A chronic mismatch of caregiver responsiveness to infant-feeding cues, such as feeding when the infant is not hungry, is hypothesized to have a role in the development of overweight by impairing an infant's response to internal states of hunger and satiation. Although this concept of mismatch or discordance has long been acknowledged in scholarly writings, a systematic assessment of the evidence supporting the role of discordant responsiveness during infant feeding in the early origins of overweight is lacking. This review was undertaken to assess evidence for this hypothesized relationship between discordant responsiveness in feeding and overweight in infancy and toddlerhood, framed within the larger social-environmental context of the infant–caregiver dyad. A systematic method was used to extract articles from three databases of the medical, psychology and nursing fields. The quality of evidence collected was assessed using Oxford University Centre for Evidence Based Medicine's level of evidence and through a narrative review. The systematic search resulted in only nine original research studies, which met a priori inclusion/exclusion criteria. Several studies provide support for the conceptual model, but most were cross-sectional or lower quality prospective studies. The need for consistent definitions, improved measures and longitudinal work is discussed. In conclusion, this review reveals preliminary support for the proposed role of discordant responsiveness in infant/child overweight and at the same time highlights the need for rigorous investigation of responsive feeding interactions in the first years of life.
Infancy is believed to be a sensitive period for the development of energy intake regulation1 and overweight.2, 3, 4, 5, 6 In the first 2 years of life, infants and toddlers are dependent on their caregivers to provide adequate and appropriate nutrition. For this reason, the potential influence of feeding dynamics on overnutrition and the development of overweight is seemingly intuitive. Scholarly work on early life origins of overweight, however, has largely ignored the possible impacts of social and behavioral transactions on infant feeding. We focus on one particular aspect of these caregiver–infant transactions: responsive feeding. In this review, we define responsiveness within the context of overweight, as involving prompt, contingent and developmentally appropriate responses to the infant's hunger and satiety cues. As the review will describe, the degree to which feeding interactions are responsive is believed to hold importance in infant development by nurturing or impeding the development of self-regulation. Self-regulation, generally defined as ‘the ability to regulate reactions…and to function more independently in a personal and social context, (pg 93)’ is said to have multi-dimensional influence on functioning, including behavioral, emotional and cognitive actions.7 The early development of this general capacity is profoundly affected by supportive caregiver–infant interactions and it is likely that these interactions have a similar impact on the development of eating self-regulation. We present a model in which a caregiver's responsiveness that is congruent with infant-feeding cues encourages the development of infant self-regulation as it relates to energy intake, and eventually obesity.
Overweight and accelerated weight gain in infancy and toddlerhood
The problem of pediatric overweight is evident before the preschool years. Data collected in the National Health and Nutrition Examination Survey 2007–2008 indicated that approximately 10% of infants and toddlers were above the 95th percentile for weight-for-length; the prevalence was as high as 14.9% in Hispanic males.8 These numbers are of concern because heavy infants are at increased risk of overweight in later stages of development.9, 10 Analysis of the CDC Pediatric Nutrition Surveillance System data revealed that overweight infants (0–11 months) were 2.9–4.3 times more likely to be overweight between the ages of 1 and 4 years than non-overweight infants.11 Accelerated weight gain during the first 2 years of life, independent of birth weight and parental weight status, has also been associated with a greater risk of overweight during childhood12, 13, 14, 15, 16, 17 and young adulthood4, 18, 19, 20, 21 (see Stettler and Iotova22 for a review). Though discussion of etiology has favored biological explanations, caregiver–infant feeding interaction, like responsiveness, is implicit when considering the potential contributing factors of infant weight gain.14, 23
Caregiving and self-regulation among infants and toddlers
Why might caregiver responsiveness be a formative dimension of feeding during the first 2 years of life? Responsiveness is a dimension of infant–caregiver interactions, which has been said to shape an infant's ability to acquire self-regulation.7 Early relationships with caregivers engender an infant's development in numerous ways, but a supportive environment to enhance the development of self-regulation of energy intake holds unique importance for childhood obesity risk. Supporting infant self-regulation of energy intake may provide the necessary conditions to facilitate development of regulatory capacity and autonomy throughout infancy and into childhood. It is believed that infants and young children are born with a nascent capacity for self-regulation that becomes actualized through cause–effect learning, which occurs when their behavior is consistently met with a prompt, developmentally appropriate response.24 Responsive caregiving has been studied for decades in domains outside of child feeding and growth, and has been positively associated with a wide range of developmental outcomes in infancy and early childhood including emotional, language and cognitive capacity, as well as the security of attachment to the mother (see Ziv et al.25 for a review). Such synchronous interactions are thought to provide fundamental support for opportunities to learn mastery and self-regulation.26 Early infant emotions, for example, stem from physical states, including temperature, sleep states and hunger.7 Newborns exhibit emotional reactivity to these physical states and express discomfort via crying, but begin to show signs of self-regulated behavior by 3 months of age.27 Examples of self-regulated behavior include turning the body or head away from undesirable stimuli and moving the hands toward the head and mouth.7 Caregivers support this developmental progression from emotional reactivity to self-regulation by their consistent, accurate and appropriate responses to infant cues, which foster expectations in the developing infant about predictable patterns of interaction.7 However, accurate interpretation by caregivers is complicated by the fact that infant crying can reflect discomfort due a variety of physical states including sleep, temperature and hunger states.28 Thus, it should be acknowledged that caregiver and child are mutually influencing one another over time through the consistency, meaning and appropriate interpretation of one another's behaviors.
Caregiver influences on the infant's development of self-regulation of feeding behavior parallels emotional development. In their monthly feeding observations of 26 mother–infant pairs over the first year of life, Ainsworth and Bell29 observed that mothers who exhibited more sensitive pacing and prompt responses to the infant during feeding had infants who cried less in early infancy and demonstrated greater attachment to their mothers at the end of the first year, compared with mothers who adhered to rigid feeding schedules. This supports the hypothesis that caregiver responsivity affects infant outcomes and development.30, 31 However, the relationship between caregiver and infant is bidirectional and can also be affected by the clarity of infant cues.32, 33 Some infants demonstrate ambiguous signals, and have fewer behaviors in general, including fewer elicited and spontaneous vocalizations, fewer smiles, less gaze and joint attention and poor signals to indicate their needs (including hunger cues). In such cases, a caregiver's responsiveness might be discordant with infant needs (due to poor clarity of cues) or might be a low level because few cues are exhibited.
To this notion, Wright and colleagues34, 35 have argued that appropriate responses to infant-feeding cues are critical for self-regulation and self-control of food intake to develop to its fullest potential. This argument is evidenced by experimental studies, which have demonstrated that infants as young as 6 weeks of age could adjust the volume of formula intake in response to its energy content to maintain daily energy intake.36, 37, 38 Observational research has also provided evidence of such a capacity, including data from 3022 children (6 to 24 month old ) participating in the Feeding Infants and Toddlers Study showing inverse relationships between the number of daily feeds and the size of feeds.39 In the context of developmental literature on responsive parenting, self-regulation of energy intake in infancy and toddlerhood is viewed as the extent to which feeding environments and interactions allow the biological potential of the child for self-regulation to be actualized.40
A potential role for responsive feeding
Although contribution of responsive feeding to self-regulation and overweight in early development has been suggested in scholarly writings for decades,35, 41, 42, 43, 44 systematic inquiry appears limited. This review was undertaken to characterize empirical support for the assertion that variation in the concordance of a caregiver's responsiveness and an infant's hunger and satiety cues has a fundamentally important role in growth by supporting energy self-regulation in infancy and toddlerhood. Figure 1 presents the conceptual framework that guided the review by articulating potential pathways through which caregiver responsiveness to infant-feeding cues influences child intake and growth. Drawing from the developmental literature, feeding responsiveness is defined for the purposes of this review as prompt, contingent and developmentally appropriate responses to the child's hunger and satiety cues.24, 45 The bidirectional nature of feeding is implicit in this definition, with the caregiver dependent on the infant to provide clear, unambiguous cues and the infant dependent on the caregiver's accurate interpretation of and appropriate response to those cues for obtaining adequate and appropriate nutrition. Given the aforementioned variance in the clarity of infant-feeding cues, caregivers may be challenged to a lesser or greater degree in their interpretation of cues. Likewise, less-responsive caregivers may challenge the infant to provide clearer cues (within a repertoire that may be relatively limited in the first months of life). Thus, when we refer to caregiver feeding responsiveness, we are referring to a characteristic of the caregiver–child dyad's level of concordance in interactions during feeding. Responsiveness to child-feeding cues has previously been included as part of an internationally recognized framework of infant-care practices oriented toward the prevention of malnutrition and the support of child mental, social and physical development.31, 44, 46 In that work, responsive feeding has been broadly conceptualized to involve the caregiver's adaptation to the child's psychomotor abilities for feeding and provision of a structured feeding environment with a goal of facilitating energy consumption to offset the effects of malnutrition.44 In contrast, this review focuses solely on the role of caregiver responsiveness to feeding cues in supporting the capacity of infants to self-regulate energy intake for optimal growth in a context of obesity risk.
Figure 1 was developed drawing from scholarly writing dating back almost a half century.35, 41, 42, 43, 44 In this model, discordant caregiver responsiveness to infant cues, specifically feeding without hunger and feeding beyond satiety, is hypothesized to impair infant-satiety response. Such impairments are suggested to promote energy intake beyond needs for growth via increases in meal size and/or feeding frequency. Overweight and rapid weight gain during infancy and toddlerhood are the main growth-related outcomes of interest. This model acknowledges feeding responsiveness as being nested within a broader social-environmental context that influences the way in which caregivers feed their children,47 including cultural influences (e.g., ethnicity/race48), policy and prevailing feeding guidance (e.g., World Health Organization Recommendations on Breastfeeding49), parental beliefs/goals/values,50 and parental knowledge of development.51 These are all the factors brought in to the feeding relationship—and all can have considerable effects on the relationship. This model acknowledges these elements, which are at the foundation of building a responsive feeding relationship. The review was initiated to assess the amount and quality of empirical evidence available to support this proposed model.
Materials and methods
We performed a systematic literature review of responsive feeding using the three major databases from three disciplines (medicine, psychology and nursing): MEDLINE via PubMed (United States National Library of Medicine, Bethesda, MD, USA), PsycINFO (American Psychological Association, Washington, DC, USA) and CINAHL (Ebscohost, Glendale, CA, USA). These databases were searched for articles listed from the database's inception to September 2009. Database limits were used to restrict the search to research in humans and infants (from birth to 24 months).
As listed in Table 1, a list of 25 search terms were generated by the research team to identify relevant articles addressing the concepts outlined in the conceptual model (Figure 1), including responsive feeding, self-regulation, feeding/hunger cues, as well as terms related to obesity, such as rapid weight gain. The phrases/terms in quotations were searched for the exact phrase/term, whereas the other terms were not put in quotations to allow for a variety of combination terms to be searched. All search terms that included the word ‘Infant’ were repeated using the term ‘Child’ to ensure an extensive search of the infant population.
Study selection and grading
Article titles and their abstracts were briefly reviewed by the first author (KI DiSantis) using the selection criteria. Those that met and questionably met the inclusion/exclusion criteria were then thoroughly reviewed for eligibility by all the authors. Inclusion criteria included the following: original research, published in English, published up to September 2009, involving healthy full-term infants, aged 0–24 months of child age, including infant feeding/eating and/or infant weight/growth outcomes, and oriented toward child overweight/rapid weight gain and/or obesity. Owing to concerns about socioeconomic and cultural consistency across studies, and the goal of focusing on overnutrition and obesity (as opposed to slow growth and malnutrition), exclusion criteria included research conducted in developing countries. In addition, given the focus on infant overweight and obesity, research that focused on growth faltering (organic or non-organic) or low birth weight was excluded. Although including growth faltering and malnutrition studies would increase the quantity of articles extracted, they likely would not have added to the content of this review as responsive feeding in these studies are investigated in relation to a very different set of outcomes, including promotion of energy intake and weight gain.
The quality assessment phase measured the level of evidence provided by each included study and utilized the Oxford University Centre for Evidence Based Medicine (CEBM) levels of evidence system.52 This system was developed to be an advancement of the Canadian Task Force on the Periodic Health Examination grading system of 1979. The ‘level’ of evidence is graded, primarily based on the rigor of the study design (e.g., randomization in interventions), sampling (e.g., response rate), longitudinal versus cross-sectional, outcome assessment (e.g., negative assessment if outcome measure is imprecise) and the follow-up rate, with ‘1a’ being the highest and ‘5’ being the lowest level. Levels of evidence were independently assigned to the included papers by two of the study authors. A consensus approach was used to resolve discordant assessments, in which a third reviewer independently graded the article and discussion by all authors was used to assign a final rating.
Systematic review results
The results of the systematic review are detailed in Table 1 and illustrated in Figure 2. The search of three databases with 25 aforementioned keywords provided 1877 hits. Owing to the fact that these databases contain subscriptions to the same journals or indexes, it was not possible to derive a unique number of hits. After elimination of redundant abstracts, 82 unique abstracts were deemed appropriate for full-abstract review based on the first author's brief review of article titles and abstracts. These abstracts were then thoroughly reviewed by all authors based on the a priori inclusion/exclusion criteria described above to determine whether they would be selected for final analysis. First, 14 articles were excluded because they were not original research studies, performed research in a population other than infants of 0–24 months and/or the research was performed with regard to malnutrition/growth faltering (Figure 2). The 68 remaining articles were assessed to determine whether some aspect of infant/child obesity was directly measured (eating behavior, intake weight and/or growth) and whether caregiver-feeding practices/behaviors were assessed (Figure 2). A total of 29 articles were excluded because some aspect of infant overweight/obesity or infant eating/intake outcomes was not directly measured. In most of these excluded cases, feeding observations or other feeding measures were used as a tool to investigate general parental responsiveness or general infant–caregiver interaction, but the observations were never presented in the context of infant eating/intake outcomes or were not directed toward obesity outcomes child overweight/obesity. Overall, 30 articles were excluded because of lack of assessment of caregiver-feeding practices/behaviors, for instance, in which feeding characteristics and infant growth might have been assessed but no aspect of responsiveness was measured. After screening all abstracts, a total of nine articles met the inclusion/exclusion criteria and were included in the systematic review (Figure 2).
The results of evidence-level grading for the nine articles, based on CEBM criteria, are provided in Table 2. All articles were graded as being a ‘2c’ level or lower level, meaning the design was cross-sectional or a lower quality cohort study or case–control study. Table 2 also provides additional details related to the quality of the research studies, including sampling characteristics, methods of measuring feeding responsiveness and a brief outcome summary. Below we briefly review these articles and their implications following the framework of the conceptual model in Figure 1.
Socioenvironmental context of discordant responsiveness
Feeding interactions take place within a wider social-environmental context, consisting of a wide range of factors like biobehavioral issues, culture, psychosocial status, physical environments and history at individual, family and community level. Mogan53 was the only article in the review that considered context, by evaluating associations of both parental and infant weight status with maternal sensitivity to infant cues. Parental weight status is an individual-level variable, but in this study they accounted for the differences between dual and caregiver households with varying combinations of parents classified as normal or overweight. Parental weight status was categorized into three groups: (1) two normal weight parents, (2) one overweight and one normal and (3) two overweight parents. Mothers and infants were observed during six feeding sessions from age 0 to 6 months to assess feeding interactions. The observations were coded using the Nursing Child Feeding Assessment Scale, which contained the following subscales: mother's sensitivity to cues, response to distress, social –and emotional growth fostering, cognitive fostering and the infant's clarity of cues and responsiveness to their mother. This measure, particularly the mother's sensitivity to cues and infant responsiveness subscales, allowed for the assessment of discordance in responsiveness in a bidirectional manner as presented in Figure 1. Although the methodology was strong, the authors did not find that these responsive feeding behaviors of the mother and/or infant differed among the parental weight groups. However, infant weight status differed significantly at 6 months of age according to parent weight group, with infants who had one or two overweight parents being more likely to be at or above the 90th percentile for weight. Mogan53 did not report on relationships between maternal feeding responsiveness and infant weight status at 6 months, so it is unclear whether any relationship existed in this sample. Although this study used observational data, which assessed the bidirectional feeding relationship, it neither found relationship between parental weight status and responsive maternal–infant feeding interactions nor did it report on the association of responsive feeding with infant weight at 6 months. Other studies of older children have connected maternal weight status with unresponsive feeding practices (i.e., restriction in feeding),54 but are outside the developmental scope of this review.
Discordant responsiveness and feeding frequency and amount
As Figure 1 illustrates, discordant responsiveness in feeding might lead to increased feeding frequency or amount. Numerous circumstances of discordance could combine to result in increased feeding frequency (including feeding an infant in the absence of hunger cues, misperception of hunger cues) and increased amounts of food (including ignoring fullness cues, or misperception of fullness cues). If this were to continue over a period of time, the risk of increased energy would increase. Kavanagh et al.55 performed a double-blinded, randomized intervention, which focused on using education on feeding responsiveness and prevention of overfeeding to reduce the risk of overfeeding (infants were 3–10 weeks at enrollment). The control group (n=21) received a 45-min educational session on general guidelines for infant feeding, including appropriate age of introduction of complementary foods, safe preparation of complementary foods, responsive feeding practices, when feeding complementary foods, and information on low-cost ways of providing nutritionally balanced meals. The intervention group (n=19) was given similar information but additionally provided with educational information on being aware of infant-satiety cues when breastfeeding or bottle feeding (e.g., understanding early versus late cues) and were encouraged to only prepare 6 ounces of formula per feeding. The outcome measures were both infant weight and length and formula intake (measured at baseline, 2 weeks after the class and at about 4 months. Kavanagh et al.55 found no differences between the intervention and control groups with regard to formula intake at any of the time points and bottle-emptying behavior, and conversely the intervention group infants were heavier and taller at 4 months. Thus, this study did not provide support for the conceptual model. Kavanagh et al.55 acknowledged shortcomings in this small study including that at baseline the intervention group infant's were taller and heavier than the control group, the bottle records were not weighed and a considerable loss to follow-up rate with only 38% of the original sample completing both the class and the assessments.
Rybski et al.56 similarly studied the associations between maternal feeding behaviors and infant intake, in a small sample of 3-day-old female, white infants. Although the study observed maternal behaviors related to responsiveness during bottle feedings (i.e., verbal interaction, eye contact, tender and caretaking touching), the purpose of the study was to observe changes in sucking behaviors and intake of formula across a 24-h period to understand the potential effects of circadian periodicity. The feeding variables (e.g., total feeding time, nutritive sucking time nutritive sucking count) were measured during six observations set at the same time for all participants across a 24-h period in a stimuli-free room, with controlled temperature and lighting. This study did not find an association between maternal feeding behaviors and infant energy intake. A potential source of bias for evaluating feeding responsiveness was that mothers were not allowed to hold their infants during any feeds, which could have limited the amount and type of (e.g., touching) behaviors a mother was engaged in during a feeding. In view of these shortcomings of these studies, additional research is needed to evaluate the effect of caregiver awareness of infant-feeding cues on infant energy intake and subsequent weight gain.
Discordant responsiveness and impaired self-regulation
As shown in Figure 1, we suggest that caregiver feeding responses that are discordant with infant hunger and fullness cues could led to impaired satiety response in the infant. Experimental and observational studies have shown that infants and toddlers possess an ability to self-regulate energy intake at and across eating occasions by adjusting food intake in response to changes in feeding frequency and the energy content of foods consumed.57, 58, 59 Responsive caregiving is believed to promote cause (i.e., infant cue) and effect (i.e., caregiver response) learning that is central to the development of self-regulation.24 Some have suggested that chronically unresponsive feeding may negatively influence what children learn about when eating should begin and end.40, 41 This systematic review, however, did not identify any research that has evaluated the assertion that feeding interactions affect infant satiation. As such, there is currently no evidence to suggest that responsiveness influences child self-regulation of energy intake in infancy or toddlerhood. The notion that feeding children in the absence of hunger and continuing to feed beyond fullness is detrimental to the development of self-regulation has appeared in scholarly writing for well over half a century. The findings of this review suggest the need for research explicitly measuring dimensions of appetite regulation, including hunger, satiety and satiation. Owing to a lack of studies and negative finding, evidence that feeding responsiveness influences infant self-regulation of energy intake remains weak.
Discordant responsiveness and accelerated weight gain and overweight
Compared with other areas of the model, a relatively greater number of studies were identified involving the association of responsiveness to infant cues with infant weight status and weight gain. Saxon et al.60 studied maternal feeding behaviors and subsequent weight gain. The feeding behaviors focused on in Saxon et al.,60 measured whether a mother reported using demand or schedule feeding in the first 6 months of life, rather than more directly assessing responsiveness as we have described. However, the authors describe demand feeding in the same manner as we have describe responsive feeding, in which the caregiver would initiate feeding in response to infant-hunger cues, rather than based on the external factor of time. Mothers retrospectively self-reported their feeding practices from birth to 6 months, and were then classified as demand or schedule feeders based on the answers to these two questions: (1) I would classify my feeding philosophy as: (answers: ‘Feeding on demand’ or ‘Feeding on a schedule’) and (2) Who would you say usually determined your baby's eating routine? (answers: ‘Me (myself/caregiver)’ or ‘Baby’). They reported that feeding style did not significantly predict infant weight gain at 2, 4 and 6 months (controlling for birth weight). Although Saxon et al.60 findings do not support a relationship between maternal feeding style and infant's future weight status as proposed in the model, a few shortcomings must be noted. Foremost, demand versus scheduled feeding does not explicitly capture ‘responsive feeding’ as we have defined it. Also breastfeeding was not controlled for during analyses, even though it has been linked to growth particularly in the first 2 years of life,61 and the two groups had differential levels of breastfeeding.60 Lastly, evaluating growth (i.e., change with time) may have been more informative than assessing weight at each time point controlling for birth weight.
Farrow and Blissett62 examined a well-researched aspect of responsive feeding, maternal control and its moderating effects on infant weight gain in the first year of life. Relative to the other studies reviewed, the maternal behavior of interest was well aligned with the definition of ‘responsive feeding’ used here. Maternal control was measured through the systematic coding of feeding observations (using the Feeding Interaction Scale,63 in which the observer rates the mother on a 1–9 scale, with 1 equaling a very controlling caregiver (e.g., mother is continuously forcing the infant to eat) and 9 equaling not controlling (e.g., mother is allowing the infant autonomy to control his or her own feeding, although supervising the infant). Infant weight was measured at 6 and 12 months (birth weight was taken from hospital records). They found that infant weight gain from birth to 6 months, and from 6 to 12 months were negatively correlated, indicating that infants appear to self-regulate their weight in the first year of life—however, this was only found in infants with low maternal control in feeding. Infants with high-maternal control in feeding had the opposite pattern—weight gain from birth to 6 months and from 6 to 12 months was positively correlated, so that a high weight gainer from birth to 6 months would continue on the path of high weight gain. This suggests that, although infants have the capability of self-regulation of energy intake and weight gain, there is an interaction with the environment (caregiver feeding control) that can modify the expression of this potential. Farrow and Blissett's62 work offers insight into the role of one aspect of responsive feeding (maternal control) and infant growth. The prospective nature of the study along with rigorous measures helps to add support to the model proposed here.
In a more recent investigation identified, Li et al.64 reported on a population-based survey of US mothers, which investigated whether breastfeeding duration and frequency, and exclusiveness and bottle-emptying practices predicted excess infant weight gain in the interval between 6 and 12 months. These feeding practices were self-reported at multiple time points from birth to 6 months. Findings revealed that infants who often emptied bottles in the first 6 months of life were 69% more likely to have excess weight gain in the second-half of infancy (6–12 months), when compared with those who rarely emptied bottles.64 Also, high breastfeeding intensity (combination of duration and frequency) resulted in significantly reduced excess weight gain. This study explores the ‘what’ of infant feeding (breastmilk versus formula), which is often focused on with relation to obesity, but also explores the ‘how,’ through the effects of bottle emptying. These findings highlight behavioral aspects of infant feeding, which affected infant weight gain in this large sample (n=1896). An issue in interpreting these findings is the extent to which bottle-emptying behaviors reflected infants' appetites versus caregivers' responsiveness to infant-feeding cues. Additional research is needed to address this issue.
Worobey et al.65 assessed the association of infant growth with maternal-feeding attitudes related to responsiveness for their association with infant growth in a sample of low income, minority women and their newborn infants. Maternal ‘pushiness’ during feeding (using the Maternal Feeding Attitudes Questionnaire66) and maternal sensitivity to infant cues (using the Nursing Child Assessment Feeding Scale67, 68) were assessed in relation to infant growth from birth to 12 months. After controlling for numerous factors (such as birth weight, gender, race/ethnicity, maternal age, maternal body mass index before pregnancy), infant weight gain between 6 and 12 months was predicted by mothers' sensitivity to satiety cues such that low-maternal sensitivity to infant cues resulted in increased weight gain. It should be noted that infant growth measures lacked standardization, as change in actual weight was investigated (i.e., weight gained from 6 to 12 months of age), no change in growth based on a standardized growth reference (i.e., weight-for-age Z-score change). As a result, the changes in growth only speak to change to one's own baseline not in reference to whether the individual's growth would trail, equal or exceed other individual's of similar gender and age.
Two studies were identified that assessed infant obesity/overweight in relationship with responsive feeding. Dubois et al.69 sampled two groups of infants; 42 normal weight 4–9 months old (25–75th percentiles and 47 overweight 4–9 months old (<90th percentile). They investigated whether these two groups differed in energy intake, infant feeding history and maternal reliance on external cues. For ‘maternal reliance on external cues’, mothers were asked open-ended questions and their answers were coded with respect to the following variables: mother usually uses external cues to initiate or terminate feedings, sometimes tries to feed more or less than the infant wants and sometimes offers food to stop infant's crying. External cues were described as time or prepared portion of formula/food and infant signals were described as sucking fingers (hunger) or turning head away (fullness). Dubois et al.69 found no differences in maternal reliance on external cues among the groups of overweight 4- to 9-month-old infants (<90th percentile) and normal infants (25–75th percentiles). However, the method of measuring maternal feeding responsiveness was not standardized as open-ended questions were used, which had not been previously validated. Baughcum et al.70 investigated the affect of maternal-feeding behaviors on weight in an infant sample (aged 11–23 months). A separate sample of preschoolers was assessed as part of this study;70 however, those results are not discussed given the focus of this review on infants and toddlers. Within the infant sample, a number of aspects of maternal feeding in which responsiveness is implicit were assessed for the association with infant weight status, including concern about the infant's weight (either over- or underweight), concern about the infant being hungry, using food to calm the infant and establishing a feeding schedule. However, no associations were identified between a particular feeding style and overweight in infants. Income was found to influence feeding behaviors, for example low-income mothers of infants reported more concern about infant hunger and feeding infant on a schedule. Thus, the findings of Baughcum et al.70 did not add support for the model. But a criticism is that the purpose of the study was in part to validate the questionnaires used—these new questionnaires might have failed to elucidate responsive feeding behaviors. For example, maternal control in feeding, commonly measured through the validated Child Feeding Questionnaire,71 has previously been connected with child overweight,72, 73 yet in this study, control was not associated with child overweight. Also measuring concurrent weight and feeding behaviors might not reflect any of the effects from feeding behaviors in infancy on later childhood overweight. Most of the evidence was gathered to support the connection between responsive feeding and weight gain/overweight as presented in the model, and of the six studies identified, three added support.
Summary of the systematic review findings
The evidence gathered was rated at 2b or lower level of the CEBM system, meaning there was a dearth of high-quality prospective work. Of the nine studies described here, only three studies (Farrow and Blissett62; Li et al.64; Worobey et al.65) revealed associations with dimensions of feeding responsiveness as described for the model. Although these studies provide preliminary support for an influence of responsive feeding on infant/toddler weight, there was a notable lack of evidence to substantiate the assertion that such a relationship is produced by excessive energy intake. Further, the methods for measuring responsive feeding as reviewed in this narrative and as presented in Table 2, generally reveal a lack of consistency in the manner in which feeding responsiveness has been operationally defined. This concern, along with the failure to consider the bidirectional nature of maternal responsiveness and infant behaviors, points to the need for further exploration.
This systematic review offered the opportunity to describe the rationale for investigating responsive feeding and overweight during infancy and toddlerhood and to assess the state of the science on this topic. The conceptual model provided a framework for identifying and thematically organizing scientific literature on this topic. The systematic approach adopted here revealed a dearth of rigorous inquiry on this topic. The three studies that provided the strongest support for the model (Farrow and Blissett62; Li et al.64; Worobey et al.65) evaluated maternal responses to satiety cues and/or bottle-emptying behaviors and their association with infant growth. Only Li et al.,64 however, assessed feeding interactions in a longitudinal manner. In general, all the articles retrieved, supportive or not, lacked prospective assessments of infant–caregiver interactions as they relate to self-regulation, growth and obesity, which was evidenced by the relatively low CEBM levels assigned to the studies gathered. This is a crucial gap in the current literature in light of recognition that feeding interactions are nested in developmental phases. So that capturing caregiver–infant interactions at one time point only provides insight into that point in development. That there was a shortage of support for the proposed model from the primarily cross-sectional studies in this systematic review should not discourage further inquiry.
Beyond the dearth of longitudinal work in this area, the assessment of the dynamic nature of caregiver–infant feeding interactions is notably lacking. Responsiveness can be conceived of as a reciprocal dimension of feeding in which children are responsible for providing clear feeding cues and the caregiver for responding in a prompt and developmentally appropriate manner. However, rather than focusing on how bidirectional aspects of feeding relate with obesity outcomes (increased energy intake, rapid weight gain overweight), studies have focused on how infant-eating behaviors or caregiver (primarily maternal)-feeding behaviors independently affect these outcomes. This is potentially because of the lack of assessment tools, which measure interactions in a dynamic manner, rather than assessing the infant or the caregiver alone. To move knowledge ahead in the area of feeding behaviors and obesity prevention, it is necessary to begin to assess the dyad. Such assessments would aid in evaluating the usefulness of the proposed model.
In addition to the alternating focus on either infant or caregiver behaviors, there was a noteworthy lack of consistency in measurement of ‘responsive feeding.’ The variation in measurement is illustrated by the variety of constructs reported on by the articles in this review (Table 2). Variables used to represent responsive feeding included demand/schedule feeding,60 bottle emptying,64 observational measures (e.g., NCAST)56 and self-report measures of caregiver behaviors.62, 70 An additional complicating factor is that previous research in responsive feeding has been largely oriented toward undernutrition and has conceptualized feeding responsiveness more broadly than is considered here for obesity prevention. Measuring responsive feeding in the framework of obesity prevention may require a different operational definition of responsive feeding than those used in the context of undernutrition. In the model explored here, discordant responsiveness leads to increases in infant intake, which could lead to chronic energy imbalance and, eventually, overweight. The relatively new interest in viewing responsive feeding in the context of child obesity risk may explain the lack of concerted study on this topic.
Why has feeding responsiveness been largely ignored in efforts to understand the development of overweight? Exploration of feeding dynamics and their role in intake and growth requires a multidisciplinary perspective involving parenting, pediatric nutrition and child development. Psychologists have long studied feeding, but viewed it primarily as a vehicle for studying parent–child relationships and developmental outcomes other than growth rather than as phenomena of interest unto itself. Nutritionists have historically viewed children's eating behavior in terms of food and nutrient intake rather than feeding and eating behavior. As a result most knowledge related to feeding pertains to what children are fed rather than how children are fed. Recognition of environmental contributions to pediatric overweight and challenges to its treatment have only recently directed scholarly interest toward understanding familial influences on the development of eating behavior.
Acknowledging that the caregiver–infant dyad and family is situated within a broader socioenvironment context encompassing among others, cultural beliefs/practices about parenting and feeding, policy, education, healthcare and childcare, it is clear that a great deal of work remains to be carried out to begin to understand the complexities of interactions among these factors and the intrapersonal and interpersonal factors we think shape dyadic feeding interactions. It is overwhelming and beyond the scope of any single study to undertake such a task, but such a broad conceptual model as proposed here is useful to the field at large, pointing to particular relationships that are un- or underexplored and highlighting areas in which interdisciplinary collaborations could be fruitful. As first steps in advancing knowledge in this area, we offer some suggestions.
The development of well-operationalized and rigorously developed measurement tools is clearly a priority for moving scientific inquiry forward in this area. Rigorous observational approaches should be considered in initial research efforts to characterize feeding responsiveness and to identify dimensions for which self-report may be possible.74 Qualitative methods may also prove fruitful to identify potential facilitators and barriers to responsiveness, such as caregiver feeding knowledge, attitudes and beliefs. Longitudinal studies will be required to draw inferences about the capacity of responsiveness to modify child nutritional and growth trajectories. Beyond closing in on the caregiver–child interactions, a multidisciplinary approach must be taken to address the complexity of both caregiver–infant interactions and obesity and how and whether responsive feeding relates to the entire growth spectrum (undernutrition to obesity). Points of exploration, include experiences of the infant and caregiver during feeding from mood and stress,75, 76 alterations in neurohormones77 and racial/ethnic differences and cultural beliefs, which might alter caregivers' attitude to optimal responsiveness, can be investigated.78, 79 Thus, efforts to understand the development of infant/child eating behavior and its importance for growth should take feeding, its goals and its context into consideration. The dependence of infants and toddlers on their caregivers to obtain nutrition and develop eating skills suggests a potentially critical role for responsive feeding in nutrition and growth from a very early point in development. A recent study of overweight children illustrates the need for such early interventions, as it found that nearly 60% of overweight children became overweight before the age of 2 years.3 As a potentially modifiable behavior, responsiveness may represent an efficacious target for early obesity prevention efforts in the future.
In conclusion, the notion that a chronic mismatch between feeding and child cues contributes to the development of overweight is not new.29, 41 On the basis of the findings of this systematic review, however, the role of feeding responsiveness in accelerated growth and overweight remains, to date, is more speculative than substantive. There is preliminary support for the proposed role of responsiveness in growth during early development, though the strength of evidence is relatively weak and the studies are few. These findings underscore that we are in the early stage of empirical research on this topic. As such, we conclude that additional rigorous investigation of feeding responsiveness is needed, particularly longitudinal studies, within the framework of early obesity prevention efforts among diverse populations.
Kral TV, Stunkard AJ, Berkowitz RI, Stallings VA, Brown DD, Faith MS . Daily food intake in relation to dietary energy density in the free-living environment: a prospective analysis of children born at different risk of obesity. Am J Clin Nutr 2007; 86: 41–47.
Stettler N, Iotova V . Early growth patterns and long-term obesity risk. Curr Opin Clin Nutr 2010; 13: 294–299.
Harrington JW, Nguyen VQ, Paulson JF, Garland R, Pasquinelli L, Lewis D . Identifying the ‘tipping point’ age for overweight pediatric patients. Clin Pediatr (Phila) 2010; 49: 638–643.
Stettler N, Kumanyika SK, Katz SH, Zemel BS, Stallings VA . Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans. Am J Clin Nutr 2003; 77: 1374–1378.
Monteiro POA, Victora CG . Rapid growth in infancy and childhood and obesity in later life—a systematic review. Obesity Rev 2005; 6: 143–154.
Wells JC, Chomtho S, Fewtrell MS . Programming of body composition by early growth and nutrition. Proc Nutr Soc 2007; 66: 423–434.
Shonkoff JP, Phillips De . From Neurons to Neighborhoods: the Science of Early Childhood Development. National Academy Press: Washington, DC, 2000.
Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM . Prevalence of high body mass index in US children and adolescents, 2007–2008. JAMA 2010; 303: 242–249.
Scaglioni S, Agostoni C, Notaris RD, Radaelli G, Radice N, Valenti M . et al. Early macronutrient intake and overweight at five years of age. Int J Obes Relat Metab Disord 2000; 24: 777–781.
Mellbin T, Vuille JC . Relationship of weight gain in infancy to subcutaneous fat and relative weight at 10 1/2 years of age. Br J Prev Soc Med 1976; 30: 239–243.
Mei Z, Grummer-Strawn LM, Scanlon KS . Does overweight in infancy persist through the preschool years? An analysis of CDC Pediatric Nutrition Surveillance System data. Soz Praventivmed 2003; 48: 161–167.
Stettler N, Zemel BS, Kumanyika S, Stallings VA . Infant weight gain and childhood overweight status in a multicenter, cohort study. Pedatrics 2002; 109: 194–199.
Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I et al. Early life risk factors for obesity in childhood: cohort study. BMJ 2005; 330: 1357.
Dennison BA, Edmunds LS, Stratton HH, Pruzek RM . Rapid infant weight gain predicts childhood overweight. Obesity 2006; 14: 491–499.
Dubois L, Girard M . Early determinants of overweight at 4.5 years in a population-based longitudinal study. Int J Obes 2006; 30: 610–617.
Karaolis-Danckert N, Buyken AE, Bolzenius K, Perim de Faria C, Lentze MJ, Kroke A . Rapid growth among term children whose birth weight was appropriate for gestational age has a longer lasting effect on body fat percentage than on body mass index. Am J Clin Nutr 2006; 84: 1449–1455.
Blair NJ, Thompson JMD, Black PN, Becroft DMO, Clark PM, Han DY et al. Risk factors for obesity in 7-year-old European children: the Auckland Birthweight Collaborative Study. Arch Dis Child 2007; 92: 866–871.
Stettler N, Kumanyika S, Katz SH, Zemel BS, Stallings VA . Rapid early infancy weight gain and obesity in African American adults. Circulation 2003; 107: E7002.
Stettler N, Stallings VA, Troxel AB, Zhao J, Schinnar R, Nelson SE et al. Weight gain in the first week of life and overweight in adulthood: a cohort study of European American subjects fed infant formula. Circulation 2005; 111: 1897–1903.
Monteiro POA, Victora CG, Barros FC, Monteiro LMA . Birth size, early childhood growth, and adolescent obesity in a Brazilian birth cohort. Int J Obes 2003; 27: 1274–1282.
Ekelund U, Ong K, Linne Y, Neovius M, Brage S, Dunger DB et al. Upward weight percentile crossing in infancy and early childhood independently predicts fat mass in young adults: the Stockholm Weight Development Study (SWEDES). Am J Clin Nutr 2006; 83: 324–330.
Stettler N, Iotova V . Early growth patterns and long-term obesity risk. Curr Opin Clin Nutr Metab Care 2010; 13: 294–299.
Ong KK, Emmett PM, Noble S, Ness A, Dunger DB, Team AS . Dietary energy intake at the age of 4 months predicts postnatal weight gain and childhood body mass index. Pedatrics 2006; 117: e503–e508.
Lamb ME, Easterbrooks MA . Individual differences in parental sensitivity: origins, components, and consequences. In: Lamb ME, Sherrod LR, (eds). Infant Social Cognition: Empirical and Theoretical Considerations. Erlbaum: Hillsdale, NJ, 1981; 127–153.
Ziv Y, Cassidy J . Maternal responsiveness to infant irritability: the contribution of Crockenberg and Smith's ‘Antecedents of mother-infant interaction and infant irritability in the first 3 months of life’. Infant Behav Dev 2002; 25: 16–20.
Feldman R, Greenbaum CW, Yirmiya N . Mother-infant affect synchrony as an antecedent of the emergence of self-control. Dev Psychol 1999; 35: 223–231.
Bronson MB . Self-regulation in Early Childhood: Nature and Nurture. The Guilford Press: New York, NY, 2000.
Green JA, Gustafson GE, Irwin JR, Kalinowski LL, Wood RM . Infant crying: acoustics, perception, and communication. Early Dev Parenting 1995; 4: 161–175.
Ainsworth MDS, Bell SM . Some Contemporary Patterns of Mother-infant Interaction in the Feeding Situation Stimulation in Early Infancy. Academic Press: New York, 1969.
Bell SM, Ainsworth MD . Infant crying and maternal responsiveness. Child Dev 1972; 43: 1171–1190.
Bornstein MH, Tamis-LeMonda CS . Maternal responsiveness and cognitive development in children. New Dir Child Dev 1989; 1989: 49–61.
Goldberg S . Social competence in infancy: a model of parent–infant interaction. Merrill-Palmer Quart 1977; 23: 163–177.
Vallotton CD . Do infants influence their quality of care? Infants' communicative gestures predict caregivers' responsiveness. Infant Behav Dev 2009; 32: 351–365.
Wright P, Fawcett J, Crow R . The development of differences in the feeding behaviour of bottle and breast fed human infants from birth to two months. Behav Process 1980; 5: 1–20.
Wright P . Development of feeding behavior in early infancy: implications for obesity. Health Bull 1981; 39: 197–205.
Fomon SJ, Filer LJ, Thomas LN, Rogers RR, Proksch AM . Relationship between formula concentration and rate of growth of normal infants. J Nutr 1969; 98: 241–254.
Fomon SJ, Filmer LJ, Thomas LN, Anderson TA, Nelson SE . Influence of formula concentration on caloric intake and growth of normal infants. Acta Paediatric Scandinavia 1975; 64: 172–181.
Kent JC, Mitoulas LR, Cregan MD, Ramsay DT, Doherty DA, Hartmann PE . Volume and frequency of breastfeedings and fat content of breast milk throughout the day. Pedatrics 2006; 117: e387–e395.
Fox MK, Devaney B, Reidy K, Razafindrakoto C, Ziegler P . Relationship between portion size and energy intake among infants and toddlers: evidence of self-regulation. J Am Diet Assoc 2006; 106 (1S): 77–83.
Wright P . Learning experiences in feeding behaviour during infancy. J Psychosom Res 1988; 32: 613–619.
Costanzo PR, Woody EZ . Domain-specific parenting styles and their impact on the child's development of a particular deviance: the example of obesity proneness. J Soc Clin Psychol 1985; 4: 425–445.
Bruch H . Eating Disorders: Obesity, Anorexia Nervosa, and the Person Within. Basic Books: New York, 1973.
Birch LL, Fisher JO . Development of eating behaviors among children and adolescents. Pediatrics 1998; 101: 539–549.
Engle PL, Bentley ME, Pelto G . The role of care in nutrition programmes:current research a research agenda. Proc Nutr Soc 2000; 59: 25–35.
Bornstein MH . Sensitive periods in development: structural characteristics and causal interpretations. Psychol Bull 1989; 105: 179–197.
Dewey K . Guiding principles for complementary feeding of the breastfed child. PAHO/WHO: Washington, DC, 2003. http://www.who.int/nutrition/publications/guiding_principles_compfeeding_breastfed.pdf.
Bronfenbrenner U . Ecology of the family as a context for human development: research perspectives. Dev Psychol 1986; 22: 723–742.
Taveras EM, Gillman MW, Kleinman K, Rich-Edwards JW, Rifas-Shiman SL . Racial/ethnic differences in early-life risk factors for childhood obesity. Pediatrics 2010; 125: 686–695.
World Health Organization U. Global Strategy for Infant and Young Child Feeding. WHO: Geneva, 2003.
Burdette HL, Whitaker RC, Hall WC, Daniels SR . Maternal infant-feeding style and children's adiposity at 5 years of age. Arch Pediatr Adolesc Med 2006; 160: 513–520.
Susin LR, Giugliani ER, Kummer SC, Maciel M, Simon C, da Silveira LC . Does parental breastfeeding knowledge increase breastfeeding rates? Birth 1999; 26: 149–156.
Centre for Evidence-Based Medicine OU. Levels of evidence and grades of recommendation. Available at: http://www.cebm.net/levels_of_evidence.asp. (Accessed 9 January 2009) 2009.
Mogan J . Parental weight and its relation to infant feeding patterns and infant obesity. Int J Nurs Stud 1986; 23: 255–264.
Francis LA, Birch LL . Maternal weight status modulates the effects of restriction on daughters' eating and weight. Int J Obes Relat Metab Disord 2005; 29: 942–949.
Kavanagh KF, Cohen RJ, Heinig MJ, Dewey KG . Educational intervention to modify bottle-feeding behaviors among formula-feeding mothers in the WIC program: impact on infant formula intake and weight gain. J Nutr Educ Behav 2008; 40: 244–250.
Rybski DA, Almli CR, Gisel EG, Powers J, Maurer Jr M . Sucking behaviors of normal 3-day-old female neonates during a 24-h period. Dev Psychobiol 1984; 17: 79–86.
Birch LL, Deysher M . Caloric compensation and sensory specific satiety: Evidence for self regulation of food intake by young children. Appetite 1986; 7: 323–331.
Birch LL, Deysher M . Conditioned and unconditioned caloric compensation: evidence for self-regulation of food intake by young children. Learn Motiv 1985; 16: 341–355.
Fomon S . Nutrition of Normal Infants. Mosby-Year Books: St Louis, Missouri, 1993.
Saxon TF, Gollapalli A, Mitchell MW, Stanko S . Demand feeding or schedule feeding: infant growth from birth to 6 months. J Reprod Infant Psychol 2002; 20: 89–100.
Kalies H, Heinrich J, Borte N, Schaaf B, von Berg A, von Kries R et al. The effect of breastfeeding on weight gain in infants: results of a birth cohort study. Eur J Med Res 2005; 10: 36–42.
Farrow C, Blissett J . Does maternal control during feeding moderate early infant weight gain? Pediatrics 2006; 118: e293–e298.
Wolke D, Skuse D . The management of infant feeding problems. In: Cooper P, Stein A (eds). Feeding Problems and Eating Disorders in Children and Adolescents. Harwood: Chur, Switzerland, 1992; 27–59.
Li R, Fein SB, Grummer-Strawn LM . Association of breastfeeding intensity and bottle-emptying behaviors at early infancy with infants' risk for excess weight at late infancy. Pediatrics 2008; 122 (Suppl 2): S77–S84.
Worobey J, Lopez MI, Hoffman DJ . Maternal behavior and infant weight gain in the first year. J Nutr Educ Behav 2009; 41: 169–175.
Kramer MS, Barr RG, Leduc DG, Boisjoly C, Pless IB . Maternal psychological determinants of infant obesity. Development and testing of two new instruments. J Chronic Dis 1983; 36: 329–335.
Barnard K . Caregiver/Parent-child Interaction Feeding Manual. University of Washington School of Nursing: Seattle, WA, 1994. NCAST Publications.
Hodges EA, Houck GM, Kindermann T . Reliability of the Nursing Child Assessment Feeding Scale during toddlerhood. Issues Compr Pediatr Nurs 2007; 30: 109–130.
Dubois S, Hill DE, Beaton GH . An examination of factors believed to be associated with infantile obesity. Am J Clin Nutr 1979; 32: 1997–2004.
Baughcum AE, Powers SW, Johnson SB, Chamberlin LA, Deeks CM, Jain A et al. Maternal feeding practices and beliefs and their relationship to overweight in early childhood. J Dev Behav Pediatr 2001; 22: 391–408.
Birch LL, Fisher JO, Castro CN, Grimm-Thomas K, Sawyer R, Johnson SL . Confirmatory factor analysis of the Child Feeding Questionnaire: a measure of parental attitudes, beliefs and practices about child feeding and obesity proneness. Appetite 2001; 36: 201–210.
Faith MS, Scanlon KS, Birch LL, Francis LA, Sherry B . Parent-child feeding strategies and their relationships to child eating and weight status. Obes Res 2004; 12: 1711–1722.
Spruijt-Metz D, Lindquist CH, Birch LL, Fisher JO, Goran MI . Relation between mothers' child-feeding practices and children's adiposity. Am J Clin Nutr 2002; 75: 581–586.
Sacco LM, Bentley ME, Carby-Shields K, Borja JB, Goldman BD . Assessment of infant feeding styles among low-income African-American mothers: comparing reported and observed behaviors. Appetite 2007; 49: 131–140.
Groer MW . Differences between exclusive breastfeeders, formula-feeders, and controls: a study of stress, mood, and endocrine variables. Biol Res Nurs 2005; 7: 106–117.
Farrow CV, Blissett JM . Is maternal psychopathology related to obesigenic feeding practices at 1 year? Obes Res 2005; 13: 1999–2005.
Swain JE, Lorberbaum JP, Kose S, Strathearn L . Brain basis of early parent-infant interactions: psychology, physiology, and in vivo functional neuroimaging studies. J Child Psychol Psychiatry 2007; 48: 262–287.
Kumanyika SK . Environmental influences on childhood obesity: ethnic and cultural influences in context. Physiol Behav 2008; 94: 61–70.
Taveras EM, Gillman MW, Kleinman K, Rich-Edwards JW, Rifas-Shiman SL . Racial/ethnic differences in early-life risk factors for childhood obesity. Pediatrics 2010; 125: 686–695.
This work was supported by the NIH K01 DK 61319-01 (Fisher), USDA 2005-55215-6 16726 (Johnson), NIH DK 56350 (Hodges), RWJF 66523 (Hodges) and Nestle Infant Nutrition (Fisher).
The authors declare no conflict of interest.
About this article
Cite this article
DiSantis, K., Hodges, E., Johnson, S. et al. The role of responsive feeding in overweight during infancy and toddlerhood: a systematic review. Int J Obes 35, 480–492 (2011). https://doi.org/10.1038/ijo.2011.3
- feeding behavior
- appetite regulation
Feeding style profiles are associated with maternal and infant characteristics and infant feeding practices and weight outcomes in African American mothers and infants
Maternal & Child Nutrition (2021)
Child attachment behavior as a moderator of the relation between feeding responsiveness and picky eating behavior
Eating Behaviors (2021)
Child Development Perspectives (2020)
‘It's not worth the fight’: Fathers' perceptions of family mealtime interactions, feeding practices and child eating behaviours