Abstract
Objectives
High birth weight (HBW ≥ 4000 g) infants are at increased risk for obesity, but research has yet to identify the mechanism for this increased risk and whether certain subsets of HBW infants are at greater risk.
Methods
This exploratory study examined child eating behaviors and maternal feeding practices and beliefs across 21 HBW and 20 normal birth weight (NBW, 2500–3999 g) infants at 7–8 months of age using maternal-report measures (n = 41) and a bottle feeding task (n = 16).
Results
HBW infants were at increased risk for high weight-for-length at 7–8 months (F (2, 38) = 6.03, p = .02) compared to NBW infants, but no statistically significance differences on weight gained per day since birth, child eating behaviors, or most maternal feeding practices and beliefs were found between HBW and NBW infants. However, HBW infants who maintained a high weight-for-length (≥ 85th percentile) at 7–8 months had a higher birth weight, gained more weight per day, and had lower maternal-reported satiety responsiveness and maternal social interactions during feedings than their HBW counterparts who were currently below the 85th percentile.
Conclusions for Practice
HBW infants did not differ from NBW infants on eating behaviors and feeding practices, but children born at HBW who maintain excess weight during infancy do differ from those infants who fall below the 85th percentile for weight-for-length. Future research should identify risk factors that longitudinally differentiate HBW infants at greatest risk for maintaining excess weight and develop early screening and intervention efforts for this subset of at-risk infants.
Keywords: Infant feeding behavior, High birth weight, Appetite, Feeding practices
Objectives
Development of obesity begins early in life with growth patterns in the first months of life predictive of obesity in later childhood (Druet et al. 2012; Ong and Loos 2006; Smego et al. 2017). Early feeding factors such as age of introduction of solid foods (Huh et al. 2011), parental feeding practices (Flores and Lin 2013), and infant appetite (van Jaarsveld et al. 2014) have been related to subsequent obesity risk. For example, vigorous feeding style, characterized by high-pressure sucking during bottle feedings, in the first month of life has been associated with greater adiposity during the toddler years (Agras et al. 1987, 1990). Additionally, infants with lower satiety responsiveness or higher food responsiveness in the first 3 months of life are heavier at both 6 and 15 months of age (van Jaarsveld et al. 2014). Additionally, maternal feeding behavior and perceptions (Faith et al. 2006), and parental concerns about child weight (Agras et al. 2004) in infancy have been related to subsequent child adiposity and weight, although evidence supporting a relationship between feeding styles and maternal beliefs and infant weight is currently insufficient (Woo Baidal et al. 2016).
Children born at high birth weight (HBW; weighing ≥ 4000 g at birth) are consistently shown to be at increased risk for obesity during childhood and adulthood (Schellong et al. 2012; Skilton et al. 2014; Yu et al. 2011). Children with HBW have a two times greater likelihood of being obese at kindergarten in comparison to children with a normal birth weight (NBW), defined as a weight at birth between 2500 and 3999 g; (Cunningham et al. 2014). This increased risk persists lifelong; among older adults (average age 65 years), higher birth weight was related to increased risk for current obesity (Phillips and Young 2000). These prior studies have demonstrated HBW, defined as an absolute cut-off of 4000 g, predicts obesity-risk even when gestational age is not accounted for. These findings may suggest that elevated birth weight is problematic even if HBW is the result of the child being post-term at later weeks gestation.
Despite the well-established increased risk for obesity among individuals born at HBW, research has yet to examine the possible mechanisms by which they are at increased risk or factors for identifying those HBW infants that will continue on a trajectory of obesity. While rates of obesity among HBW children are higher (e.g., 23% in kindergarten, 31% in eighth grade) than NBW children [e.g., 11% in kindergarten, 19% in eighth grade; (Cunningham et al. 2014)], these numbers indicate that the majority of HBW children follow a normal, healthy weight trajectory. However, research has not yet explored what puts certain HBW infants at risk and how these infants can be identified for appropriate prevention efforts.
Although research exploring risk-factors among HBW infants, specifically, is sparse, prior research has identified a number of genetic, psychosocial, and environmental factors influencing weight gain throughout development (Harrison et al. 2011). The Six-Cs Model of childhood obesity (Harrison et al. 2011) is a developmental ecological model that includes obesity-related factors at the genetic, child, family, community, national, and cultural spheres. Notably, the model includes the influence of factors at different spheres (e.g., child, family, community) on one another, including interactions between factors at different spheres. For example, the model posits that child characteristics (e.g., self-regulation) can interact with family characteristics (e.g., parent encouragement of healthy eating, parent nutritional knowledge, parent interactions). The model also categorizes factors as nutrition-related, activity-related, and personal and relational attributes (see Harrison et al. 2011 for a more comprehensive description of the Six-Cs model of childhood obesity). The current study seeks to explore personal and relational risk factors and nutrition-related risk factors at both the child and family spheres related to obesity-risk among HBW infants.
Therefore, based on the Six-Cs model and prior research findings that HBW infants are at risk for subsequent obesity, the current exploratory study hypothesized that:
HBW infants would show child characteristics (greater weight gain, greater appetite, more voracious sucking patterns when feeding) and family characteristics (parent feeding practices) associated with greater obesity-risk.
Differences in child appetite, parent feeding practices, and child eating behaviors would be found between HBW infants who maintain a high weight-for-length trajectory (≥ 85th percentile) compared to those HBW infants who follow a healthy weight trajectory (< 85th percentile).
HBW infants would display more voracious sucking patterns (e.g., more sucks per second, greater sucking pressure) during a bottle feeding task compared to NBW infants, and those HBW infants with elevated current weight-for-length would also differ from other HBW infants on these sucking patterns.
Findings from the current exploratory study have potential implications for identifying those infants most at risk for obesity. Further, behavioral factors associated with increased risk could serve as targets for future weight management prevention and intervention efforts in this at-risk population.
Methods
Participants
Participants were 20 NBW infants (i.e., defined as weighing 2500–3999 g at birth) and 21 HBW infants (i.e., defined as weighing ≥ 4000 g at birth) aged 7–8 months and their biological mothers. Infants were recruited at a range of different ages with infants not yet eligible to participate (i.e., age < 7 months) scheduled for a study visit at a later date when they would meet the age inclusion criteria. Families were informed that the objective of the study was to explore how infant eating patterns, feeding practices, appetite, and birth weight may relate to infant growth. Inclusion criteria were: (1) weight at birth ≥ 2500 g; (2) born at ≥ 38 weeks gestation; (3) being a singleton; (4) no hospitalization > 7 days following birth or extensive care through the Neonatal Intensive Care Unit (NICU); (5) no maternal-reported exposure to illicit drugs or alcohol in utero; and (6) no condition that currently impacted eating/feeding. See Table 1 for child, mother, and family demographics.
Table 1.
Participant demographics
| Child characteristics | Normal birth weight (n = 20) |
High birth weight (n = 21) |
|---|---|---|
| M (SD), range | M (SD), range | |
| Age (days) | 251.75 (16.43), 215–273 | 241.05 (15.28), 214–273 |
| Birth weight (g) | 3493.25 (278.54), 2700–3965 | 4296.05 (233.27), 4000–4690 |
| n (%) | n (%) | |
| Sex | ||
| Boy | 10 (50%) | 12 (57%) |
| Girl | 10 (50%) | 9 (43%) |
| Child ethnicity/race | ||
| Non-Hispanic Caucasian | 16 (80%) | 20 (95%) |
| Hispanic Caucasian | 2 (10%) | – |
| African American | 1 (5%) | – |
| Multiple races | 1 (5%) | 1 (5%) |
| Ever breastfed | ||
| Yes | 20 (100%) | 18 (86%) |
| No | 0 (0%) | 3 (14%) |
| Exclusively breastfed for 6 mos a | ||
| Yes | 8 (40%) | 6 (29%) |
| No | 12 (60%) | 15 (71%) |
| Current breastfeeding | ||
| Breast milk only | 12 (60%) | 11 (52%) |
| Breast milk and formula | 5 (25%) | 3 (14%) |
| No breast milk (formula only) | 3 (15%) | 7 (33%) |
| Age at introduction of solid foods | ||
| 4 months | 4 (20%) | 4 (19%) |
| 5 months | 5 (25%) | 5 (24%) |
| 6 months | 9 (45%) | 9 (43%) |
| 7 months | 1 (5%) | 3 (14%) |
| 8 months | 1 (5%) | – |
| Maternal characteristics | Normal birth weight (n = 20) |
High birth weight (n = 21) |
| M (SD), range | M (SD), range | |
| Maternal age (years) | 33.73 (4.08), 22.23–40.23 | 34.38 (4.41), 23.99–44.12 |
| Maternal pre-pregnancy weight (lbs) | 148.75 (35.29), 103.00–227.00 | 155.81 (26.18), 110.00–217.00 |
| n (%) | n (%) | |
| Maternal ethnicity/race | ||
| Non-Hispanic Caucasian | 17 (85%) | 21 (100%) |
| Hispanic Caucasian | 1 (5%) | – |
| African American | 2 (10%) | – |
| Maternal education | ||
| Graduate degree | 11 (55%) | 13 (62%) |
| College degree | 8 (40%) | 7 (33%) |
| Partial college or specialized training | 1 (5%) | 1 (5%) |
| Maternal employment | ||
| Full-time employment | 15 (75%) | 10 (48%) |
| Part-time employment | 4 (20%) | 8 (38%) |
| Not employed, not looking for work | 1 (5%) | 2 (10%) |
| Employed but on medical leave | – | 1 (5%) |
| Family characteristics | Normal birth weight (n = 20) |
High birth weight (n = 21) |
| n (%) | n (%) | |
| Family marital status | ||
| Married | 19 (95%) | 19 (90%) |
| Not married, cohabitating | 1 (5%) | – |
| Single, never married | – | 2 (10%) |
| Family income | ||
| < 75,000 | 3 (15%) | 1 (5%) |
| 75,000–99,999 | 6 (30%) | 3 (14%) |
| 100,000–124,999 | 3 (15%) | 8 (38%) |
| 125,000–174,999 | 3 (15%) | 4 (19%) |
| > 175,000 | 5 (25%) | 5 (24%) |
Defined as no formula or solid foods before 6 months
Infants were recruited using three strategies. Emails with a brief study description were sent to the email listserv of employees at the hospital where study visits were conducted and interested parents reached out to research staff using contact information provided in the email. The initial email resulted in recruitment of primarily NBW infants, so an additional email expressing an interest in HBW infants specifically was sent. Secondly, flyers with study information were circulated to parents at the daycare associated with the hospital where visits were conducted, and parents used research staff contact information from the flyers to express interest in the study. Research staff were also present at the daycare during regular child pick-up and drop-off times to provide eligible families information about the study. These recruitment strategies resulted in high enrollment of NBW infants but limited enrollment of HBW infants. Therefore, we attempted to recruit an even number of HBW and NBW infants by specifically targeting recruitment of HBW infants through a pediatrician’s practice. We selected a specific pediatrician’s practice to work with for recruitment given patients at that practice were similar demographically to the hospital employees recruited through the other study methods. Medical charts at the pediatrician’s practice were reviewed to identify HBW infants meeting study inclusion criteria. Families of infants meeting these criteria were mailed a letter about the study along with a return addressed, stamped “do not contact” postcard for the family to return if they did not want to be contacted about the study. Families not returning the card within 10 days received phone calls from research staff to further explain the study. All recruitment methods were approved by the authors’ Institutional Review Board.
Procedures
Infants and their biological mothers completed one study visit at a Midwest children’s hospital which included collection of child anthropometrics and completion of maternal-report measures of infant appetite, parent feeding practices, and infant eating behaviors. Maternal-report measures were completed on a tablet using RedCap (Research Electronic Data Capture). A subset of mothers and infants (n = 16) also completed an observed bottle feeding. The feeding was completed only with a subset of randomly selected infants due to the cost of the device used in the feeding methodology. Infants were randomly assigned using a computer-generated number list to either the no observed feeding or observed feeding group after agreeing to participate and scheduling a study visit. All participants earned a $40 gift card for participating, and participants completing the observed feeding received an additional $20 gift card. Informed consent was obtained at the beginning of the study visit prior to any data collection. Study procedures were approved by the Institutional Review Board at the hospital where data collection occurred.
Measures
Demographics Form
Mothers reported on family characteristics (e.g., income, maternal education and employment), pregnancy-related health information, and child demographics (e.g., sex, ethnicity, race).
Child Birth Weight
Birth weight was obtained from medical records at the child’s birth hospital.
Child Anthropometrics
Measurements were taken by trained research personnel from the CTRC Bionutrition core at the Schubert Research Clinic. Infant length was measured in triplicate to the nearest 0.1 centimeter using an infant length board (O’Leary, Ellard Instrumentation, LTD). Weight was measured in triplicate to the nearest 100 g with the infant in a dry diaper using a Scale Tronix 4802 digital infant/pediatric scale. The average of the three measurements for weight and length were used to calculate weight for recumbent length. Weight-for-length z-scores and corresponding percentiles for each infant were calculated using sex and age specific World Health Organization (WHO) growth charts using SAS statistical software.
Infant Feeding Questionnaire (IFQ)
The IFQ (Baughcum et al. 2001) was completed to assess maternal feeding practices and beliefs. A version of the IFQ previously adapted to assess present practices and beliefs was used in the current exploratory study (McMeekin et al. 2013). Mothers answered items using a 5-point Likert scale (“never”, “rarely”, “sometimes”, “often”, “always” for maternal feeding practices/child eating behaviors; and “disagree a lot”, “disagree a little”, “no strong feelings either way”, “agree a little”, “agree a lot” for maternal belief questions). Some items were not applicable for infants who were exclusively breastfed, and therefore a “not applicable” response option was provided for these questions. Subscales using these items were not calculated for infants with a “not applicable” response endorsed. The measure provides scores for the following: Concern about Infant Undereating and Becoming Underweight, Concern About Infant’s Hunger, Awareness of Infant’s Hunger and Satiety Cues, Concern about Infant Overeating or Becoming Overweight, Feeding Infant on a Schedule, Using Food to Calm Infant’s Fussiness, and Social Interaction with the Infant During Feeding.
Baby Eating Behaviour Questionnaire (BEBQ)
The BEBQ (Llewellyn et al. 2010) was completed to assess child appetite and eating behavior. Mothers responded on a 5-point scale: “never”, “rarely”, “sometimes”, “often”, and “always” to questions. The measure has previously been used to retrospectively report on appetite in the first 3 months of life. However for the current exploratory study, measure items were changed to the present tense for mothers to report on the child’s current appetite. Scores for “Enjoyment of Food”, “Food Responsiveness”, “Slowness in Eating”, “General Appetite”, and “Satiety Responsiveness” were calculated as the sum of all item-level scores on that subscale.
Feeding Questionnaire
Mothers reported on the infant’s early eating behavior (e.g., age of introduction of solid food, whether the child was breastfed) using a questionnaire developed for use in the current exploratory study.
Observed Feeding
A subset of infants (n = 16) completed an observed bottle feeding. Infants were fed their usual formula or breast milk by their mothers using a standardized bottle and nipple (Evenflo Classic Silicone Nipple Medium Flow) provided by the researcher. Mothers were asked not to feed their infant for at least 2 h prior to the feeding and 100% reported compliance to this request. A NFANT device (NFANT Labs, Atlanta, GA) was attached to the bottle for the feeding and measured child sucking variables. The NFANT is an US Food and Drug Administration (FDA) approved device that measures infant sucking pressure and patterns by attaching to an otherwise typical bottle and does not modify any aspect of the feeding (besides appearance of the bottle). The following variables obtained from the NFANT were used in the current analyses: length of the feeding, suck rate (i.e., sucks per second, using sucks within a burst), rate of sucking bursts (i.e., bursts per second), suck duration, burst duration, interval between sucks, and mean sucking pressure/amplitude (normalized to the maximum peak amplitude). The feeding was completed using a standardized Feeding Situation Protocol (e.g., milk/formula heated to a standardized temperature, infants fed at a specified range of angles). Mothers rated on a 5-point Likert scale (1 Not Typical to 5 Very Typical) how similar the observed feeding was to feedings at home. All feedings were rated as a 3 (typical) or greater.
Data Analyses
Data analyses consisted of three main aims: (1) examine differences in change in weight since birth, infant appetite, parent feeding practices, and infant eating behaviors between NBW and HBW infants; (2) examine differences in birth weight, change in weight, infant appetite, parent feeding practices, and infant eating behaviors between HBW infants with a current weight-for-length percentile ≥ 85th and HBW infants with a current weight-for-length percentile < 85th, (3) examine differences in bottle feeding and sucking patterns between HBW and NBW infants and between HBW infants with elevated versus normal current weight-for-length. The HBW and NBW groups did not show statistically significant differences on basic child demographics (i.e., sex, ethnicity/race), and therefore these variables were not included as covariates in analyses comparing birth weight groups (i.e., aims 1 and 2). Additionally, maternal pre-pregnancy weight and gestational weight gain were not related to infant change in weight since birth, and therefore these variables were also not included as covariates in testing aim 1. All study aims were explored using a series of ANOVAs to test for between group differences. Additionally, we sought to confirm prior findings that HBW infants are at increased risk for excessive weight gain by testing differences on current weight-for-length between birth weight groups. Maternal pre-pregnancy weight was found to be related to current weight-for-length z-scores, and so this variable was included as a covariate in a multiple regression exploring differences in current weight-for-length z-score between weight groups (i.e., NBW compared to HBW).
Results
Overall Comparison Between HBW (n = 21) and NBW (n = 20) Infants
Birth weight category was a significant predictor of infants’ current weight-for-lengths (F (2, 38) = 6.03, p = .02), with HBW infants having greater weight-for-length percentiles when controlling for maternal pre-pregnancy weight. Forty-three percent (n = 9) of HBW infants had a current weight-for-length percentile ≥ 85th, compared to only 20% (n = 4) of NBW infants. However, HBW and NBW infants did not show statistically significant differences in average change in weight (grams) per day following birth (F (1, 39) = 1.40, p = .25).
No statistically significant differences on maternal-report of infant food or satiety responsiveness, appetite, enjoyment of food, slowness in eating, or age at introduction of solid foods were found between HBW and NBW infants. Mothers of NBW and HBW infants also did not report statistically significant differences on most maternal beliefs and feeding practices, but mothers of NBW infants did endorse greater social interaction with their infants during feedings (F (1, 32) = 7.12, p = .01). See Table 2 for complete description of participant values on appetite, eating behaviors, and maternal feeding practices and beliefs by child birth weight.
Table 2.
Comparison of NBW and HBW infants on weight, eating behaviors, and maternal feeding practices and beliefs
| NBW (n = 20) |
HBW (n = 21) |
|
|---|---|---|
| M (SD) | M (SD) | |
| Weight | ||
| Current weight (kg) | 8.31 (.88) | 9.17 (.79) |
| Current length (cm) | 68.74 (2.64) | 69.98 (2.02) |
| Weight-for-length percentile | 60.00 (27.42) | 80.87 (16.21) |
| Change in weight per day (g) | 19.12 (3.08) | 20.35 (3.57) |
| Child eating behavior/appetite | ||
| Food responsiveness | 12.15 (3.30) | 11.14 (3.32) |
| Enjoyment of food | 18.05 (1.50) | 18.33 (1.65) |
| General appetite | 3.50 (.83) | 3.43 (.87) |
| Slowness in eating | 9.35 (1.81) | 9.10 (2.14) |
| Satiety responsiveness | 6.20 (1.47) | 6.19 (1.60) |
| Age of introduction of solid foods (months) | 5.50 (1.05) | 5.48 (.94) |
| Maternal feeding practices and beliefs | ||
| Concern about infant undereating or becoming underweight | 1.45 (.43) | 1.33 (.37) |
| Concern about infant hunger | 1.28 (.64) | 1.04 (.11) |
| Awareness of infant’s hunger and satiety cues | 4.39 (.27) | 4.21 (.41) |
| Concerns about infant overeating or becoming overweight | 1.43 (.43) | 1.52 (.51) |
| Feeding infant on a schedule | 2.25 (.97) | 1.98 (1.07) |
| Social interaction with the infant during feeding | 4.30 (.37) * | 3.89 (.49) * |
| Using food to calm infant’s fussiness | 2.73 (.53) | 2.60 (.70) |
Statistically significant results are given in bold
p < .05
Comparison Between HBW Infants with Current Weight-for-Length ≥ 85th Percentile (n = 9) and HBW Infants with a Current Weight-for-Length < 85th Percentile (n = 12)
HBW infants with a current weight-for-length percentile ≥ 85th had a higher average birth weight [4422.00 g compared to 4201.58 g, F (1, 19) = 5.66, p = .03] and greater average weight gain per day [22.76 oz compared to 18.55 oz, F (1, 19) = 10.50, p < .01] than their HBW peers with a current weight-for-length percentile < 85th. Mothers of HBW infants with elevated weight-for-length also reported greater concern about these infants’ hunger (e.g., felt child wanted more than just breastmilk or formula at an early age, used cereal in the child’s bottle, F (1, 17) = 4.30, p = .05). Mothers of HBW infants with a current weight-for-length ≥ 85th percentile also reported lower infant satiety responsiveness (F (1, 19) = 10.60, p < .01) and lower maternal social interaction with the infant during feedings (F (1, 17) = 9.38, p = .01) than mothers of HBW infants currently < 85th percentile. Other infant eating behaviors, appetite characteristics, and maternal feeding practices and beliefs were not significantly different between HBW infants who currently had a weight-for-length percentile ≥ 85th and those infants with a weight-for-length percentile < 85th (Table 3).
Table 3.
Comparison of HBW infants with a current weight-for-length percentile > or < the 85th
| < 85th (n = 12) |
≥ 85th (n = 9) |
|
|---|---|---|
| M (SD) | M (SD) | |
| Weight | ||
| Birth weight (g) | 4201.58* (177.03) | 4422.00* (248.42) |
| Change in weight per day (g) | 18.55 (3.14) * | 22.76 (2.64) * |
| Child eating behavior/appetite | ||
| Food responsiveness | 10.50 (3.03) | 12.00 (3.67) |
| Enjoyment of food | 18.42 (1.68) | 18.22 (1.72) |
| General appetite | 3.33 (.89) | 3.56 (.88) |
| Slowness in eating | 8.75 (2.30) | 9.56 (1.94) |
| Satiety responsiveness | 7.00 (1.35) * | 5.11 (1.27) * |
| Age of introduction of solid foods | 5.71 (.87) | 5.17 (1.00) |
| Maternal feeding practices and beliefs | ||
| Concern about infant undereating or becoming underweight | 1.44 (.44) | 1.19 (.21) |
| Concern about infant hunger | 1.00 (.00) + | 1.10 (.16) + |
| Awareness of infant’s hunger and satiety cues | 4.21 (.51) | 4.22 (.23) |
| Concerns about infant overeating or becoming overweight | 1.56 (.48) | 1.48 (.58) |
| Feeding infant on a schedule | 2.08 (1.12) | 1.83 (1.03) |
| Social interaction with the infant during feeding | 4.14 (.32) * | 3.56 (.50) * |
| Using food to calm infant’s fussiness | 2.63 (.74) | 2.56 (.68) |
Statistically significant results are given in bold
p < .05
p < .10
Observed Bottle Feeding Task
On the observed bottle feeding, HBW and NBW infants did not display statistically significant differences in the length of the feeding, suck rate, rate of sucking bursts, suck duration, burst duration, interval between sucks, or mean sucking pressure/amplitude. Regarding differences between HBW infants with a current weight-for-length percentile ≥ 85th percentile and HBW infants with a current weight-for-length < 85th percentile, trends were present for HBW infants with a current elevated weight-for-length percentile having greater mean sucking pressure/amplitude during the observed feeding (F (1, 7) = 4.19, p = .08). However, infants with a high weight-for-length did not differ statistically from other HBW infants on any other aspects of the observed bottle feeding.
Conclusions for Practice
We sought to identify infant appetite, infant eating behaviors, and maternal feeding practices or beliefs that differentiate HBW and NBW infants and explain the increased risk for subsequent obesity in HBW infants. Further, we sought to identify those HBW children who continue on high weight-for-length trajectories later in infancy. None of the infant appetite, infant eating behaviors, or maternal beliefs explored in the current exploratory study differentiated the birth weight groups, as birth weight groups differed on only one maternal feeding practice (social interaction during feedings). However, HBW infants who continue on the trajectory of having a high weight-for-length percentile at 7–8 months were found to differ on key variables from HBW infants who were below the 85th percentile weight-for-length at 7–8 months.
Interestingly, the greater risk for high weight-for-length percentile among HBW infants compared to NBW infants was not due to differences in weight trajectories postnatally (i.e., average weight change per day since birth) or behavioral differences between HBW and NBW infants. Rather, these HBW infants appear to maintain the additional weight with which they were born. Given the lack of significant behavioral differences in the current exploratory study between HBW and NBW infants, the currently assessed eating behaviors and maternal feeding practices and beliefs do not likely solely explain the mechanism by which this additional weight is maintained. Study results suggest maintenance of this additional weight could be a product of other factors not explored in the current study (e.g., increased caloric intake, metabolic differences that reduce calorie expenditure, possible differences in early physical activity or sleep patterns), which have previously been associated with increased risk for excess weight gain in infancy or childhood (Reilly et al. 2005; Woo Baidal et al. 2016). Mechanistic and physiological factors (e.g., dietary intake, resting energy expenditure) driving HBW infants to maintain a higher weight should be examined in future research. Additionally, while NBW infants were not the focus of the current study aims, an important finding from this exploratory study is that a high number (20%) of NBW infants also met criteria for elevated weight-for-length at 7–8 months of age. Future research should also explore potential mechanistic factors for excess weight and accelerated growth trajectories among all infants regardless of birth weight status.
Although most maternal feeding practices and beliefs did not differ between birth weight groups, mothers of NBW infants did report greater social interaction with their infants during feedings. Lower social interaction between infants and mothers at feedings was a consistent risk-factor in the current exploratory study as it was also found to be lower among HBW infants with a current weight-for-length percentile ≥ 85th compared to their HBW counterparts with a lower current weight-for-length. In the current exploratory study, social interaction was assessed through questions about whether the mother talked or sang to the child during feedings and whether they held the infant during bottle feedings (subscale excluded for infants only fed directly at the breast). These findings are consistent with past research that has found characteristics of mother–child interactions relate to obesity risk in childhood (Skouteris et al. 2012). If future research replicates the finding that increased social interaction at feedings is a protective factor, increasing social interaction during feedings could be one possible future target in weight management prevention and intervention for infants, including for HBW infants. Encouragingly, past research has found maternal feeding practices are amenable to change through educational intervention (Daniels et al. 2013).
Risk factors differentiating children born at HBW who continued to have high weight-for-length trajectories during later infancy were identified. HBW infants who were ≥ 85th percentile for weight-for-length at 7–8 months had an average birth weight above 4400 g, greater weight gain per day since birth, and their mothers reported lower infant satiety responsiveness and lower maternal social interaction with their infants during feedings. This pattern indicates that increased risk for obesity among HBW infants may not be universal, but that there may be a high risk subgroup. Future research should prospectively explore whether these risk factors longitudinally predict obesity risk among HBW infants, given the current exploratory study is cross-sectional in nature.
The extent of excess weight > 4000 g at birth predicted whether a HBW child had a high weight-for-length at 7–8 months. If this finding is replicated in larger samples, this finding may suggest the possibility of differential health outcomes among children within the HBW classification based on the degree of their weight > 4000 g. Interestingly while several subclassifications within the low birth weight classification exist due to differential developmental and health outcomes (i.e., extremely low birth weight, very low birth weight, low birth weight), no such differentiation of subclassifications is present for HBW infants, and all infants > 4000 g are placed into one category. Future research should continue to explore whether there are other differential health outcomes for infants of different weights within the HBW classification; if so, these findings may suggest the need for additional subclassifications.
Study findings should be interpreted within the context of study limitations. First, the current exploratory study was cross-sectional in nature with a small sample size, and therefore causal inferences about the association between variables should be limited until further investigation using a longitudinal design and larger sample size is conducted. Additionally, future longitudinal studies should begin following HBW infants starting at birth, given the current study could have missed earlier mechanistic variables given data were not collected until 7–8 months of age. Further, the pattern of non-significant findings in the current exploratory study could simply be the product of being too under-powered to detect present differences between birth weight groups. Additionally, measures of many aspects of infant appetite, infant eating behavior, and maternal feeding practices and beliefs were collected via maternal-report, and therefore may be subject to reporting bias. Given the recruitment strategies used in the current exploratory study, the sample was over-representative of families who were Non-Hispanic Caucasian and had higher incomes, mothers who had significant post-high school education, mothers who were employed, and infants who were breastfed. Future research should seek to replicate the current exploratory study findings in a larger and more diverse sample and examine how these demographic variables (e.g., maternal employment) may influence the behaviors found to be associated with increased risk for high weight-for-length in the current exploratory study (e.g., social interaction during feedings). Additionally, the classification of infants as HBW or NBW may lack precision compared to other classifications accounting for gestational age, such as weight for gestational age. However, prior research has found that HBW is a risk factor for obesity even without accounting for gestational age (Cunningham et al. 2014; Schellong et al. 2012; Skilton et al. 2014; Yu et al. 2011), and use of HBW as a classification may have greater clinical utility due to greater ease of reporting than gestational age measures, such as large for gestational age (LGA), which require more complex computations. Lastly, infants with less familiarity with bottle feedings may have responded differently to the bottle feeding task than infants who regularly fed from a bottle. This confounding factor may have influenced the results from the bottle feeding task. However, 94% of the participants completing the bottle feeding (15 of 16 participants) did currently bottle feed in their regular feeding routine (i.e., either received formula or breast milk from a bottle).
Future research should seek to replicate whether infants with birth weights well above 4000 g, continued weight gain at a high rate during infancy, lower satiety responsiveness, and lower maternal social interaction at feedings are at increased risk for elevated weight trajectories. If these findings are replicated, these risk factors may be useful screening criteria for clinicians to identify those HBW infants who may benefit from increased efforts to promote healthy weight gain. The finding that both parent (maternal social interaction) and child (satiety responsiveness) factors predicted elevated weight-for-length among HBW infants supports the importance of both parent and infant factors in the development of excess weight during infancy.
Significance.
What is already known on this subject?
High birth weight (HBW) infants are at increased risk for later obesity, but the mechanisms for this increased risk remain unclear as well as whether certain subsets of HBW infants are at greater risk.
What this study adds?
High and normal birth weight infants did not differ on eating and feeding behaviors. However, HBW infants who maintained excess weight did have a higher birth weight, lower satiety responsiveness, and lower maternal social interactions during feedings compared to other HBW infants.
Acknowledgements
We would like to acknowledge funding from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK; T32 DK063929) and the National Center for Advancing Translational Sciences (UL1 TR001425). We would also like to thank the Schubert Research Clinic, especially Suzanne Summer, Jennifer Hunt, and Molly Bowling, for completing participant anthropometric measurements. Many thanks are also due to Children’s for Children, especially Kathy Haders and the teachers and parents in the infant classrooms, for their help with participant recruitment and piloting the NFANT feeding device. We would also like to thank Pediatric Associates PSC for their help with and openness to recruitment at their practice. Thank you to Jared Connor for his contribution to measure design, and Megan Richardson for her involvement in data collection. Lastly, thank you to the families who took time to participate in both this study and also our prior pilot study using the observed feeding methodology.
Footnotes
Conflict of interest The authors declare that they have no conflict of interest.
References
- Agras WS, Hammer LD, McNicholas F, & Kraemer HC (2004). Risk factors for childhood overweight: A prospective study from birth to 9.5 years. The Journal of Pediatrics, 145(1), 20–25. 10.1016/j.jpeds.2004.03.023. [DOI] [PubMed] [Google Scholar]
- Agras WS, Kraemer HC, Berkowitz RI, & Hammer LD (1990). Influence of early feeding style on adiposity at 6 years of age. The Journal of Pediatrics, 116(5), 805–809. [DOI] [PubMed] [Google Scholar]
- Agras WS, Kraemer HC, Berkowitz RI, Korner AF, & Hammer LD (1987). Does a vigorous feeding style influence early development of adiposity? The Journal of Pediatrics, 110(5), 799–804. [DOI] [PubMed] [Google Scholar]
- Baughcum AE, Powers SW, Johnson SB, Chamberlin LA, Deeks CM, Jain A, et al. (2001). Maternal feeding practices and beliefs and their relationships to overweight in early childhood. Journal of Developmental and Behavioral Pediatrics: JDBP, 22(6), 391–408. [DOI] [PubMed] [Google Scholar]
- Cunningham SA, Kramer MR, & Narayan KM (2014). Incidence of childhood obesity in the united states. The New England Journal of Medicine, 370(17), 1660–1661. 10.1056/NEJMc1402397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Daniels LA, Mallan KM, Nicholson JM, Battistutta D, & Magarey A (2013). Outcomes of an early feeding practices intervention to prevent childhood obesity. Pediatrics, 132(1), 109. 10.1542/peds.2012-2882. [DOI] [PubMed] [Google Scholar]
- Druet C, Stettler N, Sharp S, Simmons RK, Cooper C, Smith GD, et al. (2012). Prediction of childhood obesity by infancy weight gain: An individual-level meta-analysis. Paediatric and Perinatal Epidemiology, 26(1), 19–26. 10.1111/j.1365-3016.2011.01213.x. [DOI] [PubMed] [Google Scholar]
- Faith MS, Dennison BA, Edmunds LS, & Stratton HH (2006). Fruit juice intake predicts increased adiposity gain in children from low-income families: Weight status-by-environment interaction. Pediatrics, 118(5), 2066–2075. [DOI] [PubMed] [Google Scholar]
- Flores G, & Lin H (2013). Factors predicting overweight in US kindergartners. The American Journal of Clinical Nutrition, 97(6), 1178–1187. 10.3945/ajcn.112.052019. [DOI] [PubMed] [Google Scholar]
- Harrison K, Bost KK, McBride BA, Donovan SM, Grigsby-Toussaint DS, Kim J, et al. (2011). Toward a developmental conceptualization of contributors to overweight and obesity in childhood: The Six-Cs model. Child Development Perspectives, 5(1), 50–58. 10.1111/j.1750-8606.2010.00150.x. [DOI] [Google Scholar]
- Huh SY, Rifas-Shiman SL, Taveras EM, Oken E, & Gillman MW (2011). Timing of solid food introduction and risk of obesity in preschool-aged children. Pediatrics, 127(3), 544. 10.1542/peds.2010-0740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Llewellyn CH, van Jaarsveld CH, Johnson L, Carnell S, & Wardle J (2010). Nature and nurture in infant appetite: Analysis of the gemini twin birth cohort. The American Journal of Clinical Nutrition, 91(5), 1172–1179. 10.3945/ajcn.2009.28868. [DOI] [PubMed] [Google Scholar]
- McMeekin S, Jansen E, Mallan K, Nicholson J, Magarey A, & Daniels L (2013). Associations between infant temperament and early feeding practices. A cross-sectional study of australian mother-infant dyads from the NOURISH randomised controlled trial. Appetite, 60(1), 239–245. [DOI] [PubMed] [Google Scholar]
- Ong KK, & Loos RJ (2006). Rapid infancy weight gain and subsequent obesity: Systematic reviews and hopeful suggestions. Acta Paediatrica (Oslo, Norway: 1992), 95(8), 904–908. [DOI] [PubMed] [Google Scholar]
- Phillips DIW, & Young JB (2000). Birth weight, climate at birth and the risk of obesity in adult life. International Journal of Obesity, 24(3), 281–287. 10.1038/sj.ijo.0801125. [DOI] [PubMed] [Google Scholar]
- Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, et al. (2005). Early life risk factors for obesity in childhood: Cohort study. BMJ (Clinical Research Ed.), 330(7504), 1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schellong K, Schulz S, Harder T, & Plagemann A (2012). Birth weight and long-term overweight risk: Systematic review and a meta-analysis including 643,902 persons from 66 studies and 26 countries globally. PLoS ONE, 7(10), e47776. 10.1371/journal.pone.0047776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Skilton M, Siitonen N, Würtz P, Viikari J, Juonala M, Seppälä I, et al. (2014). High birth weight is associated with obesity and increased carotid wall thickness in young adults: The cardiovascular risk in young finns study. Arteriosclerosis, Thrombosis, and. Vascular Biology, 34(5), 1064–1068. 10.1161/ATVBAHA.113.302934. [DOI] [PubMed] [Google Scholar]
- Skouteris H, McCabe M, Ricciardelli LA, Milgrom J, Baur LA, Aksan N, & Dell’Aquila D (2012). Parent-child interactions and obesity prevention: A systematic review of the literature. Early Child Development and Care, 182(2), 153–174. 10.1080/03004430.2010.548606. [DOI] [Google Scholar]
- Smego A, Woo JG, Klein J, Suh C, Bansal D, Bliss S, et al. (2017). High body mass index in infancy may predict severe obesity in early childhood. The Journal of Pediatrics, 183, 93.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- van Jaarsveld CH, Boniface D, Llewellyn CH, & Wardle J (2014). Appetite and growth: A longitudinal sibling analysis. JAMA Pediatrics, 168(4), 345–350. 10.1001/jamapediatrics.2013.4951. [DOI] [PubMed] [Google Scholar]
- Woo Baidal JA, Locks LM, Cheng ER, Blake-Lamb TL, Perkins ME, & Taveras EM (2016). Risk factors for childhood obesity in the first 1,000 days: A systematic review. American Journal of Preventive Medicine, 50(6), 761–779. [DOI] [PubMed] [Google Scholar]
- Yu ZB, Han SP, Zhu GZ, Zhu C, Wang XJ, Cao XG, et al. (2011). Birth weight and subsequent risk of obesity: A systematic review and meta-analysis. Obesity Reviews, 12(7), 525–542. 10.1111/j.1467-789X.2011.00867.x. [DOI] [PubMed] [Google Scholar]