Complementary Feeding and Childhood Adiposity in Preschool-Aged

Children in a Large Chinese Cohort

Ju-Sheng Zheng, PhD1,*, Huijuan Liu, MD2,*, Yi-Min Zhao, BSc1, Jing Li, MD2, Yu Chen, MD2, Shanlin Zhu, MD2,
Hua Chen, MD2, Tao Huang, PhD3, and Duo Li, PhD1

Objective To examine the association between the timing and type of complementary feeding and childhood
adiposity in Mainland China.
Study design During 1999-2009, 97 424 singletons were enrolled in the Jiaxing Birth Cohort, a population-based
prospective cohort study in Southeast China. Of these children, 43 848 children provided complementary diet
information and other anthropometric measurements at 1, 3, and 6 months of age and were followed up until
4-5 years of age. Obesity and overweight were identified as body mass index (BMI)-for-age z-score (SD) $2 and
between 1 and 2, respectively.
Results Among 40 510 children in the statistical analysis, 3.18% were overweight and 64.8% were fed comple-
mentary food before 3 months of age. Early introduction of complementary foods was associated with greater
BMI z-score (P-trend < .001) and higher risk of overweight (P-trend = .033). Compared with introduction of comple-
mentary foods between 4-6 months of age, before 3 months of age of introduction was associated with 11% greater
risk of overweight (OR 1.11, 95% CI 1.03-1.19). No significant association between timing of complementary
feeding and obesity was observed. Fish liver oil was the major type of complementary food associated with
adiposity. Early introduction of fish liver oil was associated with greater BMI z-score (P < .001) and greater risk
of overweight (P-trend = .004).
Conclusions Early introduction of fish liver oil is associated with greater childhood BMI and risk of overweight in
Chinese children at 4-5 years of age. (J Pediatr 2015;166:326-31).

I
n China, the prevalence of overweight/obesity is 23.23% for urban boys and 13.76% for urban girls (7-18 years of age),
respectively, in 2010.1 The prevalence of overweight/obesity during 1985-1995, 1995-2000, 2000-2005, and 2005-2010
increased 0.10%, 0.30%, 0.23%, and 0.23%, respectively.1 Accumulating evidence suggests that nutritional factors
(eg, protein and energy intake, breastfeeding) and other factors (eg, early rapid weight gain, high birth weight) during in-
fancy play roles in the development of overweight and obesity later in life.2,3 A number of observational studies suggest that
the early introduction of complementary foods is associated with a greater risk of overweight and obesity in children,4-9 but
results from other studies do not agree with this finding.10-14 The evidence from Asian countries is limited.9,10,12 In 1 pro-
spective study, investigators did not find an association between early introduction of complementary foods and childhood
obesity.10
The primary aim of the present study is to investigate the association between timing of complementary feeding and
adiposity in children of 4-5 years of age in Mainland China. In addition, the influence of the types of complementary foods
on the association was examined.

Methods
The study protocol was approved by the Ethics Committee of the College of Biosystem Engineering & Food Science at
Zhejiang University (2013013). The Jiaxing Birth Cohort was initiated in 1993 as part of a large population-based health
surveillance system in China.15,16 The cohort enrolled pregnant women who
primarily visit local clinics or Maternity and Child Health Care Hospitals in
Southeast China, Jiaxing area of Zhejiang Province. Information was ob-
tained during the period of 1999 to 2013 and follow-up information
collected until 2013. During the period, 90 066 singleton children from the
Jiaxing Birth Cohort provided complementary diet information and other
anthropometric measurements at 1, 3, and 6 months of age. Follow-up in-
formation, including weight and height, was obtained from 43 848 of these
children at 4-5 years of age (48-60 months).
1
From the Department of Food Science and Nutrition,
2
Zhejiang University, Hangzhou, China; Jiaxing Maternity
and Child Health Care Hospital, Jiaxing, China; and
3
Department of Nutrition, Harvard School of Public
Health, Boston, MA
*Contributed equally.
Funded by National Natural Science Foundation of China
(81273054), the PhD Programs Foundation of Ministry of
Education of China (20120101110107), and the National
Basic Research Program of China (973 Program:
2015CB553600). The authors declare no conflicts of
interest.

0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc.

BMI Body mass index All rights reserved.
http://dx.doi.org/10.1016/j.jpeds.2014.11.010

326
 Vol. 166, No. 2  February 2015

Children were excluded if they had any missing data on

weight or height at 4-5 years (n = 461) or had any
missing data on any kind of complementary food intro-
duction at 1, 3, or 6 months of age (n = 3338) or did
not have information for sex (n = 17). Children with
extreme values of gestational age (>44 weeks or
<33 weeks; n = 165), birth weight (>5 kg or <1.5 kg;
n = 51), or body mass index (BMI) at 4-5 years of age
(BMI z-score >5 or < 5; n = 48) also were excluded
from the study. A total of 40 510 children were included
for statistical analysis.
At each visit, parents of these children were asked whether
their children had been introduced to the following 10 kinds
of foods (yes, no, or unclear) via an in-person interview by a
trained nurse: fish liver oil, rice cereal/porridge, egg yolk, fish
paste, liver paste, tofu, animal blood, bread/steamed bun/fine
dried noodle, ground meat/soy product, and pureed noodle/
cookies. The aforementioned complementary foods are
commonly used in Southeast China.
Children were categorized into 3 groups: “complementary
feeding before and at 3 months” (if they were introduced to
any of the complementary foods before or at 3 months of
age), “complementary feeding between 4 and 6 months” (if
they were introduced to complementary foods between
4-6 months of age), and “complementary feeding after
6 months of age” if they did not receive any complementary
foods before 6 months of age.
BMI and weight commonly are used in adults to deter-
mine adiposity; however, for them to be meaningful in
children, they must be compared with a reference-
standard that accounts for the child’s age and sex. There-
fore, BMI-for-age z-scores and weight-for-age z-score
Figure 1. Timing of complementary feeding and fish liver oil
were calculated according to the World Health Organiza-
introduction and BMI z-score at 4-5 years of age in the Jiaxing
tion Child Growth Standard.17 Children were defined as Birth Cohort. Linear regression model was used to obtain
having “obesity” if the BMI z-score $2 SD and “over- P-values, adjusting for potential confounders.
weight” if the BMI z-score was between 1 and 2 SD.17
Weight gain during the first 3 months was calculated as
the difference between weight-for-age z-score at 3 months Statistical Analyses
of age and that at birth. STATA version 12 (StataCorp LP, College Station, Texas)
Confounding factors were selected according to the bio- was used to perform the statistical analyses. General maternal
logical knowledge or previous publications.2,6,9,11 Maternal and child characteristics by the timing of complementary
characteristics, including menarcheal age, educational sta- feeding were assessed by the c2 test. The association of com-
tus (>high school, high school, <high school), occupation plementary feeding time with risk of overweight and obesity
(housework, routine job, temporary job, unemployed), ce- was estimated by multinomial logistic regression models,
sarean delivery (yes or no), gestational age, and BMI (pre- with adjustment for potential confounding factors. Linear
pregnant or <18 gestational weeks), were recorded during regression was used to assess the association between comple-
the follow-up visits and were treated as confounding fac- mentary feeding time and BMI z-scores with adjustment for
tors. Paternal socioeconomic characteristics, such as potential confounders. The association between complemen-
educational status and occupation, had no impact on the tary feeding time and adiposity for any particular type of
results, so they were not included in the statistical model. complementary foods also was examined using the logistic
Alcohol drinking, smoking, or diabetes mellitus status regression model or linear regression model. A 2-tailed
were not included in the final model because of their P < .05 was considered as statistically significant.
low prevalence (<0.5%). Breastfeeding status was catego-
rized into 2 groups: never and ever breastfeeding. Ever Results
breastfeeding was defined when the infant was reported
breastfeeding at any of the visits within the first 6 months Of the 40 510 children, 64.7% of the female children
of age. and 64.9% of the male children were introduced to
327
THE JOURNAL OF PEDIATRICS  www.jpeds.com Vol. 166, No. 2

Figure 2. Introduction time of different types of complementary foods and BMI z-score at 4-5 years of age in the Jiaxing Birth
Cohort. *P < .05 indicates statistical significance compared with complementary feeding between 4 and 6 months. Linear
regression model was used to get the P-values, adjusting for potential confounders.

complementary foods within 3 months of age. Breastfeeding
status, cesarean delivery status, parity, maternal menarcheal
age, educational status, and occupation were all significantly
associated with the timing of complementary feeding
(P < .001; Table I; available at www.jpeds.com). Similar
significant associations were obtained between the original
sample and the sample in the present study (Table II;
available at www.jpeds.com). At 4-5 years of age, 2776 boys
(13.2% of all boys) and 1847 girls (9.45% of all girls) were
overweight (BMI z-score between 1 and 2 SD), and 875
boys (4.17% of all boys) and 413 girls (2.11% of all girls)
had obesity (BMI z-score $2 SD).
Early complementary feeding was significantly associated
with greater BMI z-score in the fully adjusted model (P-
trend < .001). Consistently, early fish liver oil introduction
was associated with greater BMI z-score (P-trend < .001)
(Figure 1). In addition, introduction of animal blood
after 6 months of age showed lower BMI z-score
compared with introduction between 4 and 6 months of
age (P = .044). Early introduction (#3 months of age) of
egg yolk was marginally associated with lower BMI z-score
(P = .046), compared with introduction of egg yolk
between 4 and 6 months of age. No significant association
was observed for other types of complementary foods
(Figure 2).
The introduction of complementary foods earlier than
3 months of age, compared with that between 4 and 6 months
of age, showed a significantly greater risk of overweight (OR
1.16, 95% CI 1.09-1.24) in the crude model. The positive as-
sociation remained significant after adjustment for the child
and maternal covariates (OR 1.11, 95% CI 1.03-1.19)
(Table III). Among the 10 types of complementary foods,
a significant trend between complementary feeding time
and the risk of overweight was observed only for fish
liver oil. Compared with introduction of fish liver oil
between 4 and 6 months of age, introduction of fish liver
oil earlier than 3 months (OR 1.08, 95% CI 0.99-1.17)
showed marginally significantly association with risk of
overweight (P-trend = .004) in the multivariable model
(Table IV).

Table III. Relationship between timing of complementary feeding and risk of overweight and obesity in the Jiaxing Birth
Cohort
Overweight (BMI z-score between 1 and 2 SD) Obesity (BMI z-score‡2 SD)
£3 mo 4-6 mo >6 mo P trend £3 mo 4-6 mo >6 mo P trend
Total subjects/cases 26 229/3119 13 106/1369 1175/135 26 229/864 13 106/392 1175/32
Crude OR 1.16 (1.09-1.24) 1 (ref) 1.11 (0.92-1.34) <.001 1.12 (1.00-1.27) 1 (ref) 0.92 (0.64-1.32) .037
aOR* 1.12 (1.04-1.20) 1 (ref) 1.12 (0.92-1.36) .013 1.08 (0.95-1.22) 1 (ref) 0.96 (0.66-1.39) .213
aOR† 1.11 (1.03-1.19) 1 (ref) 1.14 (0.94-1.40) .038 1.03 (0.91-1.18) 1 (ref) 0.96 (0.64-1.42) .546
*The statistical model was adjusted for age of children (mo), sex, birth weight, gestational age, cesarean delivery (yes vs no), weight gain during first 3 months (difference in weight-for-age z-score)
and breastfeeding status.
†The statistical model was adjusted for age of children (mo), sex, birth weight, gestational age, cesarean delivery (yes vs no), weight gain during first 3 mo (difference in weight-for-age z-score),
breastfeeding status, and maternal characteristics (maternal age at birth, maternal menarcheal age, maternal BMI status, maternal educational status, and maternal occupation).

328 Zheng et al
February 2015 ORIGINAL ARTICLES

Table IV. Relationship between introduction time of different types of complementary foods and risk of overweight and
obesity in the Jiaxing Birth Cohort
Overweight (BMI z-score between 1 and 2 SD) Obesity (BMI z-score‡2 SD)
£3 mo 4-6 mo >6 mo P trend £3 mo 4-6 mo >6 mo P trend
Fish liver oil
Total subjects/cases 25 545/3042 9676/1043 5289/538 25 545/840 9676/287 5289/161
Crude OR 1.12 (1.04-1.21) 1 (ref) 0.94 (0.84-1.05) <.001 1.13 (0.98-1.29) 1 (ref) 1.02 (0.84-1.24) .098
aOR* 1.08 (0.99-1.16) 1 (ref) 0.95 (0.84-1.06) .007 1.04 (0.90-1.20) 1 (ref) 1.04 (0.84-1.28) .783
aOR† 1.08 (0.99-1.17) 1 (ref) 0.93 (0.83-1.05) .004 1.05 (0.91-1.21) 1 (ref) 1.02 (0.82-1.26) .638
Rice cereal/porridge
Total subjects/cases 1577/171 33 698/3825 5235/627 1577/45 33 698/1075 5235/168
Crude OR 0.94 (0.80-1.11) 1 (ref) 1.06 (0.97-1.16) .123 0.89 (0.65-1.20) 1 (ref) 1.01 (0.86-1.20) .566
aOR* 0.99 (0.83-1.17) 1 (ref) 1.08 (0.98-1.18) .171 0.95 (0.69-1.30) 1 (ref) 1.02 (0.85-1.21) .744
aOR† 1.05 (0.85-1.30) 1 (ref) 1.13 (1.01-1.27) .083 1.04 (0.71-1.52) 1 (ref) 1.01 (0.82-1.24) .906
Egg yolk
Total subjects/cases 768/74 33 191/3818 6551/731 768/17 33 191/1057 6551/214
Crude OR 0.81 (0.64-1.03) 1 (ref) 0.97 (0.89-1.05) .956 0.67 (0.41-1.09) 1 (ref) 1.02 (0.88-1.19) .356
aOR* 0.88 (0.69-1.14) 1 (ref) 0.98 (0.90-1.08) .983 0.77 (0.47-1.26) 1 (ref) 1.04 (0.88-1.22) .414
aOR† 0.81 (0.53-1.24) 1 (ref) 0.95 (0.85-1.06) .624 0.50 (0.20-1.25) 1 (ref) 1.03 (0.85-1.26) .444
Fish paste
Total subjects/cases 530/51 24 484/2775 15 496/1797 530/14 24 484/754 15 496/520
Crude OR 0.83 (0.62-1.11) 1 (ref) 1.03 (0.97-1.10) .2 0.84 (0.49-1.43) 1 (ref) 1.10 (0.98-1.23) .079
aOR* 0.93 (0.69-1.26) 1 (ref) 1.00 (0.94-1.07) .871 0.93 (0.53-1.64) 1 (ref) 1.04 (0.92-1.17) .474
aOR† 1.29 (0.68-2.48) 1 (ref) 1.00 (0.91-1.10) .941 1.29 (0.35-4.81) 1 (ref) 1.15 (0.97-1.35) .132
Liver paste
Total subjects/cases 425/39 8559/921 31 526/3663 425/12 8559/272 31 526/1004
Crude OR 0.83 (0.60-1.17) 1 (ref) 1.09 (1.01-1.18) .008 0.87 (0.48-1.56) 1 (ref) 1.01 (0.88-1.16) .712
aOR* 0.82 (0.57-1.17) 1 (ref) 1.05 (0.97-1.14) .11 0.89 (0.49-1.62) 1 (ref) 0.92 (0.80-1.07) .366
aOR† 0.42 (0.16-1.08) 1 (ref) 1.05 (0.87-1.27) .204 0.96 (0.26-3.59) 1 (ref) 0.84 (0.61-1.16) .297
Tofu
Total subjects/cases 506/48 24 605/2786 15 399/1789 506/14 24 605/785 15 399/489
Crude OR 0.82 (0.60-1.10) 1 (ref) 1.03 (0.97-1.10) .199 0.84 (0.49-1.45) 1 (ref) 1.00 (0.89-1.12) .868
aOR* 0.86 (0.63-1.18) 1 (ref) 1.00 (0.93-1.07) .822 0.92 (0.52-1.61) 1 (ref) 0.94 (0.83-1.06) .354
aOR† 0.76 (0.38-1.53) 1 (ref) 0.98 (0.90-1.07) .833 0.96 (0.29-3.14) 1 (ref) 0.87 (0.74-1.02) .096
Animal blood
Total subjects/cases 432/41 12 107/1362 27 971/3220 432/12 12 107/378 27 971/898
Crude OR 0.82 (0.59-1.14) 1 (ref) 1.03 (0.96-1.10) .23 0.87 (0.48-1.56) 1 (ref) 1.03 (0.91-1.17) .497
aOR* 0.89 (0.63-1.25) 1 (ref) 1.01 (0.94-1.08) .61 0.97 (0.53-1.76) 1 (ref) 0.96 (0.85-1.09) .566
aOR† 0.98 (0.37-2.59) 1 (ref) 0.99 (0.88-1.11) .897 2.32 (0.51-10.6) 1 (ref) 1.02 (0.83-1.26) .964
Bread/steamed bun/fine dried noodle
Total subjects/cases 403/40 7863/836 32 244/3747 403/10 7863/238 32 244/1040
Crude OR 0.92 (0.66-1.29) 1 (ref) 1.11 (1.02-1.20) .007 0.81 (0.43-1.54) 1 (ref) 1.08 (0.94-1.25) .187
aOR* 0.96 (0.68-1.36) 1 (ref) 1.05 (0.97-1.15) .191 0.88 (0.46-1.69) 1 (ref) 0.97 (0.84-1.13) .829
aOR† 2.22 (0.88-5.12) 1 (ref) 1.07 (0.85-1.36) .766 1.14 (0.21-6.29) 1 (ref) 1.43 (0.93-2.19) .137
Ground meat/soy product
Total subjects/cases 469/41 7279/780 32 762/3802 469/12 7279/227 32 762/1049
Crude OR 0.79 (0.57-1.10) 1 (ref) 1.10 (1.01-1.19) .005 0.80 (0.44-1.44) 1 (ref) 1.04 (0.90-1.20) .394
aOR* 0.83 (0.58-1.16) 1 (ref) 1.05 (0.96-1.14) .132 0.84 (0.46-1.54) 1 (ref) 0.93 (0.80-1.09) .543
aOR† 0.52 (0.24-1.12) 1 (ref) 1.00 (0.80-1.26) .328 0.76 (0.23-2.46) 1 (ref) 0.82 (0.54-1.24) .506
Pureed noodle/cookies
Total subjects/cases 411/42 9014/994 31 085/3587 411/10 9014/286 31 085/992
Crude OR 0.91 (0.66-1.26) 1 (ref) 1.05 (0.98-1.13) .117 0.75 (0.40-1.43) 1 (ref) 1.01 (0.89-1.16) .602
aOR* 0.98 (0.70-1.38) 1 (ref) 1.02 (0.94-1.10) .626 0.85 (0.44-1.64) 1 (ref) 0.93 (0.81-1.07) .416
aOR† 2.26 (0.98-5.22) 1 (ref) 0.94 (0.79-1.10) .137 0.67 (0.11-4.20) 1 (ref) 0.92 (0.68-1.24) .579
*The statistical model was adjusted for age of children (mo), sex, birth weight, gestational age, cesarean delivery (yes vs no), weight gain during first 3 mo (difference in weight-for-age z-score),
breastfeeding status, and maternal characteristics (maternal age at birth, maternal menarcheal age, maternal BMI status, maternal educational status, and maternal occupation).
†The statistical model was adjusted for age of children (mo), sex, birth weight, gestational age, cesarean delivery (yes vs no), weight gain during first 3 mo (difference in weight-for-age z-score),
breastfeeding status, maternal characteristics (maternal age at birth, maternal menarcheal age, maternal BMI status, maternal educational status, and maternal occupation), and introduction time for
all the other types of complementary foods in the Table.

In the crude model, early complementary feeding was marginally significantly associated with lower risk of
associated with greater risk of obesity (P-trend = .037). How- obesity (OR 0.87, 95% CI 0.74-1.02).
ever, no significant association was observed in the multi-
variable model (Table III). Consistently, no significant
trend was observed between the introduction time of Discussion
any particular type of complementary foods and risk of
obesity (Table IV). For tofu, compared with 4-6 months A recent systematic review identified 23 articles that exam-
of introduction, introduction later than 6 months was ined the association between the timing of complementary
Complementary Feeding and Childhood Adiposity in Preschool-Aged Children in a Large Chinese Cohort 329
THE JOURNAL OF PEDIATRICS  www.jpeds.com
feeding and risk of overweight or obesity during childhood.
Five of them found a significant association between early
complementary food introduction and greater BMI in
childhood.18 Of note, most of these studies had a limited
sample size, ranging from 54 to 17 561. Among these
studies, a study based on the Millennium Cohort Study
had the largest sample size (17 561 participants).6 The au-
thors of this study found that early introduction of comple-
mentary foods was associated with a greater risk of obesity
for children at age 5, particularly in greater income white
families. In a second study,11 compared with 4- to 6-
months introduction of complementary food, 2- or 3-
month introduction of complementary food introduction
was positively associated with childhood obesity (OR
1.48, 95% CI 1.01-2.16) at age 7 years. Most studies with
small or moderate sample size found nonsignificant associ-
ation between complementary feeding time and risk of
childhood obesity.
In our study, it should be noted that no significant associ-
ation between complementary feeding time and obesity was
observed. It may be that the follow-up duration was not
long enough to see a significant difference in the risk of
obesity. Overweight children may later develop obesity. In
addition, the prevalence of obesity (<5% for boys, <2% for
girls) was low in the present population, which limited the
power to detect an association.
Studies in Asian countries, especially in China, are sparse.
In a small case-control (81 obesity cases, and 81 sex- and age-
matched controls) study in central of China,9 early comple-
mentary food initiation was associated with a greater risk
of obesity. In a recent prospective cohort study among
7809 Chinese children from Hong Kong, no significant asso-
ciation between early introduction of complementary foods
and childhood obesity was observed.10 However, there were
several limitations for the Hong Kong study, including recall
bias and imputation for the complementary food introduc-
tion time.
Among different types of complementary foods, the time
of introduction of fish liver oil was the only one to show
significant trend with adiposity in the present population.
One previous cross-sectional study has investigated the
relationship between different types of complementary
feeding time and childhood obesity.19 Different types of
complementary foods may exert differential effects on the
childhood adiposity, which may explain the inconsistent re-
sults for the association between introduction time of com-
plementary foods and childhood adiposity in previous
studies.18 Fish liver oil is a commonly used infant comple-
mentary food/dietary supplement in China and is intro-
duced early to children in small proportions. Fish liver oil
is extracted from fish liver and is considered to provide
vitamin A and D for newborn infants. It is often recom-
mended to take fish liver oil at 1 month of age in China.
Therefore, the present finding may have important public
health implications. The potential mechanism behind the
association may be that early introduction of fish liver oil
330
Vol. 166, No. 2
is associated with increased energy intake and rapid weight
gain during infancy. It is well known that rapid weight gain
during infancy is associated with risk of childhood over-
weight and obesity.20,21
In addition, the early introduction of animal blood was
positively associated with BMI z-score. Animal blood is
rich in many nutrients, especially heme irons. Potential
iron deficiency may occur after 6 months of age. In the pre-
sent study, animal blood was introduced to the majority of
the children after 6 months of age. One randomized trial of
iron supplementation in breast-fed infants suggested that
routine iron supplementation of breast-fed infants after
4 months of age may be good for those with low hemoglobin
but may present risks for those with normal levels of hemo-
globin.22 The majority of the children were introduced to egg
yolk between 4 and 6 months of age. The precise mechanism
for the associations between early yolk introduction
and decreased adiposity is unclear and warrants further
exploration.
There were several limitations of the study. First, dietary
habits, nutrients, and brands of complementary foods
change over time. Our dietary information was collected
from 1999 to 2009. Dietary and lifestyle factors in China
may change rapidly. Second, our cohort did have loss to
follow-up. Because the sample size was very large after
the initial inclusion, it was difficult to contact all these par-
ticipants. Parents of the children who came to visit at 4-
5 years of age may be more health-conscious and have
paid more attention to the diet of their children. However,
no significant difference of child or maternal characteristics
by timing of solid food introduction was observed between
the original sample and the sample being analyzed in this
study. Finally, our results should be interpreted with
some caution, because no significant association was
observed between timing of complementary feeding and
obesity.
The early introduction of fish liver oil was associated with
increased childhood BMI and overweight at 4-5 years of age.
More prospective studies with longer duration of follow-up
are warranted. n
We thank all of the participants involved in the study and all of the staff
working on the project. We thank Chunlei Wei, MD, and Genmen
Shen, MD (Jiaxing Maternity and Child Health Care Hospital, Jiax-
ing) for their assistance and suggestions for data collection.
Submitted for publication Jun 21, 2014; last revision received Sep 22, 2014;
accepted Nov 4, 2014.
Reprint requests: Prof. Duo Li, PhD, Department of Food Science and
Nutrition, Zhejiang University, 866 Yu-hang-tang Road, Hangzhou 310058,
China. E-mail: duoli@zju.edu.cn
References
1. Ma J, Cai CH, Wang HJ, Dong B, Song Y, Hu PJ, et al. The trend analysis
of overweight and obesity in Chinese students during 1985-2010. Zhong-
hua Yu Fang Yi Xue Za Zhi 2012;46:776-80.
Zheng et al
February 2015
2. Weng SF, Redsell SA, Swift JA, Yang M, Glazebrook CP. Systematic re-
view and meta-analyses of risk factors for childhood overweight identifi-
able during infancy. Arch Dis Child 2012;97:1019-26.
3. Pearce J, Langley-Evans SC. The types of food introduced during com-
plementary feeding and risk of childhood obesity: a systematic review.
Int J Obes (Lond) 2013;37:477-85.
4. Huh SY, Rifas-Shiman SL, Taveras EM, Oken E, Gillman MW. Timing of
solid food introduction and risk of obesity in preschool-aged children.
Pediatrics 2011;127:e544-51.
5. Schack-Nielsen L, Sorensen T, Mortensen EL, Michaelsen KF. Late intro-
duction of complementary feeding, rather than duration of breastfeeding,
may protect against adult overweight. Am J Clin Nutr 2010;91:619-27.
6. Brophy S, Cooksey R, Gravenor MB, Mistry R, Thomas N, Lyons RA,
et al. Risk factors for childhood obesity at age 5: analysis of the millen-
nium cohort study. BMC Public Health 2009;9:467.
7. Gooze RA, Anderson SE, Whitaker RC. Prolonged bottle use and obesity
at 5.5 years of age in US children. J Pediatr 2011;159:431-6.
8. Seach KA, Dharmage SC, Lowe AJ, Dixon JB. Delayed introduction of
solid feeding reduces child overweight and obesity at 10 years. Int J
Obes (Lond) 2010;34:1475-9.
9. Zhou L, He G, Zhang J, Xie R, Walker M, Wen SW. Risk factors of
obesity in preschool children in an urban area in China. Eur J Pediatr
2011;170:1401-6.
10. Lin SL, Leung GM, Lam TH, Schooling CM. Timing of solid food intro-
duction and obesity: Hong Kong’s “children of 1997” birth cohort.
Pediatrics 2013;131:e1459-67.
11. Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, et al. Early
life risk factors for obesity in childhood: cohort study. BMJ 2005;330:1357.
12. Caleyachetty A, Krishnaveni GV, Veena SR, Hill J, Karat SC, Fall CH,
et al. Breastfeeding duration, age of starting solids and high BMI risk
and adiposity in Indian children. Matern Child Nutr 2013;9:199-216.
Complementary Feeding and Childhood Adiposity in Preschool-Aged Children in a Large Chinese Cohort
ORIGINAL ARTICLES
13. Neutzling MB, Hallal PR, Ara ujo CL, Horta BL, Vieira Mde F,
Menezes AM, et al. Infant feeding and obesity at 11 years: prospective
birth cohort study. Int J Pediatr Obes 2009;4:143-9.
14. Patterson RE, Typpo JT, Typpo MH, Krause GF. Factors related to
obesity in preschool children. J Am Diet Assoc 1986;86:1376-81.
15. Li S, Moore CA, Li Z, Berry RJ, Gindler J, Hong SX, et al. A population-
based birth defects surveillance system in the People’s Republic of China.
Paediatr Perinat Epidemiol 2003;17:287-93.
16. Zheng JS, Liu H, Li J, Chen Y, Wei C, Shen G, et al. Exclusive breastfeed-
ing is inversely associated with risk of childhood overweight in a large
chinese cohort. J Nutr 2014;144:1454-9.
17. WHO Multicentre Growth Reference Study Group. WHO Child Growth
Standards: Length/height-for-age, weight-for-age, weight-for-length,
weight-for-height and body mass index-for-age: methods and develop-
ment. Geneva: World Health Organization; 2006.
18. Pearce J, Taylor MA, Langley-Evans SC. Timing of the introduction of
complementary feeding and risk of childhood obesity: a systematic re-
view. Int J Obes (Lond) 2013;37:1295-306.
19. Simon VG, Souza JM, Souza SB. Breastfeeding, complementary feeding,
overweight and obesity in pre-school children. Rev Saude Publica 2009;
43:60-9.
20. Baird J, Fisher D, Lucas P, Kleijnen J, Roberts H, Law C. Being big or
growing fast: systematic review of size and growth in infancy and later
obesity. BMJ 2005;331:929.
21. Taveras EM, Rifas-Shiman SL, Belfort MB, Kleinman KP, Oken E,
Gillman MW. Weight status in the first 6 months of life and obesity at
3 years of age. Pediatrics 2009;123:1177-83.
22. Dewey KG, Domellof M, Cohen RJ, Landa Rivera L, Hernell O,
Lonnerdal B. Iron supplementation affects growth and morbidity of
breast-fed infants: results of a randomized trial in Sweden and
Honduras. J Nutr 2002;132:3249-55.
331
THE JOURNAL OF PEDIATRICS  www.jpeds.com Vol. 166, No. 2

Table I. Study characteristics by the timing of complementary feeding in the Jiaxing Birth Cohort (n = 40 510)
Timing of complementary food introduction
n £3 mo 4-6 mo >6 mo P value
Sex
Male 20 971 13 557 (64.7) 6768 (32.3) 646 (3.08) .082
Female 19 539 12 672 (64.9) 6338 (32.4) 529 (2.71)
Birth weight, g
Low (<2500) 718 497 (69.2) 197 (27.4) 24 (3.34) .015
Normal (2500-3999) 37 205 24 018 (64.6) 12 099 (32.5) 1088 (2.92)
High ($4000) 2587 1714 (66.3) 810 (31.3) 63 (2.44)
Breastfeeding status <.001
Never 2682 1990 (74.2) 665 (24.8) 27 (1.01)
Ever 37 828 24 239 (64.1) 12 441 (32.9) 1148 (3.03)
Gestational age (wks)
Preterm delivery (<37) 1065 731 (68.6) 309 (29.0) 25 (2.35) .024
Term delivery ($37) 39 445 25 498 (64.6) 12 797 (32.4) 1150 (2.92)
Mother’s age at birth (y)
<35 39 454 25 556 (64.8) 12 760 (32.3) 1138 (2.88) .444
$35 1055 672 (63.7) 346 (32.8) 37 (3.51)
Cesarean delivery
Yes 29 761 19 496 (65.5) 9431 (31.7) 834 (2.80) <.001
No 10 667 6691 (62.7) 3638 (34.1) 338 (3.17)
Parity
First pregnancy 33 973 22 506 (66.3) 10 609 (31.2) 858 (2.53) <.001
>1 pregnancy 6537 3723 (57.0) 2497 (38.2) 317 (4.85)
Mother’s menarcheal age (y)
<14 5768 3840 (66.6) 1798 (31.2) 130 (2.25) .002
14-15 23 890 15 418 (64.5) 7772 (32.5) 700 (2.93)
>15 10 723 6896 (64.3) 3489 (32.5) 338 (3.15)
Maternal BMI status
Underweight (BMI <18.5) 7562 4971 (65.7) 2407 (31.8) 184 (2.43) .115
Normal (25 > BMI $ 18.5) 28 591 18 535 (64.8) 9205 (32.2) 851 (2.98)
Overweight (30 > BMI $ 25) 1903 1217 (64.0) 632 (33.2) 54 (2.84)
Obese (BMI $30) 196 125 (63.8) 62 (31.6) 9 (4.59)
Maternal educational status
<High school 29 359 17 715 (60.3) 10 593 (36.1) 1051 (3.58) <.001
High school 7209 5345 (74.1) 1778 (24.7) 86 (1.19)
>High school 3900 3139 (80.5) 725 (18.6) 36 (0.92)
Maternal occupation
Housework 26 553 15 987 (60.2) 9576 (36.1) 990 (3.73) <.001
Routine job 7737 5487 (70.9) 2121 (27.4) 129 (1.67)
Temporary job 2625 1960 (74.7) 636 (24.2) 29 (1.10)
Unemployed 3551 2762 (77.8) 762 (21.5) 27 (0.76)

331.e1 Zheng et al
February 2015 ORIGINAL ARTICLES

Table II. Study characteristics by the timing of complementary feeding in the original sample (n = 90 066) of the Jiaxing
Birth Cohort
Complementary food introduction time
n £3 mo 4-6 mo >6 mo P value
Sex
Male 46 479 31 074 (66.9) 14 195 (30.5) 1210 (2.60) .125
Female 43 557 29 038 (66.7) 13.4 (30.9) 1076 (2.47)
Birth weight (g)
Low (<2500) 1819 1276 (70.2) 490 (26.9) 53 (2.91) <.001
Normal (2500-3999) 82 030 54 580 (66.5) 25 361 (30.9) 2089 (2.55)
High ($4000) 6217 4283 (68.9) 1790 (28.8) 144 (2.32)
Breastfeeding status
Never 6740 4982 (73.9) 1687 (25.0) 71 (1.05) <.001
Ever 83 326 55 157 (66.2) 25 954 (31.2) 2215 (2.66)
Gestational age (wks)
Preterm delivery (<37) 2659 1880 (70.7) 727 (27.3) 52 (1.96) <.001
Term delivery ($37) 87 370 58 240 (66.7) 26 898 (30.8) 2232 (2.55)
Mother’s age at birth (y)
<35 87 413 58 409 (66.8) 26 796 (30.7) 2208 (2.53) .13
$35 2645 1724 (65.2) 843 (31.9) 78 (2.95)
Cesarean delivery
Yes 63 018 42 428 (67.3) 19 046 (30.2) 1544 (2.45) <.001
No 26 874 17 619 (65.6) 8520 (31.7) 735 (2.73)
Parity
First pregnancy 74 110 50 678 (68.4) 21 775 (29.4) 1657 (2.24) <.001
>1 pregnancy 15 956 9461 (59.3) 5866 (36.8) 629 (3.94)
Mother’s menarcheal age (y)
<14 14 038 9602 (68.4) 4161 (29.6) 275 (1.96) <.001
14-15 53 555 35 707 (66.7) 16 465 (30.7) 1383 (2.58)
>15 22 204 14 660 (66.0) 6927 (31.2) 617 (2.78)
Maternal BMI status
Underweight (BMI <18.5) 16 725 11 425 (68.3) 4951 (29.6) 349 (2.09) <.001
Normal (25 > BMI $ 18.5) 62 683 41 857 (66.8) 19 212 (30.7) 1614 (2.57)
Overweight (30 > BMI $ 25) 4419 2912 (65.9) 1384 (31.3) 123 (2.78)
Obese (BMI $30) 485 332 (68.5) 137 (28.3) 16 (3.30)
Maternal education level
<High school 62 254 38 654 (62.1) 21 577 (34.7) 2023 (3.25) <.001
High school 16 069 11 960 (74.4) 3924 (24.4) 185 (1.15)
>High school 11 743 9525 (81.1) 2140 (18.2) 78 (0.66)
Maternal occupation
Housework 57 901 36 020 (62.2) 20 009 (34.6) 1872 (3.23) <.001
Routine job 17 182 12 529 (72.9) 4366 (25.4) 287 (1.67)
Temporary job 5368 4025 (75.0) 1292 (24.10) 51 (0.95)
Unemployed 9455 7439 (78.7) 1940 (20.5) 76 (0.80)

Complementary Feeding and Childhood Adiposity in Preschool-Aged Children in a Large Chinese Cohort 331.e2