Early Peanut Introduction Awareness, Beliefs, and Practices Among Parents and Caregivers Free

A national survey was administered between January 21, 2021, and February 15, 2021, to a US general population-based sample of parents/caregivers of a child between the ages of 7 months and 3.5 years (infants born after the 2017 PPA Guidelines publication). Informed consent was obtained from all participants. The Northwestern University institutional review board approved all study activities.

The survey was developed by a multidisciplinary team of pediatricians, pediatric allergists, survey methodologists, and parents. The survey focused on peanut consumption and awareness of the PPA Guidelines and assessed knowledge, beliefs, and practices. Detailed information about the child, caregiver, family’s history of allergic disease, and demographic information was collected. The question assessing PPA Guidelines awareness was asked last to avoid biased responses to the other questions. All survey questions, including how PPA Guideline awareness and eczema diagnosis were determined, are included in the Supplemental Information.

As in previous national epidemiologic studies published by our group,^1,12  this study relied on a nationally representative household panel to support population-level inference and used a dual-sample methodology incorporating both probability- and nonprobability-based respondents.

Funded and operated by NORC at the University of Chicago, AmeriSpeak is a probability-based panel designed to be representative of the US household population. Randomly selected US households are sampled using area probability and address-based sampling, with a known, non-0 probability of selection from the NORC National Sample Frame. These sampled households are then contacted by US mail, telephone, and field interviewers (see Amerispeak in Supplemental Information).

After AmeriSpeak data collection (N = 637), study-specific base weights were derived from the final panel weight and the probability of selection from the panel under the study sample design using standardized methods employed in previous work.^1,12 

To obtain more precise estimates of our constructs of interest, we incorporated a nonprobability sample obtained from Lucid (N = 2425), via TrueNorth calibration, a hybrid calibration approach developed at NORC on the basis of small area estimation methods to explicitly account for potential bias associated with the nonprobability sample. The purpose of TrueNorth calibration is to adjust the weights for the nonprobability sample to bring weighted distributions of the nonprobability sample in line with the population distribution for characteristics correlated with the survey variables.

The instrument was pretested on pilot interviewees and then administered via Web and phone in both English and Spanish by trained personnel. For the final sample, the following standard quality control rules were employed: if respondents: (1) completed the interview in less than one-third the median duration, (2) skipped >50% of questions asked, and/or (3) provided gibberish answers to open-ended questions. These were flagged and omitted from the final sample: N = 3062 (see calibration and quality control process, Supplemental Information).

Weighted frequencies and proportions were calculated using the svyr package in R 4.1. Survey-weighted χ² statistics were calculated to test independence of key study variables across strata of interest (eg, eczema status, race/ethnicity). Covariate-adjusted weighted logistic regression models were used to facilitate statistical inference about specific study variables of interest after accounting for potential confounding variables. All hypothesis testing used a 2-sided P < .05 threshold to determine statistical significance.

Parent/caregiver self-reported race, ethnicity, and socioeconomic data are summarized in Table 1; the observed weighted distributions are representative of the US population. The age of the index child was <1 year for 12.3% of parents/caregivers, 1 to <2 years for 30% of parents/caregivers, and 2 to 3.5 years for 56% of parents/caregivers. Among biological parents with reported atopic conditions, environmental allergies were most common, followed by asthma. Regarding atopic conditions in the children, 11.3% indicated the index child had eczema, and environmental allergies were most common (18.6%).

Overall, 13.3% of parents/caregivers reported they were aware of the PPA Guidelines. Parents/caregivers who identified as being of white race, between 30 and 44 years of age, having a bachelor’s degree or higher, and/or reported an income >$60 000 were most likely to report guideline awareness (P < .001) (Table 1). Parents/caregivers who reported guideline awareness were more likely to have a personal history of atopy; specifically, mothers with FA (P = .004) or eczema (P < .001) and fathers with eczema (P = .01). Additionally, guideline-aware parents/caregivers were more likely to have an index child with an atopic condition (P < .05).

Overall, nearly half (47.7%) of US parents/caregivers believed that feeding peanuts early prevented PA, whereas 33.5% reported “Don’t know.” Additionally, 49.8% reported believing that feeding other foods, like milk and egg, in the first year decreased the risk of allergies to these foods (Table 2). These beliefs were more common among guideline-aware parents/caregivers: 76.5% reported believing peanut consumption prevented PA and 77% reported believing feeding other foods prevented FA (P < .001). Regarding when it was safe to feed peanuts 7.9% of all parents/caregivers reported before 7 months of age, 24.4% reported between 7 and 9 months, and 22.7% said between 10 and 12 months; parents/caregivers whose child had eczema had similar responses. Guideline-aware parents/caregivers felt that feeding peanuts was safe at an earlier age: 18.2% said before 6 months of age and 37.2% said between 7 and 9 months (P < .001).

Among all parents/caregivers surveyed, 47.1% reported worrying about FAs as they were preparing to feed an infant, with the greatest concern being developing a PA (35%). Guideline-aware parents/caregivers were more worried about FAs (51.6%; P < .001), but not PA specifically. Most parents/caregivers (96.4%) did not report difficulty with peanut feeding and 70% reported feeling informed or very informed about peanut introduction.

Overall, 57.8% of parents/caregivers reported their PCP discussed peanut introduction; however, 73.9% of this group reported this discussion occurred at ≥6 months of age. Among those who reported that their PCP discussed peanut introduction with them, 5.0% reported they were advised to give peanut products at age 4 months, 18.6% at 6 months, 15.9% between age 7 and 11 months, and 27% at age 1, whereas 10.8% were told to delay peanut products until age 2. There was no significant difference in the guidance provided to parents/caregivers of children with eczema versus without eczema (data not shown).

Parents/caregivers reported a wide spectrum of peanut-feeding practices: 17.2% first offered peanut-containing foods to their infant before 7 months of age, 11.8% between 7 and 8 months, and 29.9% between 9 and 12 months (Table 3). Most parents/caregivers (77.3%) reported typically feeding <2 teaspoons of peanut product per feeding during the first month of peanut introduction. Daily feeding was reported by 3.4% and a few times a week by 24.2%. Regarding birth year, 25% of parents/caregivers of children born after 2020 fed peanut-containing foods by age 6 months, compared with 20% born in 2019, 15% born in 2018, and 12% born in 2017 (P < .001). Guideline-aware parents/caregivers were more likely to give peanut-containing foods earlier, with 31% (vs 17.2% P < .001) feeding them before 7 months of age or younger; peanut containing foods were offered more frequently: 34.5% reporting several times a week and 8.4% reporting daily (P < .001). Cumulative probability of peanut introduction based on several of these factors is displayed in Fig 2.

The main reasons parents/caregivers reported for giving peanut to their infant after age 7 months was fear of reactions (32.5%) and belief that earlier introduction was not important (31.7%) (Table 3). The main reasons for feeding peanut products after age 12 months was belief that earlier introduction was not important (32%), fear of a reaction (29%), and PCP guidance (26%). Guideline-aware parents/caregivers were similarly concerned about their child having a reaction to peanut-containing foods (33.7%), but those who waited after 7 months more frequently reported it was because of doctor recommendation (36.5% vs 24.4%; P < .001).

Allergic reactions to peanuts were reported by 1.4% of all parents/caregivers during initial peanut introduction. In those infants with a reported reaction upon first ingestion, 12.9% reported the reaction occurred at <6 months of age, 37.6% at 7 to 11 months, and 49.5% at age >1 year. Reaction symptoms in infants first reacting before 12 months of age were primarily dermatologic (56.2% hives, 45.6% rash, 42.6% itching, 32.1% swelling, 28.8% flushing) and gastrointestinal (28.4% vomiting/heaving, 28.6% diarrhea, 27.7% stomach pain).

In a covariate-adjusted model, those whose PCPs recommended introduction by age 6 months were significantly more likely (odds ratio [OR] 16.09 [11.61–22.31]) to introduce peanuts by age 6 months and significantly less likely (OR 0.05 [0.02–0.10]) to delay introduction past age 1 year (Table 4). Those who reported believing “very much” that early introduction prevented PA were more likely to introduce peanuts by 6 months (OR 1.99 [1.45–2.73]) and more likely to introduce by 1 year (OR 3.42 [2.47–4.76]). Parent/caregiver history of FA also increased the likelihood of infant peanut introduction by age 6 months (OR 1.68 [1.12–2.52]) and 12 months (OR 1.33 [0.98–1.79]).

The PPA Guidelines represent a major shift in the approach to PA prevention; thus, understanding the current practices is fundamental for further promotion of early peanut introduction. Our study is the first nationally representative survey of parents/caregivers to evaluate awareness, beliefs, practices, and outcomes since their publication in 2017. We found that most parents/caregivers, including those of infants with eczema, reported being unaware of the PPA Guidelines. We also found clear differences in who reported awareness by sociodemographic factors, including caregiver-reported age, race, income, and education. Overall, less than one-fifth of US parents/caregivers are offering peanut-containing foods to their infants before 7 months of age but nearly half are offering them in the first year of life. Additionally, our study found that guidance from a PCP was a facilitator to early peanut introduction when the recommendations were in line with the PPA Guidelines. The most cited barrier was concern about the development of an allergic reaction.

After publication of the LEAP study findings, several other countries also published their own PA prevention guidelines,^13–15  with varying rates of adoption. For example, the EarlyNuts study showed that the rate of peanut introduction in an Australian population increased from 28.4% from 2007 to 2011 to 88.6% in 2018.¹⁶  This robust response is likely multifactorial, including a national campaign that encouraged regularly feeding peanut products to all infants starting at age 6 months and high access to pediatric care in Australia. Conversely, because the LEAP study only included high-risk infants, the PPA Guidelines require infants be stratified into high-risk, mild-risk, and normal-risk categories, with different instructions around peanut introduction for each group (Fig 1). This is a more complicated public health message and involves several steps, including obtaining allergy testing if an infant has severe eczema. Of note, a 2020 consensus statement from US and Canadian allergy societies recommends introduction of peanut and egg to all infants between 4 and 6 months of age without required testing.¹⁷  This guidance still relies on the pediatrician to provide counseling about early peanut introduction, but removes the need for eczema categorization and only recommends that testing be obtained when preferred by families.

Our study found that awareness around the guidelines in the United States was higher among those who self-identified as white, and reported higher income and higher education level. These metrics reflect the disproportionate access to medical care and medical information in the United States. Furthermore, parents/caregivers with a PCP who provided guideline-adherent recommendations were significantly more likely to introduce peanuts in the first year of life. However, a previous study reported that PCPs with >50% Medicaid patients were less likely to implement the PPA Guidelines and more likely to report barriers to implementation, such as less access to subspecialists and less clinic time.¹¹  These systematic barriers further contribute to the disparities around early peanut introduction; to overcome them and increase US caregivers’ awareness regardless of their access to care, a public health campaign that partners with community organizations such as early childhood programs/day cares, religious centers, and Women, Infant, Child clinics is needed.

Educating and supporting PCPs who care for infants is another important pathway to further increase peanut introduction. Although our survey showed PCP guidance to be a strong facilitator for early peanut feeding, currently, parents/caregivers reported that a quarter of PCPs are still counseling to delay introduction past age 1. Previous studies support this finding: a National Institute of Allergy and Infectious Diseases-funded national survey of pediatricians found that only 29% were fully implementing PPA Guidelines, with 68% reporting that they need additional training on them and 73% requesting practice-based aids to assist implementation. Studies involving interviews with pediatricians in Northern California¹⁸  and surveys of pediatricians in New York¹⁹  found similar themes: lack of practice-based resources, need for more professional training, and more educational handouts for caregivers. It must be stressed that support for PCPs means more than just training. During 4- and 6-month clinic visit PCPS address infant growth, development, sleep, safety, and more. Simply asking them to discuss more at these visits is insufficient; handouts and tools designed to improve PCPs’ implementation of PPA Guidelines are needed.

Public health efforts around PA prevention will also need to provide more information about allergic reactions. Like other surveys of caregivers and prospective caregivers,^10,11  our study found that US caregivers worried about food allergies and that fear of adverse reactions was the most common reason given for delaying peanut introduction. However, parents/caregivers reported actual reactions to peanuts in 1.4% of infants upon first ingestion; notably, this occurred more often in older age groups. Our study also found that the types of reactions in young infants were mild, which is consistent with several other studies^20–22  that evaluated the severity of infant reactions to a range of foods. These findings highlight a key discrepancy between perceived risk and true risk of food reactions during early peanut introduction. Future educational campaigns around PA prevention will need to focus on this discrepancy by informing them of how often these reactions occur, what they look like, and how to respond.

It is important to note that our survey found that the concept of early introduction as a prevention tool seems to be gaining traction among US caregivers. Just 3.5 years after the guidelines were released, nearly half of US caregivers/parents are introducing infant-safe forms of peanuts. Furthermore, those with younger infants were more likely to introduce peanuts earlier, a promising indicator that acceptance of including peanuts in the infant diet has increased each year. Even among those who were not guideline-aware, we found there was a general belief that early introduction of peanuts and early introduction of other foods help in preventing FA. Additionally, those who did feed peanut-containing foods to their infants reported overall high comfort. Similarly, Lai and Sicherer (2019) conducted a survey of 100 caregivers and found that 90% reported comfort with early peanut feeding.¹⁹  These data suggest that, with additional information on how to incorporate peanuts in infant-safe foods,²³  early peanut introduction may become a routine practice for many US caregivers.

This study had some limitations. The data are based on caregiver recall, and findings such as allergic reactions, presence of eczema, or pediatric guidance were not validated with other sources, such as medical records. However, our reports closely mirror other studies regarding infant reactions and other surveys of pediatric providers. Furthermore, measuring PPA Guidelines adherence requires understanding risk for each infant. Evaluating presence or absence of eczema and categorizing severity is challenging on a caregiver questionnaire; thus, fully analyzing when peanuts should have been introduced (4 months, 6 months, etc) was not possible. However, we were able to define multiple time points in our survey and describe infant peanut practices in those who did and did not report eczema in their infants.

PPA Guidelines awareness and early peanut introduction practices are increasing among US caregivers. Data from our survey suggest that the development of a public health campaign that reaches different demographic cohorts, supports PCPs, and addresses concerns about reactions in infants would increase further uptake and ultimately prevent PA development.

FA

food allergy

LEAP

Learning Early about Peanut

OR

odds ratio

PA

peanut allergy

PCP

primary care provider

PPA

Prevention of Peanut Allergy