VOLUME 16 NUMBER 11 | NOVEMBER 2008 | www.obesityjournal.org
nature publishing group
Do Differences in Childhood Diet Explain
the Reduced Overweight Risk in
Salome Scholtens1, Bert Brunekreef1,2, Henriette A. Smit2,3, Gerrie-Cor M. Gast3, Maarten O. Hoekstra4,
Johan C. de Jongste5, Dirkje S. Postma6, Jorrit Gerritsen7, Jaap C. Seidell8 and Alet H. Wijga3
Breastfeeding has been associated with a reduced risk of overweight later in life. This study investigates whether
differences in diet and lifestyle at 7 years of age between breastfed and formula-fed children can explain the
difference in overweight prevalence at 8 years of age. We studied 2,043 Dutch children born in 1996–1997 who
participated in the Prevention and Incidence of Asthma and Mite Allergy birth cohort study. Data on breastfeeding
duration and diet and lifestyle factors at 7 years were collected using questionnaires. Weight and height were
measured at 8 years. Overweight was defined according to international gender- and age-specific standards.
Compared to nonbreastfed children (15.5%, n = 316), children breastfed for >16 weeks (38.0%, n = 776) consumed
fruit and vegetables significantly more often and meat, white bread, carbonated soft drinks, chocolate bars, and fried
snacks less often. Overall, breastfed children were less likely to have an unhealthy diet (adjusted prevalence ratio:
0.77, 95% confidence interval: 0.61–0.98). The associations could only partly be explained by maternal education,
maternal overweight, and smoking during pregnancy. At 8 years, 14.5% (n = 297) of the children were overweight.
Breastfeeding for >16 weeks was significantly associated with a lower overweight risk at 8 years (adjusted odds ratio:
0.67, 95% confidence interval: 0.47–0.97), and the association hardly changed after adjustment for diet (adjusted
odds ratio: 0.71, 95% confidence interval: 0.49–1.03). Breastfed children had a healthier diet at 7 years compared to
nonbreastfed children, but this difference could not explain the lower overweight risk at 8 years in breastfed children.
Obesity (2008) 16, 2498–2503. doi:10.1038/oby.2008.403
Breastfeeding has been associated with a reduced risk of over-
weight in childhood and adulthood (1,2). An important question
is, whether this association is due to a healthier diet and lifestyle
of breastfed children compared to nonbreastfed children.
Breastfeeding initiation and continuation depends on the
choice of the mother and is associated with various mater-
nal characteristics (3–6). Studies in the Netherlands and the
United Kingdom showed that the main reason for mothers to
breastfeed their child was because they believed breastfeeding
was healthier than formula feeding (3,4). Mothers who choose
to breastfeed may, therefore, also differ in other health-related
behavior from mothers who choose to formula feed and hence
influence their children’s diet and lifestyle.
Taveras et al. showed that mothers who breastfed were less
restrictive in respect to their children’s food intake at 1 year
of age, and both breastfeeding and the less restrictive behav-
ior were associated with a lower BMI at 3 years of age (6,7).
However, the association between breastfeeding and BMI was
only minimally attenuated by the less restrictive behavior.
The aim of this study was to assess the association between
breastfeeding and children’s diet and lifestyle at 7 years of
age, and to investigate to what extent differences in diet and
lifestyle factors between breastfed and nonbreastfed children
could explain the association between breastfeeding and over-
weight at 8 years of age.
Methods And Procedures
study design and study population
For this study, we used data from 2,043 Dutch children who were
born in 1996–1997 and participated in the Prevention and Incidence
of Asthma and Mite Allergy birth cohort study. A detailed description
of the study design has previously been published (8). At baseline,
1Institute for Risk Assessment Sciences, Utrecht University, Utrecht, The Netherlands; 2Julius Center for Health Sciences and Primary Care, University Medical Center
Utrecht, Utrecht, The Netherlands; 3Centre for Prevention and Health Services Research, National Institute for Public Health and the Environment, Bilthoven, The Netherlands;
4Centre for Paediatric Allergology, Wilhelmina Children’s Hospital, Utrecht, The Netherlands; 5Department of Pediatrics, Division of Pediatric Respiratory Medicine, Erasmus
University Medical Center/Sophia Children’s Hospital, Rotterdam, The Netherlands; 6Department of Pulmonology, University Medical Centre Groningen, University of
Groningen, Groningen, The Netherlands; 7Department of Pediatric Respiratory Medicine, Beatrix Children’s Hospital, University Medical Centre Groningen, University of
Groningen, Groningen, The Netherlands; 8Institute of Health Sciences, Vrije Universiteit, Amsterdam, The Netherlands. Correspondence: Alet H. Wijga (email@example.com)
Received 3 March 2008; accepted 15 May 2008; published online 28 August 2008. doi:10.1038/oby.2008.403
obesity | VOLUME 16 NUMBER 11 | NOVEMBER 2008 2499
the study population consisted of 4,146 mothers who were recruited
from the general population during pregnancy. At 8 years of age, 3,668
children (88% of 4,146) were still participating in the Prevention and
Incidence of Asthma and Mite Allergy study; 3,518 children were
invited for the medical examination; and 2,214 children participated in
the medical examination and were weighed and measured. The main
reason for not inviting children was that they had moved and were now
living too far away from the study centers. After exclusion of twins and
children born prematurely (n = 116), children with missing data on
breastfeeding (n = 16), on all dietary data at 7 years of age (n = 32) or on
maternal education (n = 7), 2,043 children were available for analyses.
Data on breastfeeding, diet, lifestyle, and confounding factors were
collected by means of postal questionnaires, which were sent to the
parents during pregnancy, 3 months after the child was born, and from
age 1 year annually. The study protocol was approved by the medical
ethics committees of the participating institutes, and all parents gave
written informed consent.
Breastfeeding and confounder assessment
Duration of breastfeeding was assessed by questions on infant feeding
in the questionnaires administered at 3 months and at 1 year of age. At
3 months, we asked the parents how long the child was breastfed and
whether the mother was still breastfeeding. If the parents indicated in
the questionnaire at 3 months that the mother was still breastfeeding,
we used data from the 1-year questionnaire to assess the total breast-
feeding duration. The answers on these questions were combined to
one variable and categorized as “no breastfeeding,” “1–16 breastfeed-
ing,” and “>16 weeks breastfeeding.” The cutoff point of 16 weeks was
chosen, because until recently, the recommended breastfeeding dura-
tion in the Netherlands was 4–6 months, and only a small percent-
age of the mothers breastfed for at least 6 months. Breastfeeding was
defined as any kind of breastfeeding, including partial breastfeeding.
Maternal educational level was measured as the highest education com-
pleted and then divided into three categories (low, intermediate, and high
education). Employment of the mother when the child was 1 year of age
was classified as employed or unemployed. Maternal BMI (kg/m2) was
calculated from self-reported body weight (kg) and height (m) when the
child was 1 year of age, and overweight was defined as BMI ≥25 kg/m2.
Maternal smoking during pregnancy was defined as any smoking by the
mother during pregnancy after the fourth week of pregnancy.
dietary and lifestyle assessment
The questionnaire at 7 years of age contained a food-frequency ques-
tionnaire. The parents were asked to report how often in the previous
month their child consumed a certain food or drink. Response options
were: “not at all,” “less than once a week,” “at 1 or 2 days per week,” “at
3–5 days per week,” and “at 6 or 7 days per week.” In the analysis of the
consumption frequency of individual food items, daily (6–7 times per
week) consumption was compared to nondaily (less than 6 times per
week). If the food item was only consumed daily by a small number of
children (<10%), the three highest consumption frequency categories
were taken together to make a classification in weekly (1–7 times per
week) and nonweekly (less then once per week).
To create a variable that indicated whether the child had a low fruit and
vegetable consumption the response on the variables “daily fruit consump-
tion,” “daily cooked vegetable consumption,” and “weekly uncooked veg-
etable consumption” were combined. Children who scored negative on two
out of three variables were regarded as having a low fruit and vegetable con-
sumption. To create a variable that indicated whether the child consumed
unhealthy foods (i.e., snacks) frequently the responses on the variables
weekly consumption of “carbonated soft drinks,” “chocolate bars,” “fried
snacks or chips,” and “crisps or salty snacks” were combined. Children who
scored positive on two out of four variables were regarded as having high
snack consumption. Children who had a low fruit and vegetable consump-
tion in combination with a high snack consumption were considered to
have a less favorable (unhealthy) diet at 7 years of age.
Parents also reported if the child consumed any organically grown
or bred products. Furthermore, the questionnaire contained questions
regarding the number of hours a day the child watched television,
watched videos or played computer games, and the number of hours
the child played actively. The four response options that were avail-
able, ranging from less than half an hour a day till >2 h a day, were
classified as <1 h per day and >1 h per day. Watching television, watch-
ing videos, or playing computer games were categorized as “television
During the medical examination at 8 years, children were weighed
and measured in their underwear. Weight was measured to 0.1 kg and
height to 0.1 cm by trained research staff using calibrated measuring
equipment. From the weight and height measurements, BMI (weight
(kg)/height (m)2) was calculated. Overweight was defined accord-
ing to age and gender specific international standards that use cutoff
points equivalent to the 25 kg/m2 cutoff that is commonly used for
adults (9). We use the term “overweight” for the total group of chil-
dren who are either moderately overweight or obese.
Although some children had missing data on some of the dietary or
lifestyle variables, the number of missing values per variable did not
exceed 1% of the total study population. To account for missing data on
the confounding factors, that is, maternal smoking during pregnancy
and maternal overweight, these variables contained a category for
observations with missing data. Log linear regression analysis was used
to analyze the association between the exposure variable breastfeeding
and the outcome variables diet and lifestyle. Logistic regression analysis
was less feasible because the probability of the outcome was relatively
high. A logistic regression would overestimate the relative risk and pro-
duce invalid confidence limits in that case (10,11). We used the PROC
GENMOD procedure in SAS software version 9.1 (SAS Institute, Cary,
NC) to calculate the prevalence ratio and 95% confidence interval (11).
All dietary or lifestyle variables were included separately in the regres-
sion models. Maternal education, maternal overweight, maternal
smoking during pregnancy, maternal age, and maternal employment
were considered as potential confounders. Interactions between breast-
feeding and maternal education were tested.
To analyze the association between the exposure variable breast-
feeding and the outcome variable overweight at 8 years of age, logistic
regression was used. The association between breastfeeding and over-
weight at 8 years of age was additionally adjusted for the dietary factors
that were significantly associated with breastfeeding. If both a combina-
tion variable (low fruit and vegetable consumption) and the individual
variable (daily fruit consumption) were significantly associated with
breastfeeding, the combination variable was included in the model.
P values below 0.05 were considered to be statistically significant.
More than 84.5% (n = 1,727) of the mothers breastfed, and
38.0% of the mothers (n = 776) breastfed their child for >16
weeks (Table 1). Mothers who breastfed for >16 weeks were
more often highly educated and were less likely to be over-
weight and to smoke during pregnancy than mothers who did
not breastfeed or who breastfed for <16 weeks (Table 1).
Comparison of the children included in the analyses with the
excluded children showed that mothers of children included in
the analyses were more often highly educated and were more
likely to breastfeed.
At 7 years of age, children breastfed for >16 weeks consumed
brown bread, fruit, cooked vegetables, and uncooked vegeta-
bles statistically significantly more often and sugar-sweetened
VOLUME 16 NUMBER 11 | NOVEMBER 2008 | www.obesityjournal.org
milk products, meat, white bread, carbonated soft drinks,
chocolate bars, fried snacks, and crisps less often than non-
breastfed children (Tables 2 and 3). Breastfed children were
less likely the have a low fruit and vegetable consumption,
a high snack consumption, or an unhealthy diet. The asso-
ciations attenuated after adjustment for maternal education,
maternal smoking during pregnancy, and maternal over-
weight, but remained significant, except for the association
with uncooked vegetables and fruit. Breastfed children were
significantly more likely to consume organic products and
were less likely to watch >1 h of television per day than chil-
dren who were not breastfed.
In most cases, maternal education accounted for the great-
est proportion of attenuation in effect estimates of the asso-
ciation between breastfeeding and consumption frequencies
of the food items. Additional adjustment for maternal age
and maternal employment did not affect the associations. The
interaction between breastfeeding and maternal education was
not statistically significant. Stratification by maternal educa-
tion demonstrated that the associations between breastfeed-
ing and dietary and lifestyle factors were apparent in all three
education categories (Figures 1 and 2). Figures 1 and 2 show
that the percentage of children with an unhealthy diet and the
percentage of children that watched television for >1 h per day
decreased with increasing maternal education level and with
increasing breastfeeding duration.
table 1 Prevalences and means (s.d.) of various maternal
and children’s characteristics by breastfeeding duration
(n = 2,043)
2,04315.5(316) 46.6 (951) 38.0
Gender (% girls) 50.0 (1,021) 45.9(145) 49.5 (471) 52.2
25.8 (495) 30.6(90)28.6 (252) 20.6
16.1(326)21.8(68)18.3 (172) 11.2
67.3 (1,356) 61.9(192) 73.5 (686) 62.1
19.8(405)34.5(109) 21.2 (202) 12.1
42.4(866)46.5(147) 43.9 (417) 38.9
37.8(772) 19.0(60)34.9 (332) 49.0
child at 8 years
14.5(297) 20.9(66) 14.1 (134) 12.5
Mean(s.d.) Mean (s.d.) Mean (s.d.) Mean (s.d.)
Age child at diet
7.1(0.1) 7.1(0.1)7.1 (0.1)7.1
30.7(3.8)30.3(3.8) 30.5 (3.9)31.3
table 2 Prevalence of children’s consumption frequencies of
various food items at 7 years of age by breastfeeding duration
(n = 2,043)
(n = 316)
(n = 951)
(n = 776)
Brown bread 74.1(234) 80.6(766)84.9(659)
Fresh fruit49.7(157) 56.3 (535)61.9(480)
Fruit juice24.7 (78) 23.0(219)26.7(207)
Lemonade51.0(161)48.5 (461)49.2 (382)
Cookie or biscuit39.2(124) 37.2(354) 39.6(307)
60.4 (191) 58.5(556) 59.0(458)
Chocolate bars25.6(81) 19.7(187)15.2(118)
Fried snacks or
30.1 (95)26.9 (256)19.1(148)
Crisps or salty
63.0 (199)57.9(551) 54.5(423)
55.4 (175) 50.0(475)38.3 (297)
48.1(152) 41.2 (392)34.0 (264)
Unhealthy diet 29.8(94) 22.8(217)17.0 (132)
12.3 (39) 20.7 (197)33.8 (262)
Physical activity child (>1 h per day vs. <1 h per day)
49.9 (154)42.7 (402)36.8 (283)
Active playing 54.3 (171)56.0 (526)53.4 (410)
aDaily consumption: 6–7 times per week vs. less than 6 times per week; weekly
consumption: 1–7 times per week vs. less than once per week. bPulses include
beans, peas, and lentils.
obesity | VOLUME 16 NUMBER 11 | NOVEMBER 2008 2501
Breastfeeding was associated with a variety of diet and life-
style factors at 7 years of age, which could only partially be
explained by maternal characteristics. These differences in diet
and lifestyle between breastfed and nonbreastfed children did
not explain the lower overweight prevalence in breastfed chil-
dren at 8 years of age.
strengths and limitation
Strengths of this study are the prospective study design, the
large study population, and the availability of measured data
on weight and height. Also, we were able to study a large variety
At 8 years of age, 14.5% (n = 297) of the children were
overweight. Children breastfed for 1–16 weeks and children
breastfed for >16 weeks were significantly less likely to be
overweight compared to nonbreastfed children (Table 4).
Additional adjustment for factors that were significantly
associated with breastfeeding (i.e., daily consumption of
brown bread, meat and sugar-sweetened milk products, and
weekly consumption of white bread, a low fruit and veg-
etable consumption, a high snack consumption, consump-
tion of organic products, and television watching at 7 years)
hardly changed the association between breastfeeding and
table 3 Associations between breastfeeding duration and children’s consumption frequencies of various food items at 7 years
of age (no breastfeeding is reference, n = 2,043)
1–16 weeks (n = 951) >16 weeks (n = 776)
Crude PR (95% CI)b
Adjusteda PR (95% CI)b
Crude PR (95% CI)b
Adjusteda PR (95% CI)b
Skimmed or low-fat milk1.01 (0.90; 1.13)0.98 (0.87; 1.09)1.07 (0.95; 1.20) 0.98 (0.88; 1.11)
Sugar-sweetened milk products0.81 (0.69; 0.95)* 0.83 (0.71; 0.98)*0.73 (0.62; 0.87)**0.76 (0.64; 0.91)*
Meat0.91 (0.81; 1.02) 0.92 (0.82; 1.04) 0.83 (0.74; 0.94)*0.87 (0.76; 0.98)*
Brown bread1.09 (1.01; 1.17)*1.07 (1.00; 1.14) 1.15 (1.07; 1.23)**1.08 (1.01; 1.16)*
Fresh fruit1.13 (1.00; 1.28)*1.07 (0.95; 1.21) 1.24 (1.10; 1.41)*1.12 (1.00; 1.27)
Cooked vegetables 1.10 (0.91; 1.33)1.05 (0.87; 1.27) 1.42 (1.18; 1.71)**1.31 (1.09; 1.58)*
Fruit juice0.93 (0.75; 1.17)0.91 (0.72; 1.14)1.08 (0.86; 1.36)1.04 (0.83; 1.31)
Lemonade0.95 (0.84; 1.08)0.96 (0.85; 1.09)0.97 (0.85; 1.10)0.99 (0.86; 1.13)
Cookie or biscuit0.95 (0.81; 1.11) 0.92 (0.78; 1.08)1.01 (0.86; 1.19)0.95 (0.80; 1.13)
Candy or chocolate0.97 (0.87; 1.07)0.96 (0.87; 1.07) 0.98 (0.88; 1.09)0.96 (0.86; 1.07)
Full-fat milk0.85 (0.57; 1.28) 0.88 (0.59; 1.32) 1.17 (0.79; 1.73)1.27 (0.84; 1.90)
Fish 1.06 (0.89; 1.26)1.01 (0.85; 1.20) 1.17 (0.99; 1.40)1.09 (0.91; 1.30)
White bread0.72 (0.61; 0.85)**0.78 (0.67; 0.92)* 0.65 (0.55; 0.77)**0.74 (0.62; 0.89)**
Pulses 0.99 (0.84; 1.16)0.97 (0.83; 1.14) 1.07 (0.91; 1.26)1.04 (0.88; 1.23)
Uncooked vegetables1.10 (0.98; 1.25) 1.02 (0.91; 1.15) 1.22 (1.08; 1.38)*1.09 (0.96; 1.23)
Carbonated soft drinks0.79 (0.66; 0.95)*0.83 (0.70; 1.00)*0.72 (0.59; 0.87)* 0.81 (0.67; 0.99)*
Chocolate bars0.77 (0.61; 0.96)*0.83 (0.66; 1.05)0.59 (0.46; 0.76)**0.70 (0.54; 0.90)*
Fried snacks or chips0.90 (0.73; 1.09)0.99 (0.82; 1.21) 0.63 (0.51; 0.79)**0.79 (0.63; 0.99)*
Crisps or salty snacks0.92 (0.83; 1.02)0.93 (0.84; 1.03)0.87 (0.78; 0.96)* 0.89 (0.79; 0.99)*
Low fruit/vegetable consumption0.90 (0.80; 1.01) 0.93 (0.81; 1.06)0.69 (0.60; 0.79)**0.83 (0.73; 0.95)*
High snack consumption 0.86 (0.75; 0.98)*0.98 (0.88; 1.10)0.71 (0.61; 0.82)**0.83 (0.71; 0.96)*
Unhealthy diet0.77 (0.62; 0.94)*0.89 (0.73; 1.09)0.57 (0.45; 0.72)**0.77 (0.61; 0.98)*
Consumption of organic products 1.68 (1.22; 2.31)*1.39 (1.02; 1.91)*2.74 (2.01; 3.73)**1.99 (1.46; 2.70)**
Physical activity child (>1 h per day vs. <1 h per day)
Television watching0.87 (0.76; 1.00)*0.91 (0.80; 1.04)0.75 (0.65; 0.87)** 0.82 (0.71; 0.96)*
Active playing1.03 (0.92; 1.16)1.04 (0.93; 1.17)0.98 (0.87; 1.11)1.00 (0.88; 1.14)
aAdjusted for maternal education, maternal overweight, and maternal smoking during pregnancy. bPR (95% CI): prevalence ratio and 95% confidence interval. cDaily
consumption: 6–7 times per week vs. less than 6 times per week; weekly consumption: 1–7 times per week vs. less than once per week. *P < 0.05; **P < 0.001.
VOLUME 16 NUMBER 11 | NOVEMBER 2008 | www.obesityjournal.org
The children included in the study had more often mothers
who had a high educational level than the children excluded
from the analyses. However, the reported associations proba-
bly can be generalized to the total study population, because no
effect modification by maternal educational level was observed
in the associations.
Findings of other studies
Three studies have previously been published on the association
between breastfeeding and diet of children (12–14). Lande et al.
(13) reported similar associations between breastfeeding and the
consumption of meat and sugar-sweetened drinks in Norway.
However, their study population consisted of 1-year-old children
and some children still received breast milk. Cooke et al. (12)
focused on fruit and vegetable consumption among 2–6-year-old
children in London. They reported a higher consumption fre-
quency of fruit and vegetables among breastfed children, which is
in accordance with our results. Toschke et al. (14) studied physi-
cal activity, television watching, and fruit consumption among
Czech schoolchildren who were either breastfed or formula fed.
They observed that breastfed children were less likely to watch
television and consumed fruit more often, but these differences
could not explain differences in overweight prevalence in chil-
dren of 6–14 years. Victora et al. (15) did not observe an associa-
tion between breastfeeding and type of diet and physical activity
in 18-year-old men and saw no effect of these variables on the
association between breastfeeding and obesity at 18 years of age.
One study reported a considerable attenuation of the significant
association between breastfeeding and obesity at 7 years after
adjustment for confounding factors, including the dietary pattern
at 7 years of age (16).
Interpretation of the results
The observed association between breastfeeding and diet might
be explained by psychosocial as well as biological factors. As
of dietary variables. Because breastfeeding duration, diet, and
lifestyle were reported prospectively, the probability of recall
bias was eliminated.
A limitation of the study is that the food-frequency question-
naire contained a limited number of foods and drinks. However,
most foods and drinks associated with childhood overweight
were included in the questionnaire. Also, no information was
available on portion sizes, which made it not possible to obtain
detailed information on the child’s food, nutrient, and energy
intake. The assessment of the number of hours that the child
watched television and played actively was limited and did
not include a wide range of different forms of physical activity.
Watching television is an important form of inactivity at that
age and probably is an indicator of an inactive lifestyle.
Information bias is a concern because data on diet and lifestyle
of the child were reported by the parents. Mothers who breastfed
their child might have been more health conscious and might have
had more knowledge on healthy dietary behavior. This could have
influenced their responses and thus result in an overestimation of
the consumption frequency of healthy foods and an underestima-
tion of unhealthy foods in breastfed children. As the question-
naire’s main focus was on asthma and allergy and not on diet and
lifestyle, it is less likely that parents were tempted to give desirable
answers on the questions on diet and lifestyle of the child.
% Children that watched television >1 h per day70
0 weeks breastfeeding
1–16 weeks breastfeeding
>16 weeks breastfeeding
Figure 2 Percentage of 7-year-old children who watched television for
>1 h per day by breastfeeding duration and maternal education.
table 4 Adjusted associations between breastfeeding and
overweight at 8 years of age (no breastfeeding is reference,
n = 2,043)
(n = 316)
(n = 951)
(n = 776)
OR (95% CI)OR (95% CI)
Crude association—0.62 (0.45; 0.86)*0.54 (0.38; 0.76)**
— 0.66 (0.47; 0.92)*0.67 (0.47; 0.97)*
for diet and
at 7 yearsb
— 0.67 (0.48; 0.95)*0.71 (0.49; 1.03)
CI, confidence interval; OR, odds ratio.
aAdjusted for maternal education, maternal overweight, and maternal smoking
during pregnancy. bAdditionally adjusted for daily consumption of brown bread,
meat and sugar-sweetened milk products, and weekly consumption of white
bread, a low fruit and vegetable consumption, a high snacks consumption,
consumption of organic products, and television watching at 7 years of age. *P < 0.05;
**P < 0.001.
% Children with unhealthy diet
0 weeks breastfeeding
1–16 weeks breastfeeding
>16 weeks breastfeeding
Figure 1 Percentage of 7-year-old children with an unhealthy diet
(low fruit and vegetable consumption in combination with a high snack
consumption) by breastfeeding duration and maternal education.
obesity | VOLUME 16 NUMBER 11 | NOVEMBER 2008 2503 Download full-text
1. Harder T, Bergmann R, Kallischnigg G, Plagemann A. Duration of
breastfeeding and risk of overweight: a meta-analysis. Am J Epidemiol
2. Owen CG, Martin RM, Whincup PH, Smith GD, Cook DG. Effect of infant
feeding on the risk of obesity across the life course: a quantitative review of
published evidence. Pediatrics 2005;115:1367–1377.
3. Lanting CI, Van Wouwe JP, Reijneveld SA. Infant milk feeding practices in
the Netherlands and associated factors. Acta Paediatr 2005;94:935–942.
4. Hamlyn, B, Brooker, S, Oleinikova, K, Wands, S. Infant feeding 2000. A
survey conducted on behalf of the Department of Health, the Scottish
Executive, the National Assembly for Wales and the Department of Health,
Social Services and Public Safety in Northern Ireland. 2002. London, TSO.
5. Riva E, Banderali G, Agostoni C, Silano M, Radaelli G, Giovannini M et al.
Factors associated with initiation and duration of breastfeeding in Italy. Acta
6. Taveras EM, Scanlon KS, Birch L, Rifas-Shiman SL, Rich-Edwards JW,
Gillman MW et al. Association of breastfeeding with maternal control of
infant feeding at age 1 year. Pediatrics 2004;114:e577–e583.
7. Taveras EM, Rifas-Shiman SL, Scanlon KS, Grummer-Strawn LM,
Sherry B, Gillman MW et al. To what extent is the protective effect of
breastfeeding on future overweight explained by decreased maternal feeding
restriction? Pediatrics 2006;118:2341–2348.
8. Brunekreef B, Smit J, de Jongste J et al. The prevention and incidence of
asthma and mite allergy (PIAMA) birth cohort study: design and first results.
Pediatr Allergy Immunol 2002;13( Suppl 15):55–60.
9. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition
for child overweight and obesity worldwide: international survey. BMJ
10. Greenland S. Model-based estimation of relative risks and other
epidemiologic measures in studies of common outcomes and in case-
control studies. Am J Epidemiol 2004;160:301–305.
11. Spiegelman D, Hertzmark E. Easy SAS calculations for risk or prevalence
ratios and differences. Am J Epidemiol 2005;162:199–200.
12. Cooke LJ, Wardle J, Gibson EL, Sapochnik M, Sheiham A, Lawson M
et al. Demographic, familial and trait predictors of fruit and vegetable
consumption by pre-school children. Public Health Nutr 2004;7:
13. Lande B, Andersen LF, Veierod MB et al. Breast-feeding at 12 months of
age and dietary habits among breast-fed and non-breast-fed infants.
Public Health Nutr 2004;7:495–503.
14. Toschke AM, Vignerova J, Lhotska L, Osancova K, Koletzko B, Von Kries R
et al. Overweight and obesity in 6- to 14-year-old Czech children in 1991:
protective effect of breast-feeding. J Pediatr 2002;141:764–769.
15. Victora CG, Barros F, Lima RC, Horta BL, Wells J. Anthropometry and body
composition of 18 year old men according to duration of breast feeding:
birth cohort study from Brazil. BMJ 2003;327:901.
16. Reilly JJ, Armstrong J, Dorosty AR et al. Early life risk factors for obesity in
childhood: cohort study. BMJ 2005;330:1357.
17. Kramer MS, Guo T, Platt RW et al. Breastfeeding and infant growth: biology
or bias? Pediatrics 2002;110:343–347.
18. Mennella JA, Beauchamp GK. Maternal diet alters the sensory qualities of
human milk and the nursling’s behavior. Pediatrics 1991;88:737–744.
19. Birch LL. Development of food preferences. Annu Rev Nutr 1999;19:41–62.
20. Sullivan SA, Birch LL. Infant dietary experience and acceptance of solid
foods. Pediatrics 1994;93:271–277.
21. Scholtens S, Gehring U, Brunekreef B et al. Breastfeeding, weight gain
in infancy, and overweight at seven years of age: the prevention and
incidence of asthma and mite allergy birth cohort study. Am J Epidemiol
22. Heinig MJ, Nommsen LA, Peerson JM et al. Energy and protein intakes
of breast-fed and formula-fed infants during the first year of life and their
association with growth velocity: the DARLING Study. Am J Clin Nutr
23. Owen CG, Whincup PH, Odoki K, Gilg JA, Cook DG. Infant feeding and
blood cholesterol: a study in adolescents and a systematic review. Pediatrics
24. Rich-Edwards JW, Stampfer MJ, Manson JE et al. Breastfeeding during
infancy and the risk of cardiovascular disease in adulthood. Epidemiology
25. Martin RM, Gunnell D, Smith GD. Breastfeeding in infancy and blood
pressure in later life: systematic review and meta-analysis. Am J Epidemiol
Kramer et al. (17) put it in their paper, mothers who breastfeed
may be more “health conscious” or more “nutrition conscious”
and offer their child healthy foods more frequently. This
“health consciousness” can probably not fully be explained by
differences in maternal education, maternal overweight, and
maternal smoking during pregnancy. A biological explanation
for the findings of this study could be that breastfed children,
in contrast to children who were formula fed, are exposed to
flavors from the maternal diet that are transmitted to the milk.
Therefore, breastfed children are used to a variety of flavors
(18) and are more likely to accept new foods in early life, in
particular fruits and vegetables (19,20). However, the differ-
ence we observed in this study between breastfed and non-
breastfed children in the number of hours the child watched
television cannot be explained by biologic factors. This obser-
vation points again in the direction of different lifestyle choices
Breastfeeding for >16 weeks was associated with a lower
overweight risk compared to nonbreastfeeding. In spite of the
large differences in diet and lifestyle factors between breast-
fed and nonbreastfed children, as observed in this study,
these differences did not explain the lower overweight risk in
breastfed children. This finding suggests that the association
between breastfeeding and overweight could not be attributed
to the difference in diet between breastfed and nonbreastfed
children, although a study with detailed data on nutrient and
energy intake is needed to confirm our results.
The lower overweight prevalence at 8 years of age among breast-
fed children might be a consequence of a lower weight gain dur-
ing infancy. In our previously published paper we observed that
breastfed children had a lower weight gain during the first year of
life and a lower BMI at 7 years of age (21). The lower weight gain
of breastfed children in the first year of life might be caused by a
lower milk intake compared to nonbreastfed children (22).
The follow-up till 8 years of age could have been too short to
observe attenuation by recent dietary habits on the association
between breastfeeding and overweight. For disease outcomes
in adulthood that are associated with breastfeeding, such as
overweight (1,2) and cardiovascular diseases (23–25), con-
founding by dietary and lifestyle factors might be more impor-
tant due to prolonged exposure.
The results of this study show that breastfed children had a
healthier diet and lifestyle at 7 years of age than nonbreastfed chil-
dren, independent of maternal characteristics, but these differ-
ences could not explain the lower overweight prevalence at 8 years
of age in breastfed children compared to nonbreastfed children.
This study was supported by the Netherlands Organization for Health
Research and Development, the Netherlands Asthma Foundation, the
Netherlands Ministry of Health, Welfare and Sport and Numico Research,
the Netherlands. The funding source had no involvement in the study
design, in the collection of the data, data analysis, interpretation of the data,
writing the report, or in the decision to submit the paper for publication.
The authors declared no conflict of interest.
© 2008 The Obesity Society