Testing nutrient profile models using data from a survey of nutritional professionals. Public Health Nutrition, 10(4), 337-345.

Scarborough, P., Rayner, M., Stockley, L., & Black, A. (2007). Nutrition professionals' perception of the 'healthiness' of individual foods. Public Health Nutrition, 10(4), 346-353.

Tetens, I., Oberdorfer, R., Madsen, C., & de Vries, J. (2007). Nutritional characterisation of foods: science-based approach to nutrient profiling. European Journal of Nutrition, 46 (Suppl 2), 4-14.

Townsend, M. S. (2010). Where is the science? What will it take to show that nutrient profiling systems work? American Journal of Clinical Nutrition, 91(Suppl), 1109S-1115S.

Truswell, A. S. (2002). Cereal grains and coronary heart disease. European Journal of Clinical Nutrition, 56, 1-14.

U.S. Department of Agriculture. (2009). USDA National Nutrient Database for Standard Reference, Release 22. Beltsville, MD: Agricultural Research Service, Nutrient Data Laboratory Home Page, http://www.ars.usda.gov/ba/bhnrc/ndl.

U.S. Department of Agriculture. (2010a). MyPyramid Retrieved May 11, 2010, from http://www.mypyramid.gov.

U.S. Department of Agriculture. (2008). USDA Food and Nutrient Database for Dietary Studies, 3.0. Beltsville, MD: U.S. Department of Agriculture, Agricultural Research Service, Food Surveys Research Group.

U.S. Department of Agriculture. (2010b). USDA Food and Nutrient Database for Dietary Studies, 4.0. Beltsville, MD: U.S. Department of Agriculture, Agricultural Research Service, Food Surveys Research Group.

U.S. Department of Agriculture & U.S. Department of Health and Human Services. (2010). Dietary Guidelines for Americans, 2010. 7th Edition. Washington, DC: U.S. Government Printing Office.

U.S. Department of Health and Human Services, & U.S. Department of Agriculture. (2005). Dietary Guidelines for Americans, 2005. 6th Edition. Washington, DC: U.S. Government Printing Office.

Volatier, J. L., Biltoft-Jensen, A., De Henauw, S., Gibney, M. J., Huybrechts, I., McCarthy, S. N., . . . Tetens, I. (2007). A new reference method for validation of the nutrient profiling schemes using dietary surveys. European Journal of Nutrition, 46(Suppl 2), 29-36.

WHO. (2003). Diet, Nutrition and the Prevention of Chronic Disease. Report of a joint WHO/FAO Expert Consultation. WHO Technical Report Series 916. Geneva: World Health Organization.

Willett, W. C. (2002). Dietary fat plays a major role in obesity: no. Obesity Reviews, 3(2), 59-68.

End Notes

¹ Initially, soy sauce was in the top 500 commonly consumed foods, but we eliminated it because its high sodium content was unduly influencing scores for the legume food group.

Appendixes

A: Detailed Information and Nutrient Criteria for FOP Systems and Algorithms

GUIDING STARS

The Guiding Stars is a shelf label and nutrition guidance program produced for Hannaford Supermarkets. The system rates products and awards "stars": No stars mean does not meet nutritional criteria to earn a star; 1 star means "good" nutritional value; 2 stars means "better"; and 3 stars mean "best." The criteria apply to 100 kcal of the food. The system was developed by a scientific advisory panel and is owned by the Guiding Stars Licensing Company. The system is not completely transparent, the complete nutrient criteria for each star rating are not available as of this publication.

General criteria: Note that this table contains the core elements of the algorithm at their base levels per 100 calories—meeting these base levels would result in a rating of 3 "stars." Products that meet some but not all of these criteria might still rate 1 or 2 "stars" if the positive attributes outweigh the negative attributes.

Nutrients/ Components	General Foods and Beverages	Meats/Seafood/ Dairy and Nuts	Fats and Oils	Infant and Toddler Foods
Source: Guiding Stars. Guidance on understanding the science. Retrieved from http://www.guidingstars.com/wp-content/uploads/2009/11/guiding_stars_science.pdf (Accessed May 9, 2010).
Vitamins and minerals (not specified, except infant and toddler foods)	≥10% of DV of 2 per 100 kcal	≥10% of DV of 2 OR ≥20% of DV of 1 per 100 kcal	NA	≥10% of DV of 2 or ≥20% of DV of 1 per 100 kcal (vitamins A and C, calcium, iron, zinc)
Dietary fiber	≥3.75g per 100 kcal	≥1.25g per 100 kcal	NA	NA
Whole grains	≥1.5g dietary fiber per 100 kcal plus presence of whole grain key word in ingredients	NA	NA	NA
Omega-3 fatty acids	NA	NA	≥6.7g per 100 kcal	NA
DHA and EPA	NA	NA	A bonus point if contains both	NA
Saturated fat	≤1g per 100 kcal	≤1.5g per 100 kcal	≤2.2g per 100 kcal	NA
Trans fat	0g on NFP and no partially hydrogenated oil in ingredients	0g on NFP and no partially hydrogenated oil in ingredients	no partially hydrogenated oil in ingredients	NA
Cholesterol	≤15 mg per 100 kcal	≤60 mg per 100 kcal	≤15 mg per 100 kcal	NA
Added sugar	None	None	None	None
Added sodium	≤120 mg per 100 kcal	≤120 mg per 100 kcal	≤120 mg per 100 kcal	≤100 mg per 100 kcal

Heart Check Mark

The Heart Check Mark is the American Heart Association's (AHA) FOP system. The icon is awarded to products that meet the criteria. The nutrient criteria are transparent, and apply to 6 food categories and are based on reference serving amounts. The nutrient criteria are based on FDA health claim regulations, USDA extra lean meat requirements, and AHA Scientific Statements on diet/nutrition topics. The system is reviewed and updated with updates to FDA health claims.

Nutrients/ Components	Standard (Non-meat)	Standard (Meat and Seafood, i.e., Extra Lean)	Whole Oats Soluble Fiber	Whole Grain	Main Dish Products (≥6 Ounces Per Labeled Serving)	Meals or Meal-Type Products (≥10 Ounces Per Labeled Serving)
Source: http://www.heart.org/HEARTORG/GettingHealthy/NutritionCenter/HeartSmartShopping/Heart-Check-Mark_UCM_300914_Article.jsp (Accessed September 9, 2010); Criteria for main dishes and meals were obtained from: Kim Stizel, Presentation at Institute of Medicine Committee on Examination of Front of Package Nutrition Rating Systems and Symbols. April 9, 2010 at http://www.iom.edu/Activities/Nutrition/NutritionSymbols/2010-APR-08.aspx (Accessed April 29, 2010).
Positive nutrients/ components: Vitamin A, vitamin C, iron, calcium, protein or dietary fiber	≥10% of DV of 1 out of 6 nutrients per RACC	≥10% of DV of 1 out of 6 nutrients per RACC	≥10% of DV of 1 out of 6 nutrients per RACC	≥10% of DV of 1 out of 6 nutrients per RACC	≥10% of DV of 1 out of 6 nutrients per entire main dish meal	≥10% of DV of 1 out of 6 nutrients per entire main dish meal
Whole grain	NA	NA	NA	≥51% by weight/RACC	NA	NA
Minimum dietary fiber	NA	NA	NA	1.7 g/RACC of 30g 2.5 g/RACC of 45g 2.8 g/RACC of 50g 3.0 g/RACC of 55g	NA	NA
Whole oat soluble fiber (beta-glucan)	NA	NA	≥0.75g whole oat soluble fiber (Must contain beta-glucan from oat bran, rolled oats, whole oat flour, or oatrim) per RACC	NA	NA	NA
Negative nutrients/ components: Total fat	≤3g per RACC	≤5g per RACC and 100g	≤3g per RACC unless fat content is solely derived from whole oat sources	<6.5g per RACC	≤3g per 100g and ≤30% of calories	≤3g per 100g and ≤30% of calories
Saturated fat	≤1g per RACC	≤2g per RACC and 100g	≤1g per RACC	≤1g per RACC	≤1g per 100g and ≤10% of calories	≤1g per 100g and ≤10% of calories
Trans fat	≤0.5g per RACC and labeled serving	≤0.5g per RACC and labeled serving	≤0.5g per RACC and labeled serving	≤0.5g per RACC and labeled serving	≤0.5g per RACC and labeled serving	≤0.5g per RACC and labeled serving
Cholesterol	≤20 mg per RACC	≤95 mg per RACC and 100g	≤20 mg per RACC	≤20 mg per RACC	≤20 mg per 100g	≤20 mg per 100g
Sodium	≤480 mg per RACC and labeled serving	≤480 mg per RACC and labeled serving	≤480 mg per RACC and labeled serving	≤480 mg per RACC and labeled serving	≤600 mg per labeled serving	≤600 mg per labeled serving

Smart Choices Program

The Smart Choices is a FOP system that awards the icon and provides additional information on calories per serving and the number of servings per package. Nutrient criteria exist for 19 categories and are based on reference portion sizes. The criteria are transparent. The system was developed by a coalition of scientists, food manufacturers, and retailers. The system is not currently active.

*General benchmark criteria*
Nutrients to Limit	Criteria
Total fat	≤35% of calories
Saturated fat	<10% of calories
Trans fat	0g (labeled)
Cholesterol	≤60 mg per serving
Added sugars	≤25% of total calories
Sodium	≤480 mg per serving

Nutrients and Food Groups to Encourage	Criteria
Nutrients: Vitamin A, vitamin C, vitamin E, calcium, potassium, fiber, magnesium	A food must offer ≥10% Daily Value (a "good source" of at least one of these nutrients
Food groups: Fruits, vegetables, whole grains, fat-free/low-fat milk products	A food must provide at least ½ of a serving of one of these food groups

*Product categories and qualifying criteria*
Product Category	Qualifying Criteria
Source: http://www.smartchoicesprogram.com/nutrition.html (Accessed October 1, 2010).
Fruits and vegetables (with no additive)	Automatically qualify
Fruits and vegetables (with additives), 100% juice	Nutrients to limit and at least one nutrient or food group to encourage
Breads, grains, pasta	Nutrients to limit and at least one nutrient or food group to encourage
Cereals	Nutrients to limit and at least one nutrient or food group to encourage
Meat, fish, poultry	Nutrients to limit only
Meat alternatives	Nutrients to limit and at least one nutrient or food group to encourage
Seeds, nuts, nut butters	Nutrients to limit only
Cheeses and cheese substitutes	Nutrients to limit and at least one nutrient or food group to encourage
Milk, dairy products, dairy substitutes (including soy beverages)	Nutrients to limit and at least one nutrient or food group to encourage
Fats, oils, spreads (including butter)	Nutrients to limit only
Soups, meal sauces, and mixed side dishes	Nutrients to limit and at least one nutrient or food group to encourage
Entrees, sandwiches, main dishes, meal replacements	Nutrients to limit and at least one nutrient or food group to encourage
Meal	Nutrients to limit and at least one nutrient or food group to encourage, and 1.5 servings from food group to encourage
Sauces, dressing, condiments	Nutrients to limit and ≥1 nutrient or food group to encourage
Snack foods and sweets	Nutrients to limit and at least one nutrient or food group to encourage
Desserts	Nutrients to limit and ≥1 nutrient or food group to encourage
Beverages	Nutrients to limit applies to all beverages, then: if ≤20 calories/serving: nutrients to limit only if ≤40 calories/serving: at least one nutrient or food group to encourage if ≤60 calories/serving: at least one nutrient and food group to encourage (4 oz juice = 1 food group)
Water (plain and carbonated)	Automatically qualify
Chewing gum	Nutrients to limit and "sugar-free"

*Detailed criteria for nutrients to limit (per labeled serving unless otherwise noted)*
Product Category	Kcal	Total Fat	Sat Fat	Trans Fat	Chol	Added Sugars	Sodium
Notes: The calories standards for meals was set by dividing the 2000 kcal/d standard from the DGA by 3 meals (600 kcal each) and snacks (≤200 kcal).
Generic benchmarks	NA	≤35% kcal	<10% kcal	0g (labeled)	≤60 mg	≤25% kcal	≤480 mg
Fruits and vegetables (with no additive)	NA	NA	NA	NA	NA	NA	NA
Fruits and vegetables (with additives), 100% juice	NA	≤3g	≤1g	0g (labeled)	NA	≤8 kcal (or 0g for 100% juice)	≤240 mg
Breads, grains, pasta	NA	≤35% kcal	<10% kcal	0g (labeled)	NA	≤25% kcal	≤240 mg
Cereals	NA	≤35% kcal	<10% kcal	0g (labeled)	NA	≤12g	≤240 mg (<43g svg), ≤290 mg (≥43g svg)
Meat, fish, poultry	NA	≤5g per RACC (and per 100 g); no limit for fatty fish if DHA/EPA is ≥500 mg / 3 oz	≤2g per RACC (and per 100 g)	0g (labeled) (naturally occurring trans fat excluded)	≤95 mg per RACC (and per 100 g)	≤25% kcal	≤140 mg if single ingredient raw; otherwise ≤480 mg
Meat alternatives	NA	≤35% kcal	<10% kcal	0g (labeled)	≤60 mg	≤25% kcal	≤480 mg
Seeds, nuts, nut butters	NA	NA	≤28% fat kcal	0g (labeled)	NA	≤25% kcal	≤240 mg
Cheeses and cheese substitutes	NA	≤3g	≤2g	0g (labeled) (naturally occurring trans fat excluded)	≤60 mg	≤25% kcal	≤240 mg
Milk, dairy products, dairy substitutes (including soy beverages)	NA	≤3g	≤2g	0g (labeled) (naturally occurring trans fat excluded)	≤60 mg	≤12g per cup	≤240 mg
Fats, oils, spreads (including butter)	NA	NA	≤28% fat kcal	0g (labeled)	≤60 mg	≤25% kcal	≤140 mg
Soups, meal sauces, and mixed side dishes	NA	≤35% kcal OR 3g	<10% kcal	0g (labeled)	≤60 mg	≤25% kcal OR≤6g	≤480 mg
Entrees, sandwiches, main dishes, meal replacements	≤450	≤35% kcal	<10% kcal	0g (labeled)	≤90 mg	≤25% kcal	≤600 mg
Meals	≤600	≤35% kcal	<10% kcal	0g (labeled)	≤90 mg	≤25% kcal	≤600 mg
Sauces, dressing, condiments	≤100	NA	≤28% fat kcal	0g (labeled)	≤30 mg	≤25% kcal OR ≤6g	≤240 mg
Snack foods and sweets	≤160	≤35% kcal OR ≤3g	<10% kcal OR ≤1g	0g (labeled)	≤60 mg	≤25% kcal OR ≤6g	≤240 mg
Desserts	≤200	≤35 kcal OR ≤3g	<10% kcal OR ≤1g	0g (labeled)	≤60 mg	≤25% kcal OR ≤6g	≤240 mg
Beverages	≤20 ≤40 ≤60	≤3g	≤1g	0g (labeled)	≤60 mg	NA	≤140 mg
Water (plain and carbonated)	NA	NA	NA	NA	NA	NA	NA
Chewing gum	≤20	0g	0g	0g (labeled)	NA	0g	NA

*Detailed criteria for nutrients to encourage*
Product Category	Calcium	Potassium	Fiber	Magn	Vitamin A	Vitamin C	Vitamin E
Generic benchmarks	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Fruits and vegetables (with no additive)	NA	NA	NA	NA	NA	NA	NA
Fruits and vegetables (with additives), 100% juice	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Breads, grains, pasta	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Cereals	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Meat, fish, poultry	NA	NA	NA	NA	NA	NA	NA
Meat alternatives	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Seeds, nuts, nut butters	NA	NA	NA	NA	NA	NA	NA
Cheeses and cheese substitutes	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Milk, dairy products, dairy substitutes (including soy beverages)	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Fats, oils, spreads (including butter)	NA	NA	NA	NA	NA	NA	NA
Soups, meal sauces, and mixed side dishes	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Entrees, sandwiches, main dishes, meal replacements	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Meal	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Sauces, dressing, condiments	NA	NA	NA	NA	NA	NA	NA
Snack foods and sweets	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Desserts	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Beverages	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV	≥10% DV
Water (plain and carbonated)	NA	NA	NA	NA	NA	NA	NA
Chewing gum	NA	NA	NA	NA	NA	NA	NA

*Detailed criteria for food groups to encourage*
Product Category	Fruits	Veg	Whole Grains	Fat-Free/Low- Fat Milk Products
Generic benchmarks	½ svg (¼ cup)	½ svg (¼ cup)	½ svg (8 g)	½ svg (½ cup)
Fruits and vegetables (with no additive)	NA	NA	NA	NA
Fruits and vegetables (with additives), 100% juice	½ svg (¼ cup)	½ svg (¼ cup)	½ svg (8 g)	½ svg (½ cup)
Breads, grains, pasta	½ svg (¼ cup)	½ svg (¼ cup)	8 g/svg; ½ of grains whole	½ svg (½ cup)
Cereals	½ svg (¼ cup)	½ svg (¼ cup)	8 g/svg; ½ of grains must be whole	½ svg (½ cup)
Meat, fish, poultry	NA	NA	NA	NA
Meat alternatives	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Seeds, nuts, nut butters	NA	NA	NA	NA
Cheeses and cheese substitutes	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Milk, dairy products, dairy substitutes (including soy beverages)	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Fats, oils, spreads (including butter)	NA	NA	NA	NA
Soups, meal sauces, and mixed side dishes	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Entrees, sandwiches, main dishes, meal replacements	1 serving (½ cup)	1 serving (½ cup)	16 grams per serving; half of the grains must be whole	1 serving (1 cup)
Meal	1.5 servings (¾ cup)	1.5 servings (¾ cup)	1.5 servings (¾ cup)	1.5 servings (1½ cups)
Sauces, dressing, condiments	NA	NA	NA	NA
Snack foods and sweets	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Desserts	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Beverages	½ serving (¼ cup)	½ serving (¼ cup)	½ serving (8 grams)	½ serving (½ cup)
Water (plain and carbonated)	NA	NA	NA	NA
Chewing gum	NA	NA	NA	NA

Nutrient Rich Foods Index (NRFI)

The Nutrient Rich Foods Index is an overall score of nutritional quality that is applied to all foods. There are no threshold levels, but reference values for each nutrient are used as a basis for scoring. The system is transparent. It is intended for guidance and can be applied to foods, menus, or daily diets.

Nutrient	Reference Value
Saturated fat	20g
Added sugar	50g
Sodium	2,400 mg
Protein	50g
Dietary fiber	25g
Vitamin A	5,000 IU
Vitamin C	60 mg
Vitamin E	30 IU
Calcium	1,000 mg
Iron	18 mg
Magnesium	400 mg
Potassium	3,500 mg

The score is calculated with the algorithm:
(protein g/50g + fiber g/25g + Vitamin A IU/5,000 IU + Vitamin C mg/60 mg + Vitamin E IU/30 IU + calcium mg/1,000 mg + iron mg/18 mg + magnesium mg/400 mg + potassium mg/3,500 mg − saturated fat g/20g − added sugars g/50g − sodium mg/2,400 mg) × 100

The nutrient values of foods can be based on per RACC or per 100 kcal of the food.

Scores (per 100 kcal) of 5,085 foods from NHANES 1999-2002 ranged from −131 to 555.

Source: Fulgoni III, V. L., Keast, D. R., & Drewnowski, A. (2009). Development and validation of the nutrient-rich foods index: a tool to measure nutritional quality of foods. Journal of Nutrition, 139, 1549-1554.

NuVal/Overall Nutritional Quality Index (ONQI)

The NuVal system is based on an algorithm referred to as the Overall Nutritional Quality Index. The algorithm produces an overall summary score that is converted to a scale of 1 to 100. The system was developed by a scientific expert panel of 12 members from academia and health organizations. A scientific advisory board considers revisions every 2 years.

The system is complex, and the algorithm was not available for this memorandum. Some aspects of the system are noted below.

Negative Nutrients/Attributes	Positive Nutrients/Attributes
Energy density Saturated fat Trans fat Cholesterol Total sugar Added sugar Sodium Glycemic load	Protein quality Fiber Vitamins A, C, D, E, B₆, B₁₂ Folate Calcium Iron Zinc Magnesium Potassium Fat quality Omega-3 fatty acids Total bioflavonoids Total carotenoids

The algorithm is unique in its use of a trajectory score to signify how foods in the diet come together over the course of a day to move total intake of a nutrient toward, or away from, a recommended threshold value. The trajectory score addresses the following questions: how does the concentration of a nutrient in a food compare to the recommended concentration of that nutrient in the diet overall, and how does consumption of that food influence the trajectory of daily intake for recommended nutrients that are in that food?

The formula for the trajectory score is:

For positive nutrients, the threshold values for recommended intakes are based on population-weighted Estimated Average Requirements from the Dietary Reference Intakes or DGA.

The threshold scores were weighted by coefficients for the prevalence of the health conditions most strongly associated with the nutrient (based on literature and expertise of the panel), the severity of these conditions, and the strength of associations between the nutrient and the condition.

In addition to the individual nutrients in the ONQI, there are four universal adjusters that pertain to the quality of the macronutrients:

Protein quality is based on the distribution of essential amino acids, and is applied to the entire sum of the numerator. Points are assigned if rate-limiting amino acids are present in at or above the recommended level.
Fat quality is the percent of total fat that is unsaturated (polyunsaturated or monounsaturated) multiplied by the percent of total calories derived from fat, which derives to the percentage of total calories that is polyunsaturated or monounsaturated fat. This is applied to the entire sum of the numerator.
Glycemic load is used as a proxy measure for quality of carbohydrates, and is applied selectively to grain-containing foods to distinguish primarily between refined and whole-grain products as well as to foods containing added sugars. Glycemic load is applied to the denominator. Values for glycemic load are obtained from NDSR, University of Minnesota.
Energy density is entered into the formula, except for pure cooking oils, and is applied to the denominator.

The basic structure of the ONQI algorithm is:

where

W_P = weighting coefficient for prevalence of a health condition associated with that nutrient

W_S = weighting coefficient for severity of the condition associated with that nutrient

W_R = weighting coefficient for relative impact/strength of association

TS = trajectory score of a nutrient

UA1 = adjusted entry for the biological quality of fat

UA2 = adjusted entry for the biological quality of protein

ED = adjusted entry for energy density

GL = adjusted entry for glycemic load

Source: Katz, D. L., Njike, V. Y., Faridi, Z., Rhee, L. Q., Reeves, R. S., Jenkins, D. J. A., et al. (2009). The stratification of foods on the basis of overall nutritional quality: The Overall Nutritional Quality Index. American Journal of Health Promotion, 24(2), 133-143.

Food category-specific adjustments are made to ensure the system works for within category comparisons as well as across categories.

To prevent artificial inflation of ONQI values by fortification, numerator nutrients are capped for all processed foods, but are not capped for natural plant foods (i.e., fruits, vegetables, legumes, nuts, seeds, cooking grains) and meats (i.e., beef, poultry, pork, fish, seafood).

Micronutrient exceptions include vitamin A and D in fortified dairy foods. Omega-3 fatty acids and fiber cannot be added to foods in large amounts without changing the palatability or texture and are not capped.

Another example of a category-specific adjustment is that energy density is not applied to cooking oils because they are naturally energy-dense. Intrinsic sodium and sugar in fruits and vegetables are not counted in the denominator. Adjustments are made to artificially sweetened foods so that the nutrient values are not inappropriately inflated when divided by a near-zero calorie level.

Water is scored the highest score in the cold beverages category (although it would actually score poorly due to low nutrient content) due to consensus by expert panel members that water is the most highly recommended cold beverage.

For dairy, the glycemic load is only applied to dairy products to which sugar has been added.

ONQI scores can range from a low of less than 1, to a large finite number. The scores are converted to a 1 to 100 scale for consumer use. The commercial algorithm is referred to as NuVal.

Source: Katz, D. L. (2007). ONQI Version 1 Abbreviated Reference Manual. Derby, CT: Griffin Hospital.

Sensible Solution (Kraft)

Sensible Solution is Kraft's FOP system that applies the logo to products meeting category-specific criteria. Guidance for the development of nutrient criteria was provided by the Kraft Worldwide Health & Wellness Advisory Council. The nutrient criteria are transparent and published on Kraft's website.

*Product categories and qualifying criteria*
Product Category	Qualifying Criteria (Per Serving)
Source: http://www.kraftrecipes.com/kf/HealthyLiving/SensibleSolution/nutritioncriteria.aspx (Accessed October 2, 2010).
100% fruit juice	Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 120 calories Serving size of no more than 8 fluid ounces
Refreshment beverages	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 40 calories 10g added sugar
Specialty beverages	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 100 calories 2g total fat 1g saturated and trans fat 10g added sugar
Cereals	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Cereals with smaller serving sizes (standard 30 g) when served with ½ cup fat-free milk: Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable At least 8g of whole grain Provide a functional benefit AND must meet the following limits: 170 calories 30% of calories from total fat 10% of calories from saturated and trans fat 25% of calories from added sugar 360 mg sodium And contain at least 2.5g of fiber or 8g of whole grain For cereals with larger serving sizes (standard 55 g) when served with ½ cup fat-free milk: Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or 20% DV of fiber At least a half-serving of fruit or vegetable At least 16g of whole grain Provide a functional benefit AND must meet the following limits: 290 calories 30% of calories from total fat 10% of calories from saturated and trans fat 25% of calories from added sugar 480 mg sodium And contain at least 5g of fiber or 16g of whole grain
Granola and cereal bars	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 150 calories 30% of calories from total fat 10% of calories from saturated and trans fat 25% of calories from added sugar 360 mg sodium And contain at least 2.5g of fiber or 8g of whole grain or 10% protein
Cookies	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable A nutritionally significant amount of whole grain (at least 5 g) Provide a functional benefit AND must meet the following limits: 100 calories * 30% of calories from total fat * 10% of calories from saturated and trans fat 25% of calories from added sugar 290 mg sodium * Calories no more than 130 and fat no more than 35% of calories if the product contains 10% DV fiber, a nutritionally significant amount of whole grain (at least 5g) or has a functional nutrition benefit.
Nuts and nut-based snacks	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 200 calories 20g fat 2g saturated and trans fat 10% of calories from added sugar 290 mg sodium
Cracker products	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable A nutritionally significant amount of whole grain (at least 5g) Provide a functional benefit AND must meet the following limits: 100 calories * 30% of calories from total fat * 10% of calories from saturated and trans fat 25% of calories from added sugar 290 mg sodium * Calories no more than 130 and fat no more than 35% of calories if the product contains 10% DV fiber, a nutritionally significant amount of whole grain (at least 5 g) or has a functional nutrition benefit.
Salted snacks	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable A nutritionally significant amount of whole grain (at least 5 g) Provide a functional benefit AND must meet the following limits: 100 calories * 30% of calories from total fat * 10% of calories from saturated and trans fat 25% of calories from added sugar 290 mg sodium * Calories no more than 130 and fat no more than 35% of calories if the product contains 10% DV fiber, a nutritionally significant amount of whole grain (at least 5 g) or has a functional nutrition benefit.
Cheeses and products (Natural, feta, processed, cream cheese, cottage cheese, ricotta, sour cream, dips, cheese spreads, hummus)	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 100 calories 3g total fat 2g saturated and trans fat 40 mg of cholesterol 290 mg sodium 25% of calories from added sugar
Grated parmesan cheese	Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium
Convenient meal products	1) Must meet at least one Must be free of, low in, or at least 25% less, when compared to similar products in the category, in at least one of the following: calories, fat, saturated fat, sugar, or sodium Less than 35% calories from fat Meet definition of lean or extra lean OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable At least 8g whole grain Provide a functional benefit AND must meet the following limits: 250-600 calories * 35% calories from total fat 10% calories from saturated and trans fat 25% calories from added sugar 480-960 mg sodium * 60-90 mg cholesterol * * Convenient Meals category represents a broad range of products with different serving sizes and amounts of food; therefore, ranges are provided for some nutrients.
Meat and meat alternative products	1) Must meet at least one Must be free of, low in, or at least 25% less, when compared to similar products in the category, in at least one of the following: calories, fat, saturated fat, sugar, or sodium Meet definition of lean or extra lean OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable At least 8g whole grain Provide a functional benefit AND must meet the following limits: 60-110 calories * 1-4g total fat * 0.5-1.5g saturated and trans fat * 15-80 mg cholesterol * 140-480 mg sodium * At least 5g protein (10% Daily Value) * Meat and Meat Alternatives category represents a broad range of products with variable serving sizes, therefore, ranges are provided for some nutrients.
Dessert products	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 100 calories 30% calories from total fat 10% calories from saturated and trans fat 25% calories from added sugar 360 mg sodium
Salad dressings	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 80 calories 30% calories from total fat 10% calories from saturated and trans fat 25% calories from added sugar 10 mg of cholesterol 290 mg sodium
Mayonnaise and substitutes	1) Must be free of, low in, or at least 25% less in at least one of the following: calories, fat, saturated fat, sugar, or sodium OR 2) Must meet at least one: 10% or more of the Daily Value (DV) of: Vitamin A, C, E, calcium, magnesium, potassium, iron, protein, or fiber At least a half-serving of fruit or vegetable Provide a functional benefit AND must meet the following limits: 50 calories 30% calories from total fat 10% calories from saturated and trans fat 25% calories from added sugar 5 mg of cholesterol 140 mg sodium

UK Ofcom Nutrient Profiling Model

The system is a nutrient profiling scoring system developed by British Heart Foundation Health Promotion Research Group at Oxford University under commission by UK Food Standards Agency. The nutrient criteria are applied to all foods on a per 100g basis. The criteria are transparent.

This is a "simple scoring" system, where points are allocated on the basis of the nutritional content in 100g of the food or drink. There are three steps to working out the overall score for the food or drink, as follows:

Work out total "A" points

A maximum of 10 points can be awarded for each nutrient.

Total "A" points = (points for energy) + (points for saturated fat) + (points for sugars) + (points for sodium)

The following table indicates the points scored, depending on the content of each nutrient in 100g of the food or 200g of drink:

Points:	0	1	2	3	4	5	6	7	8	9	10
Energy (kcal)	≤80	>80	>160	>240	>320	>400	>480	>560	>640	>720	>800
Sat fat (g)	≤1	>1	>2	>3	>4	>5	>6	>7	>8	>9	>10
Total sugar (g)	≤4.5	>4.5	>9	>13.5	>18	>22.5	>27	>31	>36	>40	>45
Sodium (mg)	≤90	>90	>180	>270	>360	>450	>540	>630	>720	>810	>900

If a food or drink scores 11 or more "A" points, then it cannot score points for protein unless it also scores 5 points for fruit, vegetables, and nuts.

Work out total "C" points

A maximum of five points can be awarded for each nutrient/food component.

Total "C" points = (points for fruit, vegetables, and nut content) + (points for fiber [either NSP or AOAC]) + (points for protein)

The following table indicates the points scored, depending on the content of each nutrient/ food component in 100g of the food or 200g of drink:

Points:	0	1	2	3	4	5*
*If a food or drink scores 5 points for fruit, vegetable and nuts, the "A" nutrient cut-off no longer applies.
Fruit, vegetable, and nuts (%)	≤40	>40	>60	—	—	>80
NSP fiber (g)	≤0.7	>0.7	>1.4	>2.1	>2.8	>3.5
OR AOAC fiber (g)	≤0.9	>0.9	>1.9	>2.8	>3.7	>4.7
Protein (g)	≤1.6	>1.6	>3.2	>4.8	>6.4	>8.0

Work out overall score

If a food or drink scores less than 11 "A" points then the overall score is calculated as follows:

Overall score = (total "A" points) minus (total "C" points)

If a food or drink scores 11 or more "A" points but scores 5 points for fruit, vegetables and nuts then the overall score is calculated as follows.

Overall score = (total "A" points) minus (total "C" points)

If a food scores 11 or more "A" points but also scores less than 5 points for fruit, vegetables, and nuts then the overall score is calculated as follows.

Overall score = (total "A" points) minus (fiber points+fruit, vegetables, and nuts points only) [i.e., no points for protein]

A food is classified as "less healthy" where it scores 4 points or more.

A drink is classified as "less healthy" where it scores 1 point or more.

Source: Scarborough, P., Boxer, A., Rayner, M., & Stockley, L. (2007). Testing nutrient profile models using data from a survey of nutritional professionals. Public Health Nutrition, 10(4), 337-345.

Choices Programme

The Choices Programme is an FOP system that awards a logo if the food meets category-specific criteria. A stated intention of the program is to encourage product reformulation. The criteria are based on per 100g or 100 kcal amounts. The nutrient criteria were established by an independent international scientific committee of 13 scientists from 12 countries. The nutrient criteria are transparent and are based on WHO dietary recommendations for chronic disease risk reduction.

*General benchmark criteria*
Nutrient	Generic Criteria	WHO Dietary Recommendations
¹ Based on 2.4 g/day, calculated from the energy recommendation for women=2,000 kcal/day. ² Based on 25 g/day, calculated from the energy recommendation for women=2,000 kcal/day.
Saturated fat	13% of kcal or 1.1g/100g	10% of kcal
Trans fat	1.3% of kcal of 0.1g/100g	1% of kcal
Sodium	1.3 mg/kcal	1.2 mg/kcal¹
Added sugars	13% of kcal or 2.5g/100g	10% of kcal
Dietary fiber	1.3 g/100 kcal	1.3 g/100 kcal²

Detailed criteria

Basic product groups

Food groups defined on the basis of product group classifications used in more than 20 countries from 5 continents. The products from these food groups significantly contribute to the intake of essential nutrients.

Food Group	Saturated Fat	Trans Fat	Sodium	Added Sugar	Fiber	Energy
¹ The fiber source in a product must be naturally occurring in one of the main ingredients of the product group. ² Will be reduced to 24 g/100g in 3 years and to 20 g/100g in 6 years. ³ Naturally occurring trans fat from meat or milk is excluded. ⁴ Value only applies to Europe. ⁵ If all the components of the product comply with the criteria in their respective product group and the product is in line with the energy and fiber criterion for its product group, the product also complies to the criteria.
Fresh or fresh frozen fruit, vegetables, and legumes (without additives)	NA	NA	NA	NA	NA	NA
Processed fruit and vegetables	≤1.1 g/100g	≤0.1 g/100g	≤100 mg/100g	NA	≥1.3 g/100 kcal¹	NA
Fruit juices	≤1.1 g/100g	≤0.1 g/100g	≤100 mg/100g	NA	≥0.75 g/100 kcal¹	≤48 kcal/100 mL
Water (plain)	NA	NA	≤20 mg/100 mL	NA	NA	NA
Potatoes (unprocessed, without additives)	NA	NA	NA	NA	NA	NA
Potatoes (processed), pasta, and noodles	≤1.1 g/100g	≤0.1 g/100g	≤100 mg/100g	NA	≥1.3 g/100 kcal¹	NA
Rice	≤1.1 g/100g	≤0.1 g/100g	≤100 mg/100g	NA	≥0.7 g/100 kcal¹	NA
Bread	≤1.1 g/100g	≤0.1 g/100g	≤500 mg/100g	≤13% of energy	≥1.3 g/100 kcal¹	NA
Grain and cereal products	≤1.1 g/100g	≤0.1 g/100g	≤100 mg/100g	≤2.5 g/100g	≥1.3 g/100 kcal¹	NA
Breakfast cereal products	≤13% of energy	≤0.1 g/100g	≤500 mg/100g	≤28 g/100g²	≥1.3 g/100 kcal¹	NA
Meat, poultry, eggs (unprocessed)	≤1.1 g/100g OR ≤13% of energy	≤0.1 g/100g³	≤100 mg/100g	NA	NA	NA
Processed meat, meat products, and meat substitutes	≤1.1 g/100g OR ≤13% of energy	≤0.1 g/100g³	≤900 mg/100g	≤2.5g/100g	NA	NA
Fresh or fresh frozen fish, shellfish, and crustaceans	≤1.1 g/100g OR ≤30% of total fat	≤0.1 g/100g	≤100 mg/100g	NA	NA	NA
Processed fish or fish products	≤1.1 g/100g OR ≤30% of total fat	≤0.1 g/100g	≤450 mg/100g	NA	NA	NA
Milk (and milk products including milk substitutes)	≤1.4 g/100g	≤0.1 g/100g³	≤100 mg/100g	≤5 g/100g	NA	NA
Cheese (and cheese products)	≤15 g/100g	≤0.1 g/100g³	≤900 mg/100g	NA	NA	NA
Oils, fats, and fat containing spreads	≤30% of total fat⁴	≤1.3% of energy³	≤1.3 mg/kcal	NA	NA	NA
Main course (all ready-to-cook meals intended to be eaten as a main dish during lunch or dinner⁵)	≤1.1 g/100g OR ≤13% of energy	≤0.1 g/100g or ≤1.3% of energy³	≤2.2 mg/kcal	≤2.5 g/100g OR ≤13% of energy	≥1.25 g/100 kcal¹	400-700 kcal/ serving
Sandwiches/rolls (all ready-to-eat filled sandwiches/rolls)	≤1.1 g/100g OR ≤13% of energy	≤0.1 g/100g or ≤1.3% of energy³	≤1.9 mg/kcal	≤2.5 g/100g OR ≤13% of energy	≥0.8g/100 kcal¹	≤350 kcal/ serving

Nonbasic product groups

Food products that generally do not provide a substantial contribution to the intake essential nutrients but have a large product innovation potential.

Food Group	Saturated Fat	Trans Fat	Sodium	Added Sugar	Fiber	Energy
¹ Will be reduced to 20 kcal/100 mL in 3 years. Source: http://choicesprogramme.org/en (Accessed September 30, 2010).
Soups	≤1.1 g/100g	≤0.1 g/100g	≤350 mg/100g	≤2.5 g/100g	NA	≤100 kcal/100g
Meal sauces (that constitute a substantial component of the meal, portion size >35g)	≤1.1 g/100g	≤0.1 g/100g	≤450 mg/100g	≤2.5 g/100g	NA	≤100kcal/100g
Other sauces (on water basis, that constitute minor component of the meal, portion size <35g without added emulsifying agent AND have a fat content <10% w/w [e.g., ketchup, soy sauce])	≤1.1 g/100g	≤0.1 g/100g	≤750 mg/100g	NA	NA	≤100 kcal/100g
Other sauces (emulsions, that constitute a minor component of the meal, portion size <35 g, to which an emulsifying agent is added OR have a fat content ≥10% w/w [e.g., mayonnaise, salad dressing])	≤1.1 g/100g OR ≤30% total fat	≤0.1 g/100 g OR ≤1.3% of energy	≤750 mg/100g	≤13% of energy OR ≤2.5 g/100g	NA	≤350 kcal/100g
Snacks (including pastry, edible ice cream, sweet snacks, and savory snacks)	≤1.1 g/100g OR ≤13% of energy	≤0.1 g/100g OR ≤1.3% of energy	≤400 mg/100g	≤20 g/100g	NA	≤110 kcal/serving
Beverages (liquid food products, with the exception of plain water, dairy products, and fruit juices [e.g., coffee, tea, (light) soft drinks, fruit drinks])	≤1.1 g/100g	≤0.1 g/100g	≤100 mg/100g	NA	NA	≤30 kcal/100 mL¹
Bread toppings including hummus like products	≤13% of energy	≤1.3% of energy	≤400 mg/100g	≤30 g/100g	NA	NA
All other products (any food product that does not fall within any of the other product groups [e.g., baking product, seasonings, vinegar])	≤1.1 g/100g OR ≤13% of energy	≤0.1 g/100g OR ≤1.3% of energy	≤100 mg/100g OR ≤1.3 mg/kcal	≤2.5 g/100g OR ≤13% of energy	NA	NA

Keyhole Symbol

The Keyhole is a FOP system awarding an icon if the product meets the category-specific criteria. The criteria are based on per 100 g, but some nutrients are based on per 100 kcal or % of energy. The system was created by the Swedish National Food Administration to help consumers identify healthier options and to stimulate manufacturers to reformulate products. The system is also used in Denmark and Norway.

General criteria

Foods cannot contain sweeteners.
Industrially produced trans fatty acids cannot exceed 2g per 100g oil and/or other fats.
Definitions: Whole grain of cereals refers to the bran, germ, and endosperm. The grain may be ground, crushed or similarly treated; however, in this case, the original proportions of the respective kind of cereal shall be retained. Cereal refers to wheat, spelt, rye, oats, barley, maize, rice, millet and durra, and other Sorghum species.

Product categories and qualifying criteria
Product Category	Qualifying Criteria (Per 100g Unless Specified as %)
¹ Council Regulation (EC) No 2991/94 of 5 December 1994 laying down standards for spreadable fats (OJ L 316, 9.12.1994, p. 2, Celex 31994R2991). Source: Livsmedelsverket (National Food Administration). 2005. National Food Administration's Regulations on the Use of a Particular Symbol. Document LIVSFS 2005:9. Stockholm: NFA; available at http://www.slv.se/upload/nfa/documents/food_regulations/Nyckelh%c3%a5l_dec_2009_6%20eng.pdf (Accessed October 5, 2010).
Milk and corresponding fermented products (unflavored).	≤0.7g fat
Fermented milk products (flavored).	≤0.7g fat ≤9g total sugars
Vegetable products, intended for same uses as products in group 1.	≤1.5g fat ≤33% of total fat is saturated fat ≤5g total sugars ≤40 mg sodium
Products with mixture of milk and cream and intended for same uses as cream, and corresponding fermented products (may be flavored).	≤5g fat ≤5g total sugars ≤100 mg sodium
Products of vegetable origin and intended for same uses as products in group 4.	≤5g fat ≤33% of total fat is saturated fat ≤5g total sugars ≤100 mg sodium
Fresh cheese and corresponding flavored products.	≤5g fat ≤350 mg sodium
Margarine cheese (of vegetable origin and intended for same uses as products in group 8) and corresponding flavored products.	≤17g fat ≤20% of total fat is saturated fat ≤500 mg sodium
Other cheese and corresponding flavored products.	≤17g fat ≤500 mg sodium
Edible fats and blends falling within the scope of Council Regulation (EC) No 2991/94 on spreadable fats,¹ and corresponding flavored products.	≤41g fat ≤33% of total fat is saturated fat ≤500 mg sodium
Oils and liquid margarine and liquid blends.	≤20% of total fat is saturated fat ≤500 mg sodium
Untreated meat of cattle, pig, horse, sheep, goat, poultry, or game.	≤10g of fat
Untreated fish, shellfish, shell, and other mollusks.	None
a) Products containing at least 50% meat (muscle tissue), liver or blood of cattle, pig, horse, sheep, goat, poultry or game.	≤10g fat ≤5g total sugars
b) Products containing at least 50% fish, shellfish, shell, and/or other mollusks.	≤10g of fat not originating from fish ≤5g total sugars
c) Products intended as the main protein component in a meal or as a spread and containing at least 95% vegetable ingredients. The products under 13 a), b), and c) shall not be covered by coating of e.g., bread crumbs and/or eggs but may contain sauce or stock. The 50% content of the products under a) and b), and the 95% content under c) apply to the part of the product intended for consumption.	≤10g fat ≤5g total sugars
Ready-prepared products (with the exception of products in Groups 15, 16, and 17) intended to constitute a main meal and containing 400-750 kcal (per portion) and minimum 25g/100g root vegetables, leguminous plants other than peanuts, other vegetables, and/or fruit and berries (potatoes excluded).	≤30% of energy from fat, unless product contains fish consisting of >10% fat then may contain ≤40% of energy from fat (the nonfish fat must be ≤10g per portion). ≤3g refined sugars ≤400 mg sodium
Pirogues, pizzas and nondessert pies based on cereals and containing a minimum of 250 kcal (per portion) and a minimum 25 g/100g of root vegetables, leguminous plants other than peanuts, other vegetables, and/or fruit and berries (potatoes excluded). The cereal part shall contain minimum 15% whole grain calculated on the dry matter basis.	≤30% of energy from fat ≤3g refined sugars ≤500 mg sodium
Sandwiches, baguettes, wraps and similar products, based on cereals and containing a minimum 250 kcal (per portion) and a minimum 25 g/100g of root vegetables, leguminous plants other than peanuts, other vegetables, and/or fruit and berries (potatoes excluded). The cereal part shall contain minimum 25% whole grain calculated on the dry matter basis.	≤30% of energy from fat ≤3g refined sugars ≤400 mg sodium
Soups (ready-prepared products and products prepared according to the manufacturer's instructions) containing a minimum 150 kcal (per portion) and a minimum 25 g/100g of root vegetables, leguminous plants other than peanuts, other vegetables and/or fruit and berries (potatoes excluded).	≤30% of energy from fat ≤3g refined sugars ≤400 mg sodium
Fruit and berries which have not been processed; however, they may have been cleaned, sliced, chilled, deep-frozen, and thawed.	None
Potatoes, root vegetables, leguminous plants other than peanuts and other vegetables which have not been processed; however, they may have been blanched, dried, sliced, chilled, deep-frozen, thawed or preserved in water. The products may be flavored with spices.	≤1g refined sugars ≤200 mg sodium
Soft bread and bread mixes where only water and yeast is to be added, containing minimum 25% whole grain calculated on the dry matter basis. As regards bread mixes, the conditions apply to the prepared product.	≤7g fat ≤5g total sugars ≤500 mg sodium ≥5g dietary fiber
Hard bread and rusks containing minimum 50% whole grain calculated on the dry matter basis.	≤7g fat ≤5g total sugars ≤500 mg sodium ≥6g dietary fiber
Pasta (unfilled) containing minimum 50% whole grain calculated on the dry matter basis.	≤40 mg sodium (per 100g dry weight) ≥6g dietary fiber (per 100g dry weight)
Breakfast cereals and muesli containing minimum 50% whole grain calculated on the dry matter basis.	≤7g fat ≤10g refined sugars ≤13g total sugars ≤500 mg sodium ≥6g dietary fiber
Cereal flour, flakes and grains, and crushed cereals containing 100% whole grain calculated on the dry matter basis, and cereal bran.	≥6g dietary fiber
Porridge and porridge powder containing minimum 50% whole grain calculated on the dry matter basis. As regards porridge powder, the values apply to the prepared product.	≤5g fat ≤5g total sugars ≤200 mg sodium ≥6g dietary fiber

Pick the Tick/Tick Cert™ (Australia/New Zealand)

Pick the Tick is an FOP system awarding an icon if the food meets category-specific criteria. The nutrient criteria are based on per serving or per 100g or per 100 ml. The stated aim of the system is to improve the nutrition of foods and deliver better nutritional health outcomes. Criteria were developed by a Heart Foundation Criteria Working Group, comprised of experts in public health, nutrition, food technology, and food science. The process involved examination of current market nutrition levels combined with analysis of public health priorities for each food category, taking into consideration technical and market feasibilities. Nutrient criteria for some of the food categories are provided.

Examples of category-specific criteria include:

Food Category	Qualifying Criteria
Source: http://www.heartfoundation.org.au/sites/tick/Health_Professionals/Pages/TickCriteria.aspx (Accessed September 27, 2010).
Fruit juice	≤108 kcal per serving and 43 kcal per 100 mL ≥1.5g fiber from the fruit ≥98% juice (≥50% fruit juice)
Vegetable juice	≤48 kcal per serving and ≤19 kcal per 100 mL ≤120 mg sodium per 100 mL ≥1.5g fiber from the vegetable ≥98% juice (≥50% vegetable juice)
Vegetable oils	≤20% of total fat is saturated fat ≤1% of total fat is trans fat
Edible oil spreads (margarine spreads including dairy blends)	≤28% of total fat is saturated fat plus trans fat (For reduced fat spreads, defined as ≤50% total fat, ≤5g per 100 g) ≤1% of total fat is trans fat (For reduced fat spreads, ≤0.2g per 100 g) ≤400 mg sodium
Cheese (aged/ripened and processed)	≤17g saturated fat per 100g ≤750 mg sodium per 100g ≥700 mg calcium per 100g
Cheese (unripened)	≤5g saturated fat per 100g ≤400 mg sodium per 100g ≥80 mg calcium per 100g
Breakfast cereals	≤191 kcal per serving (except hot cereal 155 kcal/svg) ≤1.5g saturated fat per 100g (except mueslis ≤2.5 g/100g and hot cereal ≤2 g/100 g) No partially hydrogenated fats; or ≤0.2 g/100g trans fat ≥3g fiber per serving (except hot cereal ≥2 g/svg) OR ≥50% wholegrain content ≤400 mg sodium per 100 g(except mueslis ≤120 mg/100g and hot cereal ≤120 mg/100 g)
Bread	≤1g saturated fat per 100g (except crumpets ≤1.5 g/100 g) No partially hydrogenated fats; or ≤0.2 g/100g trans fat ≤400 mg sodium per 100g (except crumpets ≤600 mg/100 g) ≥4g fiber per 100g (except crumpets ≥2.5 g/100 g)
Sweet biscuits (cookies)	≤143 kcal per serving ≤2g saturated fat per serving No partially hydrogenated fats; or ≤0.2 g/100g trans fat ≤250 mg sodium per 100g ≥1g fiber per serving
Ready meals (canned meat meals)	≤287 kcal per serving ≤1.5g saturated fat per 100g No partially hydrogenated fats; or ≤0.2 g/100g trans fat ≤350 mg sodium per 100g ≥75g vegetables or fruit per serving OR fiber criteria ≥2g fiber per serving OR vegetable/fruit content ≥5g protein per serving
Ready meals (other)	≤526 kcal per serving ≤2g saturated fat per 100g and ≤6 g/svg No partially hydrogenated fats; or ≤0.2 g/100g trans fat ≤300 mg sodium per 100g and ≤900 mg/svg ≥75g vegetables or fruit per serving OR fiber criteria ≥3g fiber per serving OR vegetable/fruit content ≥5g protein per serving

Heart Check™ (Canada)

The Heart Check system is a FOP system that awards the icon if a food meets category-specific criteria on a per serving basis. The system was developed by registered dietitians at the Heart and Stroke Foundation. Nutrient criteria are reviewed regularly and are transparent.

These guidelines will apply to all products as of November 2010, reflecting updates to sodium criteria.

Vegetables and Fruits
Food Category	Qualifying Criteria
Fruit juices	100% fruit juice Excellent source of vitamin C (50%), vitamin A (25%), folate (25%), or source of fiber (2 g) per 250 mL serving and per on pack serving
Fresh fruit	None
Frozen fruit	100% fruit
Canned fruit	In light syrup or fruit juice
Apple and other fruit sauces	100% fruit
Dried fruit pieces	Fruit as first ingredient No added fat
Dried fruit snacks	No added sugar No added fat Source of vitamin C (5%), vitamin A (5%), folate (5%), or fiber (2 g) per 40g serving
Fresh and frozen vegetables	None
Canned vegetables (plain)	≤240 mg sodium per 125 mL serving and per on pack serving
Canned tomatoes	≤360 mg sodium per 125 mL serving and per on pack serving
Frozen and canned vegetables (seasoned, sauced)	≤3g fat ≤240 mg sodium Per 125 mL serving (110g when frozen) and per on pack serving
Tomato juice	≤480 mg sodium per 250 mL serving and per on pack serving
Vegetable juices and blends	Vitamin A and/or folate (at least 15% of DV) ≤480 mg sodium Per 250 mL sodium
Tomato paste	No added salt
Minor main entre vegetable-based sauces	≤3g fat or no added fat ≤240 mg sodium Vitamin C or vitamin A or folate (at least 5% of DV) or fiber (2 g) Per 60 mL serving and per on pack serving
Vegetable-based dips	≤3g fat or no added fat ≤240 mg sodium Vitamin C or vitamin A or folate (at least 5% of DV) or fiber (2 g) Per 30g serving and per on pack serving
Frozen fruit bars	No added sugar Vitamin C or vitamin A or folate (at least 5% of DV) Per 75 mL serving and per on pack serving

Grain Products
Food Category	Qualifying Criteria
Bread	≤3g fat OR ≤2g saturated fat and trans fat combined and ≤15% energy from the sum of saturated and trans fat ≤5% of total fat is trans fat ≥2g fiber ≤360 mg sodium Per 50g serving and per on pack serving
Bread products (e.g., bagels, pita, English muffins)	≤3g fat OR ≤2g saturated fat and trans fat combined and ≤15% energy from sum of saturated and trans fat ≤5% of total fat is trans fat ≥2g fiber ≤360 mg sodium Per 55g serving and per on pack serving
Hot breakfast cereals	≤3g fat OR no added fat ≤5% of total fat is trans fat ≥2g fiber ≤240 mg sodium ≤11g sugar (excluding sugars from pieces of fruit) except if 4g or more fiber Per 40g serving and per on pack serving
Breakfast cereals	≤3g fat OR no added fat ≤5% of total fat is trans fat ≥4g fiber ≤240 mg sodium ≤11g sugar (excluding sugars from pieces of fruit) except if 6g or more fiber Per 40g serving and per on pack serving
Very high fiber breakfast cereals (≥28g per 100 g)	≤3g fat OR no added fat ≤5% of total fat is trans fat ≥6g fiber ≤240 mg sodium Per 30g serving and per on pack serving
Flour/grains	≥2g fiber ≤240 mg sodium Per 30g serving and per on pack serving
Crackers/rusks	≤2g saturated fat and trans fat combined + ≤15% energy from sum of saturated and trans fat ≤5% of total fat is trans fat ≤3g fat ≤190 mg sodium Per 30g serving and per on pack serving
Croutons	≤3g fat ≤5% of total fat is trans fat ≥2g fiber OR ≥5% DV of vitamin A or vitamin C or calcium or iron ≤140 mg sodium Per 20g serving and per on pack serving
Rice cakes	≤3g fat (per 50 g) ≤5% of total fat is trans fat ≤140 mg sodium Per 15g serving and per on pack serving
Waffles/pancakes	≤3g fat ≤5% of total fat is trans fat ≥2g fiber ≤11g sugar ≤240 mg sodium Per 75g serving and per on pack serving
Rice (except instant)/grains (plain)	≤140 mg sodium per 45g serving and per on pack serving
Instant rice (plain)	Enriched or whole grain ≤140 mg sodium per 45g serving and per on pack serving
Pasta	Enriched or whole grain OR ≥4g fiber ≤140 mg sodium Per 45g serving and per on pack serving
Side dishes—rice, grains or potatoes (seasoned, sauced)	≤3g fat ≤5% of total fat is trans fat ≤240 mg sodium Per 140g (prepared) serving and per on pack serving
Side dishes—pasta or noodles (seasoned, sauced)	≤3g fat ≤5% of total fat is trans fat Enriched or ≤2g fiber ≤240 mg sodium Per 125 mL (prepared) serving and per on pack serving
Grain based bars	≤6g fat (maximum 7.4g per on pack serving) ≤2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans fat ≥2g fiber ≤140 mg sodium ≤50% of carbohydrates from sugars Whole grain or whole wheat as first ingredient No sweetened filling or coating Per 30g serving and per on pack serving
Muffins/muffin-style bars	≤7.4g ≤2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans fat ≥2g fiber ≤240 mg sodium ≤50% of carbohydrates from sugars Whole grain or whole wheat as first ingredient No sweetened filling or coating Per 55g muffin or 40g muffin-style bar and per on pack serving
Plain popcorn	No added salt No added fat

Milk and alternatives
Food Category	Qualifying Criteria
Milk and milk-based drinks	Lower fat (≤2%) ≥25% of DV calcium ≤240 mg sodium No added sugar Per 250 mL serving and per on pack serving
Yogurts	Lower fat (≤2%) ≥15% of DV calcium ≤140 mg sodium No added sugar Per 175g serving and per on pack serving
Yogurt-based drinks	Lower fat (≤2%) ≥15% of DV calcium ≤140 mg sodium No added sugar Per 250 mL serving and per on pack serving
Yogurt-based dips	≤3g fat ≤5% of total fat is trans fat ≤240 mg sodium Per 30g serving and per on pack serving
Fresh cheese (plain and flavored) excluding cottage and ricotta	Lower fat (≤2%) ≥15% of DV calcium ≤240 mg sodium No added sugar Per 100g serving and per on pack serving
Cheese	Lower fat (≤2%) ≥15% of DV calcium ≤240 mg sodium Per 30g serving and per on pack serving
Soy-based cheese	Lower fat (≤2%) ≥15% of DV calcium ≥5g protein ≤240 mg sodium Per 30g serving and per on pack serving
Ricotta cheese (plain)	≤3g fat ≥10% of DV calcium ≤360 mg sodium No added sugar Per 125g serving and per on pack serving
Cottage cheese (plain and flavored)	≤3g fat ≥10% of DV calcium ≤360 mg sodium No added sugar Per 125g serving and per on pack serving
Plant-based beverages (e.g., soy beverages)	Fortified/enriched ≤3g fat OR 2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans fat ≤240 mg sodium ≤13g sugar Per 250 mL serving and per on pack serving

Meat and alternatives
Food Category	Qualifying Criteria
Meats/poultry (plain, seasoned, coated)	Lean (≤10% fat) ≤5% of total fat is trans fat if fat does not originate exclusively from ruminant meat No added salt or sodium for plain meat; ≤360 mg sodium for seasoned meats Per 125g (raw) or 100g (cooked) serving and per on pack serving
Meats/poultry (with sauce)	Lean (≤10% fat) ≤5% of total fat is trans fat if fat does not originate exclusively from ruminant meat ≤360 mg sodium for seasoned meats Per 125g (raw) or 100g (cooked) serving and per on pack serving
Ground meats (plain, seasoned)	Lean (≤17% fat) ≤5% of total fat is trans fat if fat does not originate exclusively from ruminant meat ≤360 mg sodium for seasoned meats Per 100g (raw) or 60g (cooked) serving and per on pack serving
Patties, meatballs, etc.	Lean (≤10% fat) ≤5% of total fat is trans fat if fat does not originate exclusively from ruminant meat ≤360 mg sodium Per 100g (raw) or 60g (cooked) serving and per on pack serving
Sausages	Lean (≤10% fat) ≤5% of total fat is trans fat if fat does not originate exclusively from ruminant meat ≤360 mg sodium Per 55g (raw) or 75g (uncooked) serving and per on pack serving
Deli meats/ham	Lean (≤10% fat) ≤5% of total fat is trans fat if fat does not originate exclusively from ruminant meat ≤360 mg sodium Per 55g serving and per on pack serving
Fish and seafood (plain)	≤360 mg sodium per 125g (raw) or 100g (cooked) serving or per on pack serving
Fish and seafood (seasoned or coated)	Extra lean ≤7.5% fat or no added fat ≤5% of total fat is trans ≤360 mg sodium Per 125g (raw) or 100g (cooked) serving or per on pack serving
Fish and seafood (sauced)	Extra lean ≤7.5% fat or no added fat ≤5% of total fat is trans ≤360 mg sodium Per 125g (raw) or 100g (cooked) serving or per on pack serving
Canned fish and seafood (packed in broth or water)	≤5% of total fat is trans ≤360 mg sodium Per 55g serving and per on pack serving
Canned fish and seafood (seasoned, sauced)	Lean ≤10% fat or no added fat ≤5% of total fat is trans ≤360 mg sodium Per 55g serving and per on pack serving
Processed fish (e.g., crab imitation, surimi)	≤3g fat ≤5% of total fat is trans ≤360 mg sodium Per 55g serving and per on pack serving
Dried legumes	None
Frozen and canned legumes (plain)	≤360 mg sodium per 250 mL (drained) serving and per on pack serving
Canned legumes (prepared)	≤3g fat ≤5% of total fat is trans ≤360 mg sodium Per 125 mL serving and per on pack serving
Tofu (plain)	≤10g fat ≤2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤140 mg sodium Per 85g serving and per on pack serving
Vegetarian meat alternatives (seitan, veggie ground meat)	≤10% fat ≤5% of total fat is trans ≥7.5g protein ≤360 mg sodium Per 60g (cooked) serving and per on pack serving
Vegetarian terrines, spreads or pates	≤10% fat ≤2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans ≥5g protein ≤360 mg sodium Per 55g serving and per on pack serving
Eggs	None
Egg substitutes	≤3g fat ≤5% of total fat is trans ≤240 mg sodium Per 50g serving and per on pack serving
Nuts, seeds, or ready-to-eat dried legumes (e.g., soybeans)	No added salt
Nut and seed butters	Nuts or seeds as the first ingredient ≤5% of total fat is trans ≤140 mg sodium Per 15g (peanut butter) or 30g (others) and per on pack serving
Legume-based dips	≤5g fat ≤5% of total fat is trans ≤240 mg sodium Per 30g serving and per on pack serving

Oils and fats
Food Category	Qualifying Criteria
Oils	≤2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤2% of total fat is trans ≤140 mg sodium Per 10 mL serving and per on pack serving
Margarines	Non hydrogenated ≤2% of total fat is trans ≤140 mg sodium Per 10g serving and per on pack serving
Light margarines	Non hydrogenated Reduced fat (≤50% fat than regular margarines) ≤2% of total fat is trans ≤140 mg sodium Per 10g serving and per on pack serving
Salad dressings	≤2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans ≤140 mg sodium Per 15 mL serving and per on pack serving (serving on label should be 15 mL except if total fat is <30%

Combination foods
Food Category	Qualifying Criteria
Source: http://www.healthcheck.org/page/program-critieria (Accessed September 27, 2010).
Soups	≤3g fat ≤5% of total fat is trans ≤480 mg sodium ≥2g fiber OR ≥5% DV for vitamin A, vitamin C, iron, calcium, or folate Per 250 mL serving and per on pack serving
Dinners and entrees/mixed dishes	≤10g fat OR ≤15g if "low in saturated fat" ≤5% of total fat is trans ≤720 mg sodium ≥10g protein Per 250g serving and per on pack serving
Pizza	≤10g fat ≤5% of total fat is trans ≤480 mg sodium ≥10g protein Per 140g serving and per on pack serving
Vegetarian or meat pies	≤10g fat OR ≤15g if "low in saturated fat" ≤5% of total fat is trans ≤480 mg sodium ≥10g protein Per 140g serving and per on pack serving
Tofu or meat or fish with vegetables	≤10g fat ≤5% of total fat is trans ≤480 mg sodium ≥10g protein Per 140g serving and per on pack serving
Stuffed pasta	≤7.4g fat ≤5% of total fat is trans ≤480 mg sodium ≥10g protein Enriched or whole wheat pasta Per 140g serving (as sold, cooked) and per on pack serving
Stuffed meat	≤10g fat ≤5% of total fat is trans ≤480 mg sodium ≥10g protein Per 140g serving (195g with gravy or sauce) and per on pack serving
Major main entrée sauce (e.g., pasta sauce)	≤3g fat OR ≤5g if "low in saturated fat" ≤5% of total fat is trans ≤360 mg sodium ≥2g fiber OR ≥5% DV for vitamin A, vitamin C, iron, calcium, or folate Per 125 mL serving and per on pack serving
Potato and pasta salad	≤7.4g fat ≥2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans ≤240 mg sodium Per 140g serving and per on pack serving
Other salads	≤7.4g fat ≥2g saturated fat and trans fat combined + 15% energy from sum of saturated and trans fat ≤5% of total fat is trans ≤240 mg sodium Per 100g serving and per on pack serving
Dried fruit and nut mixture	No added salt
Nut and/or seed bars (with or without dried fruit	No added salt

Nutrition Score/Enhancement Program (Unilever)

Unilever developed this system to evaluate and improve the nutritional composition of all their foods and beverages. The system assigns each food to 1 of 3 categories: (1) meets global dietary recommendations; (2) meets national dietary recommendations; or (3)does not meet dietary recommendations. The nutrient criteria based on per kcal, 100 g, and % of energy. Nutrients were selected based on Joint WHO/FAO Expert Consultation on Diet, Nutrition and the Prevention of Chronic Diseases (WHO, 2003). Country-specific dietary recommendations are used for category 2 (see Nijman, 2007).

Generic benchmarks
Nutrient	Unit	Category 1 (Based on Global Dietary Recommendations)	Category 2 (Based on National Dietary Recommendations)	Category 3 (Not Meet Dietary Recommendations)
¹ Based on an average daily energy intake of 2,250 kcal.
Trans fat	% of energy	≤1	1-2	>2
Saturated fat	% of energy	≤10	10-13	>13
Saturated fat	% of total fat	≤25	25-33	>33
Sodium	mg/100 kcal	≤90	90-160¹	>160
Total sugars	% of energy	≤15	15-25	>25
Added sugars	g/100g	≤3	3-7	>7

If multiple benchmarks are available for one nutrient (saturated fat), the benchmark score in the highest category determines the final key nutrient score.

The final Nutrition Score is determined by the lowest category.

Product category-specific benchmarks were developed when it was impossible for all foods or beverages in that product category to meet a final Nutrition Score in category 1 or 2.

Product category-specific benchmarks for sodium and sugars
Nutrient	Unit	Category 1 (Based on Global Dietary Recommendations)	Category 2 (Based on National Dietary Recommendations)	Category 3 (Not Meet Dietary Recommendations)
Frozen desserts and edible ice	Added sugar, g/100g	≤8	8-17	>17
Low energy dense foods (Soup)	Sodium, mg/100g	≤200	200-360	>360
Low energy dense foods (Meal sauces)	Sodium, mg/100g	≤300	300-540	>540
Low energy dense foods (Table sauces)	Sodium, mg/100g	≤600	600-1,080	>1,080
Small portion size foods (Dressings)	Sodium, mg/100g	≤600	600-1,080	>1,080
Foods consumed as part of weight management (Meal replacement products)	Sodium, mg/100 kcal	≤130	130-240	>240

References:

Nijman, C. A. J., Zijp, I. M., Sierksma, A., Roodenburg, A. J. C., Leenen, R., van den Kerkhoff, C., et al. (2007). A method to improve the nutritional quality of foods and beverages based on dietary recommendations. European Journal of Clinical Nutrition, 61, 461-471.

WHO. (2003). Diet, Nutrition and the Prevention of Chronic Disease. Report of a joint WHO/FAO Expert Consultation. WHO Technical Report Series 916. Geneva: World Health Organization.

Netherlands Tripartite Classification Model

The Tripartite system categorizes foods into 3 levels based on nutritional quality. The system applies category-specific nutrient criteria on a per 100g basis. The nutrient criteria are transparent. The system was developed by the Netherlands Nutrition Center to aid consumers in making healthy food choices.

Nutrient criteria per 100g
Product Group	A: "Preferable"	B:"Middle Course"	C: "Exceptional"
Note: Saturated fat includes trans.
Potatoes, rice, pasta, pulses	≥3g fiber ≤1g saturated fat	2-3g fiber ≤1g saturated fat	≤2g fiber
Bread, bread substitutes, breakfast cereals	≥6g fiber ≤1g saturated fat	5-6g fiber OR Min 6g fiber Min 1g saturated fat	<5g fiber
Vegetables, fruit, and fruit juices	≥1 mg vitamin C ≥1 mcg folate ≥1g fiber ≤1g saturated fat No added sugars	≥1 mg vitamin C ≥1 mcg folate	No vitamin C
Milk and milk products	≤0.5g saturated fat ≤6g sugars	0.6-1g saturated fat OR ≤0.5g saturated fat >6g sugars	>1g saturated fat OR 0.6-1g saturated fat >6g sugars
Cheese	≤12g saturated fat ≤300 kcal	13-18g saturated fat OR ≤12g saturated fat >300 kcal	>18g saturated fat
Meat, prepared meat products, chicken, eggs	≤4g saturated fat ≤200 kcal	4-5g saturated fat OR ≤4g saturated fat >200 kcal	>5g saturated fat
Fish	≤4g saturated fat Maximum 2 portions for recommendation n-3 fatty acids ≤200 kcal	4-5g saturated fat 2-4 portions for recommendation n-3 fatty acids	>5g saturated fat >4 portions for recommendation n-3 fatty acids
Spread and cooking fats	≤16g saturated fat	17-24g saturated fat	>24g saturated fat

Thresholds for saturated fats and fiber in other products
Product Groups	"Low in Saturated Fats"	"High" in Saturated Fats	"High" in Fiber
Source: http://www.voedingscentrum.nl/nl/service/english.aspx (Accessed October 6, 2010).
Snacks, spicy filling	≤4g	>5g	NA
Sauces	≤2g	>4g	NA
Cake, pastry, nuts, savory snacks	≤6g	>6g	≥2g
Sweets, sweet filling	≤3g	>4g	≥1g
Cream	≤12g	>18g	NA
Evaporated milk	≤1g	>3g	NA

B: Examples of Calculating Scores Using the Three Proposed Options for Algorithms

This appendix provides examples of how to use each of the three proposed algorithms with one food item, 1% milk. The nutrient values for 1% milk were obtained from USDA Food and Nutrient Database for Dietary Studies (food code 11112210).

Option 1

This example uses a nutrient density algorithm, the Nutrient Rich Foods Index (described in Appendix A).

An example of calculating the score for 1% milk:

Nutrients	Nutrient Amounts per 100 kcal of 1% Milk	Divide by Reference Value	Calculation
Positive nutrients: Protein	8.1g	50g	0.16
Fiber	0g	25g	0
Vitamin A	463 IU	5,000 IU	0.09
Vitamin C	0 mg	60 mg	0
Vitamin E	0 IU	30 IU	0
Calcium	305 mg	1,000 mg	0.31
Iron	0.07 mg	18 mg	0
Magnesium	26 mg	400 mg	0.07
Potassium	359 mg	3,500 mg	0.10 Total positive points: 0.73
Negative nutrients: Saturated fat	1.5g	20g	0.08
Added sugars	0g	50g	0
Sodium	107 mg	2,400 mg	0.04 Total negative points: 0.12

The total score is calculated as positive (0.73) − negative (0.12) * 100 = 61.

Option 2

This example uses the FSA Ofcam model (described in Appendix A).

An example of calculating the score for 1% milk:

Nutrients/Components	Nutrient Amounts per 200g of 1%Milk	Points
"A" nutrients (negatives): Energy	84 kcal	1
Saturated fat	1.3g	1
Total sugar	10.4g	2
Sodium	88 mg	0 Total "A" points: 4
"C" nutrients (positives): Fruit, vegetables, and nuts	0%	0
Fiber	0g	0
Protein	6.7g	4 Total "C" points: 4

The total score is calculated as
"A" points (4) − "C" points (4) = 0.

The total score is <1 point, 1% milk is classified as "healthy."

Option 3

This example uses a threshold type system, the FDA criteria for health claims. A food must meet all of the criteria for the negative nutrients, and 10% or more of daily reference value for at least one of the positive nutrients.

Using the example of 1% milk and the FDA criteria:

Nutrients	Nutrient Amounts per RACC Serving of 1% Milk	Criteria	Meets Criteria
Negative nutrients: Fat	2.4g	<13g	Yes
Saturated fat	1.5g	<4g	Yes
Cholesterol	8 mg	<60 mg	Yes
Sodium	107 mg	<480 mg	Yes
Positive nutrients: Vitamin A	473 IU (includes fortification)	≥500 IU	No
Vitamin C	0 mg	≥6 mg	No
Iron	0.07 mg	≥1.8 mg	No
Calcium	305 mg	≥100 mg	Yes
Protein	8.2 mg	≥5 mg	Yes
Fiber	0g	≥2.5g	No

1% milk meets all of the criteria for negative nutrients and at least one of the criteria for the positive nutrients.

C: Example of Calculation of an Individual's Score for Baseline Algorithm

Table C-1. Foods Consumed by One Individual, Grams Consumed, Nutrient Values per Grams Consumed, and RACC Value of the Food
Food Code	Description	Grams Consumed	Grams per RACC	Kcal	Protein (g)	Fiber (g)	Vitamin E (mg)	Vitamin D (mcg)	Calcium (mg)	Iron (mg)	Potassium (mg)	Unsat Fat (g)	Sat Fat (g)	Sodium (mg)	Added Sugar (g)
57123000	Cheerios	31.88	30	117	3.61	3.20	0.22	1.10	130.00	10.13	194.00	1.24	0.30	212.00	1.06
11112110	Milk…2% fat	129.63	244	65	4.28	0.00	0.04	1.60	156.00	0.03	181.00	0.82	1.63	61.00	0.00
92114000	Coffee…	119.20	240	2	0.13	0.00	0.00	0.00	5.00	0.04	38.00	0.00	0.00	5.00	0.00
91101010	Sugar, white…	4.17	4	16	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	4.17
11112110	Milk, cow's, fluid, 2% fat	81.33	244	41	2.68	0.00	0.02	1.00	98.00	0.02	114.00	0.51	1.02	38.00	0.00
52215200	Tortilla, flour (wheat)	51.00	55	159	4.23	1.60	0.10	0.00	66.00	1.70	79.00	2.79	0.96	324.00	0.00
41205010	Refried beans	31.63	130	45	1.98	1.90	0.21	0.00	12.00	0.53	111.00	1.12	0.43	113.00	0.00
63105010	Avocado, raw	18.25	140	29	0.37	1.20	0.38	0.00	2.00	0.10	89.00	2.12	0.39	1.00	0.00
92410510	Soft drink, fruit…	370.00	240	148	0.19	0.00	0.00	0.00	7.00	0.41	4.00	0.00	0.00	33.00	33.26
58105105	Pupusa, bean-filled	103.00	140	161	5.24	4.80	0.14	0.00	62.00	2.91	212.00	1.24	0.26	62.00	0.00
58163410	Spanish rice	212.63	243	185	4.06	2.60	1.54	0.00	69.00	2.75	411.00	2.52	0.47	637.00	1.11
92611100	Oatmeal bev w/ milk	207.19	240	169	3.49	0.50	0.09	1.20	107.00	0.24	137.00	1.12	1.74	43.00	24.61
51161270	Roll, sweet…	93.00	55	315	5.71	1.30	0.62	0.20	18.00	2.15	73.00	6.57	2.07	176.00	20.64

Step 1: Each nutrient value is divided by the daily value to yield a fraction. The fraction was capped at 1.

Step 2: The grams consumed are divided by the grams per RACC to yield a fraction of RACC consumed.

Step 3: The kcal consumed are divided by 100 to yield a fraction of 100 kcal portions consumed.

Step 4: Sum each column of calculated values from Steps 1-3.

Using the example in TableC-1, the calculated values from Steps 1-3 and sums from Step 4 are shown in TableC-2.

Table C-2. Calculated Values from Steps 1-4
Food Code	Description	Grams Consumed	Fraction of RACC	Fraction of 100 kcal	Protein (g)	Fiber (g)	Vitamin E (mg)	Vitamin D (mcg)	Calcium (mg)	Iron (mg)	Potassium (mg)	Unsat Fat (g)	Sat Fat (g)	Sodium (mg)	Added Sugar (g)
— Means no data.
57123000	Cheerios	31.88	1.06	1.17	0.07	0.13	0.01	0.11	0.13	0.56	0.06	0.03	0.02	0.09	0.02
11112110	Milk…2% fat	129.63	0.53	0.65	0.09	0.00	0.00	0.16	0.16	0.00	0.05	0.02	0.08	0.03	0.00
92114000	Coffee…	119.20	0.50	0.02	0.00	0.00	0.00	0.00	0.01	0.00	0.01	0.00	0.00	0.00	0.00
91101010	Sugar, white…	4.17	1.04	0.16	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.08
11112110	Milk, cow's, fluid, 2% fat	81.33	0.33	0.41	0.05	0.00	0.00	0.10	0.10	0.00	0.03	0.01	0.05	0.02	0.00
52215200	Tortilla, flour (wheat)	51.00	0.93	1.59	0.08	0.06	0.01	0.00	0.07	0.09	0.02	0.06	0.05	0.14	0.00
41205010	Refried beans	31.63	0.24	0.45	0.04	0.08	0.01	0.00	0.01	0.03	0.03	0.03	0.02	0.05	0.00
63105010	Avocado, raw	18.25	0.13	0.29	0.01	0.05	0.02	0.00	0.00	0.01	0.03	0.05	0.02	0.00	0.00
92410510	Soft drink, fruit…	370.00	1.54	1.48	0.00	0.00	0.00	0.00	0.01	0.02	0.00	0.00	0.00	0.01	0.67
58105105	Pupusa, bean-filled	103.00	0.74	1.61	0.10	0.19	0.01	0.00	0.06	0.16	0.06	0.03	0.01	0.03	0.00
58163410	Spanish rice	212.63	0.88	1.85	0.08	0.10	0.08	0.00	0.07	0.15	0.12	0.06	0.02	0.27	0.02
92611100	Oatmeal bev w/ milk	207.19	0.86	1.69	0.07	0.02	0.00	0.12	0.11	0.01	0.04	0.03	0.09	0.02	0.49
51161270	Roll, sweet…	93.00	1.69	3.15	0.11	0.05	0.03	0.02	0.02	0.12	0.02	0.15	0.10	0.07	0.41
—	Sum	—	10.47	14.52	0.72	0.68	0.17	0.51	0.73	1.17	0.47	0.46	0.46	0.71	1.70

Step 5: Add the sums of the positive nutrients, multiply by 100, divide by the sum of the kcal/100, and divide by 8.

(0.72 + 0.68 + 0.17 + 0.51 + 0.73 + 1.17 + 0.47 + 0.46 + 0.46) * 100 / 14.52 / 8 = 4.22

Step 6: Add the sums of the negative nutrients, multiply by 100, divide by the sum of the kcal/100, and divide by 3.

(0.46 + 0.71 + 1.70) * 100 / 14.52 / 3 = 6.59

Step 7: Subtract the value from Step 5 − Step 6 to yield the individual's score based on 100 kcal.

4.22 − 6.59 = −2.37

To calculate the individual's score based on RACC, the sum of the grams/RACC column replaces the sum of the kcal/100 in steps 5 and 6. For the above example, the individual's score based on RACC is −3.28.

D: Calculation of Healthy Eating Index

The Healthy Eating Index (HEI) is a measure of diet quality that assesses conformance to federal dietary guidance. Food group standards are based on recommendations found in MyPyramid. The standards were created using a density approach expressed as a percentage of calories or per 1,000 calories.

HEI Components and Standards for Scoring^a
Component	Maximum Points	Standard for Maximum Score	Standard for Minimum Score of Zero
^a Intakes between the minimum and maximum levels are scored proportionately, except for saturated fat and sodium (see note 5). ^b Legumes counted as vegetables only after meat and beans standard is met. ^c Includes all milk products, such as fluid milk, yogurt, and cheese. ^d Includes nonhydrogenated vegetable oils and oils in fish, nuts, and seeds. ^e Saturated fat and sodium get a score of 8 for the intake levels that reflect the 2005 Dietary Guidelines, <10% of calories from saturated fat and 1.1 grams of sodium/1,000 kcal, respectively. Source: Guenther, P. M., Krebs-Smith, S. M., Reedy, J., Britten, P., Juan, W-Y., Lino, M., Carlson, A., Hiza, H. A., and Basiotis, P. P. (2006). Healthy Eating Index-2005. CNPP Fact Sheet No. 1. Center for Nutrition Policy and Promotion, U.S. Department of Agriculture. Available at: http://www.cnpp.usda.gov/HealthyEatingIndex.htm.
Total fruit (includes 100% juice)	5	≥0.8 cup eq per 1,000 kcal	None
Whole fruit (not juice)	5	≥0.4 cup eq per 1,000 kcal	None
Total vegetables	5	≥1.1 cup eq per 1,000 kcal	None
Dark green and orange vegetables and legumes^b	5	≥0.4 cup eq per 1,000 kcal	None
Total grains	5	≥3.0 oz eq per 1,000 kcal	None
Whole grains	5	≥1.5 oz eq per 1,000 kcal	None
Milk^c	10	≥1.3 cup eq per 1,000 kcal	None
Meat and beans	10	≥2.5 oz eq per 1,000 kcal	None
Oils^d	10	≥12 grams per 1,000 kcal	None
Saturated fat	10	≤7% of energy^e	≥15% of energy
Sodium	10	≤700 mg per 1,000 kcal	≥2,000 mg per 1,000 kcal
Calories from solid fat, alcohol, and added sugar	20	≤20% of energy	≥50% of energy

E: NDS Scores of Food Supgroupings and Selected Foods Using the Baseline Algorithm

Table E-1. Rankings of Mean Nutrient Density Scores per 100 kcal (NDS1KCAL) of Selected Subgroupings of Foods
Rank	Food Group	Mean Score	Minimum	Maximum
The means were calculated from subgroupings of 570 foods. Some subgroupings contain only a few foods (e.g., rice).
1	Dairy Milk	2.91	−2.81	11.12
2	Yogurt	−0.36	−7.16	7.96
3	Cheese	−5.45	−9.71	−1.50
4	Ice Cream	−7.26	−12.43	1.13
5	Other	−8.16	−16.06	−3.04
1	Grains Cereals, breakfast foods	−0.30	−7.08	8.23
2	Breads	−1.29	−6.96	0.87
3	Quickbreads	−1.49	−4.77	2.98
4	Crackers, salty snacks	−1.66	−8.27	3.60
5	Rice	−2.44	−2.47	−2.40
6	Mixed grain products	−2.85	−11.76	1.46
7	Cakes, cookies, pastries	−5.37	−9.79	3.40
1	Vegetables Vegetables other than potatoes	5.89	−9.90	30.62
2	Potatoes	−1.07	−3.24	1.93
3	Soup	−5.19	−10.26	0.51
4	Condiments	−18.34	−86.08	0.39
1	Fats, Oils, and Dressings Fats and oils	−3.48	−11.73	1.14
2	Salad dressings	−8.94	−33.41	−0.85
1	Sweets and Beverages Beverages	−4.63	−21.51	17.92
2	Sugars, sweets	−10.07	−17.80	2.32

Table E-2. Rankings of Mean Nutrient Density Scores per RACC (NDS1RACC) of Selected Subgroupings of Foods
Rank	Food Group	Mean Score	Minimum	Maximum
The means were calculated from subgroupings of 570 foods. Some subgroupings contain only a few foods (e.g., rice).
1	Dairy Milk	2.44	−5.82	9.27
2	Yogurt	−3.07	−19.17	8.20
3	Other	−4.96	−14.01	−0.94
4	Cheese	−5.54	−8.81	−0.65
5	Ice cream	−9.14	−17.92	1.06
1	Grains Cereals, breakfast foods	−0.42	−8.17	10.09
2	Breads	−2.20	−15.53	1.19
3	Crackers, salty snacks	−2.31	−10.70	4.18
4	Quickbreads	−2.96	−9.24	3.58
5	Rice	−4.66	−4.91	−4.46
6	Mixed grain products	−6.99	−14.00	5.75
7	Cakes, cookies, pastries	−11.15	−33.82	4.94
1	Vegetables Vegetables other than potatoes	0.75	−4.51	5.99
2	Potatoes	−1.58	−6.17	3.13
3	Condiments	−2.16	−5.80	0.18
4	Soup	−5.41	−10.71	1.17
1	Fats, Oils, and Dressings Fats and oils	−2.86	−11.76	1.36
2	Salad dressings	−4.31	−9.98	−0.92
1	Sweets and Beverages Beverages	−5.41	−19.50	3.36
2	Sugars, sweets	−12.70	−23.77	0.85

Table E-3. Nutrient Density Scores of 100 Foods Using the Baseline Algorithms (NDS1)
Food Code	Description	Nutrient Density Score per 100 Kcal (NDS1KCAL)	Nutrient Density Score per RACC (NDS1RACC)
Foodcodes and descriptions are from USDA Food and Nutrient Database for Dietary Studies (FNDDS) Version 4.0 — Means no data. ^a The same nutrient values as foodcode 71101000 were used to calculate the score except the sodium value was replaced with 5 mg per 100 grams.
11111000	Dairy Foods Milk, cow's, fluid, whole	1.54	2.30
11112110	Milk, cow's, fluid, 2% fat	3.58	4.37
11112210	Milk, cow's, fluid, 1% fat	6.88	7.05
11113000	Milk, cow's, fluid, skim or nonfat, 0.5% or less butterfat	11.12	9.27
11411100	Yogurt, plain, whole milk	−1.18	−1.62
11411200	Yogurt, plain, low-fat milk	3.18	4.51
11411300	Yogurt, plain, nonfat milk	6.51	8.20
11431000	Yogurt, fruit variety, whole milk	−7.16	−19.17
11432000	Yogurt, fruit variety, low-fat milk	−6.69	−15.36
11432500	Yogurt, fruit variety, low-fat milk, sweetened with low-calorie sweetener	2.33	5.50
11433000	Yogurt, fruit variety, nonfat milk	−4.53	−9.69
11433500	Yogurt, fruit variety, nonfat milk, sweetened with low-calorie sweetener	7.96	8.06
11511100	Milk, chocolate, whole milk-based	−2.81	−5.82
11511200	Milk, chocolate, reduced fat milk-based, 2%	−1.35	−2.57
11511400	Milk, chocolate, low-fat milk-based	−0.35	−0.56
11511300	Milk, chocolate, skim milk-based	0.66	0.93
12120100	Cream, half and half	−6.47	−2.52
12210400	Cream substitute, powdered	−16.06	−1.75
13110100	Ice cream, regular, flavors other than chocolate	−9.46	−13.02
13130300	Light ice cream, flavors other than chocolate	−5.74	−6.77
13160400	Fat free ice cream, flavors other than chocolate	1.13	1.06
14410200	Cheese, processed, American or Cheddar type	−8.51	−8.58
21101120	Meats, Poultry, and Fish Beef steak, broiled or baked, lean and fat eaten	−0.75	−1.52
21500100	Ground beef or patty, cooked, NS as to percent lean	−1.48	−3.25
22600200	Pork bacon, NS as to fresh, smoked or cured, cooked	−6.39	−5.19
24122100	Chicken, breast, roasted, broiled, or baked, NS as to skin eaten	0.87	1.45
24127110	Chicken, breast, coated, baked or fried, prepared with skin, skin/coating eaten	−0.62	−1.37
24198700	Chicken patty, fillet, or tenders, breaded, cooked	−1.72	−4.22
24198740	Chicken nuggets	−0.82	−2.07
25210210	Frankfurter or hot dog, beef	−7.51	−14.08
25230210	Ham, sliced, prepackaged or deli, luncheon meat	−8.82	−6.89
25230310	Chicken or turkey loaf, prepackaged or deli, luncheon meat	−5.92	−3.88
28140720	Chicken patty, or nuggets, boneless, breaded, potatoes, vegetable (frozen meal)	−1.09	−4.19
31105000	Eggs Egg, whole, fried	−0.49	−0.45
32105000	Egg omelet or scrambled egg, fat added in cooking	−0.15	−0.28
41205010	Legumes and Nuts Refried beans	1.25	2.32
42100100	Almonds, NFS	7.33	12.65
42202000	Peanut butter	0.62	1.17
42202150	Peanut butter, reduced fat	1.16	2.17
51101000	Grain Products Bread, white	−1.79	−2.37
51300110	Bread, whole wheat, NS as to 100%	−0.08	−0.11
51301010	Bread, wheat or cracked wheat	−0.29	−0.38
52215100	Tortilla, corn	2.98	3.58
52215200	Tortilla, flour (wheat)	−0.70	−1.21
53105260	Cake, chocolate, devil's food, or fudge, with icing, coating, or filling, made from home recipe or purchased ready-to-eat	−9.28	−27.75
53206000	Cookie, chocolate chip	−7.08	−10.06
53540400	Kellogg's Nutri-Grain Cereal Bar	−3.11	−4.68
54301000	Cracker, snack	−1.33	−2.01
54401050	Salty snacks, corn or cornmeal base, corn puffs and twists; corn-cheese puffs and twists	−1.01	−1.71
54401080	Salty snacks, corn or cornmeal base, tortilla chips	1.27	1.87
54403020	Popcorn, popped in oil, buttered	−2.45	−4.38
54403070	Popcorn, popped in oil, low-fat	0.16	0.20
54408010	Pretzels, hard	−2.74	−3.12
55101000	Pancakes, plain	−0.87	−2.16
56205010	Rice, white, cooked, regular, fat not added in cooking	−2.47	−4.46
56205110	Rice, brown, cooked, regular, fat not added in cooking	−1.97	−3.03
58106225	Pizza, cheese, regular crust	−2.66	−9.83
58132310	Spaghetti with tomato sauce and meatballs or spaghetti with meat sauce or spaghetti with meat sauce and meatballs	−0.25	−0.94
61119010	Fruit Orange, raw	5.86	3.85
61210220	Orange juice, canned, bottled or in a carton	2.60	3.05
61210250	Orange juice, with calcium added, canned, bottled or in a carton	10.15	11.45
63101000	Apple, raw	3.60	2.62
63107010	Banana, raw	3.35	4.18
63149010	Watermelon, raw	3.41	2.86
63223020	Strawberries, raw	7.49	3.36
64100110	Fruit juice blend, 100% juice	2.55	3.19
64104010	Apple juice	1.30	1.43
71101000	Vegetables White potato, baked, peel not eaten (with added salt)	−1.16	−1.17
—	White potato, baked, peel not eaten (no added salt)^a	3.11	3.14
71201010	White potato, chips	1.93	3.13
71401030	White potato, french fries, from frozen, deep fried	0.15	0.34
72125100	Spinach, raw	30.62	5.99
73101010	Carrots, raw	6.89	2.40
74101000	Tomatoes, raw	12.40	1.90
74401010	Tomato catsup	−25.28	−3.68
74402150	Salsa, red, cooked, not homemade	−17.42	−1.51
75113000	Lettuce, raw	12.89	1.53
75117020	Onions, mature, raw	4.94	1.68
75503010	Cucumber pickles, dill	−86.08	−3.10
81101000	Fats, Oils, and Dressings Butter, stick, salted	−11.55	−11.76
81101010	Butter, whipped, tub, salted	−11.73	−7.90
81103080	Margarine-like spread, tub, salted	−0.45	−0.35
83106000	Italian dressing, made with vinegar and oil	−8.98	−7.84
83107000	Mayonnaise, regular	−0.85	−0.92
83110000	Mayonnaise-type salad dressing	−3.20	−1.87
83112500	Creamy dressing, made with sour cream and/or buttermilk and oil	−3.22	−4.69
83205500	Italian dressing, reduced calorie, fat-free	−33.41	−4.71
91101010	Sweets and Beverages Sugar, white, granulated or lump	−17.22	−2.67
91107000	Sucralose-based sweetener, sugar substitute	0.00	0.00
91401000	Jelly, all flavors	−11.30	−5.71
91705010	Milk chocolate candy, plain	−9.27	−19.84
91745020	Hard candy	−10.72	−6.34
92101000	Coffee, made from ground, regular	16.66	0.52
92302000	Tea, leaf, unsweetened	9.53	0.25
92302200	Tea, leaf, presweetened with sugar	−16.24	−7.79
92410310	Soft drink, cola-type	−15.99	−14.20
92410320	Soft drink, cola-type, sugar-free	2.94	0.14
92510610	Fruit juice drink	−13.05	−14.40
92530610	Fruit juice drink, with high vitamin C	−10.93	−12.33
92560100	Gatorade Thirst Quencher sports drink	−15.49	−9.67

F: Algorithm Scores and Rankings of Major Food Groups of the Modified Algorithms

Figure F-1. Box Plot of Nutrient Density Scores per 100 Kcal for Modified Algorithm Removing Unsaturated Fat (NDS2KCAL)

Box plot shows the mean and distribution of nutrient density scores for all 570 foods and for major food groups. The mean is shown by the diamond; the left side of the boxes represents the 25th percentile of scores, and the right side of the boxes represents the 75th percentile. The minimum and maximum scores are represented at the ends of the horizontal lines.

Figure F-2. Box Plot of Nutrient Density Scores per RACC for Modified Algorithm Removing Unsaturated Fat (NDS2RACC)

Figure F-3. Box Plot of Nutrient Density Scores per 100 Kcal for Modified Algorithm Removing Added Sugars (NDS3KCAL)

Figure F-4. Box Plot of Nutrient Density Scores per RACC for Modified Algorithm Removing Added Sugars (NDS3RACC)

Figure F-5. Box Plot of Nutrient Density Scores per 100 Kcal for Modified Algorithm Adding Vitamin C (NDS4KCAL)

Figure F-6. Box Plot of Nutrient Density Scores per RACC for Modified Algorithm Adding Vitamin C (NDS4RACC)

Figure F-7. Box Plot of Nutrient Density Scores per 100 Kcal for Modified Algorithm Adding Whole Grains (NDS5KCAL)

Figure F-8. Box Plot of Nutrient Density Scores per RACC for Modified Algorithm Adding Whole Grains (NDS5RACC)

Figure F-9. Box Plot of Nutrient Density Scores per 100 Kcal for Modified Algorithm Adding Vitamin C and Whole Grains (NDS5KCAL)

Figure F-10 Box Plot of Nutrient Density Scores per RACC for Modified Algorithm Adding Vitamin C and Whole Grains (NDS5RACC)

G: Supplementary Analyses of MAXR Regressions with Whole Grains and Total Sugars

Supplementary analyses were conducted with the addition of whole grains and the replacement of added sugars with total sugars. There was only slight improvement in R² with the addition of whole grains (TablesG-1 through G-4). The plateau R² appeared to occur at eight or nine nutrient/component models. The best nine nutrient/component models had adjusted R² of 66% and 61%, on per 100 kcal and per RACC bases, respectively.

Additional analyses were performed to examine the distribution of R² among all 24, 310 eight nutrient/component models. The minimum, 25th, 50th, 75th, and 100th percentiles of R² were 0.209, 0.435, 0.485, 0.529, and 0.647 for eight nutrient/component models on per 100 kcal basis, and 0.129, 0.351, 0.421, 0.491, and 0.604 for models on per RACC basis.

We also examined the top 10 eight nutrient/component models in terms of R² to examine differences in nutrients and beta-coefficients (TablesG-5 and G-6). All of the top 10 models included fiber, unsaturated fat, saturated fat, sodium, and added sugar. All but one of the models included protein. Whole grains were not retained in the top two models, but were present in 4 or 5 of the top 10 models for both unit bases. Vitamin D replaced calcium in the second highest R models for both unit bases. Some of the top 10 models had potassium, one of the models included vitamin A, one model included vitamin B₁₂, and none of the models included vitamin E, iron, magnesium, or folic acid.

In analyses with total sugars replacing added sugars (TableG-7), the nine nutrient/component model was the plateau for R² on a per 100 kcal basis, explaining 64% of the variance in HEI scores. Total sugars did not remain in this model and, in fact, only remained in the 17 nutrient/component model. For the regressions using a per RACC basis (TableG-8), the R² appears to plateau in the 10 nutrient/component model, explaining 58% of the variance in HEI. Total sugars were retained in models with seven or more nutrient/component variables on a RACC basis. It is unclear why total sugars were apparently explaining the added sugar component of HEI scores on a RACC basis but not on a per 100 kcal basis.

Table G-1. Beta Coefficients from Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per 100 Kcal (Includes Whole Grains)
Number of Vari- ables	R- Square	Adjusted R- Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potassium	Unsat- urated Fat	Magnes- ium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Added Sugar
— Means no data.
0	0.0417	0.0413	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
1	0.3657	0.3655	—	5.3217	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2	0.4346	0.4343	—	4.6677	—	—	—	—	—	—	—	—	—	—	—	—	−1.8039	—	—
3	0.5211	0.5209	—	3.5917	—	—	—	—	—	—	—	—	—	—	—	—	−2.3941	—	−0.8843
4	0.5526	0.5523	—	3.5531	—	1.2383	—	—	—	—	—	—	—	—	—	—	−2.4270	—	−0.8187
5	0.5818	0.5815	1.4183	3.5986	—	—	—	—	—	—	—	—	—	—	—	—	−2.3479	−1.3146	−0.7915
6	0.6080	0.6078	1.5862	3.7418	—	—	—	—	—	1.7803	—	—	—	—	—	—	−2.7342	−1.4097	−0.6449
7	0.6330	0.6327	1.3749	3.5078	—	—	1.1422	—	—	2.4089	—	—	—	—	—	—	−3.0957	−1.3670	−0.5563
8	0.6476	0.6473	1.3966	3.1261	—	—	1.0008	—	—	2.5151	—	—	0.3653	—	—	—	−2.9478	−1.3347	−0.5160
9	0.6578	0.6575	1.3963	2.8113	—	—	0.8762	—	—	2.5319	—	—	0.3703	—	—	0.3109	−2.9061	−1.2995	−0.5102
10	0.6611	0.6608	1.2993	2.8792	—	0.5317	0.5864	—	—	2.5135	—	—	0.3704	—	—	0.2980	−2.8630	−1.2406	−0.5140
11	0.6631	0.6628	1.3128	2.8177	—	—	0.9412	—	1.1446	2.6444	−0.9052	—	0.2937	—	—	0.3351	−3.0013	−1.3166	−0.5043
12	0.6648	0.6644	1.2569	2.8697	—	0.3830	0.7251	—	0.9703	2.6144	−0.7798	—	0.3054	—	—	0.3223	−2.9576	−1.2720	−0.5084
13	0.6657	0.6654	1.2721	2.8259	—	0.3238	0.6984	—	1.0777	2.6792	−0.7482	—	0.3037	0.2605	—	0.3051	−2.9207	−1.2952	−0.4923
14	0.6666	0.6662	1.3258	2.8913	—	0.3458	0.6745	−0.3821	1.0792	2.6935	−0.6733	—	0.3085	0.4968	—	0.3130	−2.9243	−1.2759	−0.4813
15	0.6674	0.6669	1.3301	2.8660	0.4226	0.3448	0.6741	−0.3999	1.1062	2.5547	−0.7896	—	0.2874	0.4745	—	0.3222	−2.8995	−1.2695	−0.4847
16	0.6676	0.6672	1.3283	2.8485	0.3777	0.3133	0.6531	−0.4387	1.0920	2.5668	−0.7872	0.1833	0.2819	0.4782	—	0.3208	−2.9183	−1.2711	−0.4895
17	0.6678	0.6674	1.3607	2.8218	0.4038	0.3790	0.6507	−0.3776	1.1092	2.5525	−0.8003	0.1849	0.2807	0.4947	−0.1267	0.3192	−2.8968	−1.2770	−0.4854

Table G-2. P-Values from Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per 100 Kcal (Includes Whole Grains)
Number of Vari- ables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsat- urated Fat	Magnes- ium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Added Sugar
— Means no data.
0	0.0417	0.0413	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
1	0.3657	0.3655	—	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2	0.4346	0.4343	—	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	0.0000	—	—
3	0.5211	0.5209	—	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	0.0000	—	0.0000
4	0.5526	0.5523	—	0.0000	—	0.0000	—	—	—	—	—	—	—	—	—	—	0.0000	—	0.0000
5	0.5818	0.5815	0.0000	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	0.0000	0.0000	0.0000
6	0.6080	0.6078	0.0000	0.0000	—	—	—	—	—	0.0000	—	—	—	—	—	—	0.0000	0.0000	0.0000
7	0.6330	0.6327	0.0000	0.0000	—	—	0.0000	—	—	0.0000	—	—	—	—	—	—	0.0000	0.0000	0.0000
8	0.6476	0.6473	0.0000	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	—	—	—	0.0000	0.0000	0.0000
9	0.6578	0.6575	0.0000	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
10	0.6611	0.6608	0.0000	0.0000	—	0.0000	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
11	0.6631	0.6628	0.0000	0.0000	—	—	0.0000	—	0.0000	0.0000	0.0002	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
12	0.6648	0.6644	0.0000	0.0000	—	0.0000	0.0000	—	0.0000	0.0000	0.0004	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
13	0.6657	0.6654	0.0000	0.0000	—	0.0000	0.0000	—	0.0000	0.0000	0.0005	—	0.0000	0.0023	—	0.0000	0.0000	0.0000	0.0000
14	0.6666	0.6662	0.0000	0.0000	—	0.0000	0.0000	0.0001	0.0000	0.0000	0.0015	—	0.0000	0.0000	—	0.0000	0.0000	0.0000	0.0000
15	0.6674	0.6669	0.0000	0.0000	0.0926	0.0000	0.0000	0.0001	0.0000	0.0000	0.0004	—	0.0000	0.0000	—	0.0000	0.0000	0.0000	0.0000
16	0.6676	0.6672	0.0000	0.0000	0.1380	0.0001	0.0000	0.0000	0.0000	0.0000	0.0004	0.1193	0.0000	0.0000	—	0.0000	0.0000	0.0000	0.0000
17	0.6678	0.6674	0.0000	0.0000	0.1178	0.0000	0.0000	0.0006	0.0000	0.0000	0.0002	0.1191	0.0000	0.0000	0.0345	0.0000	0.0000	0.0000	0.0000

Table G-3. Beta Coefficients from Regression Models Using the Maximum R-square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per RACC (Includes Whole Grains)
Number of Vari- ables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potassium	Unsaturated Fat	Mag- nesium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Added Sugar
— Means no data.
0	0.0417	0.0413	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
1	0.2074	0.2071	—	3.9647	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2	0.4036	0.4033	—	4.7469	—	—	—	—	—	—	—	—	—	—	—	—	−2.0714	—	—
3	0.4732	0.4730	—	4.5970	—	—	—	—	—	—	—	—	—	—	—	—	−1.7868	—	−0.6528
4	0.5218	0.5215	—	3.1269	—	—	—	—	3.1410	—	—	—	—	—	—	—	−2.1354	—	−0.5857
5	0.5416	0.5412	—	2.8476	—	—	—	—	3.1097	1.3403	—	—	—	—	—	—	−2.7630	—	−0.6113
6	0.5654	0.5651	—	3.1202	—	—	—	—	3.3033	1.7656	—	—	—	—	—	—	−2.5134	−1.0465	−0.6170
7	0.5883	0.5880	1.4790	3.6688	—	—	—	—	—	1.6051	—	—	0.4735	—	—	—	−2.5594	−1.5006	−0.6409
8	0.6042	0.6039	1.1864	3.4482	—	—	0.9461	—	—	1.9831	—	—	0.3893	—	—	—	−2.9860	−1.4492	−0.6370
9	0.6120	0.6116	1.1924	3.1817	—	—	0.8237	—	—	1.9857	—	—	0.3866	—	—	0.2925	−2.9430	−1.4111	−0.6318
10	0.6149	0.6145	1.0171	2.8964	—	—	0.7073	—	1.0068	1.9651	—	—	0.3142	—	—	0.3023	−2.9363	−1.3732	−0.6173
11	0.6158	0.6154	1.0706	2.9918	—	—	0.7466	−0.2842	0.9935	1.9732	—	—	0.3199	—	—	0.3241	−2.9596	−1.3319	−0.6048
12	0.6173	0.6168	1.0876	3.0050	—	—	0.6287	−0.3890	0.9161	1.9895	—	0.4868	0.3113	—	—	0.3200	−2.9642	−1.3167	−0.6025
13	0.6177	0.6173	1.1059	3.0101	—	—	0.5920	−0.5888	0.9887	2.0044	—	0.4857	0.3097	0.2496	—	0.3143	−2.9617	−1.3323	−0.6019
14	0.6186	0.6181	1.1500	3.1133	0.5177	—	0.6815	−0.3721	1.1414	1.8905	−0.6801	0.4323	0.2883	—	—	0.3385	−2.9565	−1.3492	−0.6037
15	0.6189	0.6184	1.1616	3.1103	0.4934	—	0.6477	−0.5395	1.1884	1.9068	−0.6397	0.4336	0.2881	0.2075	—	0.3329	−2.9542	−1.3601	−0.6031
16	0.6191	0.6186	1.1545	3.1594	0.5125	0.1516	0.5787	−0.5341	1.0827	1.8949	−0.6280	0.3864	0.2970	0.1878	—	0.3311	−2.9371	−1.3440	−0.6033
17	0.6191	0.6186	1.1415	3.1719	0.5012	0.1242	0.5824	−0.5624	1.0721	1.9053	−0.6143	0.3833	0.2980	0.1849	0.0534	0.3317	−2.9457	−1.3390	−0.6034

Table G-4. P-Values from Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per RACC (includes Whole Grains)
Number of Vari- ables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsat- urated Fat	Mag- nesium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Added Sugar
— Means no data.
0	0.0417	0.0413	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
1	0.2074	0.2071	—	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2	0.4036	0.4033	—	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	0.0000	—	—
3	0.4732	0.4730	—	0.0000	—	—	—	—	—	—	—	—	—	—	—	—	0.0000	—	0.0000
4	0.5218	0.5215	—	0.0000	—	—	—	—	0.0000	—	—	—	—	—	—	—	0.0000	—	0.0000
5	0.5416	0.5412	—	0.0000	—	—	—	—	0.0000	0.0000	—	—	—	—	—	—	0.0000	—	0.0000
6	0.5654	0.5651	—	0.0000	—	—	—	—	0.0000	0.0000	—	—	—	—	—	—	0.0000	0.0000	0.0000
7	0.5883	0.5880	0.0000	0.0000	—	—	—	—	—	0.0000	—	—	0.0000	—	—	—	0.0000	0.0000	0.0000
8	0.6042	0.6039	0.0000	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	—	—	—	0.0000	0.0000	0.0000
9	0.6120	0.6116	0.0000	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
10	0.6149	0.6145	0.0000	0.0000	—	—	0.0000	—	0.0000	0.0000	—	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
11	0.6158	0.6154	0.0000	0.0000	—	—	0.0000	0.0139	0.0000	0.0000	—	—	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
12	0.6173	0.6168	0.0000	0.0000	—	—	0.0000	0.0006	0.0000	0.0000	—	0.0050	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
13	0.6177	0.6173	0.0000	0.0000	—	—	0.0000	0.0001	0.0000	0.0000	—	0.0047	0.0000	0.0121	—	0.0000	0.0000	0.0000	0.0000
14	0.6186	0.6181	0.0000	0.0000	0.0302	—	0.0000	0.0002	0.0000	0.0000	0.1105	0.0077	0.0000	—	—	0.0000	0.0000	0.0000	0.0000
15	0.6189	0.6184	0.0000	0.0000	0.0349	—	0.0000	0.0000	0.0000	0.0000	0.1219	0.0073	0.0000	0.0233	—	0.0000	0.0000	0.0000	0.0000
16	0.6191	0.6186	0.0000	0.0000	0.0275	0.0363	0.0000	0.0000	0.0000	0.0000	0.1243	0.0160	0.0000	0.0350	—	0.0000	0.0000	0.0000	0.0000
17	0.6191	0.6186	0.0000	0.0000	0.0294	0.0986	0.0000	0.0000	0.0000	0.0000	0.1331	0.0171	0.0000	0.0401	0.3695	0.0000	0.0000	0.0000	0.0000

Table G-5. Beta Coefficients from the Top 10 8-Nutrient/Component Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per 100 Kcal (Includes Whole Grains)
Number of Vari- ables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsat- urated Fat	Mag- nesium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Added Sugar
— Means no data.
8	0.6476	0.6473	1.3966	3.1261	—	—	1.0008	—	—	2.5151	—	—	0.3653	—	—	—	−2.9478	−1.3347	−0.5160
8	0.6462	0.6459	1.3006	3.2871	—	0.9852	—	—	—	2.2474	—	—	0.3959	—	—	—	−2.7125	−1.2367	−0.5536
8	0.6438	0.6435	1.5531	2.8613	—	—	—	—	—	2.0946	—	—	0.4276	—	—	0.3714	−2.6125	−1.3188	−0.5672
8	0.6428	0.6425	1.3743	3.2043	—	—	1.0221	—	—	2.4240	—	—	—	—	—	0.3047	−3.0568	−1.3329	−0.5512
8	0.6398	0.6395	1.2087	3.1517	—	—	0.9945	—	1.1475	2.5123	—	—	—	—	—	—	−3.0261	−1.3517	−0.5086
8	0.6393	0.6390	1.2902	3.3917	—	0.9555	—	—	—	2.1242	—	—	—	—	—	0.3165	−2.8207	−1.2414	−0.5953
8	0.6377	0.6374	1.2853	2.7882	—	—	—	—	1.6148	2.1417	—	—	—	—	—	0.3807	−2.7009	−1.3323	−0.5436
8	0.6372	0.6369	1.2668	3.5694	—	0.5930	0.8133	—	—	2.3890	—	—	—	—	—	—	−3.0458	−1.2997	−0.5603
8	0.6368	0.6365	1.7522	3.2094	—	—	1.0236	—	—	3.0447	—	—	0.4119	—	—	0.3182	−2.7854	−1.1477	—
8	0.6365	0.6362	1.3273	3.3569	—	—	0.9702	—	—	2.4179	—	0.6240	—	—	—	—	−3.1058	−1.3640	−0.5683

Table G-6. Beta Coefficients from the Top 10 8-Nutrient/Component Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per RACC (Includes Whole Grains)
Number of Vari- ables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsat- urated Fat	Magnes- ium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	WholevGrains	SaturatedvFat	Sodium	Added Sugar
— Means no data.
8	0.6042	0.6039	1.1864	3.4482	—	—	0.9461	—	—	1.9831	—	—	0.3893	—	—	—	−2.9860	−1.4492	−0.6370
8	0.6010	0.6007	1.1780	3.6412	—	0.8370	—	—	—	1.7915	—	—	0.4293	—	—	—	−2.7085	−1.3792	−0.6360
8	0.6003	0.6000	1.4400	3.3083	—	—	—	—	—	1.6680	—	—	0.4569	—	—	0.3575	−2.5742	−1.4459	−0.6340
8	0.5973	0.5970	0.9098	3.2602	—	—	0.9075	—	1.6320	1.9643	—	—	—	—	—	—	−3.0688	−1.3947	−0.6113
8	0.5968	0.5965	1.0217	2.9211	—	—	—	—	2.2568	1.6805	—	—	—	—	—	0.3736	−2.7045	−1.3609	−0.5986
8	0.5968	0.5964	1.3557	3.4904	—	—	—	—	—	1.7490	—	0.9986	0.4141	—	—	—	−2.6813	−1.4756	−0.6418
8	0.5962	0.5959	1.2036	3.5476	—	—	1.0261	—	—	2.0067	—	—	—	—	—	0.2966	−3.0721	−1.4175	−0.6283
8	0.5948	0.5945	1.1752	3.2315	—	—	—	—	1.4549	1.6401	—	—	0.3544	—	—	—	−2.6250	−1.4389	−0.6195
8	0.5932	0.5929	—	2.7850	—	—	0.9033	—	2.4461	2.1462	—	—	—	—	—	0.3154	−2.9290	−1.0388	−0.6155
8	0.5919	0.5915	1.2708	3.6709	—	—	—	—	—	1.7102	—	—	0.4580	—	0.3907	—	−2.6426	−1.4692	−0.6498

Table G-7. Beta Coefficients from Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per 100 Kcal (Total Sugar Replacing Added Sugar)
Number of Vari- ables	R- Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsat- urated Fat	Mag- nesium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Total Sugar
— Means no data.
0	0.0417	0.0413	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
1	0.3657	0.3655	—	5.3217	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2	0.4346	0.4343	—	4.6677	—	—	—	—	—	—	—	—	—	—	—	—	−1.8039	—	—
3	0.5020	0.5017	1.6699	4.2784	—	—	—	—	—	—	—	—	—	—	—	—	−2.0087	—	—
4	0.5406	0.5403	1.6797	4.2618	—	—	—	—	—	2.0661	—	—	—	—	—	—	−2.5611	—	—
5	0.5761	0.5758	1.3562	3.8857	—	—	1.3456	—	—	2.7242	—	—	—	—	—	—	−3.0194	—	—
6	0.6079	0.6076	1.7640	4.0018	—	—	1.3252	—	—	2.9597	—	—	—	—	—	—	−2.9824	−1.2048	—
7	0.6262	0.6259	1.7567	3.5363	—	—	1.1527	—	—	3.0335	—	—	0.4073	—	—	—	−2.8267	−1.1819	—
8	0.6368	0.6365	1.7522	3.2094	—	—	1.0236	—	—	3.0447	—	—	0.4119	—	—	0.3182	−2.7854	−1.1477	—
9	0.6421	0.6418	1.5703	2.9171	—	—	0.9005	—	1.0527	3.0833	—	—	0.3379	—	—	0.3293	−2.7573	−1.1494	—
10	0.6456	0.6453	1.5701	2.8548	—	—	0.7978	—	1.2268	3.1744	—	—	0.3345	0.4866	—	0.2946	−2.6953	−1.1903	—
11	0.6476	0.6472	1.6495	2.9614	—	—	0.7946	−0.5683	1.2759	3.1893	—	—	0.3374	0.8353	—	0.3091	−2.7220	−1.1721	—
12	0.6487	0.6483	1.6138	3.0438	—	0.3193	0.6447	−0.5952	1.1589	3.1670	—	—	0.3451	0.7948	—	0.3051	−2.7067	−1.1302	—
13	0.6492	0.6488	1.6546	3.0032	—	0.4143	0.6382	−0.5060	1.1725	3.1516	—	—	0.3454	0.8176	−0.1786	0.3018	−2.6790	−1.1404	—
14	0.6497	0.6492	1.6366	3.0980	0.3536	0.2837	0.7259	−0.5769	1.3451	3.0852	−0.4657	—	0.3169	0.7499	—	0.3199	−2.7346	−1.1357	—
15	0.6503	0.6498	1.6859	3.0507	0.3985	0.3939	0.7214	−0.4727	1.3704	3.0539	−0.4920	—	0.3147	0.7734	−0.2098	0.3173	−2.7011	−1.1473	—
16	0.6503	0.6498	1.6865	3.0458	0.3843	0.3842	0.7150	−0.4851	1.3668	3.0592	−0.4904	0.0574	0.3130	0.7755	−0.2105	0.3169	−2.7064	−1.1474	—
17	0.6503	0.6498	1.6815	3.0453	0.3850	0.3870	0.7165	−0.4842	1.3707	3.0516	−0.4988	0.0590	0.3144	0.7722	−0.2096	0.3174	−2.7093	−1.1504	−0.0082

Table G-8. Beta Coefficients from Regression Models Using the Maximum R-Square Option Using Nutrient or Component Values as a Percentage of Recommended Intake Levels per RACC (Total Sugar Replacing Added Sugar)
Number of Vari- ables	R-Square	Adjusted R-Square	Protein	Fiber	Vitamin E	Vitamin D	Calcium	Iron	Potas- sium	Unsat- urated Fat	Mag- nesium	Vitamin A	Vitamin C	Folic Acid	Vitamin B₁₂	Whole Grains	Saturated Fat	Sodium	Total Sugar
— Means no data.
0	0.0417	0.0413	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
1	0.2074	0.2071	—	3.9647	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
2	0.4036	0.4033	—	4.7469	—	—	—	—	—	—	—	—	—	—	—	—	−2.0714	—	—
3	0.4666	0.4663	—	3.0678	—	—	—	—	3.5502	—	—	—	—	—	—	—	−2.4323	—	—
4	0.4822	0.4819	—	2.6922	—	—	—	—	3.4790	—	—	—	—	—	—	0.4040	−2.3962	—	—
5	0.5046	0.5043	1.2483	3.5089	—	—	—	—	2.4916	—	—	—	—	—	—	—	−2.3969	−1.1935	—
6	0.5251	0.5247	1.1241	3.2833	—	—	—	—	2.6485	1.4255	—	—	—	—	—	—	−2.9348	−1.4166	—
7	0.5473	0.5469	—	3.1408	—	—	1.2390	—	3.3263	2.0305	—	—	—	—	—	—	−3.1199	−1.1449	−0.4433
8	0.5646	0.5642	1.3083	3.6994	—	—	1.2320	—	—	1.8897	—	—	0.5343	—	—	—	−3.1118	−1.5330	−0.4623
9	0.5746	0.5742	1.3130	3.3989	—	—	1.0989	—	—	1.8970	—	—	0.5347	—	—	0.3315	−3.0578	−1.4912	−0.4726
10	0.5847	0.5843	0.9766	2.8680	—	—	0.8991	—	1.8844	1.8754	—	—	0.4108	—	—	0.3504	−3.0248	−1.4245	−0.5068
11	0.5865	0.5861	1.0547	3.0040	—	—	0.9466	−0.4029	1.8360	1.8881	—	—	0.4146	—	—	0.3796	−3.0570	−1.3639	−0.4865
12	0.5883	0.5878	1.0732	3.0183	—	—	0.8171	−0.5167	1.7482	1.9064	—	0.5324	0.4050	—	—	0.3749	−3.0614	−1.3472	−0.4850
13	0.5890	0.5885	1.0956	3.0243	—	—	0.7716	−0.7666	1.8403	1.9257	—	0.5308	0.4033	0.3138	—	0.3677	−3.0575	−1.3672	−0.4866
14	0.5899	0.5894	1.1405	3.1314	0.6054	—	0.8759	−0.4981	1.9989	1.7867	−0.7423	0.4665	0.3800	—	—	0.3960	−3.0474	−1.3827	−0.4900
15	0.5906	0.5901	1.1285	3.2207	0.6355	0.2768	0.7465	−0.5144	1.8195	1.7693	−0.7180	0.3804	0.3974	—	—	0.3920	−3.0147	−1.3564	−0.4951
16	0.5910	0.5905	1.1424	3.2093	0.6050	0.2517	0.7204	−0.7016	1.8895	1.7898	−0.6754	0.3896	0.3958	0.2348	—	0.3859	−3.0146	−1.3714	−0.4954
17	0.5910	0.5904	1.1361	3.2153	0.5995	0.2385	0.7221	−0.7153	1.8843	1.7948	−0.6686	0.3881	0.3963	0.2335	0.0258	0.3862	−3.0188	−1.3690	−0.4953

H: NDS Scores of Food Subgroupings and Selected Foods Using the Final Algorithm

TableH-1. Rankings of Mean Nutrient Density Scores per 100 kcal (NDSKCAL) of Selected Subgroupings of Foods
Rank	Food Group	Mean Score	Minimum	Maximum
The means were calculated from subgroupings of 570 foods. Some subgroupings contain only a few foods (e.g., rice).
1	Dairy Yogurt	−18.86	−11.93	67.71
2	Milk	17.43	−6.63	54.73
3	Ice Cream	−16.60	−73.05	66.48
4	Cheese	−24.16	−63.95	14.51
5	Other	−29.38	−96.47	10.96
1	Grains Cereals, breakfast foods	15.76	−13.85	49.47
2	Crackers, salty snacks	15.62	−10.51	58.59
3	Breads	14.79	−31.18	41.34
4	Quickbreads	13.72	−8.36	48.52
5	Mixed grain products	2.57	−33.20	31.98
6	Cakes, cookies, pastries	1.54	−30.02	80.61
7	Rice	−4.34	−5.52	−2.96
1	Vegetables Vegetables other than potatoes	109.67	16.29	286.92
2	Potatoes	15.70	−4.02	33.46
3	Soup	−9.91	−42.11	20.95
4	Condiments	−25.46	−240.91	39.72
1	Fats, Oils, and Dressings Salad dressings	−10.76	−115.13	20.85
2	Fats and oils	−12.72	−90.35	30.83
1	Sweets and Beverages Beverages	0.93	−67.25	104.44
2	Sugars, sweets	−16.35	−55.36	31.42

Table H-2. Rankings of Mean Nutrient Density Scores per RACC (NDSRACC) of Selected Subgroupings of Foods
Rank	Food Group	Mean Score	Minimum	Maximum
The means were calculated from subgroupings of 570 foods. Some subgroupings contain only a few foods (e.g., rice).
1	Dairy Yogurt	12.29	−49.36	64.61
2	Milk	10.98	−18.83	48.83
3	Other	−14.07	−45.74	16.84
4	Ice cream	−31.29	−101.75	49.16
5	Cheese	−33.37	−71.74	2.82
1	Grains Cereals, breakfast foods	20.62	−21.94	75.63
2	Crackers, salty snacks	15.90	−18.93	70.82
3	Breads	15.30	−78.62	52.70
4	Quickbreads	12.90	−27.94	59.52
5	Mixed grain products	−3.84	−50.45	115.62
6	Cakes, cookies, pastries	−12.75	−148.62	118.62
7	Rice	−13.22	−13.52	−12.86
1	Vegetables Vegetables other than potatoes	35.88	1.04	77.12
2	Potatoes	17.66	−9.27	42.89
3	Condiments	−0.73	−18.28	19.58
4	Soup	−15.47	−54.31	34.04
1	Fats, Oils, and Dressings Salad dressings	−6.14	−26.51	10.45
2	Fats and oils	−17.82	−98.00	22.85
1	Sweets and Beverages Beverages	−5.26	−38.69	43.95
2	Sugars, sweets	−31.47	−119.52	13.15

Table H-3. Nutrient Density Scores of 100 Foods Using the Final Algorithms (NDS)
Food Code	Description	Nutrient Density Score per 100 kcal (NDSKCAL)	Nutrient Density Score per RACC (NDSRACC)
Foodcodes and descriptions are from USDA Food and Nutrient Database for Dietary Studies (FNDDS) Version 4.0. ¹ The same nutrient values as foodcode 71101000 were used to calculate the score except the sodium value was replaced with 5 mg per 100 grams.
11111000	DAIRY FOODS Milk, cow's, fluid, whole	−6.63	−18.83
11112110	Milk, cow's, fluid, 2% fat	7.62	1.47
11112210	Milk, cow's, fluid, 1% fat	28.39	22.39
11113000	Milk, cow's, fluid, skim or nonfat, 0.5% or less butterfat	54.73	40.05
11411100	Yogurt, plain, whole milk	−9.41	−21.61
11411200	Yogurt, plain, low-fat milk	27.80	30.01
11411300	Yogurt, plain, nonfat milk	54.99	60.71
11431000	Yogurt, fruit variety, whole milk	−11.93	−49.36
11432000	Yogurt, fruit variety, low-fat milk	1.44	−11.93
11432500	Yogurt, fruit variety, low-fat milk, sweetened with low-calorie sweetener	14.31	25.40
11433000	Yogurt, fruit variety, nonfat milk	11.67	11.94
11433500	Yogurt, fruit variety, nonfat milk, sweetened with low-calorie sweetener	67.71	64.61
11511100	Milk, chocolate, whole milk-based	−3.10	−16.62
11511200	Milk, chocolate, reduced fat milk-based, 2%	7.00	5.31
11511400	Milk, chocolate, low-fat milk-based	19.07	22.02
11511300	Milk, chocolate, skim milk-based	31.29	36.30
12120100	Cream, half and half	−51.65	−22.51
12210400	Cream substitute, powdered	−96.47	−10.99
13110100	Ice cream, regular, flavors other than chocolate	−34.94	−55.69
13130300	Light ice cream, flavors other than chocolate	−10.70	−18.15
13160400	Fat free ice cream, flavors other than chocolate	23.16	20.25
14410200	Cheese, processed, American or Cheddar type	−45.77	−55.06
21101120	MEATS, POULTRY, AND FISH Beef steak, broiled or baked, lean and fat eaten	7.57	−3.30
21500100	Ground beef or patty, cooked, NS as to percent lean	1.14	−17.32
22600200	Pork bacon, NS as to fresh, smoked or cured, cooked	−17.60	−22.79
24122100	Chicken, breast, roasted, broiled, or baked, NS as to skin eaten	29.07	30.92
24127110	Chicken, breast, coated, baked or fried, prepared with skin, skin/coating eaten	17.17	18.17
24198700	Chicken patty, fillet, or tenders, breaded, cooked	17.86	20.71
24198740	Chicken nuggets	19.49	29.33
25210210	Frankfurter or hot dog, beef	−33.58	−80.70
25230210	Ham, sliced, prepackaged or deli, luncheon meat	−19.94	−25.56
25230310	Chicken or turkey loaf, prepackaged or deli, luncheon meat	−4.34	−10.41
28140720	Chicken patty, or nuggets, boneless, breaded, potatoes, vegetable (frozen meal)	21.41	54.82
31105000	EGGS Egg, whole, fried	7.01	−2.87
32105000	Egg omelet or scrambled egg, fat added in cooking	7.17	−4.95
41205010	LEGUMES AND NUTS Refried beans	56.08	102.86
42100100	Almonds, NFS	74.53	114.43
42202000	Peanut butter	31.32	41.54
42202150	Peanut butter, reduced fat	31.91	45.48
51101000	GRAIN PRODUCTS Bread, white	9.49	8.82
51300110	Bread, whole wheat, NS as to 100%	27.18	32.59
51301010	Bread, wheat or cracked wheat	21.94	25.78
52215100	Tortilla, corn	48.52	59.52
52215200	Tortilla, flour (wheat)	13.73	18.04
53105260	Cake, chocolate, devil's food, or fudge, with icing, coating, or filling, made from home recipe or purchased ready-to-eat	−11.24	−52.85
53206000	Cookie, chocolate chip	−19.99	−35.24
53540400	Kellogg's Nutri-Grain Cereal Bar	20.27	24.33
54301000	Cracker, snack	7.07	2.01
54401050	Salty snacks, corn or cornmeal base, corn puffs and twists; corn-cheese puffs and twists	14.25	11.77
54401080	Salty snacks, corn or cornmeal base, tortilla chips	28.47	35.59
54403020	Popcorn, popped in oil, buttered	12.31	10.29
54403070	Popcorn, popped in oil, low-fat	44.25	56.07
54408010	Pretzels, hard	−0.41	−3.38
55101000	Pancakes, plain	12.92	25.85
56205010	Rice, white, cooked, regular, fat not added in cooking	−5.52	−13.52
56205110	Rice, brown, cooked, regular, fat not added in cooking	13.58	19.50
58106225	Pizza, cheese, regular crust	0.07	−17.70
58132310	Spaghetti with tomato sauce and meatballs or spaghetti with meat sauce or spaghetti with meat sauce and meatballs	16.28	47.52
61119010	FRUIT Orange, raw	115.15	93.81
61210220	Orange juice, canned, bottled or in a carton	50.75	55.10
61210250	Orange juice, with calcium added, canned, bottled or in a carton	91.76	98.42
63101000	Apple, raw	65.53	51.91
63107010	Banana, raw	45.35	60.68
63149010	Watermelon, raw	42.05	36.38
63223020	Strawberries, raw	128.78	82.85
64100110	Fruit juice blend, 100% juice	53.74	59.39
64104010	Apple juice	20.97	24.20
71101000	VEGETABLES White potato, baked, peel not eaten (with added salt)	17.85	18.20
-	White potato, baked, peel not eaten (no added salt)¹	34.99	36.13
71201010	White potato, chips	33.46	42.89
71401030	White potato, french fries, from frozen, deep fried	18.85	32.16
72125100	Spinach, raw	215.74	50.92
73101010	Carrots, raw	99.87	37.14
74101000	Tomatoes, raw	139.48	23.50
74401010	Tomato catsup	−63.47	−10.70
74402150	Salsa, red, cooked, not homemade	−18.25	−2.09
75113000	Lettuce, raw	147.88	18.46
75117020	Onions, mature, raw	76.39	27.49
75503010	Cucumber pickles, dill	−240.91	−9.61
81101000	FATS, OILS, AND DRESSINGS Butter, stick, salted	−90.35	−98.00
81101010	Butter, whipped, tub, salted	−88.60	−64.06
81103080	Margarine-like spread, tub, salted	7.87	−2.78
83106000	Italian dressing, made with vinegar and oil	−18.79	−26.51
83107000	Mayonnaise, regular	20.85	10.45
83110000	Mayonnaise-type salad dressing	10.15	0.14
83112500	Creamy dressing, made with sour cream and/or buttermilk and oil	12.75	0.62
83205500	Italian dressing, reduced calorie, fat-free	−115.13	−18.07
91101010	SWEETS AND BEVERAGES Sugar, white, granulated or lump	−26.86	−5.12
91107000	Sucralose-based sweetener, sugar substitute	0.00	0.00
91401000	Jelly, all flavors	−13.25	−8.58
91705010	Milk chocolate candy, plain	−35.79	−86.49
91745020	Hard candy	−17.08	−12.39
92101000	Coffee, made from ground, regular	48.61	0.95
92302000	Tea, leaf, unsweetened	−16.85	−0.45
92302200	Tea, leaf, presweetened with sugar	−26.92	−15.86
92410310	Soft drink, cola-type	−24.77	−27.31
92410320	Soft drink, cola-type, sugar-free	8.07	0.15
92510610	Fruit juice drink	−13.38	−19.95
92530610	Fruit juice drink, with high vitamin C	12.08	1.37
92560100	Gatorade Thirst Quencher sports drink	−28.56	−21.33

I: Mean Food Scores by Deciles of HEI

Table I-1. Weighted Mean Scores of Foods Consumed by Individuals in NHANES 2005-2008 by Deciles of Healthy Eating Index Scores
Subcategories of NHANES Population	Mean HEI	Median HEI	Decile of HEI 1	Decile of HEI 2	Decile of HEI 3	Decile of HEI 4	Decile of HEI 5	Decile of HEI 6	Decile of HEI 7	Decile of HEI 8	Decile of HEI 9	Decile of HEI 10
— Means no data.
Upper cutoff of HEI decile	—	—	33.73	38.75	42.97	46.71	50.83	54.80	58.97	63.86	70.62	—
Per 100 kcal Overall	51.47	50.83	−6.09	−1.76	0.75	2.80	5.62	7.46	9.77	12.84	16.95	25.60
Age groups Children (2-18 y)	50.14	49.35	−5.86	−2.19	−1.07	2.29	4.14	6.24	7.99	11.15	14.75	21.07
Adults (19+ y)	51.90	51.41	−6.17	−1.60	1.34	2.99	6.24	7.85	10.40	13.35	17.69	26.40
Older Adults (50+ y)	55.01	54.79	−6.44	−0.96	1.83	3.58	6.69	8.50	10.98	13.58	18.12	26.72
Ethnicity Non-Hispanic white	51.26	50.42	−6.66	−2.17	0.57	2.37	5.49	7.23	9.43	12.61	16.51	25.86
Non-Hispanic black	49.04	48.38	−4.86	−0.66	0.43	3.32	6.03	7.37	10.31	13.12	17.45	25.30
Mexican Americans	53.21	53.23	−4.29	−0.04	2.90	4.78	6.65	8.98	11.77	14.63	18.11	24.88
Poverty income ratio ≤1.85	50.60	50.00	−6.28	−2.15	0.54	3.19	5.69	7.74	9.85	12.97	17.30	25.06
>1.85	51.71	50.89	−6.04	−1.62	0.89	2.67	5.52	7.24	9.64	12.70	16.78	25.86
Child BMI-for-age status Obese (≥95^th pct)	49.38	48.69	−7.91	−2.05	−1.13	2.36	4.50	6.97	9.58	11.67	15.84	22.37
Overweight (≥85^th and <95^th pct)	49.70	48.30	−6.60	−3.14	−1.09	1.86	2.73	5.98	8.63	11.33	16.11	21.34
Normal weight (<85^th pct)	50.40	49.85	−5.15	−1.99	−1.05	2.37	4.37	6.09	7.56	11.03	14.15	20.79
Adult BMI status Obese (≥30)	51.06	50.28	−5.82	−1.94	1.04	3.79	6.21	7.36	10.37	13.89	17.75	25.28
Overweight (≥25 and <30)	52.40	52.22	−5.71	−0.58	1.88	2.68	6.07	7.94	10.68	12.81	18.53	27.27
Normal weight (<25)	52.21	51.60	−6.79	−2.39	1.14	2.34	6.18	8.07	9.93	13.26	16.74	26.58
LDL status Normal (<130 mg/dl)	50.52	50.12	−5.87	−1.79	0.62	3.23	5.84	7.32	10.40	13.48	17.62	27.13
Elevated (≥130 mg/dl)	50.84	50.13	−6.51	−0.96	1.16	3.24	5.45	7.69	9.71	13.25	16.96	28.96
Per RACC Overall	51.47	50.83	−12.20	−8.19	−5.05	−3.02	0.02	2.01	4.23	7.15	10.29	16.35
Age groups Children (2-18 y)	50.14	49.35	−14.74	−9.90	−8.00	−3.83	−1.37	1.25	3.41	7.44	10.63	17.11
Adults (19+ y)	51.90	51.41	−11.36	−7.55	−4.09	−2.72	0.60	2.25	4.52	7.06	10.17	16.22
Older adults (50+ y)	55.01	54.79	−10.59	−5.89	−3.36	−1.89	0.92	2.83	5.07	7.02	10.38	16.48
Ethnicity Non-Hispanic white	51.26	50.42	−12.37	−8.53	−5.07	−3.38	−0.15	1.78	3.55	6.55	9.59	15.97
Non-Hispanic black	49.04	48.38	−12.44	−7.80	−6.27	−2.84	0.34	1.60	5.46	8.12	11.19	16.83
Mexican Americans	53.21	53.23	−11.27	−6.69	−2.84	−1.08	1.37	3.62	7.02	9.98	12.45	18.41
Poverty income ratio ≤1.85	50.60	50.00	−12.64	−8.74	−5.48	−2.60	0.08	2.23	4.81	7.79	10.89	17.03
>1.85	51.71	50.89	−12.00	−7.95	−4.84	−3.15	−0.07	1.84	3.84	6.77	9.99	16.06
Child BMI-for-age status Obese (≥95^th pct)	49.38	48.69	−16.24	−8.91	−7.99	−3.51	−1.17	2.30	3.99	7.76	10.53	19.05
Overweight (≥85^th and <95^th pct)	49.70	48.30	−14.94	−11.62	−8.28	−4.39	−2.92	0.61	4.09	7.37	11.76	15.64
Normal weight (<85^th pct)	50.40	49.85	−14.31	−9.71	−7.94	−3.77	−1.08	1.08	3.19	7.40	10.40	17.19
Adult BMI status Obese (≥30)	51.06	50.28	−11.07	−7.99	−4.38	−2.07	0.28	1.49	3.87	7.01	9.86	15.05
Overweight (≥25 and <30)	52.40	52.22	−11.44	−6.15	−3.69	−2.64	0.79	2.16	4.76	6.79	10.28	16.87
Normal weight (<25)	52.21	51.60	−11.80	−8.58	−4.16	−3.58	0.32	2.78	4.67	7.13	10.17	16.52
LDL status Normal (<130 mg/dl)	50.52	50.12	−11.31	−8.15	−5.02	−2.68	0.03	1.67	4.26	7.42	10.22	16.86
Elevated (≥130 mg/dl)	50.84	50.13	−11.68	−6.93	−4.29	−2.06	0.13	1.68	3.59	6.50	9.02	17.48

Topics

Food & Nutrition