Secondary Logo

Journal Logo

Original Articles: Nutrition

Nutritional and Safety Concerns of Infant Feeding Trends

DiMaggio, Dina M.; Du, Nan; Porto, Anthony F.

Author Information
Journal of Pediatric Gastroenterology and Nutrition: May 2022 - Volume 74 - Issue 5 - p 668-673
doi: 10.1097/MPG.0000000000003401


What Is Known/What Is New

What Is Known

  • The American Academy of Pediatrics (AAP) has issued a warning about new infant feeding practices including informal breast milk sharing, use of imported European infant formula, toddler formulas in infants and homemade formulas.

What Is New

  • This is the first study to evaluate the prevalence of these new infant feeding practices and identify reasons why parents are choosing them.

Nutrition during the first 1000 days of life has been shown to play a critical role in a child's health and development (1). Although most professional societies, suchas the American Academy of Pediatrics (AAP) and North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN), recommend exclusive breastfeeding for the first six months of life (2,3), suitable alternatives include U.S. Food and Drug Administration (FDA) registered formula or donor breast milk (DBM) from established human milk banks. The FDA ensures that all infant formula sold in the United States meets strict nutritional and safety standards (4–6). The AAP recommends against the use of informal human milk sharing, with a risk of infectious contamination of the unpasteurized milk or contamination with medications, illicit drugs, or cow's milk (2,7). The AAP, also does not recommend alternative formula choices, such as non-FDA reviewed imported European infant formula (EIF), toddler formulas (TF) or homemade infant formula (HIF) for infants that do not meet the infant Formula Act (iFA) and associated regulations (8–10). These formulas present safety concerns and do not provide any benefits over FDA reviewed infant formulas (9).

Little is known on the US prevalence of these four infant feeding practices. There are reports of infant death, along with infants presenting with severe electrolyte disturbances, bacterial contamination, hypothyroidism, potential iron deficiency anemia (11), and inadequate nutrition obtained from these alternative feeding choices (12–15). The purpose of this study, therefore, is to better establish the prevalence of these infant feeding practices across the United States that pose nutritional concerns and gain insight into the parental reasoning for selecting those choices.


Study Design

An anonymous, cross-sectional, voluntary electronic survey was sent to all active prescribers to the Yumi listserv, a baby food subscription company. A link to the survey was emailed to a group of active customers in April 2021 and two subsequent reminder emails were sent two weeks apart in May 2021. Basic demographics and utilization of infant feeding practices were collected.

A unique 44-question survey was developed for the study. The total number of questions answered per participant was dependent on their specific responses. The survey was modeled on the National Health and Nutrition Examination Survey and prior research study assessing general infant formula use in the United States (16,17). The survey contained questions on demographics, including parental socioeconomic status, highest level of education obtained, ethnicity, infant's sex and gestational age and infant feeding practices. Participants answered questions with Yes/No, multiple choice, or open-ended answers. The content and phrasing of questions in the survey were developed in an iterative manner. The survey was modified based on input from three pediatric faculty educators and three pediatricians who ensured the content and construct validity of the survey items. Pilot testing of the survey for readability, clarity, and functionality was conducted on a small focus group of pediatricians and subscribers of the Yumi listserv. Feedback from the focus group was used to guide final survey revisions before distribution.


Frequency of each infant feeding practice was summarized as a percent of total users. For questions with missing data, we utilized the complete case analysis, where we only used the information that was answered by parents. Nonresponse rates ranged from 5% to 10% for each question. Chi-square and Fisher exact tests were used to examine associations between categorical variables. Significance was established at α = 0.05, and statistical analyses were performed using SAS (Cary, NC). Study was approved by the Yale University Institutional Board.


A total of 2429 of 11,903 possible responses were received (total response rate: 20.4%); 114 responses were removed due to completing <50% of the survey or if the age of the child was >3 years to reduce recall bias. Responses were collected from all 10 zip-code regions in the United States; 59% (1435) of the households had a salary <$150,000 and 81% (1739) had at least a bachelor's degree. 86% (1983) of the infants were born full-term and 10.0% had a cow's milk protein allergy (CMPA). The mean age of the infant at time of survey completion was 10.3 months old, with a median of 9 months (Table 1). Initially, 60% (1399) infants were exclusively breastfed on day of birth, which transitioned to 40% (928) exclusively breastfeeding by 1 year.

TABLE 1 - Demographics
Demographics Total survey participants No. (%)
Patients 2315
Male 1287/2315 (55.6%)
Age (mean (SD)) 10.3 (4.9)
Age (median, mo) 9
Preterm 322/2315 (13.9%)
 White 1486/2157 (68.9%)
 Asian 326/2157 (15.1%)
 Black 160/2157 (7.4%)
 Other 185/2157 (8.6%)
 Hispanic 302/2157 (14.0%)
Highest level of education
 <College graduate 576/2157 (19.4%)
 >Bachelors 1739/2157 (80.6%)
Household income
 $0–49,999 228/2157 (10.6%)
 $50,000–99,999 494/2157 (22.9%)
 $100,000–149,999 480/2157 (22.2%)
 $150,000–199,999 346/2157 (16.0%)
 >$200,000 609/2157 (28.2%)
Private insurance 1782/2157 (82.6%)
Cow's milk protein allergy 232/2157 (10.0%)
Baby first fed
 Breast milk 1399/2315 (60.4%)
 Formula 104/2315 (4.5%)
 Combination 812/2315 (35.1%)
First year of life feeding
 Breastmilk only 928/2315 (40.1%)
 Formula Only 279/2315 (12.1%)
 Combination of breastmilk + formula 1108/2315 (47.8%)
Prevalence of infant feeding trends
Donor milk user (n = 2036) 160/2036 (7.9%)
 From unregulated source 58/2036 (2.8%)
Toddler formula User (n = 1387) 71/1387 (5.1%)
Homemade formula (n = 1387) 31/1387 (2.2%)
European infant formula (n = 1387) 192/1387 (13.8%)
At least one infant feeding trend (n = 2315) 404/1387 (17.5%)

Eighteen percent of the families followed at least one of the four infant feeding practices; 21% of those utilizing at least one of the practices were premature. Respondents were more likely to have tried any trend if their infant was premature (P< 0.001), had a household income >$200,000 (p < 0.001), or a bachelor's education (p < 0.001) (Table 2).

TABLE 2 - Infant feeding practices characteristics
Tried at least one trend
No Yes P value
n = 1911 n = 404
Birth age ∗∗∗<0.001
 <37 wk 237 (12.4%) 85 (21.0%)
 >37 wk 1666 (87.2%) 317 (78.5%)
 I do not know 8 (0.4%) 2 (0.5%)
Sex 0.513
 Female 853 (44.6%) 176 (43.6%)
 Male 1051 (55.0%) 225 (55.7%)
 Prefer not to say 7 (0.4%) 3 (0.7%)
Race 0.118
 Asian/Pacific Islander 256 (14.4%) 70 (18.6%)
 Black 137 (7.7%) 23 (6.1%)
 Other 149 (8.4%) 36 (9.6%)
 White 1239 (69.5%) 247 (65.7%)
Hispanic 0.413
 No 1526 (85.7%) 329 (87.5%)
 Yes 255 (14.3%) 47 (12.5%)
Household income ∗∗∗<0.001
 <200,000 1311 (73.6%) 237 (63.0%)
 >200,000 470 (26.4%) 139 (37.0%)
Highest level of education ∗∗0.003
 With bachelors and higher 1415 (79.4%) 324 (86.2%)
 Without bachelors 366 (20.6%) 52 (13.8%)
 Private 1495 (83.9%) 322 (85.6%) 0.719
 Public 278 (15.6%) 53 (14.1%)
 None 8 (0.5%) 1 (0.3%)
Baby's first feed 0.367
 Breastmilk 1166 (61.0%) 233 (57.7%)
 Combination of formula + BM 658 (34.4%) 154 (38.1%)
 Formula 87 (4.6%) 17 (4.2%)
P value <0.05. ∗∗P value <0.01. ∗∗∗P value <0.001.

Donor Breast Milk

Eight percent (160/2036) of respondents reported using DBM. Sixty-nine percent of those using DBM acquired it from the hospital and 36% of parents obtained it from unregulated sources including from someone that they knew (30%), or over the Internet (6%). Some families would purchase DBM from both unregulated and regulated sources. Respondents were more likely to use DBM if infants were premature (p< 0.001).

European Infant Formulas

Approximately 14% (192/1387) of the respondents were using EIF; 24% of users of EIF reported that the infant had CMPA.

Forty percent (76/192) of the EIF users did not speak with their pediatrician on their use of this formula. Of the respondents that did speak with their pediatrician, 33% (39/116) felt that their physician was not or only slightly informed about EIF. Information about EIF was obtained from either a parent/friend recommendation (51%, 98/ 192), Internet source (40%, 77/192), or a medical professional (12.5%, 24/192). EIF was purchased predominantly from Internet stores/third party vendors (93.8%, 180/192). The most important reason for choosing EIF was the perception that these formulas contained “better” ingredients, had an organic option, less ”synthetic ingredients,” and that the European Union (EU) had stricter formula standards (89%, 73%, 65%, and 69%, respectively). Participants’ concerns about EIF use included: formula labels being in a different language (23%, 44/192), delay in notification of potential formula recalls (15%, 29/192), different measuring scoop size (9.4%, 18/192), and risk of overmixing/undermixing formula (7%, 13/192). Participants reported they would use an FDA reviewed infant formula if there were enhanced US regulation of the formula industry (66%, 127/192), and organic (52%, 100/192), and grass-fed options (40.6%, 78/192). Respondents were more likely to use EiF if infants had initially exclusively breastfed (p = 0.02), had higher household annual income >$200,000 (p < 0.001), a bachelor's education (p < 0.001), private insurance (p = 0.03) or of white race (p < 0.001).

Toddler Formulas

Five percent (71/1387) of the respondents using formula were using TF fortheirinfants <1 year of age. Approximately41% (30/71) did not discuss using TF for their infants with their pediatrician. Of those families that did, 10% felt their pediatrician was not informed or only slightly informed on the use of TF for their infant. The predominant source of information about TF was from “other Internet source” (21%, 15/71), parent/friend recommendation (23%, 16/71) or medical professionals (17%, 12/71). Typically, the respondents purchased the TF from an Internet store (55%, 39/ 71) or from an in-person store (41%, 29/71). The most important reason for choosing TF was the perception that these formulas contained “better” ingredients, was an organic option, and contained less ”synthetic ingredients” (55%, 42%, 36%, respectively). Respondents were more likely to use TF if infant was female (p < 0.001) and of “other” race (mixed race, Hispanic, and Native American) (p = 0.016).

Homemade Formulas

Two percent of the respondents using formula were making their own HIF. The respondents obtained the recipe from the Internet (35.5%, 11/31), a medical professional (19.4%, 6/31) or parent/friend recommendation (16.1%, 5/31). Approximately 58% (18/31) of the HIF users did not tell their pediatrician about using HIF. Respondents were more likely to use HIF if infants were of non-white race (p < 0.001),


At least 18% of parents that participated in our national survey were following at least one infant feeding practice that posed potential nutritional and safety concerns. Use of imported European formulas was the most used of these infant feeding practices (14%), followed by use of DBM (8%), toddler formula in infants (5%), and HIF (2%). Characteristics of those who were more likely to utilize one of these feeding practices was trend specific (Table 3).

TABLE 3 - Summary of non-AAP recommended infant feeding practices, prevalence, parent/infant characteristics & safety concerns
Infant feeding practice and associated prevalence and parental/ infant characteristics Safety concerns
Informal milk sharing • Risk of bacterial and viral contamination
 • 8% of respondents reported using DBM • Risk of contamination with environmental toxins
  o 36% obtained from unregulated source  o Cow's milk
 • More likely to use if premature  o Herbs
 o Illegal drugs
 o Medications
 o Mercury
 o Pesticides
Imported European infant formula (EIF) • Imported outside the normal “chain of control,” and bypasses the safety regulations of the infant formula act (IFA)
 • 14% of the respondents were using EIF   o May not be shipped or stored correctly
  o Purchased predominantly from third party vendors (94%)   o If an EIF recall occurs in the country of manufacture, it might not reach a US consumer in a timely manner
 • More likely to use if   o Labels that are not in English may lead to incorrect mixing of formula
  o exclusively breastfed  • EIF is available in stages with different nutrient levels
  o household income >$200,000  • EIF has different scoop sizes and mixing instructions (1scoop:1 ounce of water) than most US infant formulas (2 scoops:1 ounce of water)
  o at least a bachelor's education  • Imported hypoallergenic EIF contains only partially hydrolyzed protein which should not be used for the treatment of cow's milk protein allergy
  o private insurance and • Does not need to meet the nutrition and labeling requirements of the IFA
  o from white race  o Does not have to meet the same strict requirements to prevent infection or have annual FDA inspections
Toddler formula (TF)   o Does not need to show sufficient biological quality of proteins
 • 5% of respondents were using TF for their infants < 1 year of age   o Does not need to show that the formula supports normal physical growth in infants when fed as a sole source of nutrition
  o Purchased from Internet stores (55%) or in-person store  • Most toddlers do not need to use these formulas
(41%)  • Nutritional composition is not well defined
 • More likely to use if from other race (mixed race, native American, etc.)
Homemade infant formula (HIF) • Lacks essential nutrients for proper infant growth
 • 2% of the respondents using formula were making their own HIF  • Clinical reports of complications due to malnutrition
  o Obtained recipe from internet or friend (52%)   o Mineral deficiency including: vitamin D-deficient rickets,
 • More likely to use if from non-white race severe hypocalcemia, and hypothyroidism
  o Cardiorespiratory failure
  o Status epilepticus
• Risk of bacterial contamination when unpasteurized milk is used
• Risk of infection from unsterile storage
AAP = American Academy of Pediatrics.

Informal Breast Milk Sharing

There are several donor human milk banks with established safety standards in the United States that are either part of the Human Milk Banking Association of North America (HMBANA) or are from for-profit commercial human banks. Thirty-six percent of DBM users in our survey acquired the milk from unregulated sources. Current safety regulation is lacking over direct known “mother to mother” milk sharing and from unscreened donors over the Internet. Informal milk sharing from a single donor may lead to a higher risk of infectious contamination than pasteurized milk from pooled donors since the potential contaminate is not diluted out through the donor pool (7). Risks of bacterial and viral contamination occurs if DBM is not collected in a sterile manner, properly stored, or transported under appropriate conditions (18). Besides infectious risks of informal milk sharing, there is risk of contamination with environmental toxins such as pesticides, mercury, medications, herbs, illegal drugs and cow's milk (7,19).

Non-U.S. Food and Drug Administration Reviewed Imported European Infant Formula

Infant formulas commercially available in the United States must be registered with the FDA and meet the nutrition and safety requirements of the IFA (20). When non-FDA reviewed formulas are purchased from third-party sellers, they are imported outside the normal “chain of control,” and bypass the safety regulations of the IFA (21 U.S.C. 350a) and associated regulations (21 CFR 106 and 107) (4–6). The formulas may not be shipped or stored correctly (20). Incorrect storage and exposure to extreme temperatures may lead to degradation of certain nutrients. Also, if an EIF recall occurs in the country of manufacture, it may not reach a US consumer in a timely manner. EIF also come in stages, 0–6 months and 6-

12 months. If an infant is fed the wrong stage, a baby may receive an improper amount of nutrients for their age. Additionally, the powdered formula is mixed at different ratios than US infant formulas and scoops hold smaller weights than US formula scoops (20). Labels that are not in English may lead to parents incorrectly reading not only the formula's expiration date, but also instructions on how to properly prepare the formula. If parents cannot read the label correctly or incorrectly interchange EIF and US formula scoops and inadvertently mix the formula too dilute or too concentrated, electrolyte imbalances, seizures and poor weight gain in infants can occur.

Finally, hypoallergenic (HA) EIF that is being imported can contain either partially or extensively hydrolyzed protein as HA formulas in Europe are marketed to prevent and not treat allergy. In addition, NASPGHAN recommends that formula fed infants with CMPA be treated with an extensively hydrolyzed protein or amino acid protein formula (21). Nearly one in four families who were using EIF had an infant with CMPA. This difference in definition may lead to infants with CMPA consuming an HA EIF, an improper treatment of CMPA (20).

Use of Toddler Formula in Infants

TF in the United States is regulated as a food, but does not need to meet the requirements of the IFA. Therefore, TF does not need to demonstrate sufficient biological quality of proteins nor does it need to ensure normal physical growth of an infant. Manufacturers of TF are inspected every 3–5 years instead of annually by the FDA and are not required to test powder formula for contamination with microorganisms. In addition, TF do not meet the labeling requirements of the IFA. Hence, it is difficult to know if a specific TF contains all the essential nutrients needed for adequate growth of infants. TF are designed as an alternative to cow's milk or breast milk and marketed to improve the nutritional status in toddlers by adding nutrients thought to be low in the diet (22). Most toddlers, however, do not need these formulas. The nutritional composition of TF is quite variable. Due to this variability, these formulas may not meet the nutrition requirements for infants, including vitamins A and C, B vitamins and iron.

Use of Homemade Infant Formulas

The FDA, CDC, and the AAP recommend against HIF since it may lack essential nutrients for proper infant growth (9,23). The FDA has received reports of infants hospitalized and infant death from consuming HIF, and the CDC has issued a warning on infants fed homemade alkaline diet formula causing vitamin D-deficient rickets, severe hypocalcemia, cardiorespiratory failure, and hypo-thyroidism (12–14). Since many of these recipes call for unpasteurized cow or goat's milk and formulas may be prepared or stored in an unsterile manner, there is high risk of bacterial contamination.


Limitations of this study include potential recall, selection and non-response biases since it was a cross-sectional survey and not a prospective survey conducted through an online baby food delivery service. In addition, our participants had a high socioeconomic background which limits the generalizability of the results. Given the survey design, we were also not able to independently confirm the participants’ answers. Furthermore, the survey was performed during the coronavirus disease 2019 (COVID-19) pandemic which may have impacted some of the participants’ answers.

The AAP has clear guidelines on infant nutrition, recommending exclusive breastfeeding for at least 6 months or FDA reviewed infant formula or DBM from an established milk bank as alternatives. Yet, our study found that nearly one in five parents are using at least one non-AAP recommended infant feeding practice. Families were more likely to have tried a trend if they had yearly income >200,000, education level higher than a bachelor's degree or if their infant was premature. Parents may not report use of these feeding practices with their health care provider. Therefore, it is vital for pediatricians, pediatric gastroenterologists and registered dietitians to ask their patients how they are specifically feeding their infants and be aware of these feeding practices that may pose significant health risks. Breastfeeding parents should be advised on safe sources of breast milk and formula feeding mothers deserve more information on infant formulas. Health care professionals should provide education to parents on these popular trends and combat the wide array of harmful misinformation readily available on the Internet to protect the health of our infants. Furthermore, we need more formalized infant nutritional training for pediatric residency programs so health care professionals can feel prepared to address these feeding practice concerns.


1. Schwarzenberg SJ, Georgieff MK. Committee on N advocacy for improving nutrition in the first 1000 days to support childhood development and adult health. Pediatrics 2018; 141:e20173716.
2. Breastfeeding SO. Breastfeeding and the use of human milk. Pediatrics 2012; 129:e827–e841.
3. Nutrition CO, Agostoni C, Braegger C, et al. Breast-feeding: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr 2009; 49:112–125.
4. Electronic Code of Federal Regulations. Infant formula requirements pertaining to current good manufacturing practice, quality control procedures, quality factors, records and reports and notifications. In: Food, Durg, Administration in Department of Health, Human Services, editor. 21. Washington; 2018.
5. Electronic Code of Federal Regulations. Labeling. In: Food, Durg, Administration in Department of Health, Human Services, editor. 21. Washington; 2018.
6. Electronic Code of Federal Regulations. Nutrient specifications. In: Food, Durg, Administration in Department of Health, Human Services, editor. 21. Washington; 2015.
7. Nutrition CO, Breastfeeding SO, Fetus CO, et al. Donor human milk for the high-risk infant: preparation, safety, and usage options in the United States. Pediatrics 2017; 139:e20163440.
8. Fuchs G. Imported infant formula not reviewed by FDA may pose health risks. AAP News. May 11, 2018 ed: AAP; 2018.
9. Abrams SA. Is homemade baby formula safe? from the American Academy of Pediatrics. Accessed at Published March 5, 2021. Accessed February 10, 2022.
10. Lott M, Callahan E, Welker Duffy E, et al. Healthy beverage consumption in early childhood: recommendations from key national health and nutrition organizations. Consensus Statement. Available online: Published September, 2021. Accessed February 10, 2022.
11. Food, Drug Administration Able Groupe Recalling Products labeled as infant formula. Formulas have insufficient iron levels as per requirements for infant formula in the U.S., and products do not meet other FDA requirements. US Food and Drug Administration Alerts, Advisories & Safety Information. Available online: Published August 8, 2021. Accessed February 10, 2022.
12. Calello DP, Jefri M, Yu M, et al. Notes from the field: vitamin D-deficient rickets and severe hypocalcemia in infants fed homemade alkaline diet formula — three states, August 2020-February 2021. MMWR Morb Mortal Wkly Rep 2021; 70:1124–1125.
13. Food, Drug Administration. FDA Advises Parents and Caregivers to Not Make or Feed Homemade Infant Formula to Infants. US Food and Drug Administration Alerts, Advisories & Safety Information. Available online: Published February 24, 2021. Accessed February 10, 2022.
14. Vieira MA, Kube PK, van Helmond JL, et al. Recipe for disaster: homemade formula leading to severe complications in 2 infants. Pediatrics 2021; 148:e2021050947.
15. Jones K. A nutritional and marketing analysis of milk-based toddler drinks on the market in the United States. Chapel Hill: University of North Carolina; 2019.
16. Johnson CL, Paulose-Ram R, Ogden CL, et al. National health and nutrition examination survey: analytic guidelines, 1999–2010. Vital Health Stat 2013; 2:1–24.
17. DoHaH Services. Breastfeeding Among US Children Born 2009–2015. Atlanta, Georgia: CDC National Immunization Survey; 2018.
18. Keim SA, Hogan JS, McNamara KA, et al. Microbial contamination of human milkpurchased via the Internet. Pediatrics 2013; 132:e1227–e1235.
19. Keim SA, Kulkarni MM, McNamara K, et al. Cow's milk contamination of human milk purchased via the Internet. Pediatrics 2015; 135:e1157–e1162.
20. DiMaggio DM, Du N, Scherer C, et al. Comparison of imported European and US infant formulas: labeling, nutrient and safety concerns. J Pediatr Gastroenterol Nutr 2019; 69:480–486.
21. Koletzko S, Niggemann B, Arato A, et al. Diagnostic approach and management of cow's-milk protein allergy in infants and children: ESPGHAN GI Committee practical guidelines. J Pediatr Gastroenterol Nutr 2012; 55:221–229.
22. Romo-Palafox MJ, Pomeranz JL, Harris JL. Infant formula and toddler milk marketing and caregiver's provision to young children. Matern Child Nutr 2020; 16:e12962.
23. Food and Drug Administration DoHaHS. Infant Formula: Safety Do's and Dont's. US Food and Drug Administration Alerts, Advisories & Safety Information 2021.

donor breast milk; European infant formula; homemade formulas; toddler formula

Copyright © 2022 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition