The design and development of Food Compass involved four main steps: (1) assessing existing Nutrient Profiling Systems (NPS), dietary guidelines, health claims, and diet-health relationships, (2) selecting attributes, (3) developing the scoring principles and algorithm, and (4) testing and validating. This work was performed using published reports as well as deidentified, publicly available data in NHANES.

Assessing Existing NPS, Dietary Guidance, and Diet-Health Relationships

We first assessed the current scientific landscape of major NPS and related food front of pack labels, as well as national and international dietary guidance and found more than 100 reported NPS with certain similarities and many differences in components, scoring principles and thresholds, and design.4,9

For guidelines and official recommendations, we used a recent systematic review of about 90 national and international dietary guidelines.2 Unlike NPS, dietary guidelines primarily focus on foods and encourage certain beneficial food groups, such as vegetables, fruits, and whole grains. Guidelines also encourage the consumption of certain food groups to help achieve nutrient targets, as well as limitation of certain foods and nutrients.

We also assessed the US FDA nutrient content requirements for health claims, including for general health claims, as well as health outcomes for nutrients, ingredients, and other food characteristics were linked to health outcomes in observational studies or randomized trials, have been prioritized in population diet pattern scores, or were of emerging public health interest.18-21

Selecting Attributes

Table of the Domains and Attributes of the Food Compass Score. Across 9 domains, 54 individual attributes were assessed per 100 kcal (418.4 kj) of food product, with scoring from 0 to 10 for beneficial attributes, -10 to 0 for harmful attributes, and -10 to 10 for attribute ratios which could range from harmful to beneficial.  Each domain receives a score, calculated as the average of all attributes in that domain (or for food ingredients, as the sum, given contents of ingredients are interdependent).  For vitamins and minerals, the domain score was calculated from the highest (absolute value, i.e. negative or positive) 5 attribute scores; and for specific lipids, from the highest 3 attribute scores.  Attributes with emerging evidence for health impacts (5 additives, fermentation and frying as processing methods) were scored using half weights.  All domains scores are summed, using equal weights for the first 6 domains and half weights for the latter 3 domains. The final Food Compass score (FCS) is scaled across all food and beverage items to range from 1 (least healthful) to 100 (most healthful). 

Nutrient Ratios Unsaturated: Saturated fat ratio 
Fiber:Carbohydrate ratio
Potassium:Sodium ratio
Vitamins (top 5) Vitamin A
Thiamin (B1)
Riboflavin (B2)
Niacin (B3)
Vitamin B6
Folate (B9)
Cobalamin (B12)
Vitamin C
Vitamin D
Vitamin E
Vitamin K
Choline
Minerals (top 5) Calcium
Phosphorus
Magnesium
Iron
Zinc
Copper
Selenium
Sodium
Potassium
Iodine
Food-based Ingredients Fruits
Vegetables, non-starchy
Beans & legumes
Whole grains
Nuts and seeds
Seafood
Yogurt
Plant oils
Refined grains
Red or processed meat
Additives Added sugar
Nitrites
Artificial sweeteners, flavors, or colors
Partially hydrogenated oils
Interesterified or hydrogenated oils
High fructose corn syrup
Monosodium glutamate (MSG)
Processing NOVA classification
Fermentation
Frying
Specific Lipids (0.5 weight) (top 3) Alpha-linolenic acid (ALA)
EPA + DHA
Medium-chain fatty acids (MCFA)
Dietary cholesterol
Trans fats
Fiber & Protein (0.5 weight) Total fiber
Total protein
Phytochemicals (0.5 weight) Total flavonoids
Total carotenoids

We considered several standards for assessing each attribute, including content per 100 kcal, per 100 g, per liter, and per serving size.4 We decided to score per 100 kcal (418.4 kj) to facilitate the use of a single scoring algorithm for a diverse range of items, from a single small item to a food with mixed ingredients or a large mixed dish or meal, as well as among items that differ greatly in bulk. Scoring per 100 kcal was also considered valuable for scaling up to compare diverse combinations that may be sold and consumed together, for example to score an entire shopping basket, an entire diet, or an entire portfolio of foods being sold by a particular vendor.

Developing Scoring Principles and Algorithm

Attributes could be scored in four ways. Most were scored based on a linear 10-point scale, from 0 to 10 for attributes considered to have a positive overall health impact, and from -10 to 0 for attributes considered to have an adverse overall health impact.

For attributes with a defined Dietary Reference Intake (DRI), such as vitamins and minerals, the target level (maximum points) was set at 25% of the adult DRI value for a 2000 kcal/day diet, which most consistently distinguished foods and beverages with higher vs. lower levels of these nutrients and was generally similar to the 95% percentile value of content across all foods and beverages reported in NHANES 2015-16. For attributes without DRIs, the target level was set based on the 95th percentile value of foods and beverages consumed by the U.S. population (based on 2015-16 NHANES data).

For attributes that were ratios of positive vs. adverse factors (e.g., ratio of unsaturated to saturated fat), scoring was on a log-linear scale from -10 to 10 points to represent the full range of the ratio, with reference targets for lowest and highest points based on the 5th and 95th percentile values of foods consumed by the U.S. population.

For attributes for which information was generally binary (e.g., presence or absence of preservatives, artificial sweeteners, flavors, or colors, fermentation, or frying), scoring was binary (-10, 0), with half weights for most of these factors based on still emerging evidence for health impacts. The NOVA processing attribute was based on assigning scores ranging from -10 to 0 for the four categories: ultra-processed foods; processed foods; processed culinary ingredients; and unprocessed or minimally processed foods.

To prevent any single attribute from dominating a food’s score and to provide a more holistic assessment of overall health impact, identified relevant attributes were grouped into 9 domains that represented different health-relevant aspects of foods: major nutrients, vitamins, minerals, food ingredients, etc. Each domain’s score was calculated as the average of its attribute scores (for food ingredients, the sum), and then the domain scores were summed to calculate the summary score for each food. The same scoring principles and algorithm were used for all foods and beverages.

For attributes that were ratios of positive vs. adverse factors (e.g., ratio of unsaturated to saturated fat), scoring was on a log-linear scale from -10 to 10 points to represent the full range of the ratio, with reference targets for lowest and highest points based on the 5th and 95th percentile values of foods consumed by the U.S. population (based on 2015-16 NHANES data).

Testing and Validating

For testing and validation, Food Compass was applied to all foods and beverages reported in NHANES 2015-16, utilizing the USDA’s Food and Nutrient Database for Dietary Studies (FNDDS) 2015-2016, with information from the Food Pattern Equivalents Database (FPED) 2015-2016 and the 2010 USDA Flavonoid database. Scored foods and beverages were those reported as consumed by children and adults (a total of 8,032 products), in their form as consumed. We excluded infant formula, baby foods, specialized dietary foods such as nutritional supplements for athletic performance or treatment for health conditions, alcohol, and beverages providing <5 kcal per 100g (e.g., unsweetened tea or coffee, diet soda).

Food Compass was explicitly developed for content validity: inclusion of nutrients, food ingredients, and other dietary characteristics of public health concern. We further assessed face validity by evaluating Food Compass scores across FNDDS food categories and subcategories, considering both the distributions of the scores and the specific examples of foods scoring higher, middle, and lower in each category. We assessed convergent and discriminant validity by comparing Food Compass to the foods classified within the NOVA classification system, the Health Star Rating, and Nutri-Score.38,39


2. Herforth, A. et al. A Global Review of Food-Based Dietary Guidelines. Adv Nutr 10, 590-605, doi:10.1093/advances/nmy130 (2019).

4. Labonte, M. E. et al. Nutrient Profile Models with Applications in Government-Led Nutrition Policies Aimed at Health Promotion and Noncommunicable Disease Prevention: A Systematic Review. Adv Nutr 9, 741-788, doi:10.1093/advances/nmy045 (2018).

9. Pomeranz, J., Mozaffarian, D. & Micha, R. Mandating front-of-package food labels in the U.S. – What are the First Amendment obstacles? Food Policy 86, Epub 19 June 2019, doi:https://doi.org/10.1016/j.foodpol.2019.05.005 (2019).

18. Micha, R. et al. Etiologic effects and optimal intakes of foods and nutrients for risk of cardiovascular diseases and diabetes: Systematic reviews and meta-analyses from the Nutrition and Chronic Diseases Expert Group (NutriCoDE). PLoS One 12, e0175149, doi:10.1371/journal.pone.0175149 (2017).

19. U.S. Department of Health and Human Services and U.S. Department of Agriculture. (2015).

20. World Health Organization. Comprehensive Implementation Plan on Maternal, Infant, and Young Child Nutrition.  (2019). <https://apps.who.int/iris/bitstream/handle/10665/113048/WHO_NMH_NHD_14.1_eng.pdf?ua=1>.

21. Elizabeth, L., Machado, P., Zinocker, M., Baker, P. & Lawrence, M. Ultra-Processed Foods and Health Outcomes: A Narrative Review. Nutrients 12, doi:10.3390/nu12071955 (2020).

33. U.S. Department of Agriculture et al. What We Eat in America, NHANES 2015-2016. (2020). <www.ars.usda.gov/nea/bhnrc/fsrg>.

34. European Food Safety Authority. The food classification and description system FoodEx 2 (revision 2). EFSA Supporting Publications 12, 804E (2015).

35. Neal, B., Sacks, G., Shahid, M., Taylor, F. & Huffman, M. FoodSwitch: State of the Food Supply.  (2019). <www.georgeinstitute.org >.

36. Department of Economic and Social Affairs of the United Nations Secretariat & Statistics Division. Classification of Individual Consumption According to Purpose (COICOP).  (2018). < http://unstats.un.org/unsd/classifications>.

37. Mozaffarian, D. Dietary and Policy Priorities for Cardiovascular Disease, Diabetes, and Obesity: A Comprehensive Review. Circulation 133, 187-225, doi:10.1161/CIRCULATIONAHA.115.018585 (2016).

38. Access to Nutrition Initiative. (2021). <https://accesstonutrition.org/>.

39. Chantal, J., Hercberg, S. & World Health Organization. Regional Office for, E. Development of a new front-of-pack nutrition label in France: the five-colour Nutri-Score. Public health panorama 03, 712-725 (2017).