If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to: Elizabeth Dunford, PhD, Carolina Population Center, University of North Carolina at Chapel Hill, CB # 8120 Carolina Square, Chapel Hill, NC 27516-3997.
Food Policy Division, The George Institute for Global Health, University of New South Wales, Sydney, New South Wales, AustraliaDepartment of Nutrition, Gillings Global School of Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
Carolina Population Center, The University of North Carolina at Chapel Hill, Chapel Hill, North CarolinaDepartment of Nutrition, Gillings Global School of Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
Carolina Population Center, The University of North Carolina at Chapel Hill, Chapel Hill, North CarolinaDepartment of Nutrition, Gillings Global School of Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
Food additives have been used mainly in the past century to perform specific functions in foods. Some types of food additives have been linked to adverse health outcomes, yet there is little research examining food additives in the US food supply.
Objective
To examine the proportion of products purchased by US households containing four common technical food additives using time-specific food composition data and examine whether purchases have changed over time.
Participants/setting
Nielsen Homescan Consumer Panels, 2001 and 2019.
Main outcome measures
The proportion of packaged food products containing common types of food additives purchased by US households was determined overall and by food category.
Statistical analysis performed
Differences were examined using Student t test; P value < 0.001 was considered significant.
Results
Between 2001 and 2019, the proportion of food products purchased by US households that contained additives increased from 49.6% to 59.5% (P < 0.001). The proportion of carbonated soft drinks purchased containing flavors decreased, with a subsequent increase in purchases containing nonnutritive sweeteners. Baby foods showed a 20% increase in the proportion of purchases containing additives and >15% increase in the proportion of purchases containing three or more additives.
Conclusions
There is convincing evidence that US household purchases of common types of technical food additives are increasing. Despite some positive changes such as a decrease in the use of added flavors in carbonated soft drinks, across most food categories an increase in purchases of all types of products containing additives was observed. In particular the finding that purchases of baby food products containing additives have increased substantially is crucial in informing future research in this area and warrants further investigation.
Research Question: Has the prevalence and volume purchased of packaged food and beverages containing food additives with specific technical effects changed between 2001 and 2019 in the United States?
Key Findings: The proportion of food products purchased by US households that contained additives increased from 49.6% in 2001 to 59.5% in 2019. The proportion of carbonated soft drinks purchased containing flavors decreased, with a subsequent increase in purchases of these products containing nonnutritive sweeteners. Baby foods showed a 22% increase in the proportion of purchases containing additives and >15% increase in the proportion of purchases containing three or more additives. Baby food also had the largest percent increase in purchases containing additives between 2001 and 2019 out of all major food categories examined.
Unhealthy diets are underpinned by the overconsumption of packaged food and beverage products. Over the past decade concern for ultra-processed foods (UPFs), which generally contain many additives added to enhance palatability, have been associated with increased bodyweight and other nutrition-related noncommunicable diseases and mortality.
Association of ultra-processed food consumption with cardiovascular mortality in the US population: long-term results from a large prospective multicenter study.
Ultra-processed foods, incident overweight and obesity, and longitudinal changes in weight and waist circumference: the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).
Ingredients characteristic of UPFs include food additives (eg, colors, flavors, flavor enhancers, emulsifiers, thickeners, and artificial sweeteners) designed to make the final product more palatable.
Food additives have been used predominantly over the past century to perform specific functions in foods. In its broadest sense, a food additive is defined as any substance that is added to food. Legally, in the United States, the term additive refers to "any substance the intended use of which results or may reasonably be expected to result—directly or indirectly—in its becoming a component or otherwise affecting the characteristics of any food." Food additives are commonly used as coloring agents, flavoring agents, preservatives and sweeteners. Although many additives are undoubtedly useful for increasing shelf life and food safety and have no negative health associations, there is evidence showing a number of food additives are linked to adverse health outcomes
and that high consumption of UPFs (which in turn generally contain more additives compared with their less-processed counterparts) potentially linked to addictive eating behavior.
Exposure to the common food additive carrageenan leads to glucose intolerance, insulin resistance and inhibition of insulin signalling in HepG2 cells and C57BL/6J mice.
In addition, there is some evidence from experimental research demonstrating that the interaction of different additives may lead to negative health effects, although research in this area is still in its infancy.
On balance, technological advances have resulted in the introduction of food additives with no known negative health consequences to the US food supply that has allowed for longer shelf life, protection against microbial growth and numerous other benefits to the population.
There is little research examining sources and types of food additives consumed or purchased by the US population. Research globally is also lacking, with only a few studies from France,
Deconstructing the supermarket: systematic ingredient disaggregation and the association between ingredient usage and product health indicators for 24,229 Australian foods and beverages.
examining the types of food additives in national food supplies. A small number of US-based studies exist that examine artificial color use in US foods and beverages.
Despite this, it is yet unknown to what extent US households are purchasing packaged foods and beverages containing common food additives, and there has been no research examining whether there have been changes over time.
The food composition databases underpinning dietary surveys are often unable to identify the unique ingredients in individual food and beverage products that are purchased and consumed by US adults and children.
US consumers purchase >400,000 different packaged food and beverage products annually, and the specific products purchased are constantly changing due to product introduction and reformulation.
Hence, most studies to date that have examined food additives in the United States have been limited by out-of-date product-specific food composition data, and no research has been undertaken examining changes that may have occurred over time, with the exception of a recent US study examining changes in purchases of products containing nonnutritive sweeteners (NNS) that showed purchases increased between 2001 and 2018.
The objective of this study was to examine the proportion of food and beverage products purchased by US households that contain four common types of technical food additives (colors, flavors, preservatives, and NNS) using time-specific food composition data, and to examine whether changes have occurred over time. Food group sources of food additives were also examined.
Subjects and Methods
Study Design and Population
The Nielsen Homescan panel is an ongoing nationally representative longitudinal survey in which participating households record all Universal Product Code (UPC) purchases on a daily basis using a handheld scanner.
Households also report sociodemographic and household information and are sampled and weighted to be nationally representative. The Homescan dataset is used frequently by researchers to examine food consumption and purchasing patterns.
For this study Homescan data from 2001 and 2019 were used. These years were chosen as they represent the oldest and newest available datasets for which this analysis could be undertaken. By utilizing secondary de-identified Homescan data, this work was exempted from institutional review board review.
Linkage of UPC Food Products with Nutrition Facts Panel Data and Ingredient Information
Each UPC captured in Nielsen Homescan 2001 and 2019 was matched with Nutrition Facts Panel data and ingredient information (when available) using commercial nutrition databases (Gladson, Label Insight, Product Launch Analytics, USDA National Nutrient Database for Standard Reference, and Mintel Global New Products Database) in a time-relevant manner. These commercial databases contain national brands and private label items at the barcode level, and data are updated regularly as new products enter the market. Further details regarding matching these commercial datasets have been published previously,
and involve matching and merging of data using UPCs, product descriptions, nutrient information, and ingredients lists. A team of registered dietitians reviewed the Homescan purchases and assigned all products to major and minor food categories. Homescan does not include bulk items without barcodes, thus foods and beverages without a barcode or a Nutrition Facts Panel were excluded (eg, unpackaged fresh fruits and vegetables, fresh meats, and loose bread and bakery products without a barcode). It should be noted that these foods are much less likely to incorporate food additives than packaged food and beverage items and are consistent with the NOVA definition of minimally processed foods.
Identification of Food Additives
A list of industrialized food additives utilized in UPFs and their technical functions was compiled from the Food and Drug Administration (FDA) Substances Added to Food inventory, previously known as Everything Added to Foods in the United States.
Four common classes of technical food additives were selected to examine in this study (ie, preservatives, NNS, color additives, and flavors). In brief (using the FDA’s Substances Added to Food inventory), colors were defined as substances authorized by a regulation in 21 CFR Part 73, 74, or 82. Flavors were defined as substances with a Flavour and Extracts Manufacturers Association or Joint Food and Agriculture/World Health Organization Expert Committee on Food Additives number, excluding those that are not listed as being a flavoring agent or flavor enhancer and excluding those in the 182.10 category (herbs and spices). Preservatives were defined as substances identified as an “antimicrobial agent,” and NNS were defined as substances listed as having both an NNS function under the Substances Added to Foods inventory and that also appear in the FDA’s list of approved high-intensity sweeteners. Three additional additives (luo huan guo, sucralose, and steviol glycosides) were also included in the definition of NNS because they appear in FDA’s list of approved high-intensity sweeteners but not under the Substances Added to Food inventory as NNS. To remain consistent with existing literature on NNS, sugar alcohols were also included as NNS (sorbitol, xylitol, lactitol, mannitol, erythritol, and maltitol). All remaining substances were included as “other” additives. Figure 1 lists the search terms used to define each class of food additives examined in this study.
Figure 1Search terms used to identify color, flavor, preservative, and nonnutritive sweetener additives in US packaged food and beverage products. aJECFA = Food and Agriculture Organization/World Health Organization Expert Committee on Food Additives. bFEMA = Flavor and Extract Manufacturers Association of the United States. cRefer to: https://www.cfsanappsexternal.fda.gov/scripts/fdcc/?set=FoodSubstances.
Colors search terms
Flavors search terms
Preservatives search terms
Nonnutritive sweeteners search terms
acetic acid acetone alcohol sda-3a alcohol, denatured formula 23a aluminum hydroxide ammonium bicarbonate ammonium carbonate ammonium hydroxide ammonium phosphate, dibasic ammonium phosphate, monobasic ammonium sulfate ammonium sulfite annatto, extract (bixa orellana l.) astaxanthin benzoin benzyl alcohol beta-apo-8'-carotenal beta-carotene butterfly pea flower extract butyl alcohol calcium carbonate calcium hydroxide canthaxanthin capsicum (capsicum spp.) capsicum, oleoresin (capsicum spp.) caramel carmine (coccus cacti l.) carrot, oil (daucus carota l.) castor oil (ricinus communis l.) cetyl alcohol citric acid citrus red no. 2 cochineal extract (coccus cacti l.) copals, manila corn endosperm oil corn steep liquor cottonseed flour, partially defatted, cooked, toasted cyclohexane damar gum (shorea dipterocarpaceae) dehydrated beets dextrose diethylene glycol distearate dioctyl sodium sulfosuccinate dipotassium phosphate dried algae meal edta, calcium disodium edta, disodium ethoxyquin ethyl acetate ethyl alcohol ethyl cellulose ethylene dichloride ethylene glycol distearate ethylene glycol monoethyl ether fd&c blue no. 1 fd&c blue no. 1, aluminum lake fd&c blue no. 2 fd&c blue no. 2, aluminum lake fd&c green no. 3 fd&c red no. 3 fd&c red no. 40 fd&c red no. 40, aluminum lake fd&c red no. 40, calcium lake fd&c yellow no. 5 fd&c yellow no. 5, aluminum lake fd&c yellow no. 5, calcium lake fd&c yellow no. 6 fd&c yellow no. 6, aluminum lake fd&c yellow no. 6, calcium lake ferric oxide ferrous gluconate ferrous lactate fruit juice grape color extract grape skin extract haematococcus algae meal hexane hydroxypropyl cellulose invert sugar iron oxide isobutyl alcohol isopropyl alcohol japan wax malt syrup (malt extract) methyl alcohol methylene chloride mica molasses (saccharum officinarum l.) naphtha orange b paprika (capsicum annuum l.) paprika oleoresin (capsicum annuum l.) phaffia yeast phosphoric acid polyethylene glycol (m w 200-9,500) polyglycerol esters of fatty acids polysorbate 60 polysorbate 65 polysorbate 80 polyvinyl acetate polyvinyl alcohol polyvinylpyrrolidone potassium bicarbonate potassium carbonate potassium hydroxide potassium phosphate, tribasic potassium sulfate potassium sulfite propylene glycol riboflavin rosin (pinus spp.) and rosin derivatives rosin, adduct with fumaric acid, pentaerythritol ester rosin, limed saffron (crocus sativus l.) salts of fatty acids shellac, purified silicon dioxide sodium bicarbonate sodium carbonate sodium copper chlorophyllin sodium hydroxide sodium phosphate, dibasic sodium phosphate, monobasic sodium phosphate, tribasic sodium sulfate sodium sulfite sodium tripolyphosphate sorbitan monooleate sorbitan monostearate soy leghemoglobin spirulina extract sucrose sulfur dioxide sulfuric acid sulfurous acid tagetes meal & extract talc terpene resins, natural terpene resins, synthetic titanium dioxide tomato lycopene trichloroethylene turmeric (curcuma longa l.) turmeric, oleoresin (curcuma longa l.) ultramarine blue vegetable juice xanthophyll zinc oxide
All substances with a JECFAa or FEMAb number were included in the flavors search terms. More than 3,000 exist and hence too many to list in a single documentc
The volume of foods purchased overall (in grams), by food and beverage status and by food group was determined overall and for all food additives, colors, flavors, preservatives, and NNS. Differences over time in the percent volume purchased overall and in each major and minor food category were examined through t tests using household projection weights provided by Nielsen. Bonferroni multiple comparison adjustment was applied and a P value < 0.001 was considered significant.
Results
Overall Results
The number of products purchased overall between 2001 and 2019 increased by 17%, from 355,870 in 2001 to 414,629 in 2019 (Table 1). The mean number of additives in purchased food and beverage products overall increased from 3.7 in 2001 to 4.5 in 2019 (P < 0.001). Between 2001 and 2019, the percent volume of products purchased by US households that contained additives increased from 49.6% to 59.5% (P < 0.001) (Table 2). Overall, there was a significant increase in the proportion of products purchased that contained one additive (4.9% to 7.1%; P < 0.001), two additives (5.5% to 7.5%; P < 0.001), and three or more additives (44.6% to 51.2%; P < 0.001), with a subsequent decrease in the proportion of products containing zero additives (45.0% to 34.2%; P < 0.001) (Table 3).
Table 1Number of unique products purchased by US households and mean number of additives per product purchased, by food category (2001 and 2019)
Volume is in grams or milliliters as provided on product packaging.
purchased among US households (weighted means) of foods and beverages containing additives (n = 34,925 households in 2001 and n = 61,467 households in 2019)
Volume is in grams or milliliters as provided on product packaging.
of food and beverage products purchased by US households with zero to 3+ additives by subcategory (n = 34,925 households in 2001 and n = 61,467 households in 2019)
The overall increase in additives was driven primarily by food purchases (50.9% to 63.7%; P < 0.001). There were significant increases in volume purchases of food products containing colors (39.1% to 47.9%; P < 0.001), flavors (34.2% to 41.7%; P < 0.001), preservatives (26.7% to 34.5%; P < 0.001), and NNS (1.2% to 2.8%; P < 0.001).
Smaller significant change was observed overall for the proportion of beverages containing additives (48.6 to 56.0%; P < 0.001) (Figure 2). Although beverage products had a significant increase in volume purchases of products containing colors (41.8% to 46.5%; P < 0.001) and NNS (12.1% to 16.5%; P < 0.001), a significant decrease was observed in beverages containing flavors (40.4% to 36.3%; P < 0.001) and preservatives (44.2% to 40.7%; P < 0.001) (Figure 2).
Figure 2Changes in the proportion of US household food and beverage purchases between 2001 and 2019 containing additives. University of North Carolina calculation based in part on data reported by NielsenIQ through its Homescan Services for all food categories including beverages for 2018 across the US market.
Authors’ calculations based in part on data reported by NielsenIQ through its Homescan Services for all food categories, including beverages for 2018 across the US market. The conclusions drawn from the data are those of University of North Carolina and do not reflect the views of NielsenIQ. NielsenIQ is not responsible for and had no role in, and was not involved in, analyzing and preparing the results reported herein.
Frozen entrées, appetizers, and pizza was the subcategory with the highest mean number of additives in both 2001 and 2019 (11.1 and 10.5, respectively), followed by carbonated soft drinks (7.4 and 8.0). Between 2001 and 2019 the majority of categories and subcategories showed an increase in the volume of purchases containing all types of additives (Table 2). Table 3 shows changes in the mean number of additives in each food category and subcategory. Baby food showed the largest increase in the proportion of purchases containing three or more additives (28.7% to 48.2%; P < 0.001), followed by meat, poultry, fish and mixtures (37.3% to 53.9%; P < 0.001) and grain products (59.0% to 73.6%; P < 0.001). Only one category (fats and oils) showed an increase in purchases of products containing zero additives (47.6% to 54.7%; P < 0.001). More than half of all food categories had a larger proportion of purchases of products containing three or more additives compared with zero additives.
There was a significant decrease in purchases of carbonated soft drinks containing flavors between 2001 and 2019 (80.3% to 75.4%; P < 0.001) (Table 2). Despite this significant decrease, in 2019 more than 75% of carbonated soft drinks purchases still contained added colors and more than 95% contained at least one additive. Of all beverage subcategories, carbonated soft drinks had the highest mean number of additives in both survey years; 7.4 in 2001 and 8.0 in 2019 (Table 1). Almost every subcategory under sweets and snacks and beverages had an increase in the mean number of additives in purchased products, as well as an increase in the proportion of purchases containing NNS (Table 1 and Table 2).
Discussion
UPFs represent a growing proportion of the US diet, and consumption of industrial food additives has been identified as a growing concern alongside consumption of UPFs.
To the knowledge of the authors, this is the first study to examine the extent to which US consumers are purchasing food and beverage products containing industrial food additives, with the exception of one recent study examining changes in purchases of packaged foods containing NNS showing increases between 2001 and 2018.
Results showed not only an increase between 2001 and 2019 in the proportion of food and beverage purchases containing additives but also an increase in those containing three or more additives. Along with these findings there was a more than 10% decrease in the proportion of products purchased that contained zero additives.
The overall results align for the most part with existing research. For example, a recent study examining sensory-related industrial additives in US food and beverage products found that most contained at least one additive, and more than one-third had three or more additives.
The same study also found that more than 80% of sweets, nonalcoholic beverages, and ready-to-eat cereals contained additives, and in the present study it was observed that products purchased from these three categories had the highest mean number of additives in both 2001 and 2019. Another study conducted in France found that 54% of packaged foods and beverages contained at least one additive, with beverages and sweets having the highest number of additives.
Deconstructing the supermarket: systematic ingredient disaggregation and the association between ingredient usage and product health indicators for 24,229 Australian foods and beverages.
Estimates from the aforementioned studies are slightly higher overall than what was observed in the present study, with our observation that in 2019 just over 50% of all products purchased by US households contained additives.
With US dietary guidelines recommending lowered consumption of processed foods and a higher consumption of less processed foods, it’s concerning that there has been an increase in purchases of products containing additives since 2001. Of specific concern, baby foods (which include products such as baby formula, purées, and infant cereals) showed a 20% increase in the proportion of purchases containing additives, a 15% increase in the proportion of purchases containing three or more additives between 2001 and 2019, as well as an increase in colors, flavors, and NNS. To the knowledge of the authors there is currently no research examining the extent to which food additives are used in baby foods in the United States, however with research suggesting a negative association between some food additives and adverse health effects in young children,
as well as the findings in the present study, there is cause for concern. It is important to note that although certain additives may, for this category in particular, provide certain benefits (such as for preservative effects), it is unlikely that the large increases seen for added colors (>10% increase) and flavors (>20% increase) are providing nutritional and health benefits and warrants further investigation.
Results showed that purchases of food and beverage products containing added colors have increased between 2001 and 2019. Added colors are a widely studied food additive in relation to adverse behavioral effects,
and may be cause for concern, particularly in beverages that are marketed to, and consumed by, children. A US-based study from 2016 examined 810 food and beverage products from a single grocery store that were marketed to children and found that 43% contained at least one food color.
In 2019, more than 50% of carbonated soft drink purchases still contained added colors and more than 95% still contained at least one additive. Soft drinks had on average four additives in 2019 and in fact every beverage subcategory had an increase in the mean number of additives purchased by US households between 2001 and 2019.
This study found an increase in total food products purchased containing NNS from 12.1% to 16.5%, and in carbonated soft drinks specifically from 28.0% to 32.7%. This fits with existing research that has demonstrated the increasing use of NNS in beverages available in the Untied States. This increase is shown to be occurring in tandem with a reduction in purchases of beverages containing caloric sweetener (sugar) and an increase in purchases of beverages containing both caloric sweetener and NNS.
It will be important to examine whether the move toward products containing both caloric sweetener and NNS (and in some cases, color additives as well) have any effect on population health. Despite negative health having been demonstrated for a number of specific additives, both colors and NNS in particular have been negatively associated with changes to gut microbiota,
and little is known about whether such effects are further affected by the combination of multiple additive types in a product. Clinical research in this area is still in its infancy, with few studies examining the interactive health effects of different types of food additives.
A strength of this study was the use of nationally representative food purchase data. An important point to raise is that purchase data cannot be equated with intake due to unknown intrahousehold distribution, plus foods acquired through other sources.
This study is not without limitations. One challenge in the use of Homescan data is that estimates of household purchases might not be comparable with per capita intake from dietary intake surveys such as the National Health and Nutrition Examination Survey. For example, in a given household all purchases of products containing additives could be consumed by a single member of the household, rather than being consumed by all household members. Homescan also does not capture purchases from fast-food chains and other restaurants, which would undoubtedly have resulted in an underestimate of purchased products containing additives in the present study. Another potential limitation is that each ingredient was not further examined to determine if each listed ingredient contained additives. In a study on added sugar, Ng and colleagues
developed a technique to use a commercial database that included the subingredients of each ingredient to estimate the amount of added sugar in each US food product. In that study they found added sugar in many of these ingredients above and beyond the added sugar terms that were indicated in the ingredients list. This indicates that the present study could potentially have underestimated the number of additives purchased by US households.
Conclusions
Given the importance of understanding what components make up UPFs in the US diet, this study found convincing evidence that US household purchases of packaged foods containing food additives is increasing. When an increase of about 10% of the volume of food and beverage products purchased from retail outlets contain additives, this is a topic of concern to nutrition and dietetics practitioners. Despite some positive changes such as a decrease in purchases of carbonated soft drinks containing added flavors, across most food categories an increase in purchases of all types of additives was observed. In particular, our finding that purchases of baby food products containing additives has increased substantially is crucial in informing future research in this area and warrants further investigation.
Association of ultra-processed food consumption with cardiovascular mortality in the US population: long-term results from a large prospective multicenter study.
Ultra-processed foods, incident overweight and obesity, and longitudinal changes in weight and waist circumference: the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).
Exposure to the common food additive carrageenan leads to glucose intolerance, insulin resistance and inhibition of insulin signalling in HepG2 cells and C57BL/6J mice.
Deconstructing the supermarket: systematic ingredient disaggregation and the association between ingredient usage and product health indicators for 24,229 Australian foods and beverages.
E. Dunford is adjunct faculty, University of North Carolina Chapel Hill, Chapel Hill, and a research fellow, Food Policy Division, The George Institute for Global Health, University of New South Wales, Sydney, Australia.
D. Miles is director, research programming, Carolina Population Center, University of North Carolina Chapel Hill, Chapel Hill.
B. Popkin is a professor, Department of Nutrition, Gillings Global School of Public Health, University of North Carolina at Chapel Hill, and with the Carolina Population Center, University of North Carolina Chapel Hill, Chapel Hill.
Article info
Publication history
Published online: January 31, 2023
Accepted:
November 18,
2022
Received:
April 12,
2022
Publication stage
In Press Corrected Proof
Footnotes
STATEMENT OF POTENTIAL CONFLICT OF INTEREST No potential conflict of interest was reported by the authors.
FUNDING/SUPPORT Funding for this work comes primarily from Arnold Ventures with additional support from the Carolina Population Center National Institutes of Health grant no. P2C HD050924.
ACKNOWLEDGEMENTS The authors thank the UNC Global Food Research Program team of registered dietitians and research associates (Bridget Hollingsworth, Jessica Ostrowski, and Julie Wandell) for their support in identifying food additive search terms.
AUTHOR CONTRIBUTIONS E. Dunford and B. Popkin designed and conducted this research. D. Miles analysed data. E. Dunford drafted the manuscript. E. Dunford, D. Miles, and B. Popkin had joint primary responsibility for final content.
01193-5&id=gr1.jpg){kind=link}