Perspective: Structure-Function Claims on Infant Formula

Abstract

In the context of a food product label, the term “claim” refers to information that attributes value to the product. The term extends to many different types of information, from product identity, descriptors of intended use, and identification of characteristic properties to the physiologic effects in the body of substances in the food, including the reduction of risk of disease. Food labeling, which includes claims, provides information that consumers want and use to improve their diets. Consumers prefer short statements on the front label claims to longer, more detailed information, including ingredients statements and a nutrition panel. Three types of claims are permitted in the United States. Nutrient content claims describe the level of the nutrient in the food relative to an established daily value, e.g., “Excellent source of choline,” and are subject to composition limits for other nutrients, such as total fat, saturated fat, and cholesterol. Health claims describe the relation between a food substance and the risk of disease, e.g., “Adequate calcium and vitamin D throughout life, as part of a well-balanced diet, may reduce the risk of osteoporosis.” They must undergo a premarket evaluation by the FDA to ensure that there is significant scientific agreement about the relation in question. The third type of claim, structure-function (SF) claims, has recently come under scrutiny, particularly regarding their use on infant formula. Such claims represent a food's effect on the structure or function of the body for maintenance of good health and nutrition. These claims must be truthful and not misleading, but are not subject to premarket approval before use. The purpose of this perspective is to describe the origins and unique niche of SF claims, and to comment on recent proposals to further regulate such claims on infant formula.

Background

The Food Drug & Cosmetic Act (FD&C Act) (1) defined drugs as “articles (other than food) intended to affect the structure or any function of the body.” Statements about the effect of a food (or the nutritive substances in the food) on the structure or function of the body can be made on the food's label without causing the food to be classified as a drug.

The functions of most nutrients considered essential to humans were discovered in animals (2–5). Many of the discoveries were made in the years leading up to the passage of the FD&C Act in 1938, perhaps explaining why the FD&C Act exempted foods from the drug definition. The medical and scientific communities were churning out discovery after discovery about substances in food that had a special role in the functioning of the body. In each weekly issue of The Journal of the American Medical Association (JAMA) during the 6-mo period leading up to passage of the FD&C Act, there was ≥1 article about these newly discovered substances—substances that affected the structure or function of the body and were not drugs (Table 1). Introducing a forthcoming series on vitamins by highlighting the recent chemical syntheses of vitamin C and riboflavin, the editors alluded to the explosion of knowledge: “there are more substances with vitamin properties than there are letters in the alphabet” (6).

TABLE 1.

JAMA articles leading up to the FD&C Act, 1938¹

Reference	Authors and title
Volume 110
(6):414–419	Goodhart R, Jolliffe N. Effects of vitamin B (B1) therapy on the polyneuritis of alcohol addicts.
(8):577–579	Fishbein M. New series of articles on vitamins.
(8):584	New series of articles on the vitamins. (editorial)
(9):645–649	Nelson EM. The components of the vitamin B complex.
(10):727–732	Williams RR. The chemistry of thiamin (vitamin B1).
(11):805–812	Cowgill GR. The physiology of vitamin B1.
(12):870–877	Schour I. Calcium metabolism and teeth.
(13):953–956	Strauss MB. The therapeutic use of vitamin B, in polyneuritis and cardiovascular conditions. Clinical indications.
(14):1081–1084	Spies TD, Aring CD. The effect of vitamin B1 on the peripheral neuritis of pellagra.
(14):1105–1111	Booher LE. Chemical aspects of riboflavin.
(15):1188–1193	Hogan AG. Riboflavin: physiology and pathology.
(15):1196	Cystine and methionine. (editorial)
(16):1278–1280	Sherman HC, Lanford CS. Riboflavin: dietary sources and requirements.
(17):1338–1341	Lewis JM, Barenberg LH. The relationship of vitamin A to the health of infants. Further observations.
(17):1370–1372	Federal food and drug bill advanced.
(18):1441–1446	McCay CM. Other factors: less well known vitamins.
(19):1620	Sydenstricker VP. Present status of nicotinic acid in the treatment of pellagra.
(20):1665–1672	Sebrell WH. Vitamins in relation to the prevention and treatment of pellagra.
(21):1748–1751	Palmer LS. The chemistry of vitamin A and substances having a vitamin A effect.
(22):1831–1837	Mattill HA. Vitamin E.
(23):1896–189	Wortis H, Liebmann J, Wortis E. Vitamin C in the blood, spinal fluid and urine.
(23):1920–1925	Booher LE. Vitamin A requirements and practical recommendations for vitamin A intake.
(24):1993–1994	Knapp AA. Vitamin D complex in keratoconus. Etiology, pathology and treatment of conical cornea: preliminary report.
(25):2065–2066	Schmidt HL, Sydenstricker VP. Nicotinic acid in the prevention of pellagra. A preliminary note.
(25):2072–2080	Bessey OA, Wolbach SB. Vitamin A: physiology and pathology.
(26):2150–2155	Bills CE. The chemistry of vitamin D.
(26):2157–2158	Vitamin C and complement. (editorial)
Volume 111
(1):28–29	Relationship of vitamins to enzymes. (editorial)
(2):144–154	Clausen SW. The pharmacology and therapeutics of vitamin A.
(3):245–252	Munsell HE. Vitamin A: methods of assay and sources in food.
(4):324–326	The new federal food and drug act at last.
(6):528–530	Nelson EM. The determination and sources of vitamin D.
(7):614–619	Shohl AT. Physiology and pathology of vitamin D.
(8):703–711	Jeans PC, Stearns G. The human requirement of vitamin D.
(11):1009–1016	Cowgill GR. Human requirements for vitamin B1.
(12):1098–1101	King CG. The physiology of vitamin C.
(13):1179–1187	Park EA. The use of vitamin D in the prevention and treatment of disease.
(15):1376–1379	Dalldorf G. The pathology of vitamin C deficiency.
(16):1462–1464	King CG. The chemistry of vitamin C.
(17):1555–1565	Abt AF, Farmer CJ. Vitamin C: pharmacology and therapeutics.
(19):1753–1764	Smith SL. Human requirements of vitamin C.

¹FD&C Act, Food Drug & Cosmetic Act; JAMA, The Journal of the American Medical Association.

Nutrient content claims and health claims were introduced by the Nutrition Labeling and Education Act (NLEA) (7), but the NLEA did not address structure-function (SF) claims on foods. Absent statutes and regulations specifying standards of evidence, the legal standard which SF claims must achieve is that they are truthful and not misleading, per section 403(a)(1) of the FD&C Act (1). Manufacturers are responsible for maintaining the supportive information for SF claims, but there is no requirement to provide such information to the FDA in advance of marketing foods labeled with SF claims, as is required for other food claims by the NLEA.

The only other legislation addressing SF claims was the 1994 Dietary Supplement Health and Education Act (DSHEA) (8). The DSHEA defined 4 types of Statements of Nutritional Support, 2 of which use the words “structure” and “function”. Dietary supplements can make a claim that “describes the role of a nutrient or dietary ingredient intended to affect the structure or function in humans” or “characterizes the documented mechanism by which a nutrient or dietary ingredient acts to maintain such structure or function.” The FDA promulgated regulations that identified a safe haven for SF claims that did not fall into any of the 10 criteria¹ that can constitute a drug claim (9). SF claims on dietary supplements may pertain to the nutritive value in the supplements like SF claims on foods, but may also pertain to nonnutritive functions (e.g., the functions of ginger and caffeine). Manufacturers of dietary supplements must notify the FDA of their SF claims and label the product with the disclaimer, “This statement has not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease”. Nevertheless, SF claims on dietary supplements are subject to the same standards as they are for food: they must be truthful and not misleading.

Since the DSHEA, guidance documents that explain the FDA's thinking but do not have the force of law have been published on SF claims. In 2001, the Federal Trade Commission (FTC), which monitors the advertising of claims on foods and supplements through a memorandum of understanding with the FDA, published Dietary Supplements. An Advertising Guide for Industry, which laid out its view on the substantiation of claims on dietary supplements (10). The FTC stated that “structure/function claims, within the broader category of statements of nutritional support, refer to representations about a dietary supplement's effect on the structure or function of the body for maintenance of good health and nutrition.” The FTC's guidance established “competent and reliable evidence” as the standard for substantiation, defined by what experts thought was adequate. At about the same time, the FDA considered its conceptual approach to SF claims on conventional foods and identified the need to address substantiation (11). In 2002, the FDA published guidance on SF claims on dietary supplements (12), then in 2008 it published Guidance for Substantiation of Dietary Supplement Claims (13), which further clarified through examples how to use the 10 criteria that differentiate SF claims from drug claims set forth in regulation.

In September 2016, the FDA published Substantiation for Structure/Function Claims Made in Infant Formula Labels and Labeling: Guidance for Industry. Draft Guidance (DG) (14). This was the FDA's first guidance regarding SF claims on conventional foods, albeit limited to infant formula. Until this document, there had been no discussion of SF claims in infant formula legislation, regulations, or guidance. The Infant Formula Act (IFA), passed in 1980 (15), was focused on nutrient composition and nutrient testing. Amendments in 1986 established good manufacturing practices, quality factors, and labeling requirements, among other things. Neither the IFA nor the 1986 amendments addressed claims. The NLEA (7) exempted infant formulas from mandatory nutrition labeling, limited nutrient content claims that could be made on infant formula [21 Code of Federal Regulations (CFR) 101.13(b) (16)], and permitted formulas to bear health claims like other foods [21 CFR 101.14(e)(5) (16)], but did not address SF claims on either general foods or infant formulas. Recently completed rulemaking on good manufacturing practices and quality factors for infant formula also did not address SF claims (17). Finally, at the same time that the FDA published the DG, it published a separate document, Guidance on Labeling of Infant Formulas, which discussed health claims and nutrient content claims, but was silent on SF claims (18). It is curious that the FDA's first foray into the substantiation of SF claims on conventional foods applied only to a group of foods that had its own ample statutory and regulatory history, in which SF claims had never been mentioned.

Shortly after the publication of the DG, a viewpoint coauthored by (among others) the former Director of the Center for Food Safety and Applied Nutrition and the former Deputy Commissioner of the FDA was published in JAMA calling attention to the need for regulation of SF claims on infant formula and decrying the lack of available evidence for such claims (19). These recent developments call for greater understanding of the origin, purpose, and context of SF claims.

Raising the Substantiation Bar

As mentioned, guidance documents do not have the force of law, but represent the agency's most recent thinking. The DG (14) is open to public comment at any time, though the FDA also formally solicited comments (20). There are a number of elements in the DG that merit examination. 1) The DG describes the type and quality of evidence that manufacturers should have in order to substantiate SF claims. The DG states that the FDA intends to apply the “competent and reliable evidence” standard in a manner that is consistent with the FTC's standard and the FDA's approach for the substantiation of dietary supplement claims. Such an assessment would precede any regulatory action, such as a warning letter, if the FDA were to consider an SF claim to be potentially false or misleading. However, the DG introduces new terms and deviates from previously published guidance in significant ways. 2) The FDA recommends that the endpoints for SF claims be limited to those recognized and accepted by authoritative bodies. 3) The FDA describes a reluctance to extrapolate findings from the study of one formulation to other, similar formulations, even though infant formula has comprehensive requirements for nutrient levels and the population that is fed infant formula is rather homogeneous.

Quality and Type of Evidence

In the DG (14), the FDA indicated its interest in maintaining consistency with the FTC, i.e., using competent and reliable evidence. However, the information described in the DG goes far beyond the FTC's reliance on the opinion of experts toward the type of evidence required for drug claims. In the DG, the FDA writes, “competent and reliable scientific evidence means evidence that includes findings from well-designed and controlled intervention studies…with and without the constituent of interest,” and “substantiation for structure/function claims in infant formula labeling rely primarily on the results of infant feeding intervention studies.” This position disregards other factors described by the FTC as determinants of evidence needed for substantiation, such as the type of claim, the benefit of a claim, the cost of generating the evidence, the consequences of a false claim, and the amount of evidence that experts in the field believe is reasonable. The DG frequently describes SF claims in terms of benefit (literally, “structure/function benefit”), a novel construction that modifies the statutory language in the FD&C Act of “affect” to “benefit.” The DG goes to some length to detail the need for randomized controlled trials (RCTs) that establish the benefit, but does not take into account the 2015 US District Court decision that multiple RCTs are not the correct standard for SF claims on dietary supplements (21), because the FTC itself had stated the standard was flexible. Finally, the emphasis on intervention trials for SF claims deviates from the standard for significant scientific agreement for health claims, which acknowledges that “the scientific evidence supporting a substance/disease relationship may have to be derived wholly or in part from observational studies” (22).

Evidence to substantiate drug efficacy (RCTs) differs from evidence needed to establish a nutrient function. Drugs are not ordinarily normal body constituents; nutrient intervention typically elevates the level of a constituently present substance. Assessment of drug efficacy requires short-time effects to remedy pathology; nutrient interventions, particularly those relating to standard infant formulas², involve infants who are not ill, but need optimum nutrition for growth and development. The use of RCTs for SF claims is limited by the need for very large numbers and long-term follow-up, which is frequently complicated by high attrition rates. Others have noted the limitations of applying drug efficacy study designs to nutrition studies, and have proposed alternatives that use all the available evidence (23), which is especially important for bioactive compounds (24, 25) that are candidates for SF claims. Finally, the DG emphasis on intervention studies for SF claims is more stringent than the FDA uses in its evaluation of health claims, specifying studies to be done on US infants using a specific formula matrix, and assessing clinical meaningfulness; whereas the evaluation of totality of evidence for health claims more broadly considers number, type, and quality of studies, and reproducibility and consistency of findings (26).

Relative claims on conventional foods that suggest benefit or use any of the adjectives “restore,” “support,” “maintain,” “raise,” “lower,” “promote,” “regulate,” or “stimulate” in the context of normal physiologic function may be SF claims (14), and some but not all SF claims are of this type. Consider “DHA is important for visual function.” Experimental studies produced DHA-deficient monkeys. Improvements in their electroretinogram after the administration of supplemental DHA were demonstrated (27). These findings were repeated in numerous species and confirmed in humans when formula-fed infants were supplemented with DHA (28), so the biological relation between the structure (DHA) and the function (the visual function) is established as truthful and not misleading.

Endpoints

The DG recommends that endpoints for SF claims be limited to those recognized and accepted by authoritative bodies. The FDA writes, “Examples of documentation that an endpoint is recognized and accepted by qualified experts or an authoritative scientific body include (a) the opinion of an “expert panel” that is specifically convened for this purpose by an authoritative body such as the National Academy of Sciences, or (b) the opinion or recommendation of a federal government scientific body with relevant expertise, such as the National Institutes of Health or the Centers for Disease Control and Prevention.” This level of agreement about the endpoint is rare in “emerging science” and sets a standard higher than that proposed for DRIs for bioactive substances (24) or even drugs. The authoritative body standard is akin to that required for health claims that, by statute, require premarket approval. Nutrient content claims also require a daily value to be established by regulation, a process that relies on authoritative body recommendations. However, Congress has not established premarket review for SF claims for foods or dietary supplements. Premarket notifications for dietary supplements that contain new dietary ingredients require evidence that the supplement is reasonably expected to be safe under its labeled conditions of use, but the premarket review does not include the review of documentation of SF claims. Finally, the authoritative body standard is inconsistent with the standard that the FTC uses to judge the adequacy of evidence for SF claims: “A guiding principle for determining the amount and type of evidence that will be sufficient is what experts in the relevant area of study would generally consider to be adequate” (10). For the FTC, experts in the relevant area of study need not be constituted into an expert panel at the level of the National Academies of Sciences, Engineering and Medicine or at the federal government level. It is ironic that the FDA proposes to use only recognized endpoints for SF claims on infant formula, when almost all nutrient DRIs for infants were set as Adequate Intakes, based on the nutrient composition of human milk, not predefined endpoints (29). Adequate Intakes are used in place of RDAs when the evidence is insufficient to establish a requirement level. The standard the FDA proposes for the evaluation of SF claims on infant formulas is stricter than the standard it accepts for the required nutrient composition of infant formulas.

On the other hand, many SF claims are the result of investigations into the composition of human milk. Functions of nucleotides and β-carotene in support of the immune system (30), inositol in ocular function (31), taurine in infant metabolic conjugation and osmotic regulation (32), oligosaccharides as prebiotic and anti-infective agents (33), and DHA in visual acuity (see below) are among the contributions to pediatric nutrition science from the study of human milk composition, providing evidence to support SF claims.

Extrapolation

The FDA describes a reluctance to extrapolate findings from the study of a component in one formula to other formulas. The FDA writes, “For studies relating to specific nutrients or other constituents, a structure/function benefit demonstrated for the constituent in one matrix (e.g., cow milk-based formula) may not be generalizable to other matrices (e.g., soy protein isolate-based formula) because the beneficial outcome may vary between matrices due to different interactions within each matrix.” For SF claims that describe the role of a substance in the body or the mechanism of action, there is no dependence on the food from which the substance was derived. In contrast, the FTC writes that “advertisers need to evaluate whether it is appropriate to extrapolate from the research to the claimed effect.” The stringent composition requirements for infant formula and the rather homogeneous population that uses infant formula should enable easier extrapolation of results across formulations than is possible for dietary supplements, when the FTC expresses that extrapolations may be appropriate.

The Call for New Regulation

Hughes et al. (19) called for tighter regulation of SF claims on infant formula and noted that the FDA has no authority to enforce the guidance, but did not point out the statutory protection of SF claims by the FD&C Act (7) and the DSHEA (8). Perhaps the emergence of SF claims on infant formulas has created new regulatory needs, though the DSHEA safeguarded SF claims when it could have established requirements similar to the NLEA for nutrient content claims and health claims. The authors (19) state, “For many claims there is no evidence available to the public, and when the results of randomized trials are made public, we learn that they are limited by small sample sizes, poor follow-up, and provide unpersuasive results.” It is certainly not the case that there is no information available to the public. Hughes et al. (19) cite a Cochrane review from 2011 (which is publicly available) to support the interpretation that studies on DHA “did not find any benefit” to cognitive development. Yet, the most recent Cochrane review on DHA and cognitive outcomes concluded that the “Majority of the RCTs have not shown beneficial effects of LCPUFA [long-chain PUFA] supplementation on the neurodevelopmental outcomes of term infants” and “The beneficial effects on visual acuity have not been consistently demonstrated” (34). By implication, some studies did find beneficial effects of LCPUFAs on neurodevelopmental outcomes and visual acuity of term infants. The 2017 Cochrane Review found 9 RCTs on visual acuity after LCPUFA supplementation of term infants, of which 4 reported improvement and 5 reported no significant change (34).

Under its compliance program, the FDA annually collects infant formula labels and promotional brochures that usually include references to published scientific reports (35). Consequently, the FDA is aware of the SF claims made on the labels of infant formulas, but has not issued any warning letters for unsubstantiated claims. One possible reason for this is that the legal standard requires the FDA to prove that the claim is not truthful or is misleading, which is difficult when there is some supportive evidence to the contrary. The DG sets expectations for the type and conclusiveness of science to support claims, but does not change the legal standard of truthful and not misleading. Congress could create premarket requirements for SF claims, though Hughes et al. (19) note that it is not likely that Congress would take any such action. Another possible explanation for the absence of warning letters is that the review of claims substantiation is labor intensive and a low priority for the FDA when there is no apparent risk to health related to unreviewed claims. Hughes et al. (19) propose something like a voluntary GRAS (generally recognized as safe) notification, which uses outside experts to review the science. This would ease the review burden on the agency and is closer to the FTC standard of “expert opinion” than the “authoritative body” imprimatur described in the DG. A voluntary premarket review system would need to benefit the FDA, industry, and consumers. Premarket review of claims might reduce the number of SF claims regardless of the extent of scientific support; the regulation of health claims has reduced their prevalence (36).

The standard for assessing substantiation needs further discussion as well, possibly through comments to the FDA on the DG. What evidence should be considered and how conclusive must the evidence be? The Hughes et al. (19) criticism of the evidence supporting SF claims for DHA in infant formula relied on selected evidence and did not include numerous publicly available reports (including meta-analyses or comprehensive reviews) that concluded that there were positive benefits from DHA. One uncited review coauthored by FDA staff reported that DHA/arachidonic acid (ARA) supplementation had beneficial effects on neurodevelopment in only 2 out of 12 studies that used the Bayley Scale of Infant Development, but benefits were reported in 8 out of 11 studies using other endpoints (37). An uncited opinion from the European Food Safety Authority considered that there was sufficient evidence to establish a cause and effect relation between DHA and normal brain development. It suggested specific language for the claim, i.e. “DHA contributes to normal brain development” (38), and required the mandatory addition of DHA to all infant formula. A recent authoritative body report citing 3 meta-analyses found that visual acuity was improved by DHA supplementation to term infants, and found the overall strength of the evidence to be moderate (39).

Similarly, Hughes et al. (19) cite one survey of infant formulas bearing claims on colic and gastrointestinal symptoms that stated that the claims had insufficient evidence, and that drug evidentiary standards (clinical evidence) had been used for their comparisons (40). Regarding colic and gastrointestinal symptoms, there is no Cochrane Review available, only a proposal (41) that cites many studies reporting the effects of formula modifications to reduce colic symptoms and notes a 2012 systematic review (publicly available on Google Scholar) which determined, among other things, that “Evidence from this systematic review suggests… changing to a hydrolysed protein formula may reduce colic in formula-fed infants” (42).

There is some similarity between SF claims for which evidence of effects has accrued but for which there is still some variability in its interpretation and qualified health claims for which evidence has not yet reached the level of significant scientific agreement. The FDA uses enforcement discretion to permit qualified health claims provided they use mandatory qualifying language. SF claims commonly use less definitive language and qualify the claim—e.g., “helps support,” “may help reduce” (Table 2)—an acknowledgement that the supporting science is not conclusive. Consumers have a rational interpretation of claims (43); they are at the same time skeptical of qualified claims such as “may reduce” or “helps reduce”, while finding them more credible than broad or absolute claims (24, 25). Guidance from the FDA on how to properly qualify the wording of SF claims could be more helpful for consumers than creating new standards for substantiation.

TABLE 2.

Claims on US formulas

Claim	Structure	Function
Bones and teeth
Vitamin D to support bone health and development	Vitamin D	Bone health and development
Supports strong bones & teeth	Entire formula	Bones & teeth
Vision/eye; brain/cognition
Brain & Eye, DHA & ARA	DHA & ARA	Brain & eye
Brain-nourishing DHA and choline to help support your infant's brain development	DHA & choline	Brain development
DHA and Choline—brain-nourishing nutrients also found in breast milk	DHA & choline	Brain nourishment
DHA, Iron & choline to help nourish the brain	DHA, iron & choline	Brain nourishment
OptiGRO. DHA for BRAIN. LUTEIN for EYES. VITAMIN E for DEVELOPMENT	DHA, lutein, vitamin E	Brain, eyes development
Cognitive, MFGM Fosters Cognitive Development	MFGM	Cognitive development
Neurocomplete. Helps support a healthy brain, 4 areas pediatricians check for most. Cognitive, Motor, Social, Communication	Entire formula	Cognitive function, motor function, social function, communication
Digestion
Small, easy-to-digest COMFORT PROTEINS	Proteins	Digestion
Gentle nutrition tailored for infants	Entire formula	Digestion
Easy-to-digest protein	Protein	Digestion
Easy-to-digest gentle protein	Protein	Digestion
Gentle nutrition when moms choose to introduce formula	Entire formula	Digestion
Easy-to-Digest	Entire formula	Digestion
Complete Nutrition for Delicate Tummies	Entire formula	Digestion
Similac Tummy Care Guarantee	Entire formula	Digestion
Small, easy-to-digest COMFORT PROTEINS	Proteins	Digestion
Gentle nutrition when moms supplement	Entire formula	Digestion
Small, easy-to-digest COMFORT PROTEINS	Proteins	Digestion
MILK & LACTOSE-FREE nutrition for digestive comfort	Lactose free	Digestion
COMFORTING PROBIOTICS to support digestive health	Probiotics	Digestion
Gentle Protein & Prebiotics	Protein, prebiotics	Digestion
COMFORTING PROBIOTICS to reduce excessive crying	Probiotics	Digestion, crying
Helps reduce fussiness	Entire formula	Digestion, fussiness
Helps reduce fussiness, gas & crying within 24 hours	Entire formula	Digestion, fussiness, gas & crying
Clinically proven blend of two prebiotics to soften stools	2 prebiotics	Digestion, soften stools
Significantly reduces frequent spit-up	Entire formula	Digestion, spit up
Compound claims
Triple Health Guard Helps Support Growth, Brain & Eye, Immune system	Entire formula	Growth, brain & eye, immune system
Triple Benefits. Growth. Easy to Digest. Brain & Eye Immune function	Entire formula	Growth, Easy to Digest. Brain & Eye
Natural Defense Dual Prebiotics	Prebiotics	Immune function
Helps support newborn immune health	Entire formula	Immune health
Lactoferrin Supports Baby's Immune System	Lactoferrin	Immune system
Non GMO, Ingredients not genetically engineered	Entire formula	Ingredients
BioGaia, PROBIOTICS L. reuteri	Probiotics	Ingredients
Complete nutrition including DHA	DHA	Ingredients
Plant-sourced protein	Protein	Ingredients
Soy formula, milk-free and lactose-free	Entire formula	Ingredients
Unique blend of DHA & probiotics	DHA & probiotics	Ingredients
Unique blend of DHA, prebiotics & probiotics	DHA, prebiotics & probiotics	Ingredients
Benefits of adding rice cereal without the hassle	Entire formula	Ingredients
Claims FDA excluded from SF claims
Our closest formula to early breast milk	Entire formula	Not SF claim
Experts Recommend Vitamin D	Vitamin D	Not SF claim
Our closest formula to breast milk. With DHA & ARA	DHA and ARA	Not SF claim
Inspired by mature breast milk	Entire formula	Not SF claim
Our closest to breast milk	Entire formula	Not SF Claim
Complete nutrition for your baby's first year		Not SF Claim
Modeled after the complete nutrition of breastmilk	Entire formula	Not SF claim
Modeled after the changing protein levels in breastmilk	Entire formula	Not SF claim
Complete nutrition including DHA	DHA	Not SF claim
NEW Formula!	Entire formula	Not SF claim

ARA, arachidonic acid; GMO, genetically modified organism; MFGM, milk fat globule membrane; SF, structure-function.

Hughes et al. (19) raised 2 possible adverse consequences of SF claims: they may dissuade breastfeeding and increase cost. Surveys on the promotion of foods for older infants and young children in 4 lower income countries have recently been reported (44, 45) that suggest an association between the promotion of all types of commercial products and their use, but they did not examine the SF claims on the product labels. Pereira et al. (46) collected comprehensive information on product labels, including colors, logos, and ad mascots, but did not report on the SF claims on the labels of formulas for older infants. Another recent report reviewed the role of promotion of formula on breastfeeding (47). Although there is evidence that gift bags containing formula samples may shorten the duration of breastfeeding (48), there are no studies that have examined the impact of infant formula or other breast milk substitutes marketing relative to other determinants of suboptimal breastfeeding (49). Pacifier use has a similar-sized effect as discharge packs [both Rosenberg et al. (48) and Kair and Colaizy (50) report ORs of 0.71], suggesting that multiple variables need to be considered to address confounding. SF claims have not been reported in any study to date as having an effect on breastfeeding discontinuation. Major surveys probing discontinuation of breastfeeding do not even include formula labeling among the many reasons thought to bear on a mother's decision (51–53). Even had surveys associated SF label claims with shortened or discontinued breastfeeding, the question of causality would still remain; it is quite possible that a mother would not give much thought to labeling statements on a formula until she had already determined to feed formula (i.e., reverse causality).

Are current SF claims on US formulas so persuasive that they might alter the trajectory of breastfeeding? An informal survey of claims on major brand labels found 51 distinct claims (Table 2). Ten were of the type “closer to breast milk,” which the FDA states in the DG are not SF claims. Eight were statements of ingredient content, which the FDA states may be used on food labels, such as “X mg of EPA and DHA omega-3 fatty acids per serving” (54). The remaining claims can be subdivided into 4 functional categories: bone health (n = 2), immune function (n = 3), visual acuity/brain development (n = 7), and digestion (n = 19), with 2 claims being combinations of the above. Of the 2 more common claims, evidence for visual acuity was discussed above; and some evidence exists in support of each formula modification that has been made to support digestion. Given the inherently better digestibility of breast milk compared with formulas (55), it seems unlikely that a mother would abandon breastfeeding simply because a formula claimed to help digestion. As a statement to differentiate between formulas, however, such labeling can help formula-using consumers with product selection, as well as educate them about food digestion, the role of the intestinal microbiome, and factors that influence nutrient absorption.

Hughes et al.’s (19) point about the value of ingredients is more difficult to assess. What value does one place on neurodevelopment? Victora et al. (56) estimated substantial increases in monthly income associated with longer breastfeeding, with 72% of the effect being estimated to be mediated via higher IQ. Improved nutrition of older infants and young children, similarly, has been related to lifelong benefits, without attribution to individual nutrients (57). What is the value of improved visual acuity after DHA supplementation of formula fed infants, and is visual acuity a marker for functions, which are perhaps not clinically defined, in other neural tissues that have a high DHA content? Will a mother continue to purchase a more expensive specialty formula labeled for fussiness if there is no improvement. Few studies have assessed the economic value of bioactive substances added to functional foods (58).

The International Context

The Codex Guidelines on Nutrition and Health Claims (59) defines nutrient function claims, other function claims, or claims about the reduction of risk of disease. Other function claims are like SF claims in that they “concern specific beneficial effects of the consumption of foods or their constituents… on normal functions or biological activities of the body. Such claims relate to a positive contribution to health or to the improvement of a function or to modifying or preserving health.” In Europe, “general function” claims are a subset of health claims that resemble SF claims; Europe established a premarket approval process during which experts review the substantiation evidence. As of January 2018, two hundred sixty-one claims have been approved from the 2319 submitted (60). In 2015, Japan established a voluntary notification system for a new category of claims, foods with functional claims, to expand information about foods for consumers. Claims supported by a clinical study or systematic review but not requiring a clinical study on the product itself as required by Foods for Specified Health Uses (FOSHU) are posted on a government website (61). These approaches to assure substantiation of SF claims for the general food supply are quite recent. The value these regulatory structures bring to substantiation of SF claims for general foods could inform the FDA's efforts to establish expectations for substantiation of SF claims broadly, including the special case of infant formula.

The Codex Standard for infant formula allows that “other ingredients may be added in order to provide substances ordinarily found in human milk and to ensure that the formulation is suitable as the sole source of nutrition for the infant or to provide other benefits [emphasis added] that are similar to outcomes of populations of breastfed babies” and optionally added ingredients must be added in “sufficient amounts… to achieve the intended effect, taking into account levels in human milk” (62). It is ironic that the Codex allows substances to be added to formula when the benefits are like those observed among breastfed infants, whereas the DG's proposed standard of evidence for SF claims makes no reference to breastfed infants, only RCTs among formula-fed infants.

Conclusion

The origin of SF claims was based in scientific discovery, learning about substances in foods that affect the function of the body and potentially benefit health. As in 1938, there is abundant research of this type ongoing today; the 2017 issues of the American Journal of Clinical Nutrition contain published articles on functions of oligosaccharides, fiber types, isoflavones, choline metabolites, alpha linolenic acid, catechins, flavanols, and branched-chain fatty acids, among other substances. SF claims are an important way to educate consumers about emerging science and stimulate further research. Questions of substantiation of claims raised by the DG and the viewpoint by Hughes et al. should be discussed more generally for conventional foods, while recognizing the unique status of SF claims afforded by statute, explicitly (FD&C Act, DSHEA) or by silence (IFA, NLEA), and should align closely with the FTC standard. Consistent standards can ensure that SF claims across food types are truthful and not misleading, and protect consumers’ access to new scientific learnings about the functions of substances in foods.

Notes

Author disclosure: JCW consults on regulatory matters to companies that manufacture foods for special dietary uses, including infant formula, and companies that supply ingredients to manufacturers of infant formula.

Abbreviations:

ARA

arachidonic acid

CFR

Code of Federal Regulations

DG

Substantiation for Structure/Function Claims Made in Infant Formula Labels and Labeling: Guidance for Industry. Draft Guidance

DHA

docosahexaenoic acid

DSHEA

Dietary Supplement Health and Education Act

FD&C Act

Food Drug & Cosmetic Act

FTC

Federal Trade Commission

IFA

Infant Formula Act

LCPUFA

long-chain polyunsaturated fatty acids

NLEA

Nutrition Labeling and Education Act

RCT

randomized controlled trial

SF

structure-function