Summary¹

Complementary feeding refers to the introduction of foods other than human milk or formula to the infant’s diet that occurs once human milk or formula alone is insufficient to meet the nutritional needs of the infant. The World Health Organization and the American Academy of Pediatrics recommend that complementary feeding begins at about 6 months of age, and not before 4 months of age, with continued breastfeeding until 2 years of age or older. Around 6 months of age, the nutritional needs of the infant may no longer be met by human milk or formula alone, which is why introducing complementary foods at this time is recommended. Healthy complementary feeding behaviors for infants and young children under age 2 years are essential for age-appropriate growth and social, emotional, and cognitive development.

Existing guidance on what to feed infants and young children includes introducing nutrient-dense, developmentally appropriate foods, including foods rich in iron, zinc, and vitamin D, and avoiding foods high in added sugars and sodium. The existing research on how to feed infants and young children recognizes that complementary feeding ideally occurs through ongoing and reciprocal interactions between the caregiver and the developing child. Responsive feeding is an approach to feeding that is sensitive to the child’s hunger and fullness cues, as well as to the child’s emotional and developmental needs. A recent National Academies of Sciences, Engineering, and Medicine (the National Academies) consensus study report

___________________

¹ This summary does not include references. Citations for findings presented in the summary appear in the subsequent chapters of the report.

titled Feeding Infants and Children from Birth to 24 Months summarized existing recommendations on what and how to feed infants and young children.

A number of federal programs, such as the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and the Child and Adult Care Food Program (CACFP), as well as other public health initiatives support healthy complementary feeding practices for families with low income. Opportunities exist to initiate new interventions or expand existing efforts by identifying new and complementary interventions to improve nutrition during this time. The existing literature contains many studies that have examined interventions aimed at improving nutrition for infants and young children. However, many were conducted in controlled clinical or research settings, and their effectiveness in and scalability to practical, real-life settings have not been assessed. Gaps in the existing literature include:

1. What interventions can potentially impact infant nutrition and feeding behaviors and/or outcomes outside of controlled clinical and research settings;
2. What interventions can be implemented at a community or state level; and
3. If/how interventions could complement existing federal-level programs.

THE COMMITTEE’S TASK

The Centers for Disease Control and Prevention (CDC) requested that the National Academies conduct a scoping review to identify promising complementary feeding interventions. The National Academies’ Health and Medicine Division convened the Committee on Complementary Feeding Interventions for Infants and Young Children Under Age 2 whose members had expertise in epidemiology, public health nutrition, dietetics/community nutrition, infant and child feeding practices and nutritional requirements, federal food and nutrition programs (along with equity and access to these programs), early childhood education, communications, and systematic reviews (see Appendix A for committee biographies). The committee was asked to conduct a scoping review and assessment of the peer-reviewed published literature and other publicly available information on interventions aimed at improving infant and young child feeding behaviors. CDC requested that the review be limited to developed countries or U.S.-specific contexts² and interventions occurring in the following three settings:

___________________

² For purposes of the scoping review at the request of the sponsor to limit the review to U.S.-specific contexts, the committee limited the scoping review to interventions that occurred in high-income countries as classified by the World Bank.

1. health care systems that influence feeding (e.g., health care centers, federally qualified health care centers, doctors’ offices, health clinics),
2. early care and education (ECE) settings, and
3. university cooperative extension (CE) programs that include nutrition and feeding for young children.

The committee was asked to produce a report that summarizes the available evidence and provides information on possible interventions that could be scaled up or implemented at a community or state level. CDC requested that the report be limited to research addressing: (1) what to feed (e.g., avoiding foods and beverages with added sugars; offering a variety of foods, textures, and flavors; consuming nutrient-dense foods) and (2) how to feed (e.g., using hunger and satiation cues to guide feeding; repeated exposures to foods; other responsive feeding practices). CDC also requested that the committee map the existing interventions by setting and describe factors that may be needed to scale the interventions (e.g., financial or human resources, barriers and facilitators, measurable and standardized indicators), reach underserved populations, and complement federal-level programs. As requested by the sponsor, interventions aimed at influencing breastfeeding and timing of introduction of complementary foods are outside the scope of this report.

THE COMMITTEE’S APPROACH

The committee interpreted its task as a request to scope the literature and use expert judgment to identify promising interventions, or aspects of interventions, that positively impact infant and young child feeding behaviors and practices related to what and how to feed infants and young children between 6 and 24 months of age. The committee acknowledged that scoping reviews provide an overview of evidence and differ from systematic reviews, which involve in-depth analysis of the strength of evidence. The scoping review targeted interventions conducted in U.S.-specific contexts (defined by the committee as high-income settings) in health care settings, ECE settings, and CE settings. At the direction of the study sponsor, the committee also included interventions that complemented existing federal programs targeting at-risk children under 2 years of age and their caregivers that may influence infant and young child feeding behaviors (e.g., WIC and home visiting programs) and interventions conducted in other settings that otherwise met the criteria for inclusion.

The committee developed a literature search strategy, screening protocol, and predetermined extraction criteria. The committee described factors considered when reviewing the evidence, commented on the

strengths and weaknesses of each individual article, noted the potential for scalability of the intervention, and use its collective expert judgment when identifying potentially informative studies.

The committee identified a small number of interventions that are the most broadly informative for developing new initiatives. While none are without methodologic limitations or universally effective in improving all outcome measures, they provide valuable models that may serve as a framework for future scaled interventions. These designated “informative studies” were chosen based on consideration of three key characteristics:

1. The quality of methods, considering the design, quality of measurement, study size, and other features;
2. Evidence of effectiveness, namely evidence that at least some key outcomes were influenced by the intervention; and
3. The potential for scalability, including whether the program could be widely implemented in the United States, considering such features as acceptability to recipients, suitability for diverse populations, feasibility within existing U.S. systems, and financial cost.

The committee used its expert judgment to identify informative studies and did not quantify the extent to which each study met each characteristic. In addition, other studies offered “informative intervention elements,” defined as clear suggestions of features that should be considered for incorporation into any effective and scalable infant and young child feeding program. For each of the informative study and informative intervention elements, the committee provided a brief description of the study and the insights that the study contributed toward the committee’s conclusions. The committee also described factors needed to scale interventions to the community or state levels and noted the ability of the intervention to reach underserved populations,³ reduce inequities, and complement federal-level programs such as WIC and home visiting programs.

___________________

³ At the request of the sponsor, the committee was asked to note the ability of the intervention to reach higher risk populations. Throughout the report, the committee will use the term “underserved populations” to refer to populations that have been systematically denied a full opportunity to participate in aspects of economic, social, and civil life. Specifically, in the U.S. context, this may include families that are Asian, Black, Hispanic/Latino, Indigenous, and/or Native Hawaiian or Pacific Islander; have low incomes; are located in rural or urban areas with limited access to healthy foods; speak languages other than English in the home; or experience other systematic discrimination or disadvantage. When possible, specific language about the characteristics of study populations has been used in sections of the report describing study findings, and additional details about characteristics of study populations are available in Appendix E.

SCOPING REVIEW FINDINGS

The committee identified 83 publications from 58 studies that met the criteria for inclusion in the scoping review. The number of studies varied across settings, with 16 in health care settings (27 publications), 5 in ECE (5 publications), 1 study (1 publication) in CE systems, 2 in WIC settings (4 publications), 12 in home visiting programs (18 publications), and 23 studies (28 publications) in other settings.

A noteworthy feature of the literature is that only about one-third of the publications were from studies based in the United States (29 publications). Given that the scoping review was limited to higher-income countries, most of the studies occurring outside the United States took place in Europe (29 publications), with a sizable number in Australia (15 publications), and smaller numbers in New Zealand, Canada, Israel, and South Korea.

The studies used one or more of the following modalities:

1. Live (e.g., in-person sessions [group and individual])
2. Remote-live interactive (e.g., via telephone [human to human])
3. Remote-tech interactive (e.g., interactive apps or websites, two-way texting)
4. Remote-tech noninteractive (e.g., video, books, noninteractive websites, one-way texting)

Most studies used live (69 publications) and/or remote-tech noninteractive modalities (66 publications).

The scoping review focused on outcomes related to “what to feed” and “how to feed” in infants and young children. Tables S-1 and S-2 describe intervention targets related to, respectively, what to feed and how to feed infants and young children.

Health Care

The committee identified 16 studies (27 publications) occurring within the health care setting that met the inclusion criteria. Included studies were conducted at hospitals, clinics, or medical offices; by health care setting personnel conducting home visits; or in a structure considered to be a medical model. The 27 publications within the health care setting were from 10 countries. These studies were intended to take advantage of the family’s connection with their health care provider to alter caregiver knowledge and behavior, which subsequently alters the child’s eating behavior and improves nutritional intake. All 16 studies delivered the intervention through some form of counseling with anticipatory guidance (i.e., guidance to assist caregivers in understanding the expected growth and development of their children). The approaches to counseling varied

TABLE S-1 Number of Publications Targeting What to Feed by Outcome and Setting

among studies, with the most common intervention including individual counseling by health professionals (e.g., physicians, midwives, nurses, or dietitians) in the context of routine preventive obstetric or pediatric visits.

Key Findings

Overall, the heterogeneity of the intervention and assessment tools resulted in a variety of study outcomes and few consistent findings. Within the health care setting, counseling interventions generally led to parents reporting healthier diets and improved responsive feeding behaviors. One key finding is that the more intensive interventions (e.g., multi-week interventions involving multiple in-person education sessions) were not shown to be more effective than less-intensive interventions (e.g., participation in a Facebook group, week-long interventions).

Early Care and Education

The committee reviewed five studies (five publications) conducted in ECE settings that met the inclusion criteria. One of the five studies was conducted within the United States; the others were in the United Kingdom, Spain, and Belgium. All five studies involved child care centers or nurseries, ECE settings that typically involve multiple child care providers caring for multiple children in “classroom”-type settings. None

TABLE S-2 Number of Publications Targeting How to Feed by Outcome and Setting

occurred in family child care homes, where typically a single provider cares for a relatively small number of children. The mode of intervention varied between the five studies, four of which were education interventions, with three delivered in person and one via a website. The fifth was an in-person tasting experience for the infant.

Key Findings

Repeated exposure to novel vegetables while at child care can be effective in helping young children like to eat vegetables, at least over a relatively short period of time (e.g., several months) in this age group; however, the long-term impacts on dietary intakes—including outside of child care—are not known. Providing training to child care providers may influence what young children eat in child care, but the only study that had this intervention focus has limited utility because it focused primarily on breastfeeding and the timing of introduction of complementary foods, issues that are outside the scope of this report. Both educational interventions that aimed to reach parents through child care centers achieved positive outcomes.

University Cooperative Extension

The committee reviewed one study (one publication) that described an intervention done within the U.S. CE system. This study examined the effects of four nutrition education lessons followed by structured reinforcements and is unique in that the children remained in Early Head Start (a federally funded type of ECE), while parents took lessons in groups to increase feeding knowledge and self-efficacy. The parents also received structured reinforcement via home visits.

Key Findings

The intervention increased both parental knowledge and self-efficacy, although there was no impact on toddler self-regulation. Both parental knowledge and self-efficacy are potential mediators of feeding behaviors. Furthermore, the participants stated that they learned something new and changed their behavior based on the classes and home lessons.

Special Supplemental Nutrition Program for Women, Infants, and Children

The committee reviewed two studies (four publications) in the WIC setting that met the inclusion criteria, and all were conducted in the United States. The studies are important in that their primary focus was

on modalities that may be instrumental in supportive behavior change related to complementary feeding, namely interactive texting and the distribution of video content to be viewed in the home.

Key Findings

An interactive texting study demonstrated that messaging was a highly accepted method of receiving education among women with low incomes served by WIC (English and Spanish speakers). While most participants receiving the intervention reported that the text messages were useful and led them to make changes in the way they fed their infants, the impacts on measured behavior changes were mixed. The video intervention led to a greater increase in knowledge and behavior, which suggests that short, culturally relevant videos have the potential to affect both knowledge and behavior change among WIC participants.

Home Visiting

For the home visiting setting, the committee reviewed 12 studies (18 publications) that met the inclusion criteria. Six of the identified studies were conducted in the United States. Three of the U.S. studies were conducted in the context of home visiting models meeting the U.S. Department of Health and Human Services (HHS) criteria to be designated as evidence based. Outside of the United States, two studies were completed in New Zealand, and the remaining four studies were done in the Netherlands, France, the United Kingdom, and Australia. Most of the home visiting studies (10 studies) tested interventions in which nutrition-focused education, counseling, and/or skill-building and goal-setting activities and resources (e.g., recipes) were delivered to caregivers one on one in the home setting. Studies differed based on who delivered the education or counseling and the frequency and duration of the visits.

Key Findings

The home visiting interventions had consistent significant impact on increasing caregivers’ use of responsive feeding practices, especially related to reducing pressure on the child to eat.

Other

The committee reviewed 23 studies (28 publications) that met the inclusion criteria and focused on what or how to feed but did not take place

in the context of the settings mentioned above (i.e., health care, ECE, CE, WIC, or home visiting). These publications spanned 11 countries, with the most prevalent being within the United States (9 publications). Six were from Norway, three were from the United Kingdom, and the rest were from Australia, Canada, France, Germany, the Netherlands, New Zealand, Sweden, and South Korea. Fifteen studies investigated the impacts of early (e.g., when introducing complementary foods or through human milk) and repeated exposures to new foods, with the aim of increasing child intake of the novel foods. One study tested counter-marketing messages about unhealthy beverages with parents of young children. Seven studies tested more comprehensive interventions delivered to parents virtually.

Key Findings

Repeated exposure to vegetables when introducing solids may be effective in helping young children eat vegetables, at least over a relatively short time period (e.g., several months); however, the impacts on dietary intake diminish over longer periods of time. Findings from one study suggest that reinforcing feeding exposure with an educational component (e.g., reading children’s books about vegetables) may also improve intake among young children. A single viewing of a video on unhealthy beverages can change parent attitudes over the short term, but the degree of counter-marketing required for the change in attitudes to persist and translate into changes in parental behavior and child dietary intake is unknown. Findings from the more comprehensive interventions suggest that technology-based interventions, including the use of phone calls, texting, and websites, show promise in being potentially scalable and having positive impacts on what and how to feed young children, at least over the short term in the first year of life and with relatively well-educated mothers.

INFORMATIVE EXAMPLES

The committee identified three informative studies and six informative intervention elements based on its expert judgment on the quality of methods, evidence of effectiveness, and potential for scalability.

Informative Studies

Among the informative studies, one was conducted in the health care setting, and two involved home visiting (see Box S-1). One informative study was the INfant Feeding, Activity, and Nutrition Trial (INFANT), a

randomized controlled trial (RCT) conducted in health care settings in Australia. Noted for its rigorous design, INFANT assessed some implementation metrics during the efficacy trial, and showed benefits, at least through parent self-report. Based on the promising results from this initial efficacy work, a translational trial (scale-up) informed by estimates from the published efficacy work is being conducted. Another informative study was the Intervention Nurses Start Infants Growing on Healthy Trajectories (INSIGHT) RCT, which involved a home visiting model. With a strong theoretical framework and description of visit fidelity, this trial provides important insights for future implementation of home visiting models. The final informative study was the Family Spirit Nurture RCT, which had a rigorous design, strong theoretical framework, attention to

community co-design and cultural sensitivity, positive impact on what to feed and how to feed outcomes for a Navajo community, and the potential for scalability within the existing Family Spirit home visiting program network.

Informative Intervention Elements

Among the six studies with informative elements, or features, to be considered in the design of an effective and scalable program, two were in the ECE setting, one was in a health care setting, one in a WIC setting, one in home visiting, and one in an “other” setting (web-based) (see Table S-3).

TABLE S-3 Informative Intervention Element Studies and Takeaways

NOTE: See Chapter 5 for citations and further description of the informative intervention element studies.

CONSIDERATIONS FOR SCALING INFANT AND YOUNG CHILD FEEDING INTERVENTIONS

The committee identified several interventions that improved self-reported outcomes regarding what and how to feed infants and young children, but their scalability and generalizability to diverse populations is unknown. When designing an intervention that offers promise to be viable and effective for wide dissemination, considering the following factors are important for scaling and facilitating broader implementation in large or diverse populations (see Box S-2). Factors related to developing programs, monitoring progress, making adaptions, and obtaining funding that are important when scaling infant and young child feeding interventions were identified by the committee.

COMMITTEE CONCLUSIONS ON EFFECTIVE AND SCALABLE INTERVENTIONS

Building on Existing Infrastructure and Ecosystems

The committee concluded that the expansion and harmonization of existing supports for complementary feeding in the United States across settings would facilitate families with young children receiving consistent messages about complementary feeding and responsive feeding. Across health care, ECE, CE, WIC, and home visiting settings, there are distinct opportunities—and challenges—to harnessing existing programs and harmonizing strategies across these settings (see Box S-3).

Collection and Application of Standardized Indicators

The committee concludes that the development of consensus complementary feeding impact outcomes and measurement tools and the integra-

tion of these tools into the everyday operations of the settings discussed in the report will support systemwide changes to improve complementary feeding and infant and early childhood nutrition in the United States. At least some of these outcomes and tools should be objective measures of effectiveness. Improved interagency collaboration within and across the settings described in this report and across states to capture, share, and report the same key impact outcomes would allow for more effective evaluation and improve U.S. complementary feeding efforts. Improved harmonization of process outcomes documenting service implementation could also strengthen implementation and improvement efforts. Efforts to develop innovative data strategies to facilitate data sharing across settings and minimize administrative burden would be well placed. Monitoring and evaluation efforts that include identifying disparities in access to services, program impact, and outcomes measures by race/ethnicity, parental education level, and socioeconomic status could be used to drive program refinements that improve health equity.

Reaching Underserved Populations

The committee concludes that the successful implementation of complementary feeding interventions in underserved populations will require partnership, collaboration, and community engagement with the target populations throughout the research, implementation, and scale-up processes. The most effective interventions will flexibly adapt to the needs and input of the target communities, while adhering to the evidence base.

Additional Considerations

Food Insecurity

The committee concludes that for the anticipatory guidance (i.e., information provided with the intent to prevent poor health outcomes), nutrition education, and brief targeted feeding interventions reviewed in this report to be most effective, they should also assess and address food insecurity.

Food Marketing

The committee concludes that counter-marketing and mass media communications strategies directed at families with young children is a promising intervention deserving of evaluation over longer time periods.

Nutrition Literacy

The committee concludes that the nutrition literacy of caregivers and professionals should be considered in the design of interventions aimed at improving child feeding practices behaviors. Nutrition literacy for caregivers of young children includes understanding the developmentally appropriate nutrition needs of and what and how to feed young children 6‒24 months old, as well as skills to identify credible sources to combat misinformation. In addition, the age-appropriate feeding practices and eating behaviors of young children and the dietary habits of families should be addressed to sustainably improve the dietary intakes of young children.

GOING FORWARD: FINAL CONSIDERATIONS

Based on this scoping review, the committee did not identify one specific intervention that could be immediately scaled in the United States. Collectively, the evidence-based interventions identified were based on messaging that is highly consistent with the “what to feed” aspects of the 2020‒2025 Dietary Guidelines for Americans focusing on children under 2 years of age and with the “how to feed” aspects of the responsive feeding framework.

The strongest model for translating findings from an RCT to widespread implementation of an intervention in the community was provided by the INFANT study. The intervention was family-centered and relied on group sessions of parents, taking advantage of existing programs and contact opportunities through the health care system and diverse health professionals, including dietitians, nurses, and parenting skills instructors. It is possible to envision an INFANT-like program in the United States delivered through WIC, CE, ECE (including Early Head Start), home visiting, or well-child visits, and it could be particularly effective if these systems were empowered to coordinate this effort with each other through a multiagency coordinating body that included agencies such as CDC, the U.S. Department of Agriculture, HHS, and the Administration for Children and Families. The Family Spirit Nurture home visiting intervention also holds promise as an intervention that could be readily scaled in the United States, pending the findings from the second RCT.

Families with infants and young children can benefit from access to credible information on best feeding practices. Since healthy eating behaviors are established early in the life course, widespread implementation of best practices for feeding infants and young children can establish an important strong foundation for U.S. population health in the long term,

in concert with continued efforts later in the life course. The implementation and sustainment of large-scale, effective, population-wide infant and young child feeding programs across the settings identified in this report would represent progress toward this goal. Intergovernmental agency collaboration on workforce development, internet and technology infrastructure, evaluation and program improvement, and efforts to increase funding levels and options and reimbursement for this work will be essential. It will also be important for both public and private health insurance actors to be engaged with this effort.

The committee concludes that there were several interventions with promising elements that could be part of a multicomponent constellation of interventions delivered across settings and systems to address what and how to feed infants and young children. It is critically important that the settings and corresponding systems that were examined—health care, ECE, CE, home visiting, and WIC—be included in such an effort. No one system is currently equipped or adequately funded to reach all children up to age 24 months in the United States, but the existing complementary feeding supports available for some families in the United States could be expanded and harmonized so that all families with young children receive consistent messages about complementary feeding and responsive feeding across multiple settings. The effective scaling of any intervention requires consideration of implementation science and equity principles. Securing permanent funding for program implementation; supporting personnel recruitment, training, and retention; and considering integration of virtual options across the settings highlighted in this report will be key for sustainability.