This is a preprint copy of the text prior to print review.
For the full published work please see:
Brown, A., & Lee, M. (2012). Breastfeeding during the first year promotes satiety responsiveness in children aged 18–24 months. Pediatric obesity, 7(5), 382-390.
Corresponding Author
Dr. Amy Brown
College of Human and Health Sciences
Swansea University, UK
SA2 8PP
Email:
Breastfeeding during the first year promotes satiety responsiveness
Abstract
Aim: Breastfeeding may reduce childhood risk of overweight. One explanation for this is that the baby-led nature of breastfeeding promotes appetite regulation as the infant has increased control of amount consumed. However the relationship between breastfeeding and later child-eating style is largely unexplored. The aim of this study was to examine the association between infant milk feeding and later child appetite responsiveness.
Methods: 298 mothers reported breastfeeding duration and exclusivity up to six months postpartum when their infant was aged 6 – 12 months old. In phase two mothers completed the satiety responsiveness and food responsiveness scales of the child eating behaviour questionnaire and the Child Feeding Questionnaire. Infant birth and current weight were collected.
Findings: Infants who were breastfed for a longer duration were rated as more satiety responsive (p = .001) although no difference was seen for feeding method at birth. Compared to infants who were formula fed from birth, at least six weeks breastfeeding was required for increased satiety responsiveness to emerge. This relationship was independent of current maternal child-feeding style. Food responsivity was unrelated to any breastfeeding behaviour.
Conclusions: Breastfeeding may promote satiety responsiveness potentially through the baby-led nature of feeding.
Key words: Breastfeeding; Formula-feeding; Satiety responsiveness; Child eating behaviour; Maternal control
Background
Breastfeeding may be an important factor in limiting the risk of overweight and obesity in childhood. Breastfeeding reduces chances of childhood overweight by 15 – 30%, with a longer duration of breastfeeding and exclusive breastfeeding offering the greatest potential reduction (Arenz, Ruckerl, Koletzko & von Kries, 2004; Owen, Martin, Whincup, Smith & Cook, 2005; Harder, Bergmann, Kallischnigg & Plagemann, 2005). Explanations for the protective effect of breastfeeding include slower weight gain (Dewey, Heinig, Nommsen, Peerson & Lonnerdal, 1993; Kramer et al., 2002) and a lower energy intake during the first year (Garza & Butte, 1990, Heinig, Nommsen, Peerson, Lonnerdal & Dewey, 1993). Rapid early weight gain has been associated with later risk of obesity (Ong & Loos, 2006). Breast and formula milk also have different contents and concentrations which may explain differences in growth patterns (Michels, Willett, Graubard, Vaidya, Cantwell, Sansbury & Forman, 2007).
Differences in feeding style between breast and formula fed infants may also affect later weight. In older children a responsive eating style is protective against obesity and overweight. Children who can recognise internal satiety cues and do not respond to external cues to eat when satiated are more likely to be a healthy weight (Benton, 2004). However maternal child feeding style has been shown to affect this in children over the age of two years. Mothers who exert high levels of control over their child’s eating, such as restricting access to foods or pressurising the child to eat are more likely to have a child with eating and weight issues (Ventura & Birch, 2008).
The origins of this self regulation may begin in infancy. Infants who are breastfed may be given greater opportunity to self regulate the amount of energy they consume compared to formula fed infants which in turn may encourage stronger satiety responsiveness and positive weight gain trajectories (Bartok & Ventura, 2009). Whereas formula fed infants are usually presented with meals of a constant volume and energy content, breast milk is not a uniform product, changing in energy content and other properties across a 24-hour period (Nommsen, Lovelady, Heinig, Lönnerdal & Dewey, 1991). Breastfed infants respond to this variation, adapting their intake of milk and feeding patterns accordingly (Dewey & Lonnerdal, 1986) by, for example consuming lower volumes if the milk is higher in fat content (Tyson, 1992). Moreover, differences in the feeding style of mothers who breast or formula feed have been noted.Formula feeding is open to greater levels of caregiver manipulation as the caregiver can view and track amount consumed, and encourage the infant more readily to finish a feed if desired (Dewey, 2001; Fomon et al. 1975). Mothers who formula feed are more likely to use greater manipulation over their infants feeding pattern such as scheduling feeds and encouraging intake of milk compared to mothers who breastfeed (Brown, Raynor & Lee, 2010). Moreover, mothers who breastfeed during the first year exert lower levels of control over their child’s diet when eating solid foods suggesting an overall feeding style low in control (Blissett & Farrow, 2007; Farrow & Blissett, 2008; Farrow & Blissett, 2006).
Breastfed infants may therefore be given greater opportunity to self regulate their intake of energy according to internal hunger and satiety cues (Woolridge, Ingram, & Baum, 1990) which may encourage self regulatory eating behaviours when consuming a solid diet. Data examining this relationship is however sparse. Infants who are fed directly from the breast in early infancy are less likely to empty a cup or bottle in later infancy compared to those who are formula fed or mixed fed (Li, Fein & Gummer-Strawn, 2010). One study also explored child eating behaviour at three to six years with retrospective reports of breastfeeding duration. Infants who were directly breastfed were rated as significantly more satiety responsive than those fed expressed breastmilk in a bottle. No significant difference was found between direct breastfeeding and exclusive formula feeding but sample size was small with a limited number of formula feeders (n = 22). The study also relied on retrospective reports of breastfeeding using medical charts and maternal recall when the child was aged three to six years (DiSantis, Collins, Fisher & Davey, 2011). Both these studies point to the suggestion that breastfeeding may promote a responsive feeding style in later infancy and childhood.
The aim of the current study was therefore to further examine the association between breastfeeding duration and reported child appetite responsiveness during the second year postpartum where children are eating a solid diet.
Method
Participants
All aspects of this study have been performed in accordance with the ethical standards set out in the 1964 Declaration of Helsinki. Approval for this study was granted by a Psychology Research Ethics Committee. All participants gave informed consent prior to inclusion in the study.
The data was collected as part of a longitudinal study examining early influences of feeding style and behaviours during the first year postpartum upon later child eating style and weight. In phase one six hundred and four mothers with an infant aged six to twelve months (mean age 8.34 months) reported data on breastfeeding duration, timing of introduction to complementary foods and experiences of weaning. Phase two data was collected when the infants were now aged 18 – 24 months. Consent was sought from mothers for potential follow up at phase one with three hundred and twenty mothers completing the follow up questionnaire. After exclusion criteria (child health problems or severe issues with weight such as failure to thrive, failure to give consent or incomplete survey entry) two hundred and ninety eight mothers remained in the full analysis (49.5% of original sample). No significant difference was found in maternal age, education or breastfeeding duration in phase one for those mothers who completed the follow up or did not.
Measures
Data was collected using an online questionnaire designed and hosted using SurveyMonkey.com. Mothers who consented to follow up at stage one were sent a link to complete the second part of the study online or offered a paper copy. 94.96% of participants completed the survey using their online link.
Participants completed five scales of the Child Eating Behaviour Questionnaire (CEBQ) [Food responsiveness’, ‘Enjoyment of food’, ‘Satiety responsiveness’, ‘Slowness in eating’ and ‘Food fussiness’] as part of the wider study (Wardle, Guthrie, Sanderson & Rapoport, 2001).Two scales relating to appetite control; Satiety responsiveness and Food Responsiveness are examined here. The ‘Food responsiveness’ scale measures desire of the child to eat in response to food stimuli regardless of how hungry they are. The ‘Satiety responsiveness’ examines ability to regulate intake of food in relation to satiety.
Participants also completed a copy of the Child Feeding Questionnaire (Birch, Fisher, Grimm-Thomas, Markey, Sawyer & Johnson, 2001) specifically the scales of restriction, pressure to eat, monitoring and perceived parental responsibility.
In phase one participants reported whether the infant was breast or formula fed at birth, breastfeeding duration up to time of questionnaire, timing of any use of supplementary formula and timing of introduction to complementary foods.
In phase two participants reported any further breastfeeding duration and whether they were still breastfeeding at the time of questionnaire.
Data analysis
Data analyses were carried out using SPSS v13, SPSS UK Ltd. Data was checked for normal distribution and found adequate. The CEBQ (Wardle et al, 2001) is typically used for preschool aged and older children. Therefore principal components analysis using varimax rotation was performed on the scales to ensure that the original factor structures held within this new sample and age range (Tabachnik and Fidell, 2006). Using a threshold of 0.5 (Nunnally, 1978) and retaining factors with eigenvalues over 1, factors produced mirrored those on the original questionnaires. Cronbach’s alpha was also computed for items loading above the threshold onto each scale and found to be over 0.7 for each scale. Therefore the CEBQ was scored as per original instructions.
Breastfeeding duration was reported in weeks up to current time point (possible two years). Distribution of duration was abnormal with a high proportion of mothers ceasing breastfeeding in the first few days or breastfeeding for at least six months(Kolmogorov-Smirnov = .238 = .000). Therefore breastfeeding duration data was transformed and the natural logarithms computed used in order to correct for the skewed distribution.
MANCOVA examined differences in child eating behaviour according to feeding mode at birth, whether infants were exclusively breastfed for six months and current breastfeeding controlling for maternal age and education. Pearson’s r correlations explored the association between breastfeeding duration, timing of introduction of formula, milk feeding behaviour at 6 – 12 months and timing of introduction to complementary foods. Maternal child feeding style factors were used as covariates. A series of t tests then compared child eating behaviour for those infants formula fed at birth versus duration of breastfeeding (two, four, six, twelve, eighteen and twenty six weeks).
Results
Two hundred and ninety eight mothers remained in the analysis after exclusion criteria. The mean age of the sample was 29.37 (SD:5.55) with a mean number of years in education of 13.05 (SD:2.26) [Table one].
Breastfeeding duration and Child Eating Behaviour.
Mothers reported whether they breastfed (N= 251) or formula fed (N = 47) at birth. No significant differences in food responsiveness and satiety responsiveness were seen between these two groups. However, significant associations were seen between breastfeeding duration and child eating behaviour. A longer duration of breastfeeding was significantly associated with increased satiety responsiveness (Pearson’s r = .134, p = 0.01) but not food responsiveness.
As mothers who breastfeed during the first year have been shown to exert lower levels of control over their child’s diet during solid feeding (which in turn can affect child eating behaviour), the five maternal control scales were placed as a covariate. The relationship between breastfeeding duration and satiety responsiveness remained significant but reduced (Pearson’s r = .118, p = .030).
Current breastfeeding at 18 – 24 months was also examined. Thirty three mothers were still breastfeeding (11.1%). Mothers who were currently breastfeeding reported higher levels of child satiety responsiveness compared to those who were not [t (294) = 2.262, p = .021] independently of maternal control. However when they were compared to mothers who breastfed for at least six months but were no longer breastfeeding (n = 70), there was no significant difference between the two groups for either behaviour.
Considering the influence of specific breastfeeding duration, a series of time point comparisons were performed comparing different lengths of breastfeeding duration to formula feeding from birth. In comparison one, infants who were formula fed from birth (n=47) were compared to those who were breastfed for up to two weeks (n= 42). No significant difference was found in either food responsiveness or satiety responsiveness between these two groups. Similarly, no difference was found when comparing infants who were formula fed from birth to those who were breastfed for three to five weeks (n=31).
However when mothers who breastfed for six to eight weeks (n =36) were compared to those who formula fed, a significant difference in satiety responsiveness was found. Infants who were breastfed for six to eight weeks were significantly more satiety responsive than those formula fed from birth [t (81) = 2.753, p = .007]. A significant difference also occurred between the formula group and those who breastfed for 9 – 12 weeks (n=20) [t(65) = 2.108, p = .038], 13 – 26 weeks (n= 19) [t (64) = 2.357, p = .019] and 26 weeks plus (n=103) [t (158) = 3.307, p = .0001]. No significant difference was found for any group for food responsiveness.
Exclusivity and child eating behaviour
Formula use and timing of timing of introduction to complementary foods were also examined. Mothers who breastfed but also introduced supplementary formula reported age of infant in weeks when formula was first introduced. 78.8% of the sample had used formula at some stage before six months postpartum (n = 198).Infants who had ever been mixed fed were significantly less responsive than those who had only received breast milk (n = 53) but were significantly more responsive than those formula fed from birth [F (1, 285) = 5.385, p = 0.005]. Furthermore, an earlier introduction of formula wassignificantly associated with lower satiety responsiveness (Pearson’s r = .124, p = 0.39). No significant relationship was found with food responsiveness.
Timing of introduction to complementary foods was also significantly associated with greater satiety responsiveness (Pearson’s r = .198, p = 0.00) but not food responsiveness. This difference remained when breastfeeding duration was controlled for. In line with this, exclusive breastfeeding was examined as per World Health Organisation guidelines to follow exclusive breastfeeding for the first six months postpartum. Mothers who exclusively breastfed were identified by breastfeeding duration of twenty six weeks or more, no introduction of supplementary formula during this period and an introduction of complementary food at twenty six weeks or later. Thirty six mothers (12.1%) of the sample were classed as having exclusively breastfed. Children who were exclusively breastfed had significantly higher levels of satiety responsiveness compared to those who were not [t (292) = 2.262, p = .024]. No significant difference occurred for food responsiveness.
Breastfeeding, infant weight and child eating behaviour
Mothers self reported child weight at the time of the questionnaire. No significant assocaition was seen between child weight and child eating behaviour, nor between child weight and breastfeeding duration.
Discussion
This paper reports a significant association between breastfeeding duration during the first year and later child satiety responsiveness measured at 18 – 24 months. It examines infant feeding mode at birth, breastfeeding duration and exclusivity finding that satiety responsiveness increases in relation to both duration and exclusivity. Previous research has shown that children who have greater levels of satiety responsiveness are at a lower risk of childhood overweight (Webber, Hill, Saxton, Van Jaarsveld, & Wardle). The findings are thus important in considering how breastfeeding may play a protective role against obesity and highlight the importance in promoting breastfeeding to new mothers.
Breastfeeding duration was significantly associated with satiety responsiveness echoing previous findings by DiSantis et al (2011). Infants who were breastfed and for a longer duration were rated as more able to regulate their appetite in accordance with energy intake. This fits well within a discourse which suggests that experience of breastfeeding enables infants to learn to regulate their appetite. As breastmilk changes in energy density over the course of the day (Nommsen et al., 1991) and amount consumed cannot be easily tracked, infants may learn to match their intake more closely to need. In comparison, formula fed infants are more likely to be presented with feeds at regular times and of regular volumes and energy density leading them to learn to respond to external cues of intake rather than internal satiation. Moreover, mothers who breastfeed are more likely to follow infant demands and cues of satiety whilst mothers who formula feed are more likely to manipulate energy intake (Brown et al., 2011). Over time, this combination of factors may promote stronger internal appetite regulation in the breastfed infant.
This relationship occurred independently of current maternal child-feeding style. Research has shown that mothers who breastfeed during the first year are more likely to adopt a feeding style low in control during solid feeding (Blissett & Farrow, 2007; Farrow & Blissett, 2006; Farrow & Blissett, 2008). Thus, breastfeeding duration may have been an artefact of the impact of later maternal control on child eating style. However the relationship remained significant suggesting that breastfeeding has an independent influence. Potentially the experience of baby-led breastfeeding encourages mothers to adopt a feeding style low in control (Taveras, Scanlon, Birch, Rifas-Shiman, Rich-Edwards & Gillman, 2004) or alternatively mothers who are generally low in control are more likely to breastfeed (Brown et al, 2011). The effect is likely to be additive but highlights the importance of early infant feeding choices.