Impact of breastfeeding duration on age at menarche.

Abstract

The study aims to assess the relation between breastfeeding duration and age at menarche. Analysis was based on a cohort of 994 Filipino girls born in 1983-1984 and followed up from infancy to adulthood by the Cebu Longitudinal Health and Nutrition Survey. The main outcome was self-reported age at menarche. Cox regression was used to investigate the relation between duration of exclusive and any breastfeeding with age at menarche with adjustment sequentially for specific sets of known socioeconomic, maternal, genetic, and prenatal confounders. The estimated median of age at menarche was 13.08 years. After adjustment for potential confounders of the association of breastfeeding with age at menarche, exclusive breastfeeding duration retained an independent and significant association with age at menarche. An increase in 1 month of exclusive breastfeeding decreases the hazard of attaining earlier menarche by 6% (hazard ratio = 0.94, 95% confidence interval: 0.90, 0.98). Any breastfeeding duration was not associated with age at menarche. Although this is the first longitudinal study that reveals a negative association between exclusive breastfeeding and early menarche, the relation is still elusive. Further longitudinal studies within different contexts are warranted to assess the generalizability of these findings.

Full text

Page 1

American Journal of Epidemiology
ª The Author 2011. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of
Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Vol. 173, No. 9
DOI: 10.1093/aje/kwq496
Advance Access publication:
March 23, 2011
Original Contribution
Impact of Breastfeeding Duration on Age at Menarche
Ban Al-Sahab*, Linda Adair, Mazen J. Hamadeh, Chris I. Ardern, and Hala Tamim
* Correspondence to Ban Al-Sahab, School of Kinesiology and Health Science, Bethune College, York University, 4700 Keele
Street, Toronto, Ontario, Canada M3J 1P3 (e-mail: bsahab@yorku.ca).
Initially submitted January 7, 2010; accepted for publication June 7, 2010.
The study aims to assess the relation between breastfeeding duration and age at menarche. Analysis was based
on a cohort of 994 Filipino girls born in 1983–1984 and followed up from infancy to adulthood by the Cebu
Longitudinal Health and Nutrition Survey. The main outcome was self-reported age at menarche. Cox regression
was used to investigate the relation between duration of exclusive and any breastfeeding with age at menarche
with adjustment sequentially for specific sets of known socioeconomic, maternal, genetic, and prenatal con-
founders. The estimated median of age at menarche was 13.08 years. After adjustment for potential confounders
of the association of breastfeeding with age at menarche, exclusive breastfeeding duration retained an indepen-
dent and significant association with age at menarche. An increase in 1 month of exclusive breastfeeding de-
creases the hazard of attaining earlier menarche by 6% (hazard ratio ¼ 0.94, 95% confidence interval: 0.90, 0.98).
Any breastfeeding duration was not associated with age at menarche. Although this is the first longitudinal study
that reveals a negative association between exclusive breastfeeding and early menarche, the relation is still
elusive. Further longitudinal studies within different contexts are warranted to assess the generalizability of these
findings.
breast feeding; longitudinal studies; menarche; nutrition; Philippines
Abbreviation: CLHNS, Cebu Longitudinal Health and Nutrition Survey.
Editor’s note: A related article by Kramer et al. appears
on page 978, an invited commentary on both articles is
published on page 984, and the response by Kramer et al.
to the commentary is on page 988. In accordance with Jour-
nal policy, Al-Sahab et al. were asked if they wanted to
respond to the commentary but chose not to do so.
Menarche is a critical biomarker in the reproductive life
offemales (1, 2). It servesas an intermediate health outcome
that affects women’s wellbeing at later stages of life (3).
Currently, it is gaining more attention as a considerable
body of evidence suggests that the age at menarche has de-
clined in the past century (3–7). It has been shown that
menarcheal age in the United States, as well as Europe,
has decreased at a rate of 2–3 months per decade (5, 7).
Early menarche is among the few established risk factors
for breast cancer (8–11). It has also been associated with
metabolic syndrome (12), teenage depression (13), and
overweight (14, 15).
Only recently have studies begun to investigate the effect
of early life events on the timing of puberty, with the ma-
jority of studies focusing on birth weight, birth length, ges-
tational age, and ponderal index (1, 3, 4, 16). Although
nutrition in early development and childhood has a signifi-
cant effect on the timing of menarche (17), it is surprising
that little attention has been given to the potential role of
breastfeeding. Breastfeeding has been shown to be inversely
associated with weight gain during childhood (18–20),
which in itself is a risk factor for early age at menarche
(3, 21).
Although no study has previously investigated the asso-
ciation between breastfeeding and menarcheal age as its
primary objective, 3 studies examined this association as
part of determining the predictors of menarche (22–24).
One of these studies by Novotny et al. (22) revealed that
971 Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from

Page 2

formula-fed girls were 2.7 (95% confidence interval: 1.12,
6.38) times more likely to attain early menarche than
breastfed babies. The other 2 studies, however, failed to
detect significant associations between breastfeeding and
menarcheal timing (23, 24). Several notable limitations such
as small sample size, study design, long-term recall, and
inconsistency in breastfeeding definitions limit the general-
izability of these findings. Addressing these limitations, the
present study aims to assess the relation between breastfeed-
ing duration in the first months of life with age at menarche
among a cohort of Filipino girls who were followed from
infancy to adulthood.
MATERIALS AND METHODS
Sample
The analysis of this study was based on the Cebu Longi-
tudinal Health and Nutrition Survey (CLHNS) conducted by
the Carolina Population Center, University of North Caro-
lina at Chapel Hill, the Nutrition Center of the Philippines,
and the Office of Population Studies, University of San
Carlos. The study was initiated in 1981 with the aim of
determining prospectively the feeding patterns of infants
in the Metro Cebu Area, Philippines.
CLHNS initially recruited all pregnant women who gave
birth in a 12-month span (May 1983–April 1984) from 33
randomly selected communities in Metro Cebu (17 urban
and 16 rural). The inclusion criteria for the baseline sample
of pregnant women were as follows: 1) residency in one of
the 33 sample communities; 2) pregnancy termination be-
tween May 1, 1983, and April 30, 1984; and 3) delivery of
the sample child in a sample community or any health fa-
cility located in the Metro Cebu Area. All pregnant women
were interviewed during the second or third trimester of
pregnancy and at delivery. Singleton births were subse-
quently followed, and mothers and infants were assessed
bimonthly for 24 months. Follow-up studies were conducted
in 1991, 1995, 1998, 2002, and 2005. In this study,data from
the first 3 follow-up surveys, when girls were 8–9, 11–12,
and 14–15 years of age, were used. Detailed data on the
health and wellbeing of the mother and the child, household
characteristics, environmental conditions, and community
setting were collected by trained personnel. CLHNS has
been described previously (25, 26).
Age at menarche
The analysis sample for the present study includes all
girls who participated in the 1998 survey and had complete
data on breastfeeding. The main outcome of the study was
age at menarche, self reported in the 1995 and 1998 surveys
as the month and year of the occurrence of the first men-
strual cycle. For the purpose of this study, age at menarche
was used as a continuous and as a categorical variable. The
categorical variable was grouped into early (<12.1 years),
average (12.1–14.1 years), and late (>14.1 years) menarche.
The age limit for early and late menarche was determined as
1 standard deviation away from the estimated mean age at
menarche (3, 4). Premenarcheal girls who were 14 years of
age or older at the time of the 1998 interview were classified
in the late menarche group (3, 6, 21).
Breastfeeding
The durations of exclusive breastfeeding and any breast-
feeding were the main independent variables in the study.
Exclusive breastfeeding was defined as the intake of breast
milk alone without any other type of milk, liquid, or food
(27). Its duration was calculated as the infant’s age when
exclusive breastfeeding was terminated. Based on the 7-day
recall of feeding patterns, this information was reported bi-
monthly by the mother. Any breastfeeding, on the other
hand, was defined as the intake of breast milk regardless
of other type of food or drink intake (18, 28). The duration
of any breastfeeding, reported bimonthly by the mother for
the first 24 months of life, was defined as the infant’s age in
days when any breastfeeding had been permanently
stopped. Any breastfeeding duration for infants still breast-
feeding at 24 months was calculated as the last age when
breastfeeding status was known. To ensure the quality of
food intake and breastfeeding data, all personnel undertook
a 2-week training program in food models and measurement
of diets (29).
Covariates
A comprehensive list of covariates that have been found
to predict age at menarche was adjusted for in this study.
Socioeconomic status was represented through type of res-
idence (urban vs. rural), income, number of household
members, and mother’s highest educational level as col-
lected from the 1983–1984 baseline questionnaire. Maternal
characteristics including age at delivery, history of previous
births, skinfold thickness during pregnancy, and postpartum
body mass index were also examined from the baseline and
birth surveys. Triceps skinfold thickness was measured in
triplicate during the third trimester of pregnancy. Baseline
data on maternal age at menarche and maternal height were
derived from the baseline survey and used as proxy mea-
sures of genetic factors (3). Perinatal measures were deter-
mined from the birth survey. Birth weight was measured by
birth attendants after delivery. Babies missing this informa-
tion (12.5%) had their birth weight substituted by theweight
value measured by the survey staff within the first 8 days of
life. Birth length was measured by the survey staff using
custom-made measuring boards that complied with the stan-
dards of the US Centers for Disease Control and Prevention
(25). Gestational age was estimated from the date of the
mother’s last menstrual period. If this date was missing, if
the infant was low birth weight, or if the mother experienced
pregnancy complications, the Ballard method performed by
trained nurses was used instead to calculate gestational age
(3). All anthropometric measures, for both the child and
mother, were conducted by well-trained staff.
Statistical analysis
All analyses were performed by using SPSS, version
17.0, statistical software (SPSS, Inc., Chicago, Illinois).
972Al-Sahab et al.
Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from

Page 3

Initially, Kaplan-Meier analysis was used to estimate the
median age at menarche. Accordingly, the distribution of
early, average, and late menarche was calculated. This 3-
level categorical variable was used only for assessing the
characteristics of the population. Differences in the distri-
bution of categorical variables or means of continuous vari-
ables were tested by using the chi-square and analysis of
variance, respectively. Cox regression was used to investi-
gate the association of breastfeeding with age at menarche
while adjusting sequentially for different sets of covariates.
Age at menarchewas defined in this analysis as a continuous
variable taking into consideration girls who did not attain
their first menarche at the time of outcome assessment.
These observations were treated as censored data. Adjusted
hazard ratios with 95% confidence interval were reported
for the Cox regression models. Statistical significance for all
analyses was set at P < 0.05.
RESULTS
Of the 1,447 singleton females born alive in 1983–1984,
994 (68.7%) females were interviewed and completed the
1998 survey. The vast majority of the girls (94.9%) had
started menstruating at the time of the follow-up survey.
Only 51 girls were premenarcheal. By use of Kaplan-Meier
analysis, the estimated median age at menarche was 13.08
years. Half of the girls attained their menarche between the
ages of 12.42 and 13.92 years. The proportions of early,
average, and late maturers were 15.5%, 65.0%, and 19.5%,
respectively. Of the 994 girls, 974 girls had complete infor-
mation on breastfeeding. The mean duration of exclusive
breastfeeding in this population was 2.72 months (standard
deviation ¼ 1.89), while the mean duration of any breast-
feeding was 13.91 months (standard deviation ¼ 7.98).
Table 1 presents the characteristics of early, average, and
late maturers. The mean duration for exclusive and any
breastfeeding was significantly shorter among early matur-
ing girls as compared with average and late maturing girls.
Early maturers were also more likely to be firstborns and to
live in an urban community. Income and maternal educa-
tional level were higher among girls with early menarche in
contrast to other girls. Moreover, the mother’s age at men-
arche was significantly lower for girls attaining menarche at
a younger age. The mother’s postpartum body mass index
and triceps skinfold thickness during pregnancy were also
higher among early maturers than their counterparts. Fi-
nally, birth length was significantly different across the 3
menarche groups.
By use of Cox regression analysis, different model spec-
ifications are demonstrated in Table 2 for the impact of
Table 1. Characteristics of Early, Average, and Late Maturers in the Metro Cebu Area, Philippines, 1995–1998a,b
Early (<12.1 Years) Average (12.1–14.1 Years)Late (>14.1 Years)
Mean (SD)No.% Mean (SD)No.% Mean (SD)No.%
Exclusive breastfeeding, months (n ¼ 974)
Any breastfeeding, months (n ¼ 974)
Household income at baseline,
per 100 pesos/week (n ¼ 965)
No. of household members at baseline
(n ¼ 974)
Mother’s highest educational level at
baseline, years (n ¼ 974)
Mother’s age at menarche, years (n ¼ 973)
Mother’s height, cm (n ¼ 973)
Mother’s age at delivery, years (n ¼ 969)
Birth weight, kg (n ¼ 956)
Baby’s length at birth, cm (n ¼ 972)
Gestational age, weeks (n ¼ 919)
Mother’s body mass index after
birth, kg/m2(n ¼ 965)
Mother’s triceps skinfold thickness,
mm (n ¼ 974)
Place of residence (n ¼ 974)
Urban
2.4 (1.8) 2.6 (1.8)3.2 (2.0)
11.8 (8.3)13.8 (7.9) 16.1 (7.3)
3.1 (3.5)2.5 (2.5) 2.4 (2.4)
5.3 (3.0)5.7 (2.7) 5.8 (2.8)
8.8 (3.9) 7.3 (3.5)6.3 (3.3)
13.6 (1.5) 14.0 (1.5)14.6 (1.6)
150.8 (4.8) 150.6 (5.0)151.5 (5.1)
26.0 (5.2)26.9 (6.2)26.5 (5.6)
3.0 (0.5) 3.0 (0.4) 3.0 (0.4)
49.3 (1.9)49.1 (2.1)48.7 (2.2)
38.6 (3.3) 38.7 (2.8)38.5 (3.1)
22.7 (2.6) 22.1 (2.6)21.4 (2.1)
13.8 (4.0) 12.7 (3.9)11.7 (3.6)
124 81.0473 75.3113 58.5
Rural 29 19.0155 24.7 8041.5
First child (n ¼ 974)
No 10166.0 495 78.816183.4
Yes52 34.013321.2 3216.6
Abbreviation: SD, standard deviation.
aNumber and frequency for categorical variables.
bMean and standard deviation for continuous variables.
Impact of Breastfeeding on Age at Menarche973
Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from

Page 4

breastfeeding duration on age at menarche, with adjustment
for specific sets of known confounders. The sole effect of
breastfeeding variables on age at menarche has been as-
sessed in model 1. Shorter durations of both exclusive and
any breastfeeding were associated with earlier menarche. In
model 2, socioeconomic variables and maternal character-
istics were added to the analysis; in this case, only exclusive
breastfeeding remained significantly associated with men-
archeal timing. Urban residence, mother’s educational level,
postpartum body mass index, and being the first child were
also found to be associated with early age at menarche.
Model 3 added variables related to genetic factors and infant
characteristics. Urban residence and maternal education
sustained their significance. The negative association be-
tween exclusive breastfeeding and menarcheal timing was
not altered either (hazard ratio ¼ 0.94, 95% confidence in-
terval: 0.90, 0.98). Moreover, mother’s age at menarche and
birth length emerged as potential predictors for age at men-
arche, whereby older maternal age at menarche and shorter
birth length were associated with lower hazard ratios of
menarche. No significant interactions were found between
any and exclusive breastfeeding variables in any model.
DISCUSSION
The present study aimed to explore the association of
breastfeeding with menarcheal age. The estimated median
age at menarche for the Filipino cohort analyzed in this
study was 13.08 years. After adjustment for potential con-
founders of the association of breastfeeding with age at
menarche, exclusive breastfeeding duration retained an in-
dependent and significant association with age at menarche.
An increase in 1 month of exclusive breastfeeding decreased
the hazard of attaining earlier menarche by 6%. This asso-
ciation is a novel finding in the present study. Results sug-
gest that nutrition in early development and childhood plays
an important role in determining the timing of menarche.
One of the often cited pathways between breastfeeding
and age at menarche is weight gain during childhood. The
tempo of growth during infancy and childhood has been
associated with breastfeeding and age at menarche. Breast-
feeding in early life may prevent or delay weight gain during
childhood (18–20). According to a study conducted on
a Danish cohort, longer durations of any breastfeeding de-
creased the weight gain during the first year of life (18).
Babies breastfed for less than 20 weeks gained 317.4 g more
during the first year of life than thosewho breastfed for more
than 40 weeks (18). As explained by Ong et al. (30), the
growth patterns for 3-month-breastfed infants were slower
than those for formula-fed infants for the first 5 years of life.
Rapid weight gain during infancy has also been shown to
increase the risk of attaining earlier menarche (3, 21). Using
the same data set as the present study, Adair (3) revealed that
faster growth in weight and/or length at 6 months of age also
predicted young menarcheal age. Although the mechanism
Table 2.
Known Confounders in the Metro Cebu Area, Philippines, 1995–1998
Model Specifications for the Impact of Breastfeeding Duration on Age at MenarcheaWith Sequential Adjustment for Specific Sets of
Model 1
(Breast Feeding Effect)
Model 2Model 3
Model 1 1 Socioeconomic
and Maternal Related
Variables
Model 2 1 Genetic and
Infant Characteristics
Variables HRb
95% CIHR 95% CIHR 95% CI
Exclusive breastfeeding, months0.93**
0.89, 0.960.95*
0.91, 0.990.94**
0.90, 0.98
Any breastfeeding, months 0.99*
0.98, 0.991.00 0.99, 1.011.00 0.99, 1.01
Urban residence at baseline 1.25**
1.07, 1.47 1.24*
1.05, 1.47
Total household income at baseline, 100 pesos/week0.990.96, 1.021.00 0.96, 1.03
No. of household members at baseline 0.98 0.96, 1.010.970.94, 1.00
Mother’s education at baseline, years1.03**
1.01, 1.051.02*
1.00, 1.05
First child1.25*
1.04, 1.49 1.200.99, 1.44
Mother’s age at delivery, years
Mother’s body mass index after birth, kg/m2
1.01 0.99, 1.02 1.010.99, 1.02
1.05**
1.01, 1.09 1.030.99, 1.07
Mother’s triceps skinfold thickness, mm1.010.99, 1.03 1.020.99, 1.04
Mother’s age at menarche, years0.91**
0.87, 0.95
Mother’s height, cm0.99 0.97, 1.00
Birth weight, kg0.83 0.67, 1.02
Birth length, cm 1.07**
1.02, 1.11
Gestational age, weeks 1.010.98, 1.03
Abbreviations: CI, confidence interval; HR, hazard ratio.
* P < 0.05; **P < 0.01.
aAge at menarche was used as a continuous variable whereby premenarcheal girls’ data were considered as censored.
bHazard ratios greater than 1 and less than 1 signify higher and lower risks of attaining earlier menarche, respectively.
974Al-Sahab et al.
Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from

Page 5

explaining the link between infant weight gain and age at
menarche is still unclear, it has been speculated that a rise in
leptin among rapid weight gainers triggers the onset of pu-
berty (31). Similarly, the increased concentration of insulin-
like growth factor I due to weight gain during infancy may
also promote the initiation of puberty (31).
Until recently, childhood obesity has been thought to be
the link as well between menarche and breastfeeding. Child-
hood obesity has previously been associated with early age
at menarche (1, 4, 32, 33). Given that a critical proportion of
body fat (17%–22%) may be required to trigger menarche
(34–36), it is not surprising that, in multivariable analyses,
a lower expected birth weight ratio and higher body mass
index at 8 years of age were found to be the strongest pre-
dictors of early menarche (4). However, based on the clin-
ical trials by Kramer et al. (37–39), it is suggested that
breastfeeding has no impact on childhood weight. Accord-
ing to Kramer, previous findings on the protective role of
breastfeeding on childhood weight are attributed mainly to
unmeasured confounders (39).
The present study assessed the total effect of exclusive
breastfeedingonage atmenarche.Althoughdata onpostnatal
growth, childhood anthropometrics, and childhood energy
intake and diet was collected in the CLHNS, it was not con-
sidered in the analysis. These variables are believed to lie on
the causal pathway between breastfeeding and age at menar-
che.Thedirectpathway, therefore,through which breastfeed-
ing can influence age at menarche is still elusive. Because the
infant’s ability to develop physiologic changes in response to
the environment extends to the postnatal period (40), nutri-
tional programming is plausible and may cause long-term
consequences in later life (41). In 2004, Singhal et al. (42)
provided evidence on the positive programming effect of
breast milk on the lipoprotein profile, while the programming
of undernutrition in early life may delay the development of
insulin resistance for preterm babies (43).
The study findings suggest that girls exclusively breastfed
for longer durations had a decreased hazard ratio for menar-
che than their counterparts. One previous cross-sectional
study found that formula-fed girls were 2.7 (95% confidence
interval: 1.12, 6.38) times more likely to attain early menar-
che than breastfed babies (22); however, this study was sub-
ject to recall bias, as mothers of the study participants were
asked to report their breastfeeding practices 9–14 years prior.
Moreover, the sample size was limited (n ¼ 349), and exclu-
sive breastfeeding and duration of breastfeeding/bottle feed-
ing were not considered in the analysis (22). Two other
longitudinalstudies byKaraolis-Danckert etal.(23)andBlell
et al. (24) aimed to assess the general predictors of menarche
and puberty. As one of the predictors, breastfeeding in both
studies was not found to be associated with menarche (23,
24). More specifically, Blell et al. (24) examined the effect of
duration of partial breastfeeding on age at menarche that was
reported by women aged 49–51 years. Other than the long
recall period for age at menarche, only 276 women of the
1,142 original cohort were analyzed in the study (24). The
study by Karaolis-Danckert et al. (23), on the other hand,
investigated the effect of full breastfeeding at 4 months on
age at take-off of the pubertal growth spurt and age at peak
height velocity among girls and boys. Similar to the study by
Blell et al. (24), the sample size analyzed (n ¼ 215) was at
least 5 times less than the original cohort (23).
Interestingly, in the present study, age at menarchewas not
associated with the duration of any breastfeeding, suggesting
that the exclusivity of breastfeeding is more important. How-
ever, the reason behind the superiority of exclusive breast-
feeding over any breastfeeding in terms of age at menarche is
not clear. It can be speculated that the difference in effects
betweenthe2typesofbreastfeedingisattributedtothenature
ofcomplementaryfoods,levelsofenergyintake,weightgain,
and health status of the baby. These factors are relatively
dependent on geographic areas and cultural variations. This
association may be more pronounced in developed countries
where complementary foods are more nutritive and energy
dense than in developing countries. Understanding this asso-
ciation in different contexts is thus highly warranted.
Several limitations of the current study warrant mention.
As expected, the present study lost participants to follow-up.
Of the 1,447 singleton girls born alive in 1983–1984, 974
(67.3%) were considered in the analysis. In terms of socio-
economic and maternal related characteristics, girls with
missingdataweresomewhatmorelikelytobeurbanresidents
and to have mothers with a higher educational level. More-
over, girls excluded from the analysis had shorter durations of
both, any and exclusive breastfeeding. Therefore, the gener-
alizability of the results to the original cohort must be made
with caution. In addition to the above, the study is subject to
information bias due to possible measurement errors in the
outcome, exposure, and confounding variables. Age at men-
arche was self-reported by the girls. Menarche data, however,
were collected close to the time of the event, reducing the
likelihood of recall bias. Errors in reporting breastfeeding
practices are also possible as mothers might not remember
accurately the timing of liquid or food introduction and/or
breastfeeding termination. To ensure the quality of food in-
take and breastfeeding data, on the other hand, all personnel
undertook a 2-week training program in food models and
measurement of diets. Residual confounding due to unmea-
sured variables like diet during pregnancy, intrauterine life,
and genetic factors also stands out as another limitation.
Although this is the first longitudinal study that reveals
a negative association between exclusive breastfeeding and
early menarche, the relation is still elusive. The results sug-
gest that duration of exclusive breastfeeding (rather than any
breastfeeding) reduces the chances of attaining early menar-
che. Thus, nutrition during infancy could be an ‘‘early win-
dow’’ that might program a child’s future growth and
development (31, p. 140). It is also among the few childhood
modifiable risk factors that might prevent or delay diseases in
adulthood.Nevertheless,thepotentialmechanismforthe pro-
gramming of breastfeeding on age at menarche remains un-
clear. Further longitudinal studies within different contexts
are warranted to assess the generalizability of these findings.
ACKNOWLEDGMENTS
Author affiliations: School of Kinesiology and Health
Science, York University, Toronto, Ontario, Canada (Ban
Impact of Breastfeeding on Age at Menarche 975
Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from

Page 6

Al-Sahab, Mazen J. Hamadeh, Chris I. Ardern, Hala
Tamim); and Department of Nutrition, School of Public
Health and Medicine, University of North Carolina at
Chapel Hill, Chapel Hill, North Carolina (Linda Adair).
Funding for the CLHNS project design, data collection,
and database development was provided by the Ford Foun-
dation; the US National Academy of Sciences, the National
Institutesof Health (grants
HD18880, P01-HD28076, and R01-HD23182); the Nestle
Coordinating Center for Nutrition Research; Wyeth Interna-
tional; and the US Agency for International Development.
Conflict of interest: none declared.
R01-HD19983A,R01-
REFERENCES
1. Sloboda DM, Hart R, Doherty DA, et al. Age at menarche:
influences of prenatal and postnatal growth. J Clin Endocrinol
Metab. 2007;92(1):46–50.
2. Thomas F, Renaud F, BeneficeE, et al. International variability
of ages at menarche and menopause: patterns and main de-
terminants. Hum Biol. 2001;73(2):271–290.
3. Adair LS. Size at birth predicts age at menarche [electronic
article]. Pediatrics. 2001;107(4):e59.
4. Tam CS, de Zegher F, Garnett SP, et al. Opposing influences
of prenatal and postnatal growth on the timing of menarche.
J Clin Endocrinol Metab. 2006;91(11):4369–4373.
5. Ersoy B, Balkan C, Gunay T, et al. Effects of different socio-
economic conditions on menarche in Turkish female students.
Early Hum Dev. 2004;76(2):115–125.
6. Windham GC, Bottomley C, Birner C, et al. Age at menarche
in relation to maternal use of tobacco, alcohol, coffee, and tea
during pregnancy. Am J Epidemiol. 2004;159(9):862–871.
7. Petridou E, Syrigou E, Toupadaki N, et al. Determinants of age
at menarche as early life predictors of breast cancer risk. Int J
Cancer. 1996;68(2):193–198.
8. Magnusson CM, Persson IR, Baron JA, et al. The role of re-
productive factors and use of oral contraceptives in the aeti-
ology of breast cancer in women aged 50 to 74 years. Int J
Cancer. 1999;80(2):231–236.
9. Rockhill B, Moorman PG, Newman B. Age at menarche, time
to regular cycling, and breast cancer (North Carolina, United
States). Cancer Causes Control. 1998;9(4):447–453.
10. Hamilton AS, Mack TM. Puberty and genetic susceptibility to
breast cancer in a case-control study in twins. N Engl J Med.
2003;348(23):2313–2322.
11. Maskarinec G, Zhang Y, Takata Y, et al. Trends of breast
cancer incidence and risk factor prevalence over 25 years.
Breast Cancer Res Treat. 2006;98(1):45–55.
12. Frontini MG, Srinivasan SR, Berenson GS. Longitudinal
changes in risk variables underlying metabolic syndrome X
from childhood to young adulthood in female subjects with
a history of early menarche: the Bogalusa Heart Study. Int J
Obes Relat Metab Disord. 2003;27(11):1398–1404.
13. Kaltiala-Heino R, Kosunen E, Rimpela ¨ M. Pubertal timing,
sexual behaviour and self-reported depression in middle ado-
lescence. J Adolesc. 2003;26(5):531–545.
14. Adair LS, Gordon-Larsen P. Maturational timing and over-
weight prevalence in US adolescent girls. Am J Public Health.
2001;91(4):642–644.
15. Wattigney WA, Srinivasan SR, Chen W, et al. Secular trend of
earlier onset of menarche with increasing obesity in black and
white girls: the Bogalusa Heart Study. Ethn Dis. 1999;9(2):
181–189.
16. Romundstad PR, Vatten LJ, Nilsen TI, et al. Birth size in re-
lationto ageatmenarcheand adolescentbody size: implications
for breast cancer risk. Int J Cancer. 2003;105(3):400–403.
17. Gluckman PD, Hanson MA. Evolution, development and
timing of puberty. Trends Endocrinol Metab. 2006;17(1):7–12.
18. Baker JL, Michaelsen KF, Rasmussen KM, et al. Maternal
prepregnant body mass index, duration of breastfeeding,
and timing of complementary food introduction are associated
with infant weight gain. Am J Clin Nutr. 2004;80(6):
1579–1588.
19. Ong KK, Emmett PM, Noble S, et al. Dietary energy intake at
the age of 4 months predicts postnatal weight gain and child-
hood body mass index [electronic article]. ALSPAC Study
Team. Pediatrics. 2006;117(3):e503–e508.
20. Singhal A, Lanigan J. Breastfeeding, early growth and later
obesity. Obes Rev. 2007;8(suppl 1):51–54.
21. dos Santos Silva I, De Stavola BL, Mann V, et al. Prenatal
factors, childhood growth trajectories and age at menarche. Int
J Epidemiol. 2002;31(2):405–412.
22. Novotny R, Daida YG, Grove JS, et al. Formula feeding in in-
fancy is associated with adolescent body fat and earlier menar-
che. Cell Mol Biol (Noisy-le-grand). 2003;49(8):1289–1293.
23. Karaolis-Danckert N, Buyken AE, Sonntag A, et al. Birth and
early life influences on the timing of puberty onset: results
from the DONALD (DOrtmund Nutritional and Anthropo-
metric Longitudinally Designed) Study. Am J Clin Nutr.
2009;90(6):1559–1565.
24. Blell M, Pollard TM, Pearce MS. Predictors of age at menar-
che in the Newcastle Thousand Families Study. J Biosoc Sci.
2008;40(4):563–575.
25. Cebu Longitudinal Health and Nutrition Survey. Chapel Hill,
NC: Carolina Population Center, University of North Carolina
at Chapel Hill; 2008. (http://www.cpc.unc.edu/projects/cebu).
(Accessed May 8, 2008).
26. Underlying and proximate determinants of child health:the
Cebu Longitudinal Health and Nutrition Study. Am J Epide-
miol. 1991;133(2):185–201.
27. World Health Organization. Up to what age can a baby stay
well nourished by just being breastfed? Geneva, Switzerland:
World Health Organization; 2008. (http://www.who.int/
features/qa/21/en/index.html). (Accessed April 21, 2008).
28. Adair LS, Popkin BM, Guilkey DK. The duration of breast-
feeding: how is it affected by biological, sociodemographic,
health sector, and food industry factors? Demography. 1993;
30(1):63–80.
29. Zohoori N, Popkin BM, Fernandez ME. Breast-feeding pat-
terns in the Philippines: a prospective analysis. J Biosoc Sci.
1993;25(1):127–138.
30. Ong KK, Preece MA, Emmett PM, et al. Size at birth and early
childhood growth in relation to maternal smoking, parity
and infant breast-feeding: longitudinal birth cohort study
and analysis. ALSPAC Study Team. Pediatr Res. 2002;52(6):
863–867.
31. Dunger DB, Ahmed ML, Ong KK. Early and late weight gain
and the timing of puberty. Mol Cell Endocrinol. 2006;254–
255:140–145.
32. Cooper C, Kuh D, Egger P, et al. Childhood growth and age at
menarche. Br J Obstet Gynaecol. 1996;103(8):814–817.
33. Anderson SE, Dallal GE, Must A. Relative weight and race
influence average age at menarche: results from two nationally
representative surveys of US girls studied 25 years apart. Pe-
diatrics. 2003;111(4 pt 1):844–850.
34. Frisch RE, McArthur JW. Menstrual cycles: fatness as a de-
terminant of minimum weight for height necessary for their
maintenance or onset. Science. 1974;185(4155):949–951.
976Al-Sahab et al.
Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from

Page 7

35. Apter D, Viinikka L, Vihko R. Hormonal pattern of adolescent
menstrual cycles. J Clin Endocrinol Metab. 1978;47(5):
944–954.
36. Van der Spuy ZM. Nutrition and reproduction. Clin Obstet
Gynaecol. 1985;12(3):579–604.
37. Kramer MS, Matush L, Bogdanovich N, et al. Health and
development outcomes in 6.5-y-old children breastfed
exclusively for 3 or 6 mo. Am J Clin Nutr. 2009;90(4):
1070–1074.
38. Kramer MS, Matush L, Vanilovich I, et al. Effects of pro-
longed and exclusive breastfeeding on child height, weight,
adiposity, and blood pressure at age 6.5 y: evidence from
a large randomized trial. Am J Clin Nutr. 2007;86(6):
1717–1721.
39. Kramer MS, Matush L, Vanilovich I, et al. A randomized
breast-feeding promotion intervention did not reduce child
obesity in Belarus. J Nutr. 2009;139(2):417S–421S.
40. Barker DJ. The origins of the developmental origins theory.
J Intern Med. 2007;261(5):412–417.
41. Gluckman PD, Cutfield W, Hofman P, et al. The fetal, neo-
natal, and infant environments—the long-term consequences
for disease risk. Early Hum Dev. 2005;81(1):51–59.
42. Singhal A, Cole TJ, Fewtrell M, et al. Breastmilk feeding and li-
poprotein profile in adolescents born preterm: follow-up of a pro-
spective randomised study. Lancet. 2004;363(9421):1571–1578.
43. Singhal A, Fewtrell M, Cole TJ, et al. Low nutrient intake and
early growth for later insulin resistance in adolescents born
preterm. Lancet. 2003;361(9363):1089–1097.
Impact of Breastfeeding on Age at Menarche977
Am J Epidemiol. 2011;173(9):971–977
at York University Libraries on May 1, 2011
aje.oxfordjournals.org
Downloaded from