Infant Feeding and Weight in the First Year of Life in Babies of Women
with Eating Disorders
NADIA MICALI, MD, MRC PSYCH, EMILY SIMONOFF, MD, FRC PSYCH, AND JANET TREASURE, PHD, FRC PSYCH
studied. Prospectively collected data on feeding difficulties at age 1 and 6 months, breast-feeding during the first year, and
weight and conditional growth at age 9 months were compared for infants of women with a self-reported history of an eating
disorder (anorexia nervosa or bulimia nervosa) and women with and without other severe psychiatric disorders.
The women with eating disorders were more likely to breast-feed. Infants of women with anorexia nervosa were at
higher risk for feeding difficulties between age 0 and 6 months compared with those of women without psychiatric disorders,
after controlling for relevant confounders. Women with other psychiatric disorders reported more feeding difficulties than
those without psychiatric disorders. Infants of bulimic women were significantly more likely to be overweight and to have faster
growth rates at age 9 months compared with controls.
Maternal eating disorders affect infant feeding and growth in the first year. Health professionals should be alert
to these potential effects. (J Pediatr 2008;xx:xxx)
Mothers with eating disorders tend to experience difficulties feeding their children, including refusal of solids, more
scheduled feedings, and other problems.8Although the literature has mainly investigated children as opposed to infants, those
studies focusing on infants have found similar differences in feeding patterns, including
lower rates and shorter duration of breast-feeding,9faster sucking,10and lower weight.11
Most previous studies, however, have relied on small clinical samples and have not
included a control sample. Thus, it is of interest to investigate infants of women with
eating disorders in a large population-based longitudinal cohort, the Avon Longitudinal
Study of Parents and Children (ALSPAC).
We aimed to determine whether rates and duration of breast-feeding, rates and
persistence of feeding difficulties, and weight and growth outcomes at age 9 months
differed between infants of women with eating disorders and infants of women with and
without other psychiatric disorders in a prospective population-based cohort. We included
women with other psychiatric disorders as a comparison group to study whether the
effects of a maternal eating disorder were specific or could be attributed to having any type
of psychiatric disorder.
To examine feeding patterns and growth in the first year of life in infants of women with eating disorders in a
Women and their infants (n ? 12 050) from the Avon Longitudinal Study of Parents and Children were
eeding difficulties are relatively common in infancy1,2and are likely to persist into childhood.3,4An infant’s feeding
patterns have an immediate impact, causing poor growth, for example,5as well as long-lasting effects on well-being, such
as influences on weight gain6and cognitive development.7
The ALSPAC is a longitudinal, population-based, prospective study of women
and their children.12All pregnant women living in the geographical area of Avon, UK
ALSPAC Avon Longitudinal Study of Parents and
Body mass index
Relative risk ratio
From the Child and Adolescent Psychiatry
Department, Institute of Psychiatry (N.M.,
E.S.), and Eating Disorders Research Unit,
Department of Academic Psychiatry, Guy’s
Hospital (J.Y.), King’s College London, Lon-
Funded by the National Alliance for Schizo-
phrenia and Depression and the National
Eating Disorders Association. The funding
agencies had no involvement with study
design, data collection, analysis, and inter-
pretation of data; the writing of the report;
and the decision to submit the manuscript
for publication. The authors declare no
conflicts of interest.
Submitted for publication Feb 26, 2008; last
revision received May 22, 2008; accepted
Jul 2, 2008.
Reprint requests: Dr Nadia Micali, King’s
College London, Institute of Psychiatry, De-
partment of Child and Adolescent Psychi-
atry, De Crespigny Park, London SE5 8AF,
UK. E-mail: email@example.com.
0022-3476/$ - see front matter
Copyright © 2008 Mosby Inc. All rights
ARTICLE IN PRESS
who were expected to give birth between April 1, 1991 and
December 31, 1992 were invited to participate in the study.
All of the women who agreed to participate gave informed
and written consent. The sample has been shown to be
representative of the British population. A total of 14 663
women were enrolled by the 9th week of pregnancy; data
were obtained on 14 472 of these women by postal ques-
tionnaires. Women were excluded from the current study if
they had not answered the questionnaire after approxi-
mately 12 weeks (n ? 2019). The questionnaire elicited
recent or past history of psychiatric problems, including
depression, schizophrenia, and alcoholism (n ? 1148 total,
including 45 with drug addiction, 83 with alcoholism, 4
with schizophrenia, 938 with severe depression, and 231
with other psychiatric problems); anorexia nervosa (self-
report; n ? 247); and bulimia nervosa (self-report; n ?
194). Women reporting a history of both eating disorders
(n ? 82) were included in the anorexia nervosa group,
given the comparable baseline characteristics with women
with a history of anorexia only. Only singleton births and
infants who were alive at age 1 year (n ? 12 050) were
included. Sociodemographic data, including maternal
height and weight at the time of conception, were obtained
during pregnancy. The study design was approved by the
Institute of Psychiatry Ethics Committee (Ref. 110/02),
the ALSPAC Law and Ethics Committee, and the local
research ethics committees.
The women were asked about breast-feeding at 1, 6,
and 15 months postnatally. Feeding data at age 1 and 6
months were obtained by questionnaire. The questions were
specifically developed for this study to determine the most
common feeding behaviors and difficulties. At age 1 month,
the occurrence of weak sucking, overly fast drinking, and
exhaustion with feeding was assessed.
At both 1 month and 6 months, each mother was asked
whether her infant took only small quantities at each feed, was
slow in feeding, and was unsatisfied or hungry after feeding
(at 6 months the mother indicated occurrence of these prob-
lems at age 0 to 3 months, age 4 to 6 months, or age 0 to 6
months). At 6 months, the mother was asked whether the
infant refused solids and whether a feeding routine had been
established (again at age 0 to 3 months, 4 to 6 months, or 0
to 6 months).
Duration of feeding difficulties was established by com-
bining data from the 1-month and 6-month questionnaires to
generate a composite outcome. The data collected at both
time points were consistent. If a mother reported a feeding
difficulty on both the 1-month and 6-month questionnaires,
then difficulties were coded as early-onset persistent. If a
mother reported difficulties on only 1 of the 2 questionnaires
(either the 1-month or 6-month), then these were coded as
transient. We conceptualized transient feeding difficulties as
being less severe and possibly having less impact on the infant
and mother.4This was corroborated by preliminary analyses
on the effect of persistent feeding difficulties on child growth
and development. Consequently, in this study we focused on
early-onset persistent feeding difficulties.
Missing data on feeding outcomes and on the con-
founders tested reduced the sample to 11 261 (93.5%) at 1
month and between 8567 (71.1%) and 9976 (82.8%) at 6
months. Attrition was associated with maternal age, parity,
and employment. To take into account the possible con-
founding role of infant developmental problems due to the
association between developmental problems, feeding diffi-
culties and poor growth,13,14we included in the analyses a
mother-completed version of the Denver Developmental
Scale for the child at age 6 months.15
Data on weight were obtained from the children’s
health records. We analyzed weight data collected at age 9
months (with these weights corrected by the child’s age at
measurement and converted to z-scores16) and conditional
growth scores between birth and 9 months (with the weights
converted to z-scores, which took into account birth weight,
sex, and regression from the mean, using 1990 British growth
reference data17). Infants whose weight for age (ie, z-score ?
1.645 standard deviation [SD] from the mean) was below the
5th percentile were classified as “underweight,”18those whose
weight for age was above the 95th percentile17were classified
as “overweight.” The same method was applied to growth
z-scores to define “poor” and “rapid” growth. Missing data on
weight z-scores and relevant confounders reduced the sample
for these analyses to 8213 (68.1%).
Data on breast-feeding were analyzed using the Cox
regression method19to test group differences in the rates of
women who ceased breast-feeding at different time points in
the first year of infant life. Group comparisons used paramet-
ric (1-way analysis of variance) and nonparametric tests as
appropriate, after testing for normality. Linear regression
models tested for predictors of continuous outcomes.
Multinomial and binary logistic regression models exam-
ined predictors of categorical and binary outcomes, respec-
tively. Potential covariates likely to influence outcomes
(based on the literature) were first tested in univariate
models and were included in multivariate models when
significant at the 5% level. The final model accounted for
the main effects of each covariate. Testing for selective
attrition in the index groups found none. Overall attrition
was related to maternal age and education. The models
built to investigate the relative importance of maternal
group in predicting feeding problems and growth were
based on cases with complete data on all variables. All
analyses were performed using Stata version 9 for Windows
(StataCorp, College Station, Texas). All statistical tests
were 2-tailed. A P value ? .05 was considered significant.
2 Micali, Simonoff, and TreasureThe Journal of Pediatrics • Month 2008
ARTICLE IN PRESS
Maternal sociodemographic and infant data were com-
pared across the groups. The women with a history of eating
disorders were slightly older and less likely to be married than
the controls. The women with other psychiatric disorders
were less likely to be married, to be employed full time, and
to be primiparous; their infants were slightly lighter at birth
compared with those of the controls (Table I; available at
The women with a history of eating disorders were
more likely to start breast-feeding compared with those in the
2 other groups: 83%, compared with 76% of the general
population controls (odds ratio [OR] ? 1.5; 95% confidence
interval [CI] ? 1.2 to 2.0; P ? .002) and 72% of women with
other psychiatric disorders (OR ? 1.9; 95% CI ?1.4 to 2.6;
P ? .001). The women with eating disorders were less likely
to stop breast-feeding during the first year of infant life
compared with the general population controls (hazard ratio
[HR] ? 0.8; 95% CI ? 0.7 to 0.9; P ? .05) and those with
other psychiatric disorders (?2? 5.8; P ? .01). In contrast,
the women with other psychiatric disorders were more likely
to stop breast-feeding than the controls (HR ? 1.1; 95%
CI ? 1.0 to 1.2; P ? .05) (Figure).
Among the women with eating disorders, those with
bulimia nervosa were most likely to continue breast-feeding
(HR ? 0.8; 95% CI ? 0.6 to 0.9; P ? .05). These differences
persisted after adjusting for confounders previously identified
as affecting breast-feeding, including maternal age, parity,
ethnicity, and education and infant sex.
Feeding difficulties were relatively common across all
groups in the first month of infant life (Table II; available at
www.jpeds.com). The women with other psychiatric disor-
ders reported higher rates of weak sucking and exhaustion
with feeding compared with the controls, but comparable
rates to the women with eating disorders. The rate of overly
fast drinking were comparable in the infants of women with
other psychiatric disorders and controls and higher than that
in the infants of women with eating disorders (Table II).
The women with anorexia nervosa reported more early-
onset persistent (at both 1 month and 6 months) feeding
difficulties in all domains studied except refusal to take solids
compared with controls, but comparable with those with
other psychiatric disorders (Table III). In contrast, infants of
the women with bulimia nervosa differed from those of con-
trols in the rate of refusal to take solids and from those of the
women with anorexia nervosa in the rate of being unsatisfied/
hungry after feeding (Table III). The women with other
psychiatric disorders reported more difficulties in all feeding
domains investigated at 1 month and 6 months compared
Infant factors significantly related to feeding difficulties in
univariate analyses were birth weight, gestational age, and sex;
significant maternal factors were parity, employment status, and
cohabitation status. After these variables were entered into the
multivariate analyses, the differences among groups persisted for
most of the outcomes studied (Table III).
Weight and Growth at 9 Months
All feeding difficulties assessed at age 1 month and 6
months except no established feeding routine affected weight
z-scores at 9 months. Slow feeding, small quantity of feeds,
and refusal of solids were positively related to lower weight
and negatively associated with higher weight. Being unsatisfied/
hungry after feeding predicted higher weight at age 9 months
and thus was retained as an individual predictor. We generated
those feeding difficulties with a similar effect on weight (infants
with slow feeding, small quantity feeds, or refusal of solids were
coded as 1; those with none of these difficulties, as 0).
In univariate analyses, the feeding difficulties variable
predicted lower weight (relative risk ratio [RRR] ? 1.2; 95%
CI ? 1.2 to 1.3; P ? .001) and growth z-scores (RRR ? 1.2;
95% CI ? 1.1 to 1.3; P ? .001) at age 9 months. Being
unsatisfied/hungry after feeding predicted higher weight
(RRR ? 1.1; 95% CI ? 1.0 to 1.2; P ? .05) and growth
z-scores (RRR ? 1.2; 95% CI ? 1.1 to 1.3; P ? .001).
A maternal history of bulimia nervosa was predictive of
the infant being overweight in univariate analyses (RRR ?
1.8; 95% CI ? 1.1 to 3.1; P ? .05). When all other univariate
predictors of weight at age 9 months (ie, child sex, maternal
height, ethnicity, and breast-feeding during the first
6 months) were included in the multivariate model, maternal
bulimia was still predictive of the infant being overweight
(Table IV). The infants of the women with bulimia nervosa
were heavier and grew more rapidly between birth and
9 months compared with those born to control women and
those born to women with other psychiatric disorders (?2?
4.0, P ? .05 and ?2? 8.4, P ? .05, respectively) in univariate
analyses. These effects persisted after relevant covariates (ma-
Figure. Kaplan-Meier survival estimates by group.
Infant Feeding and Weight in the First Year of Life in Babies of Women with Eating Disorders3
ARTICLE IN PRESS
ternal height and ethnicity; infant sex, birth weight, and
gestation) were included (Table IV).
To explore the relationship between feeding problems
and developmental problems, we investigated whether the
effect of the studied predictors was relevant for all children,
including those with developmental problems. We tested the
model on the whole sample of children and then after exclud-
ing those children who scored ?2 SD below the mean on the
Denver Development scale.20The same relationships among
the variables were seen in both cases.
We investigated the role of maternal body mass index
(BMI) at conception on infant weight at age 9 months. We
found that maternal BMI at conception significantly predicted
in univariate analysis but did not affect the RRR for infant
overweight or rapid growth in multivariate analysis.
The present study compared breast-feeding rates and du-
ration, feeding difficulties, and weight and growth in the first
year of life in infants of women with a history of eating disorders
and women with and without other severe psychiatric disorders.
The women with eating disorders (particularly those
with a history of bulimia nervosa) were more likely than the
Table III. Feeding problems at 1 and 6 months: RRR with 95% CIs
Unadjusted RRRAdjusted RRR†
Feeding problems at 1
month (n ? 11 261)
Drinking too fast
1.3 (1.1-1.6)*** 0.9 (0.6-1.3)
1.2 (0.8-1.7)1.1 (0.7-1.7)
a?b?c Exhaustion with feeding
Feeding problems between
0 and 6 months
Slow feeding (n ? 9540) 2.1 (1.3-3.4)*
Small quantity feeding
(n ? 9745)
(n ? 8935)
Refused to take solids
(n ? 9511)
No established feeding
routine (n ? 8567)
1.0 (0.8-1.3)1.1 (0.8-1.5)1.3 (1.2-1.5)*1.1 (0.8-1.4) 1.1 (0.8-1.5) 1.3 (1.1-1.4)*
1.3 (0.7-2.3)1.9 (1.5-2.5)*
1.7 (1.1-2.6)*** 1.7 (1.0-2.8)
1.3 (0.7-2.4) 1.8 (1.4-2.4)*
a?b?c1.7 (1.1-2.6)** 1.7 (1.0-2.8)
2.4 (1.5-3.9)* 0.8 (0.4-1.5)2.2 (1.7-2.8)*2.5 (1.5-4.0)*0.8 (0.4-1.6) 2.1 (1.6-2.7)*a?b***; a?c;
0.6 (0.1-2.6)3.1 (1.4-6.8)** 2.4 (1.6-3.6)* 0.6 (0.1-2.5) 2.9 (1.3-6.4)** 2.6 (1.7-3.8)*b?a; a?c;
2.4 (1.2-4.9)** 1.0 (0.3-3.3)1.5 (1.0-2.4)*** 2.2 (1.1-4.5)*** 0.9 (0.3-3.1)1.6 (1.0-2.5)*** a?b?c
*P ? .001, **P ? .01, ***P ? .05; the general population acted as the reference sample.
†Adjusted for corrected birth weight, sex, parity, maternal employment, cohabition status, and breast-feeding at 1 month (for data relating to 1 month) and at 6 months (for data
relating to 6 months).
Table IV. Z-scores for weight and growth at 9 months: Adjusted† multivariate RRRs and 95% CIs
Univariate predictors Underweight RRR Overweight RRRSlow growth RRRRapid growth RRR
Maternal psychiatric history
Anorexia nervosa (n ? 208)
Bulimia nervosa (n ? 166)
Other psychiatric disorders (n ? 961)
Feeding problems in first 6 months of life
Feeding problems score‡
Hungry/not satisfied with feeding
Breast-feeding in first 6 months of life
*P ? .001, **P ? .05; the general population acted as the reference sample.
†The model includes maternal height, ethnicity, sex of the child, birth weight, and gestation.
‡This includes slow feeding, small quantity feeds, no established feeding routine, refusal to take solids.
4 Micali, Simonoff, and TreasureThe Journal of Pediatrics • Month 2008
ARTICLE IN PRESS
others to commence and continue breast-feeding. Previous
studies have reported contrasting findings on breast-feeding
in women with eating disorders. One community study found
that women with an active eating disorder were less likely to
breast-feed than controls,12whereas another study found sim-
ilar rates of breast-feeding in women with eating disorders
and controls, although the former group reported more dif-
ficulties with breast-feeding.20These discrepant findings may
be due to differences in case definition or case mix.
Interestingly, a study on this same sample (ALSPAC)
found that pregnant women with greater concerns about their
body shape and weight were less likely to express an intention
to breast-feed.21That study did not investigate maternal
eating disorders, however, and concerns about weight and
body shape were common in the pregnant women without
eating disorders in ALSPAC.22Another possible reason for
these discrepant findings is that many women change their
minds about breast-feeding after giving birth. Our findings
can be explained in several ways. Previous studies have sug-
gested that some mothers with bulimia nervosa worry that
they might “binge” on their child’s food,23and thus they may
decide to prolong breast-feeding to avoid cooking or other-
wise preparing food.
Infant Feeding Difficulties
The women with a history of anorexia nervosa reported
a higher rate of persistent feeding difficulties during the first
6 months compared with the women without psychiatric
disorders, but a similar rate as the women with other psychi-
atric disorders. The infants of women with bulimia nervosa
had higher rates of refusal to take solids compared with the
infants of controls; this behavior has been related to longer
duration of breast-feeding in the literature.24
Previous studies have found associations between ma-
ternal eating disorders and childhood feeding.11A study on
objective sucking rate in 2- and 4-week-old infants reported
higher rates in daughters of women with eating disorders
compared with controls.10Unlike our study, this study did not
distinguish between eating disorders. Even though maternal
psychopathology has been linked to controlling and restrictive
feeding patterns25and persistent food refusal,26no previous
study has investigated rates of feeding problems in infants of
women with other psychiatric disorders.
Weight and Growth at 9 Months
Overall, an infant being underweight at age 9 months
was predicted by feeding difficulties in the first 6 months of
life, whereas being overweight at age 9 months was predicted
by dissatisfaction/hunger after feeding. A maternal history of
bulimia nervosa predicted more rapid growth and overweight
at age 9 months over and above the presence of feeding
difficulties. This finding has not been reported previously.
Possible explanations for this are that these infants are genet-
ically at risk for higher weight, or that these women overeat
during pregnancy, affecting their infants’ appetite regulation
after birth. A third possibility is that a mother exerting greater
control during feeding predicts greater infant weight gain in
the first year of life.27Our findings differ from those of
Stein,15who found that infants of women with current or past
bulimia, an eating disorder not otherwise specified, or a sub-
threshold eating disorder weighed less at age 12 to 14 months
compared with infants of control and depressed women. Our
women with eating disorders may not be comparable with
those in Stein’s study, or, alternatively, the effect in the latter
may be driven by women with active illness.
Interestingly, although the women with anorexia ner-
vosa reported high rates of their infants “not being satisfied/
hungry after feeding” (predictive of overweight at age 9
months), these infants were not overweight at 9 months.
Possible explanations for this finding include misinterpreta-
tion of infant behavior as “not being satisfied” or maternal
restriction of infant intake, leading to increased hunger.
The high rate of feeding difficulties reported by the
women with other psychiatric disorders was not reflected in
poor growth or low weight of their infants. Possible explana-
tions for this are that the strong concerns about feeding
reported by this group reflect a lower threshold for general
concern or an altered interpretation of their infants’ cues.
Our study has some limitations. Only maternal report,
which can be subject to bias, was used to assess feeding
difficulties. This could be particularly relevant for the index
mothers. However, the fact that feeding during the first 6
months affected infant weight (also in the index groups)
suggests that these maternal reports were reliable. Another
possible limitation is the attrition rate in the cohort. However,
covariates related to overall attrition (ie, maternal sociodemo-
graphic data) were included in all multivariate analyses. No
selective attrition was noted for the women with eating dis-
Information on maternal eating symptoms contempo-
raneous to the infant feeding problems was not available. The
women were identified as having an eating disorder based on
self-report. This may have led to an underestimation of the
prevalence of eating disorders, given the tendency of screen-
ing measures for eating disorders in community samples to
miss cases. However, a recent study found that self-reported
diagnosis was of comparable accuracy to widely used eating
disorder screening tools in general population studies.28
Given the nature of this study, the women with eating
disorders were able to have children; thus, it is possible that a
proportion of these women may have suffered from a milder
disorder compared with clinical samples. In this case, our
findings are likely to be an underestimate rather than an
overestimate of the rate of feeding difficulties.
A great strength of our study lies in its large sample size,
ready availability of prospectively collected data, and general
population sample. These features allow inferences about
causation that have been problematic in previous studies due
to their retrospective nature or the nature of the samples
Infant Feeding and Weight in the First Year of Life in Babies of Women with Eating Disorders5
ARTICLE IN PRESS
The present study has important implications for the
prevention of feeding problems, as well as poor growth and
later overweight. Feeding problems in infancy often persist
into school age.5Associations among persistent early feeding
difficulties, poor development, and later behavioral problems
have been documented.29Moreover, feeding problems are a
cause of great stress for parents.30
Pediatricians need to be aware of the associations
among maternal eating disorder, infant feeding difficulties,
and later weight outcomes to adequately support and advise
parents and help prevent later negative outcomes. Further
insight is needed into the mechanisms behind higher rates of
feeding difficulties in infants of women with other psychiatric
disorders. Studies on childhood overweight have found that
rapid growth and overweight during the first months of life
are strong predictors of later overweight.8Therefore, close
follow-up of children of women with eating disorders is very
We thank all of the families who participated in this study, the
midwives who helped recruit them, and the entire ALSPAC team,
including the interviewers, computer and laboratory technicians,
clerical workers, research scientists, volunteers, managers, recep-
tionists, and nurses. We also thank Dr Sam Leary for her help
with the data. The UK Medical Research Council, the Wellcome
Trust, and the University of Bristol provide core support for
Int J Eat Disord 1991;10:395-405.
Chatoor I. Feeding disorders in infants and toddlers: diagnosis and treatment.
Child Adolesc Psychiatr Clin North Am 2002;11:163-83.
Dahl M, Rydel AM, Sunderlin C. Children with early refusal to eat: follow-up
during primary school. Acta Paediatr 1994;83:54-8.
Jacobi C, Agras WS, Bryson S, Hammer LD. Behavioral validation, precursors,
and concomitants of picky eating in childhood. J Am Acad Child Adolesc Psychiatry
Wright CM, Parkinson KN, Drewett RF. How does maternal and child feeding
behavior relate to weight gain and failure to thrive? Data from a prospective birth cohort.
Ong KK, Emmett PM, Noble S, Ness A, Dunger MD, and the ALSPAC Study
Team. Dietary energy intake at the age of 4 months predicts postnatal weight gain and
childhood body mass index. Pediatrics 2006;117:503-8.
Skuse D, Pickles A, Wolke D, Reilly S. Postnatal growth and mental develop-
ment: evidence for a “sensitive period.” J Child Psychol Psychiatry 1994;35:521-45.
Lindberg L, Bohlin G, Hagekull B. Early feeding problems in a normal sample.
disorders. Clin Child Fam Psychol Rev 2002;5:1-19.
Larsson G, Andersson–Ellstroem A. Experiences of pregnancy-related body shape
changes and of breast-feeding in women with a history of eating disorders. Eur Eat
Disord Rev 2003;11:116-24.
Agras S, Hammer L, McNicholas F. A prospective study of the influence of
eating-disordered mothers on their children. Int J Eat Disord 1999;25:253-62.
Stein A, Murray L, Copper P, Fairburn CG. Infant growth in the context of
maternal eating disorders and maternal depression: a comparative study. Psychol Med
Golding J, Pembrey M, Jones R. ALSPAC, the Avon Longitudinal Study of
Parents and Children, I: study methodology. Paediatr Perinat Epidemiol 2001;
Sullivan PB, Lambert B, Rose M, Ford-Adams M, Johnson A, Griffiths P.
Prevalence and severity of feeding and nutritional problems in children with neurological
impairment: the Oxford Feeding Study. Dev Med Child Neurol 2000;42:674-80.
Schreck KA, Williams K, Smith AF. A comparison of eating behaviors between
children with and without autism. J Autism Dev Disord 2004;34:433-8.
Frankenburg WK, Dodds JB. Denver developmental screening test. J Pediatr
Blair PS, Drewett RF, Emmett PM, Ness A, Emond AM. Family, socioeconomic
and prenatal factors associated with failure to thrive in the Avon Longitudinal Study of
Parents and Children (ALSPAC). Int J Epidemiol 2004;33:839-47.
Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in
overweight among US children and adolescents, 1999-2000. JAMA 2002;288:1728-32.
Wilcox WD, Nieburg P, Miller DS. Failure to thrive: a continuing problem of
definition. Clin Pediatr (Phila) 1989;28:391-4.
Cox DR. Regression models and lifetables. J R Stat Soc B 1972;34:187-220.
Evans J, le Grange D. Body size and parenting in eating disorders: a comparative
study of the attitudes of mothers towards their children. Int J Eat Disord 1995;18:39-48.
Barnes J, Stein A, Smith T, Pollock JI, and the ALSPAC Study Team. Extreme
attitudes to body shape, social and psychological factors and a reluctance to breast-feed.
J R Soc Med 1997;90:551-9.
Micali N, Treasure J, Simonoff E. Eating disorder symptoms in pregnancy: a
longitudinal study of women with recent and past eating disorders and obesity. J Psy-
chosom Res 2007;63:297-303.
Fahy T, Treasure J. Children of mothers with bulimia nervosa. BMJ
Northstone K, Emmett P, Nethersole F, and the ALSPAC Study Team. The
effect of age of introduction to lumpy solids on foods eaten and reported feeding
difficulties at 6 and 15 months. J Hum Nutr Diet 2001;14:43-54.
Farrow C, Blissett J. Is maternal psychopathology related to obesigenic feeding
practices at 1 year? Obes Res 2005;13:1999-2005.
Coulthard H, Harris G. Early food refusal: the role of maternal mood. J Reprod
Infant Psychol 2003;21:335-45.
Farrow C, Blissett J. Does maternal control during feeding moderate early infant
weight gain? Pediatrics 2006;118:293-8.
Keski-Rahkonen A, Sihvola E, Raevuori A, Kaukoranta J, Bulik CM, Hoek HW,
et al. Reliability of self-reported eating disorders: optimizing population screening. Int
J Eat Disord 2006;39:754-62.
Motion S, Northstone K, Emond A, and the ALSPAC Study Team. Persistent
early feeding difficulties and subsequent growth and developmental outcomes. Ambul
Child Health 2001;7:231-7.
Douglas J. Behavioural approaches to the assessment and management of feeding
problems in young children. In: Southall A, Schwartz, editors. Feeding Problems in
Children. Oxford, UK: Radcliffe Medical Press; 2000. p. 41-58.
Patel P, Wheatcroft R, Park RJ, Stein A. The children of mothers with eating
6 Micali, Simonoff, and TreasureThe Journal of Pediatrics • Month 2008
ARTICLE IN PRESS
Table I. Sociodemographic data for mothers and infants Download full-text
(n ? 247)
(n ? 194)
(n ? 1148)
(n ? 10 461)
Maternal age at delivery, years,
Multiparity, % (OR, 95% CI)
Caucasian ethnicity, % (OR, 95% CI)
Employment (full-time or part-time
employment or full-time
education, training vs.
unemployed, housewives or
retired at enrollment), % (OR,
Marital status (married vs all
others), % (OR, 95% CI)
Birth weight, g, mean (SD)
Gestational age, weeks, mean (SD)
Male sex, % (OR, 95% CI)
29.1 (5.0)**28.3 (4.6) 28.1 (5.5)28.2 (4.8)
52.4 (0.9, 0.6-1.2)
97.0 (0.8, 0.4-1.7)
47.2 (0.9, 0.7-1.2)
51.9 (0.9, 0.7-1.2)
97.3 (0.9, 0.4-2.2)
48.7 (1.0, 0.8-1.4)
59.4 (1.2, 1.0-1.3)*
98.0 (1.2, 0.8-1.9)
38.1 (0.6, 0.5-0.7)*
65.7 (0.5, 0.4-0.7)*67.5 (0.6, 0.4-0.8)* 62.5 (0.5, 0.4-0.5)*78.6 (reference)
51.8 (1.0, 0.7-1.2)
49.0 (1.1, 0.8-1.4)
53.2 (0.9, 0.8-1.0)
3431 (532) (reference)
39.5 (1.8) (reference)
*P ? .001, **P ? .01 for comparisons between groups and the general population.
Table II. Feeding problems at 1 and 6 months
(n ? 234)
(n ? 182)
(n ? 1053)
(n ? 9792)
Feeding problems at 1 month
Weak sucking, n (%)
Drinking too fast, n (%)
Exhaustion with feeding, n (%)
Feeding problems between 0 and 6 months
Slow feeding (n ? 9540)
199 150838 8353
Small quantity feeding (n ? 9745)
Not satisfied/hungry after feeding (n ? 8935)
Refusal to take solids (n ? 9511)
No established feeding routine (n ? 8567)
Infant Feeding and Weight in the First Year of Life in Babies of Women with Eating Disorders 6.e1
ARTICLE IN PRESS