In this review, we explain how the US Food and Drug Administration (FDA) used its evidence-based review system to evaluate the scientific evidence for a qualified health claim for 100% whey-protein partially hydrolyzed infant formula (W-PHF) and reduced risk of atopic dermatitis (AD). The labeling of health claims, including qualified health claims, on conventional foods and dietary supplements require premarket approval by the FDA. Health claims characterize the relationship between a substance (food or food component) and disease (eg, cancer or cardiovascular disease) or health-related condition (eg, hypertension). To determine whether sufficient evidence exists to support the qualified health claim, the FDA evaluated human intervention studies that evaluated the role of W-PHF in reducing the risk of AD. The FDA concluded there is little to very little evidence, respectively, to support a qualified health claim concerning the relationship between intake of W-PHF and a reduced risk of AD in partially breastfed and exclusively formula-fed infants throughout the first year after birth and up to 3 years of age. In addition, the FDA required a warning statement be displayed along with the health claim to indicate to consumers that partially hydrolyzed infant formulas are not hypoallergenic and should not be fed to infants who are allergic to milk or to infants with existing milk allergy symptoms.
- whey-protein partially hydrolyzed infant formula
- atopic dermatitis
- evidence-based review
- qualified health claim
- AD —
- atopic dermatitis
- CI —
- confidence interval
- FDA —
- US Food and Drug Administration
- SSA —
- significant scientific agreement
- W-PHF —
- 100% whey-protein partially hydrolyzed infant formula
Atopic Dermatitis and 100% Whey-Protein Partially Hydrolyzed Infant Formula
Atopic dermatitis (AD), a form of eczema that generally begins in early infancy, is a complex disease, both multifactorial and heterogeneous with regard to etiology and aggravating factors.1 The distribution of the rash varies with age and involves the cheeks and outer surfaces of the arms and legs in infancy, the inner surfaces of the arms and legs in the young child, and inner surfaces of the arms and legs, hands, and feet in teenagers and young adults.2 In severe cases, scratching the skin can cause redness, swelling, cracking, scaling, and occasionally oozing clear fluid and crusting.2 Although AD is identified by a constellation of symptoms by using standard diagnostic criteria,3 it is difficult to measure the incidence and prevalence of AD because many people self-report symptoms but have not received a diagnosis from a doctor. The National Institute of Arthritis and Musculoskeletal and Skin Diseases reports an estimated 10% to 20% of infants and young children experience symptoms of the disease.4
The pathophysiology of AD involves immunoglobulin E-mediated hypersensitivity reactions, both Th1 and Th2 lymphocyte cell-mediated processes, as well as mutations or decreased function of key structural proteins in the skin involved in moisture retention (eg, filaggrins).5 The multifactorial etiology of AD includes predisposing genetic factors, dietary influences, bacterial skin colonization, and environmental exposures from irritants such as secondhand smoke and house dust mites, as well as mediated allergens such as pollen.6 The early introduction of solid foods in infants younger than 6 months of age and the consumption of cow’s milk formula versus breastfeeding are cited as dietary risk factors.2
It has been hypothesized that hydrolyzing cow milk protein into smaller peptides will reduce its allergenicity, and thereby reduce the risk of AD in infants fed formula milk. In this review, we explain how the US Food and Drug Administration (FDA) used its evidence-based review system to evaluate the scientific evidence for a qualified health claim petition for 100% whey-protein partially hydrolyzed infant formula (W-PHF) and reduced risk of AD.7
The Nutrition Labeling and Education Act of 19908 mandated that the FDA authorize health claims that meet the significant scientific agreement (SSA) standard on the labeling of conventional foods and dietary supplements. Health claims are statements that describe the relationship between a substance (food or food component) and a disease (eg, cardiovascular disease) or health-related condition. A health-related condition is a condition (eg, hypertension) that is essentially indistinguishable from a disease (eg, coronary heart disease), is a surrogate marker for risk of a specific disease (eg, serum cholesterol concentrations for coronary heart disease), or both. The labeling of conventional foods and dietary supplements with health claims require premarket approval by the FDA.
Health claims were first authorized by Congress under the SSA standard. The SSA standard is a rigorous standard that requires a high level of confidence in the validity of a substance–disease relationship.9 Court decisions dealing with health claims for dietary supplements raised First Amendment issues that resulted in the establishment of qualified health claims for the labeling of conventional foods and dietary supplements.10 When credible evidence falls short of the SSA standard, health claims with qualifying language about the level of scientific evidence (ie, qualified health claims) are issued through letters of enforcement discretion.10 When a letter of enforcement discretion has been issued, the FDA does not object to the use of the claim specified in the letter provided that the products that bear the claim are consistent with the stated criteria. SSA and qualified health claims pertain to disease risk reduction in the US population or a target subgroup (eg, women or elderly) that does not have the disease that is the subject of the claim.
The FDA’s Evidence-Based Review of Health Claims
The FDA conducts a thorough review of the scientific evidence supporting a health claim when evaluating authorized SSA or qualified health claims. In 2009, the FDA published final guidance on the evidence-based review system for the scientific review of health claims.9 The evidence-based system is applied during the evaluation of the scientific evidence for a given substance–disease relationship in a health claim. The FDA reviews all evidence related to the health claim (eg, supportive and not supportive) that the petitioner is required to provide in support of its claim. Through a literature search, the agency identifies any additional studies that are considered to be relevant to the health claim petition. The FDA focuses its review on reports of human intervention and observational studies because scientific conclusions about the substance–disease relationship in humans can only be drawn from such studies.9 In addition to individual reports of human studies, the agency also considers other types of data and information in its review such as meta-analyses, review articles, and animal and in vitro studies. These other types of data and information may be useful to assist the agency in understanding the scientific issues about the substance, the disease or health-related condition, or both, but cannot by themselves support a health claim relationship.
Health claims involve reducing the risk of a disease in people who do not already have the disease that is the subject of the claim. Thus, the FDA considers evidence from studies in individuals diagnosed with the disease stated in the health claim only if it is scientifically appropriate to extrapolate to individuals who do not have the disease. The FDA evaluates the individual reports of human studies to determine whether any scientific conclusions can be drawn from each study. Studies that lack critical criteria, such as a control group or statistical analysis, are eliminated from further review because scientific conclusions about the health claim relationship cannot be drawn from them.11,12
The FDA rates the remaining human intervention and observational studies for methodological quality. This quality rating is based on several criteria related to study design (eg, use of a placebo control versus a nonplacebo controlled group), data collection (eg, type of dietary assessment method), the quality of the statistical analysis, the type of outcome measured (eg, disease incidence versus validated surrogate end point), and study population characteristics other than relevance to the US population (eg, selection bias and whether important information about the study subjects [eg, age, smoker versus nonsmoker] was gathered and reported).9
Finally, the FDA evaluates the results of the remaining studies and the strength of the total body of publicly available evidence. The agency conducts this rating evaluation by considering the study type (eg, intervention, prospective cohort, case control, cross-sectional), the methodological quality rating previously assigned, the quantity of evidence (number of the various types of studies and sample sizes), whether the body of scientific evidence supports a health claim relationship for the US population or target subgroup, whether study results supporting the proposed claim have been replicated,13 and the overall consistency14 of the total body of evidence. Based on the totality of the scientific evidence, the FDA determines whether such evidence is credible to support the substance–disease relationship for an authorized SSA health claim or a qualified health claim. If there is any credible evidence, the agency determines the qualifying language that reflects the level of scientific evidence to support the relationship.
Safety Review of the Substance
W-PHFs have been safely and lawfully marketed in the United States for decades. However, W-PHFs are not considered hypoallergenic and may cause allergic reactions in one-third to one-half of milk allergic infants,15–17 and thus are not safe for all populations. Hypersensitivity reactions in milk allergic infants represent a significant medical concern because they may range from cutaneous (eg, urticaria and worsening eczema) to severe gastrointestinal reactions (eg, food protein induced enterocolitis syndrome and eosinophilic gastrointestinal disease) or life-threatening anaphylaxis. Although rare, death from cow’s milk anaphylaxis has been reported in voluntary registries of fatal food anaphylaxis cases.18–20 Therefore, W-PHF should not be fed to infants who are known to be allergic to milk or who have existing milk allergy symptoms.
End Point for AD
The FDA identified the following end point to use in identifying a reduced risk of AD for the purposes of a health claim: incident cases of AD. The FDA identified no validated surrogate end points to use in assessing AD risk reduction.4 Diagnosis of AD is based on a combination of historic and morphologic findings because there are no single distinguishing features of AD.1 As there is no objective laboratory biomarker for this disease, the Hanifin-Rajka criteria remain a standard for the diagnosis of AD and are frequently used in randomized controlled clinical trials.3 Physical examination by using the Hanifin-Rajka criteria must include 3 or more basic findings: pruritus, typical morphology, and distribution; facial and extensor involvement in infants and children; chronic or chronically relapsing dermatitis; and/or personal or family history of atopy (asthma, allergic rhinitis, AD).3
Evaluation of Intervention Studies
The petition requested a qualified health claim for the relationship between the consumption of W-PHF and a reduced risk of AD. Specifically, the requested qualified health claim identified (1) the target population as healthy infants who are not exclusively breastfed and have a family history of allergy and (2) the duration of the reduced risk of disease as throughout the first year after birth and up to 3 years of age.
The FDA evaluated 20 reports of intervention studies addressing the relationship between the consumption of W-PHF and a reduced risk of AD. Scientific conclusions could not be drawn from 16 of these 20 reports for the reasons discussed below.
Three studies21–23 were a republication of another study being evaluated24–27 for the substance and disease relationship. Because these republications provided no new data or information pertinent to the qualified health claim, the original publications were relied upon for review.
Two studies did not select infants who were healthy and had a family history of allergy: the population identified by the qualified health claim.24,31 Because the petition focused on infants at risk for developing AD, these publications were not relied upon for review.
Six studies24,25,31–34 did not definitively diagnose incident cases of AD in the study’s subjects based on a health history and physical examination.3 These studies did not provide the criteria used for evaluating AD, thus it was not possible to determine whether results reported were cases of AD. Extensive research has been performed, but interpretation of the literature is complicated by the lack of standardization with regard to diagnosis, measures of severity, and the lack of an objective test to measure the activity of AD.1 Because these studies did not definitively diagnose AD, no scientific conclusions could be drawn from them concerning the incidence of AD in the subjects.
One study by Chirico et al35 did not specify that W-PHF, the substance of the claim, was used in the study nor did supplemental information in the petition confirm that the formula used was the substance of the claim. Partially hydrolyzed infant formulas made from cow’s milk may contain casein in addition to whey-protein. Because the formula was not specifically described as W-PHF, the FDA could not determine that the formula used in the study was the substance of the claim.7 Thus, scientific conclusions could not be drawn from this study about the relationship between W-PHF and reduced risk of AD.
One study by De Seta et al36 was a randomized study not designed with AD as a primary outcome. Infants at risk for atopy were randomly assigned to W-PHF (n = 23) or conventional cow’s milk formula (n = 39) for 6 months. Information on blinding of the study, compliance with formula consumption, and weaning recommendations were not reported. Therefore, the FDA could not determine if infants consumed their assigned formulas and adhered to the study protocol. In addition, no information was provided about factors that could influence AD (eg, house pets, dust mites, smoking in the home, weaning foods). Known modifiers of disease risk need to be collected and adjusted to minimize bias so that the substance–disease relationship is accurately measured.9 Furthermore, results for cumulative incidence of AD and cow’s milk protein allergy intolerance were combined and reported together (eg, at 24 months, there were 3 cases of AD and cow’s milk protein allergy intolerance combined in the W-PHF group). Thus, the FDA could not determine the independent role of the substance in reducing the risk of disease.9 Statistical analysis between the W-PHF and cow’s milk formula groups was also not reported by the authors. Due to the shortcomings described above, this study was so deficient in methodological quality that it was considered to be of low-quality design. Based on the above reasons, scientific conclusions could not be drawn from this study about the relationship between W-PHF and reduced risk of AD.
Four intervention studies26,37–39 published in 6 reports (2 studies27,40 have follow-up analysis to the initial study report) were available from which scientific conclusions could be drawn about the relationship between the consumption of W-PHF and a reduced risk of AD for which the petition requested a qualified health claim. These studies are discussed below.
Vandenplas et al37,40 was a randomized, partially blinded 6-month intervention study of moderate methodological quality. A total of 58 Belgian infants with family history of atopy (2 first-degree relatives) were randomly assigned to receive either W-PHF (n = 28) or cow’s milk infant formula (n = 30) (control) exclusively for 6 months. The incidence of AD, referred to as eczema by the authors, was determined by physical examinations by using the Hanifin-Rajka criteria3 conducted every 6 months from the age of 6 months to 5 years. Direct statistical comparisons of the incidence of eczema between the 2 groups were not reported by the authors. However, an analysis of the Vandenplas et al40 data by Szajewksa and Horvath41 revealed no significant difference between the 2 groups when relative risk and 95% confidence interval (CI) were calculated from the cumulative incidence data. At 0 to 12 months of age, the relative risk was 0.46 and CI was 0.13 to 1.60. At 0 to 36 months of age, the relative risk was 1.07 and the CI was 0.43 to 2.67.
Chan et al38 conducted a randomized, single-blind 4-month intervention study of moderate methodological quality with 110 Singaporean infants with a family history (first-degree relative) of atopy. The exclusively formula fed infants received either W-PHF (n = 53) or cow’s milk based formula (n = 57) (control). Infants were monitored by a physician for AD at 3, 6, 12, 18, 24, and 30 months. Compliance with the intervention was not addressed, and there were no restrictions with regard to consumption of weaning foods. Factors that could exacerbate AD such as socioeconomic status, house pets, and use of air-conditioning were considered; however, other factors, such as timing and types of weaning foods consumed, were not addressed. There was a significantly lower incidence of AD for the treatment group compared with the control group from 3 months of age (odds ratio = 0.20; χ2P = .011) up to 2 years of age (odds ratio = 0.37; χ2P = .019).
von Berg et al26,27 was a high quality, randomized, double-blind 4-month intervention trial that compared the effect of differently hydrolyzed infant formulas with cow’s milk formula on allergic diseases including AD in infants with family history of atopy. Mothers were encouraged to exclusively breastfeed for at least 4 months. If mothers chose not to breastfeed exclusively or not at all, infants were randomly assigned to W-PHF (n = 241) or cow’s milk formula (control) (n = 256). Information on the quantity of breast milk fed and duration of breastfeeding was not reported. The intervention consisted of study formula for the first 4 months of age and recommendations on introduction and type of weaning foods after 4 months. AD was determined by clinical examinations conducted at 1, 4, 8, 12, and 36 months of age. The incidence of AD from birth to 1 year of age was 22 in the W-PHF group and 38 in the cow’s milk formula group. There was a significantly lower incidence of AD when infants consumed the W-PHF compared with the control group (odds ratio = 0.56; CI = 0.32–0.99) when adjusted for AD in family history, gender, and maternal smoking after birth. At follow-up at 3 years of age, the incidence of AD from birth to 3 years of age was 34 in the W-PHF group (n = 229) and 55 in the cow’s milk formula group (n = 245).27 The significant effect of W-PHF on reduced risk of AD persisted at 3 years of age (odds ratio = 0.60; CI = 0.37–0.97 when adjusted for AD in family history, gender, and maternal smoking after birth).27
Marini et al39 conducted a 5- to 6-month intervention study of moderate methodological quality with 279 Italian infants with parental history of atopy. The intervention included randomization to formula, instruction on maternal diet and weaning foods, as well as environmental advice. Exclusively formula-fed infants received W-PHF (n = 48) or conventional cow’s milk formula (n = 47) (control group). Data were reported for infants who received breast milk in addition to W-PHF (n = 32) and breast milk in addition to cow’s milk formula (n = 28). AD was determined by clinical examinations conducted at 3, 6, 12, 24, and 36 months of age. Compliance with maternal and infant diet recommendations and environmental advice was collected. Statistical comparison of cumulative incidence of AD for the treatment and control groups was not reported by the authors. However, an analysis of the incidence data from 0 to 12 and 0 to 36 months39 by Alexander and Cabana42 revealed no significant difference between the treatment group and control group at 1 year of age (relative risk = 0.48; CI = 0.13–1.78) and at 3 years of age (relative risk = 0.42; CI = 0.14–1.26).
Strength of the Scientific Evidence
Throughout the First Year After Birth
Of the 4 studies previously discussed, only 2 studies26,38 supported a beneficial relationship between the consumption of W-PHF during the first 4 months after birth and reduced risk of AD throughout the first year after birth (Table 1). One was a large high quality intervention study (n = 241 in the treatment group and n = 256 in the control group),26 whereas the other was a smaller moderate quality intervention study (n = 53 in the treatment group and n = 57 in the control group).38
Two studies37,39 did not support a beneficial relationship between the consumption of W-PHF during the first 4 months after birth and risk of AD throughout the first year after birth. Both were smaller moderate quality intervention studies (n = 48 and 28 in the treatment group and n = 47 and 30 in the control group, respectively). Because there are only 2 intervention studies that support the substance–disease relationship throughout the first year after birth, whereas 2 other intervention studies do not support that relationship, there is little evidence from which to conclude that a risk reduction relationship actually exists during this time frame.
Up to 3 Years of Age
Only 1 study revealed a beneficial relationship when feeding a W-PHF for the first 4 months after birth and reduced risk of AD up to 3 years of age27 (Table 1). Although the study was a large high quality intervention study (n = 229 in the treatment group and n = 245 in the control group), these findings at 3 years of age have not been replicated, and replication of scientific findings is important to substantiate results.9
Two studies39,40 did not support a beneficial relationship of feeding a W-PHF for the first 4 months after birth and reduced risk of AD up to 3 years of age. Both were smaller moderate quality intervention studies (n = 48 and 28 in the treatment group and n = 47 and 30 in the control group, respectively). Because there is only 1 intervention study that supports the substance–disease relationship throughout the first year after birth and up to 3 years of age, whereas 2 intervention studies do not support that relationship, there is very little evidence from which to conclude that a risk reduction relationship actually exists during this time frame. The agency concluded that the relationship between feeding a W-PHF for the first 4 months after birth and a reduced risk of AD throughout the first year after birth and up to 3 years of age is uncertain.
The FDA concluded that there is little to very little credible evidence for a qualified health claim about W-PHF and a reduced risk of AD7 and that the qualified health claim must be appropriately worded so as to not mislead consumers. Because the relationship between W-PHFs and the reduced risk of AD is uncertain, the agency issued a letter of enforcement discretion for the use of 4 qualified health claims.7 An example of 1 of the claims is provided below:
“Very little scientific evidence suggests that, for healthy infants who are not exclusively breastfed and who have a family history of allergy, feeding a 100% Whey-Protein Partially Hydrolyzed infant formula from birth up to 4 months of age instead of a formula containing intact cow’s milk proteins may reduce the risk of developing atopic dermatitis throughout the 1st year of life and up to 3 years of age.”
Furthermore, the FDA required that the following warning statement be included when using the above qualified health claim.
“Partially hydrolyzed formulas should not be fed to infants who are allergic to milk or to infants with existing milk allergy symptoms. If you suspect your baby is already allergic to milk, or if your baby is on a special formula for the treatment of allergy, your baby’s care and feeding choices should be under a doctor’s supervision.”
The FDA concluded that the use of bold type as set forth above is necessary, in light of the significant public health risk that would be created by the feeding of these formulas to infants who are allergic to milk or to infants with existing milk allergy symptoms. Furthermore, the fact that the articulation of a relationship between the consumption of W-PHF and a reduced risk of developing the allergic disease of AD could mislead consumers to think that these formulas are an appropriate choice for such infants.
Since the FDA issued the letter of enforcement discretion on W-PHF and reduced risk of AD, a study of partially hydrolyzed whey infant formula and risk of allergic disease has been published.43 The study by Lowe et al43 revealed that infants fed W-PHF were not at lower risk of allergic manifestations, defined by the presence of eczema and food reactions, in the first 2 years after birth compared with conventional cow’s milk formula. Eczema was identified by parent reports of doctor diagnosed eczema or any rash that was treated by a topical steroid, and its incidence was evaluated as a secondary outcome. This study has not been evaluated by using the FDA’s evidence-based review system; however, based on preliminary appraisal, it does not seem to contradict the conclusions made by the FDA based on the credible evidence used to evaluate the qualified health claims for W-PHF and reduced risk of AD.
As scientific information and consumer consumption patterns are subject to change, the FDA intends to evaluate new information that becomes available to determine whether it necessitates a change in this decision. For example, scientific evidence may become available that will support a qualified health claim for 1 or more claims that were denied; or no longer support the use of the above qualified health claims; or may raise safety concerns about the substance that is the subject of the claims.
- Accepted April 10, 2012.
- Address correspondence to Carolyn S. Chung, PhD, Office of Nutrition, Labeling, and Dietary Supplements, Center for Food Safety and Applied Nutrition, US Food and Drug Administration, 5100 Paint Branch Parkway, College Park, MD 20740. E-mail:
Dr Chung conducted the scientific review and wrote the article. Drs Yamini and Trumbo edited the article.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
- ↵Leung D, Rhodes A, Geha R, et al. Atopic Dermatitis. In: Fitzpatrick T, Eisen A, Wolff K, Freedberg I, Austen F, eds. Dermatology in General Medicine. 4th ed. New York: McGraw-Hill, Inc; 1993:543–564
- Hanifin JM,
- Rajka G
- ↵National Institute of Arthritis and Musculoskeletal and Skin Diseases. National Institute of Arthritis and Musculoskeletal and Skin Diseases, Handout on Health: Atopic Dermatitis. 2009
- Akdis CA,
- Akdis M,
- Bieber T,
- et al.,
- European Academy of Allergology and Clinical Immunology/American Academy of Allergy, Asthma and Immunology
- ↵US Food and Drug Administration. 100% Whey-Protein Partially Hydrolyzed Infant Formula and Reduced Risk of Atopic Dermatitis. 2010. 12-15-0011. Available at: www.fda.gov/Food/LabelingNutrition/LabelClaims/QualifiedHealthClaims/ucm256731.htm. Accessed May 23, 2011
- ↵The Nutrition Labeling and Education Act of 1990. Public Law No. 101-535. 2011
- US Food and Drug Administration
- ↵US Food and Drug Administration. Consumer Health Information for Better Nutrition Initiative. Task Force Final Report. 7-10-2003. 2011. Available at: www.fda.gov/Food/LabelingNutrition/LabelClaims/QualifiedHealthClaims/QualifiedHealthClaimsPetitions/ucm096010.htm. Accessed May 1, 2011
- ↵Spilker B. Guide to Clinical Studies. 1st ed. New York, NY: Raven Press; 1991
- ↵Federal Judicial Center. Reference Manual on Scientific Evidence. 2nd ed. 2000. Available at: www.fda.gov/Food/LabelingNutrition/LabelClaims/QualifiedHealthClaims/QualifiedHealthClaimsPetitions/ucm096010.htm. Accessed May 1, 2011
- ↵Wilson EB. An Introduction to Scientific Research. 1st ed. New York, NY: Dover Publishing; 1992
- Exl BM,
- Deland U,
- Secretin MC,
- Preysch U,
- Wall M,
- Shmerling DH
- Vandenplas Y,
- Malfroot A,
- Dab I
- von Berg A,
- Filipiak-Pittroff B,
- Krämer U,
- et al.,
- GINIplus study group
- von Berg A,
- Koletzko S,
- Grübl A,
- et al.,
- German Infant Nutritional Intervention Study Group
- von Berg A,
- Koletzko S,
- Filipiak-Pittroff B,
- et al.,
- German Infant Nutritional Intervention Study Group
- Copyright © 2012 by the American Academy of Pediatrics