Price, Nutrition, Time, and Other Trade-Offs: A Web-Based Food Value Analysis Application to Compare Foods at Different Levels of Preparation and Processing

Muth, Mary K. PhD; Karns, Shawn A. BA; Zmuda, Michal BS; Coglaiti, Michaela C.; Koyanagi, Mark MS; Duffey, Kiyah PhD; Dunn, Carolyn PhD; Jensen, Helen H. PhD; Gregory, Christian PhD

Nutrition Today:
doi: 10.1097/NT.0000000000000039
Nutrition Research
Abstract

Consumers choose to eat different forms of foods based on a wide variety of factors such as price, taste, nutrition, and convenience and, in doing so, make trade-offs among them. A Web-based application for use by nutrition educators was developed to help individuals compare foods prepared from home recipes with those for other forms of food (eg, frozen, canned, dry mix). Foods with a home-recipe form in US Department of Agriculture databases were selected to represent a range of commonly consumed entrées, baked goods, side dishes, fruits, vegetables, desserts, and beverages. Multiple US Department of Agriculture and commercial databases along with other public data sources were used to construct prices, nutrient values, food groups and components, preparation and cooking times, shelf life, and food safety concerns for foods in the database. Per-serving and per-100-g values were constructed for 100 individual foods with a home recipe and 1 or more other forms. The data are available in a Web-based application, located at http://www.foodvalueanalysis.org, allowing comparisons of individual foods or a daily diet constructed from foods in the database. Nutrition educators can use the application to advise individuals in selecting foods to consume to meet dietary guidelines while taking into consideration cost, preparation time, food preparation skills, and individual preferences. For example, the application can be used to evaluate differences in prices of fresh or processed foods, whether home recipe or processed foods are less costly when taking into consideration the value of preparation time, and the differences in nutrients across different forms of foods.

Author Information

Mary K. Muth, PhD, is the director of the Food and Nutrition Policy Research Program at RTI International, Research Triangle Park, North Carolina, and leads studies to analyze the impacts of food safety, nutrition, food assistance, food labeling, food marketing, and other types of policies and regulations.

Shawn A. Karns, BA, is a research analyst, Food and Nutrition Policy Research Program at RTI International, Research Triangle Park, North Carolina.

Michal Zmuda, BS, is a research programmer/analyst at RTI International, Research Triangle Park, North Carolina.

Michaela C. Coglaiti, is a social science researcher, Food and Nutrition Policy Research Program at RTI International, Research Triangle Park, North Carolina.

Mark Koyanagi, MS, is a research programmer/analyst at RTI International, Research Triangle Park, North Carolina.

Kiyah Duffey, PhD, is a research assistant professor in the Department of Human Nutrition, Foods, and Exercise at Virginia Tech, Blacksburg.

Carolyn Dunn, PhD, is a professor in the Department of 4-H Youth Development and Family and Consumer Services at North Carolina State University, Raleigh.

Helen H. Jensen, PhD, is a professor in the Department of Economics at Iowa State University, Ames.

Christian Gregory, PhD, is an economist at the US Department of Agriculture, Economic Research Service, Washington, DC.

This work was supported by the Project Committee on Food Value Decisions of the North American Branch of the International Life Sciences Institute. ILSI North America is a public, nonprofit foundation that provides a forum to advance understanding of scientific issues related to the nutritional quality and safety of the food supply by sponsoring research programs, educational seminars and workshops, and publications. ILSI North America receives support primarily from its industry membership.

The opinions expressed herein are those of the authors and do not necessarily represent the views of the funding organization or the US Department of Agriculture, Economic Research Service.

The authors have no conflicts of interest to disclose.

Correspondence: Mary K. Muth, PhD, RTI International, 3040 Cornwallis Rd, PO Box 12194, Research Triangle Park, NC 27709 (muth@rti.org).

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License, where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially.

Article Outline

Consumers choose to eat different forms of foods based on a wide variety of factors such as price, taste, nutrition, and convenience and, in doing so, make trade-offs among them. Food choices are strongly affected by what a household can afford and whether household members have the necessary food preparation skills to prepare a food from raw or partially prepared ingredients.1,2 In addition, a consumer’s need for foods that are convenient and easy to prepare also affects food choices.3 An increasing portion of the American population lacks confidence in their cooking skills, and many have competing demands that limit time for food preparation.

In the 2010 Dietary Guidelines for Americans, concerns were raised regarding consumption of processed foods because of their excessive sodium, solid fats, refined grains, and added sugars.4 However, a recent Institute of Medicine report recommended that the US Department of Agriculture (USDA) Food and Nutrition Service recognize the cost-time trade-offs involved in procuring and preparing a nutritious diet when considering the adequacy of the Supplemental Nutrition Assistance Program allotment.5 In particular, time constraints of households, particularly those with a single working head of household, may lead households to purchase value-added or prepared foods to ensure an adequate diet.5 Other research has also shown that household time resources significantly affect how much time is allocated to preparing foods6 and that the total cost of foods in the Thrifty Food Plan are substantially higher when time costs are included in the calculation.7,8

The average American diet is too high in added fats and sugars and falls short in fruits, vegetables, and dairy.9–13 Furthermore, many consumers need to change the mix of food groups they consume to better align with dietary guidance.10 Barriers to adherence with dietary guidance include food preferences, limited food budget, lack of time, and lack of food preparation skills.1,14–16 Lack of cooking skills and eating more foods prepared outside the home are considered barriers to an eating pattern that is consistent with dietary guidance.2,17

Data to make objective comparisons across a range of home-recipe foods and other forms of foods have not previously been available in 1 location. With the availability of government and proprietary databases, comprehensive data on food attributes can be linked for use by nutrition educators in a Web-based application. The Food Value Analysis application, available at www.foodvalueanalysis.org, was designed for use by nutrition educators to help consumers meet dietary guidelines while considering the real and perceived barriers to consuming a healthy diet, including budgets, time, food preparation skills, and shelf life and food safety concerns. It can be used to address questions such as the following:

* Taking into consideration the value of preparation and food costs, which is cheaper—homemade or frozen lasagna?

* Would buying whole carrots and cutting them up instead of buying baby carrots save money?

* Taking into consideration food costs alone, which is cheaper—home recipe, dry mix, or frozen macaroni and cheese?

* What are the nutrient differences across home recipe, canned, and dry-mix chicken noodle soup?

* What are the food safety concerns when preparing meatloaf and mashed potatoes using a home recipe versus preparing a frozen meal?

The free Food Value Analysis application available at www.foodvalueanalysis.org helps consumers make trade-offs while meeting the Dietary Guidelines.

Back to Top | Article Outline

DEVELOPMENT OF THE WEB-BASED APPLICATION

The Web-based Food Value Analysis application was developed based on a review of available data sources for core values that influence a consumer’s choice of a particular form of the food. Foods included in the application were selected based on whether a home-recipe form of the food is available from USDA databases to avoid subjective judgments about which recipe to use to calculate the values and because nutrients, food groups, and, in many cases, prices could be obtained from existing publicly available data sources and linked by food code. Once foods were selected, food values were calculated as described below. Although other values, such as energy use to prepare foods (kilowatts) or potential food waste, may be of interest, lack of existing data sources at the food code level prevented their inclusion.

Back to Top | Article Outline
Selection of Foods

The foods were selected to represent a range of commonly consumed entrées, entrée components, fruits, vegetables, grains and starches, baked goods, desserts, condiments, and beverages (Table 1). The home-recipe form of a food was selected from the USDA Food and Nutrition Database for Dietary Studies (FNDDS) or the USDA National Nutrient Database for Standard References (SR). In some cases, 2 or more FNDDS or SR codes were used to create a home-recipe form of a food. The processed or prepared forms of foods were selected from the FNDDS, SR, and Gladson Nutrition Database (GND) to match a home-recipe form in the USDA databases. Up to 3 processed forms were included based on the available data. Newer types of processed foods were obtained primarily from the GND. The version of the application described in this article includes 243 foods in 100 categories where each category comprises a home-recipe form and 1 or more other forms.

Back to Top | Article Outline
Data Sources and Calculations

Table 2 provides a list of variables and data sources used in developing the Food Value Analysis application along with key adjustments or calculations. We describe these variables and associated calculations below.

The database contains 100 categories for comparing home recipes and other forms of foods.

Back to Top | Article Outline
Serving Sizes

Serving sizes for foods were selected to be closest to the standard serving sizes defined in the FNDDS or the Reference Amounts Commonly Consumed (RACC). When the FNDDS and RACC gram weights for serving sizes differed, the serving sizes of commercially available products were used as the basis to select the closest size to the FNDDS serving size. (For example, the gram weight of a cookie was 14 g in FNDDS and 30 g for the RACC. Several standard packaged cookies had serving size weights ranging from 11 to 13 g, so a 13-g serving size was used for all cookies.) For foods with clear units (eg, enchilada or stuffed pepper), the standard unit provided (eg, 1 enchilada or half a stuffed pepper) was used. For consistency, the serving size of the home-recipe form was applied across all other forms of the same food.

Back to Top | Article Outline
Home Recipes

All home-recipe forms of foods were obtained from either the USDA FNDDS (version 5.0) or SR (version 24) databases. In some instances, a combination of FNDDS and SR data was used to create a home-recipe form of the food because it was not available within a single code (eg, garlic bread, meatloaf with mashed potatoes and vegetables, and cheese pizza). These foods were prepared in a test kitchen, and the ingredients (eg, French bread, butter, and garlic powder) were weighed. These individual ingredient weights were then used to calculate a nutrient profile for the entire food product using the individual ingredient nutrient profiles.

Recipes were also used to determine preparation and cooking time and to assign food safety concerns and shelf life (as described below). The FNDDS-SR links file was used to determine the individual food components in the recipe. This file lists each individual ingredient for which nutrient values were used to create the composite nutrient profile for the FNDDS home-recipe food. If nutrient data for a food came from the SR file, no recipes were available from the USDA databases; thus, recipes and numbers of servings were obtained from Betty Crocker Cookbook: 1500 Recipes for the Way You Cook Today; Better Homes and Gardens New Cookbook, 15th Edition or FoodNetwork.com.18–20 For packaged foods with added ingredients, food labels were used to determine additional ingredients for assigning food safety concerns and shelf life.

Back to Top | Article Outline
Nutrients and Daily Values

Nutrient data were obtained from the FNDDS or SR when available (about 93% of the total foods in the application and 100% of home-recipe forms) and calculated for the selected home-recipe serving size. For processed forms of foods that did not have a food code match in either FNDDS or SR, composite nutrient data were calculated from the average of the 3 top-selling products using the GND. However, only nutrients listed on the Nutrition Facts Panel are included in the GND.

Daily values (DVs) were calculated from the Reference Daily Intakes, which are based on Recommended Dietary Allowances (RDAs). Recommended Dietary Allowances are the established daily dietary intake levels of a nutrient considered sufficient to meet the requirements of 97.5% of healthy individuals in each life stage and gender group. To calculate the percentage DV, we divided the nutrient value per serving of food by the Reference Daily Intakes and multiplied by 100. Daily values were calculated only for nutrients with an established RDA. Polyunsaturated fat, monounsaturated fat, total sugars, and most micronutrients have no established RDA. Percent DV was not calculated for protein because the US Food and Drug Administration has ruled that scientific evidence suggests that protein intake is not a public health concern.21

Back to Top | Article Outline
Food Groups

Food group data were obtained from the MyPyramid Equivalents Database 2.0 supplemented with the Center for Nutrition Promotion and Policy (CNPP) Addendum to the MyPyramid Equivalents Database, which provides food group data for an additional 820 food codes. (Note: These data were not yet available for the MyPlate system.) From the 7 major food groups available (grains, vegetables, fruits, meats and beans, milk, oils, and extras), data for several important subgroups were also included. Specifically, in addition to total grains, we included whole grains, and from the meat group, we separated eggs, soybean products, and nuts and seeds. Values obtained from these databases were adjusted to the home-recipe serving size for the food.

Back to Top | Article Outline
Food Prices

Food prices were obtained from the USDA CNPP Food Prices Database, 2003–2004, if available for the FNDDS code. If food prices were not available in the CNPP Food Prices Database, national average prices for 2010 were calculated using household purchase data from The Nielsen Company’s Homescan panel, a nationally representative sample of US households. All prices were adjusted to 2011 values using the Bureau of Labor Statistics’ consumer price index for “food at home.”22

For foods that require no additional ingredients and have no or minimal moisture loss or gain during preparation and cooking, weighted average prices were calculated after applying The Nielsen Company’s proprietary demographic population weights to the food purchase data for households in the Homescan panel. For home-recipe foods or packaged foods requiring additional ingredients, we added average prices of all of the ingredients (adjusted for refuse losses) from the Homescan database on a 100-g basis, adjusted for moisture losses or gains, and converted the 100-g price to a per-serving price based on the gram weight of the serving size. We obtained measures of refuse losses and moisture losses and gains from SR and Matthews and Garrison.23 In cases where refuse losses or moisture gains and losses were not available from these sources, we developed estimates through direct measurement by preparing foods in a test kitchen.

Back to Top | Article Outline
Food Preparation and Cooking Time

Food preparation and cooking times for home recipes were obtained using the Betty Crocker Cookbook, Better Homes and Gardens New Cookbook and FoodNetwork.com’s “Recipes and Cooking.”18–20 For foods prepared from packaged foods, preparation and cooking times were obtained from package instructions. If preparation involves opening a package and putting the food in a microwave, preparation time is shown as “minimal” in the application. Because of minor differences across food labels, we developed general assumptions for specific activities based on food package instructions such as 7 minutes for heating a frozen meal, 4 minutes for heating a canned entrée, and 3 minutes for preparing a dry mix with 1 or 2 ingredients. Foods without published preparation and cooking times were prepared in a test kitchen to measure preparation and cooking times. The application displays the assumed method used for cooking or heating the food (eg, microwave vs conventional oven).

Preparation time is assigned a dollar value using an average hourly wage rate entered by the user. We calculated the value of preparation time on the basis of active, hands-on preparation time. Cooking time (eg, baking lasagna in an oven) or other preparation time (eg, soaking dried beans) was not included in the value calculation under the assumption that the food preparer could be engaged in other activities during that time. We first calculated the total time cost of preparing the entire recipe and then divided the total time cost by the number of servings in the recipe. This allows users to view the total value of their time for preparing the entire recipe and the per-serving value of time. By adding the per-serving direct cost to the per-serving value of time, individuals can compare the true total cost of a food across different forms.

Back to Top | Article Outline
Food Safety

Food safety concerns were assigned based primarily on information available from the Partnership for Food Safety Education, FoodSafety.gov, and the USDA Food Safety and Inspection Service. For each ingredient in a food, we identified food safety concerns associated with raw meat; raw poultry; raw fish or shellfish; raw eggs; raw fruits or vegetables prepared without cooking; frozen or refrigerated foods with meat, poultry, fish, or eggs; deli meats; and refrigerated dough. For example, when preparing recipes with raw meat, consumers would need to keep the meat refrigerated or frozen until ready for use, avoid cross-contamination with other foods, and use a thermometer to check internal temperature during cooking. Specific temperatures for refrigeration and cooking are provided for each food with a food safety consideration.

Back to Top | Article Outline
Shelf Life

Shelf life estimates were obtained from Cooperative Extension publications including the Colorado State University Cooperative Extension,24 Texas Agricultural Extension Service,25 and Virginia Cooperative Extension.26 For home-recipe foods or processed foods using fresh ingredients, shelf life was determined for the ingredient with the shortest shelf life. Shelf life for packaged foods with no additional ingredients was calculated as the length of time until the use-by or sell-by date as indicated on product labels at a local grocery store. Shelf life was assigned assuming a consumer stores the food in the same form it is in when purchased. For example, we assumed a consumer refrigerates a food that is purchased as a refrigerated food rather than storing the food at room temperature or placing the food in a freezer.

Back to Top | Article Outline
Web Application Structure and Development

The Food Value Analysis Web site is structured with the following pages:

* home page—provides an overview and links to use the application

* search page—allows users to search and select a food to display

* browse page—shows all foods organized by type and allows users to select a food to display

* food information page—shows food values for all forms of the selected food

* sources pages—describes and provides links to data sources

* FAQs (frequently asked question) page—provides information on uses, sources, and future plans

Within each broad category of foods shown in Table 1, users can select individual foods to view on the food information page. By default, the home-recipe form of the food is shown on the left, and other forms are shown on the right. The food information page shows the serving size information in common measures and weights. It also allows users to enter an average wage rate for individuals and toggle between values on a per-serving or per-100-g basis. In addition, users can hide or expand the forms of the food displayed and output the selected data to a PDF or Microsoft Excel file. Of the 243 foods in this version of the application, 159 are based on food codes from FNDDS, 58 from SR, 16 from GND, and 10 from a combination of these sources.

Back to Top | Article Outline
An Example Food Comparison

The Figure shows an example comparison of home-recipe versus canned meat ravioli on a per-serving basis, defined as one cup. If desired, users can select to view values on a per-100-g basis. Values displayed include economic measures, shelf life of the most perishable ingredient, food safety considerations, nutrients from the Nutrition Facts Panel (with percent DV for nutrients with established DVs), and food group information if available from USDA databases. Users can select to expand the nutrient display to include all nutrients available for the food (limited to those nutrients included in FNDDS). Additional information regarding the values in the display can be viewed at any time by hovering over the adjacent i symbol. Users can submit questions and provide feedback on the application by clicking on a link at the bottom of each page of the Web site.

The Food Value Analysis application is a tool nutrition educators can use to help consumers make wise decisions.

Back to Top | Article Outline

CONCLUSIONS AND FUTURE RESEARCH

The Food Value Analysis application is designed for use by nutrition educators as they assist consumers in selecting foods for consumption based on individual preferences, budget constraints, time availability, food preparation skills, and shelf life and food safety concerns. The application provides a means to begin a discussion about trade-offs involved in making food choices across multiple dimensions. A key constraint in developing such an application is selecting and obtaining data for a home-recipe version of a food. Although we relied on USDA data for the home recipe, it is important to acknowledge that consumers have access to a wide variety of sources for recipes and frequently adapt recipes (as well as packaged foods with added ingredients) based on their own taste preferences and dietary needs. In particular, they may alter the amount of salt or the type of fat used in a recipe. Also, they may alter the time and cost of preparing a food by using processed or prepared components (eg, by preparing spaghetti and meatballs using pasta sauce in a jar, boxed spaghetti, and frozen meatballs instead of preparing any of these from scratch).

In addition to the values included in the application, other attributes of foods may influence a consumer’s choice of foods such as potential waste, local or organic origins, type of packaging, and energy use for cooking. In some cases, an individual may not have an adequately equipped kitchen or cooking skills required to prepare all forms of a food, thus limiting their choices. Furthermore, the location of grocery stores, availability of transportation, and household schedules influence an individual’s ability to buy and prepare foods. The application does not account for other values that may be of interest because of the inherent difficulty in measuring these values and linking them to each food code.

Foods sold in grocery stores with some degree of preparation or processing may help Americans with limited time or food preparation skills consume a diet that is more consistent with dietary guidance. One nutrient, in particular, however, may prove problematic in light of current food processing formulas and regulatory requirements (such as for ready-to-eat meats). Many processed food products, such as bread, processed meats, and sauces, have sodium levels that make it difficult to consume a diet within suggested limits for sodium.27 However, in some cases, foods formulated to have lower sodium levels can help individuals consume less sodium.

An analysis was recently completed to compare nutrients, costs, and time of example daily diets constructed mainly from foods in the application. Results of this effort provides a broader view of the multiple trade-offs involved in selecting foods on a daily basis. As the USDA data sets are updated over time, the data in the application can be updated. In addition, the foods included in the database can be expanded in cases where a home-recipe version is available from the USDA databases, and alternative forms of foods can be added using the proprietary GND.

Back to Top | Article Outline

REFERENCES

1. Drewnowski A, Darmon N. Food choices and diet costs: an economic analysis. J Nutr. 2005; 135: 900–904.
2. van der Horst K, Brunner TA, Siegrist M. Ready-meal consumption: associations with weight status and cooking skills. Public Health Nutr. 2011; 14 (2): 239–245.
3. Ragaert P, Verbeke W, Devlieghere F, Debevere J. Consumer perception and choice of minimally processed vegetables and packaged fruits. Food Qual. 2004; 15: 259–270.
4. US Department of Health and Human Services, Department of Agriculture. Dietary Guidelines for Americans, 2010. 7th ed. Washington, DC: GPO; 2010.
5. Institute of Medicine. Supplemental Nutrition Assistance Program: Examining the Evidence to Define Benefit Adequacy. Washington, DC: The National Academies Press; 2013.
6. Mancino L, Newman C. Who Has Time to Cook? How Family Resources Influence Food Preparation. Economic Research Report No. 40. Washington, DC: US Department of Agriculture, Economic Research Service; 2007.
7. Davis GC, You W. The Thrifty Food Plan is not thrifty when labor cost is considered. J Nutr. 2010; 140 (4): 854–857.
8. Davis GC, You W. Not enough money or not enough time to satisfy the Thrifty Food Plan? A cost difference approach for estimating a money-time threshold. Food Policy. 2011; 36: 101–107.
9. 9. Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System. Updated March 19, 2013. http://www.cdc.gov/brfss/index.htm. Accessed February 20, 2013.
10. 10. Cook A, Friday J. Pyramid servings intakes in the United States, 1999–2002. Agricultural Research Service, US Department of Agriculture Web site. Published March 2005. http://www.ars.usda.gov/sp2UserFiles/Place/12355000/foodlink/ts_3-0.pdf. Accessed December 20, 2012.
11. Cotton PA, Sugar AF, Friday JE, Cook S. Dietary sources of nutrients among US adults, 1994 to 1996. J Am Diet Assoc. 2004; 104: 921–930.
12. 12. National Cancer Institute. Risk factor monitoring and methods. Table 5a: mean intake of added sugars & percentage contribution of various foods among US population, by age, NHANES 2005–06. Updated May 21, 2010. http://riskfactor.cancer.gov/diet/foodsources/added_sugars/table5a.html. Accessed December 20, 2012.
13. Putnam J, Kantor LS, Allshouse J. Per capita food supply trends: progress toward dietary guidelines. Food Rev. 2000; 23 (3): 2–14.
14. Bisogni CA, Jastran M, Shen L, Devine CM. A biographical study of food choice capacity: standards, circumstances, and food management skills. J Nutr Educ Behav. 2005; 37 (6): 284–291.
15. French S. Pricing effects on food choices. J Nutr. 2003; 133 (3): 841S–843S.
16. Pelletier JE, Laska MN. Balancing healthy meals and busy lives: associations between work, school, and family responsibilities and perceived time constraints among young adults. J Nutr Educ Behav. 2012; 44: 481–489.
17. Guthrie JF, Lin BH, Frazao E. Role of food prepared away from home in the American diet, 1977–78 versus 1994–96: changes and consequences. J Nutr Educ Behav. 2002; 34 (3): 140–150.
18. Better Homes and Gardens Books (Firm). Better Homes and Gardens New Cookbook. 15th ed. Des Moines, IA: Meredith; 2008.
19. Betty Crocker Cookbook: 1500 Recipes for the Way You Cook Today. 11th ed. Hoboken, NJ: Betty Crocker; 2011.
20. 20. FoodNetwork.com. Recipes and cooking. Web site. http://www.foodnetwork.com/recipes-and-cooking/index.html. Accessed February 20, 2013.
21. 21. Food and Drug Administration. How to understand and use the nutrition facts label. Food and Drug Administration Web site. Updated March 2, 2013. http://www.fda.gov/food/ResourcesForYou/Consumers/NFLPM/ucm274593.htm#nopercent. Accessed January 1, 2013.
22. 22. US Department of Labor, Bureau of Labor Statistics. Consumer price index. http://www.bls.gov/cpi/#tables. Accessed July 31, 2012.
23. Matthews RH, Garrison YJ. Food Yields Summarized by Different Stages of Preparation. Agriculture Handbook 102. Washington, DC: US Department of Agriculture; 1975.
25. 24. Kendall P, Dimond N. Food storage for safety and quality. No. 9.310. Colorado State University Cooperative Extension Web site. Published June 2012. http://www.ext.colostate.edu/pubs/foodnut/09310.html. Accessed February 20, 2013.
26. 25. van Laaneen P. Safe food home storage. Texas Agricultural Extension Service Web site. http://nchfp.uga.edu/how/store/texas_storage.pdf. Accessed February 20, 2013.
27. 26. Boyer R. Food storage guidelines for consumers. Virginia Cooperative Extension Web site. Published 2013. http://pubs.ext.vt.edu/348/348-960/348-960.html. Accessed February 20, 2013.
28. Webster JL, Dunford EK, Neal BC. A systematic survey of the sodium contents of processed foods. Am J Clin Nutr. 2010; 91: 413–420.
© 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins