Scandinavian Journal of Public Health, 2010; 38: 208–219
STUDY DESIGN ARTICLE
Lifestyle during pregnancy: Neurodevelopmental effects at 5 years of
age. The design and implementation of a prospective follow-up study
ULRIK SCHIØLER KESMODEL1, METTE UNDERBJERG1,2, TINA RØNDRUP KILBURN1,
LEIV BAKKETEIG3, ERIK LYKKE MORTENSEN1,4, NILS INGE LANDRØ5,
DIANA SCHENDEL6, JACQUELYN BERTRAND6, JAKOB GROVE1, SHAHUL EBRAHIM6&
1School of Public Health, Department of Epidemiology, University of Aarhus, Denmark,2Vejlefjord Children’s Neurocenter,
Stouby, Denmark,3Institute of Public Health, Research Unit of Epidemiology, University of Southern Denmark, Denmark,
4Institute of Public Health, University of Copenhagen, Denmark,5Center for the Study of Human Cognition, Department of
Psychology, University of Oslo, Norway,6Centers for Disease Control and Prevention (CDC), Atlanta, Georgia, USA, and
7Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA, USA
Background: It has been suggested that even mild exposure to alcohol, caffeine, smoking, and poor diet may have adverse
long-term neurodevelopmental effects. In addition, there is evidence that timing of high exposures (e.g. binge drinking) can
have particularly negative effects. This paper describes the design and implementation of The Lifestyle During Pregnancy
Study addressing major methodological challenges for studies in this field. The study examines the effects of lifestyle during
pregnancy on offspring neurodevelopment. Methods: In 2003, we initiated a prospective follow-up of 1750 mother–child
pairs, sampled on the basis of maternal alcohol drinking patterns from The Danish National Birth Cohort (DNBC), a study
of 101,042 pregnancies enrolled 1997–2003. Data collection in the DNBC involved four prenatal and postnatal maternal
interviews, providing detailed information on maternal alcohol drinking patterns before and during pregnancy, caffeine
intake, smoking, diet, and other lifestyle, medical, and sociodemographic factors. Results: At the age of 5 years, the children
and their mothers participated in a comprehensive assessment of neurobehavioural development focusing on global
cognition, specific cognitive functions, and behaviour. Two new tests assessing attention and speed of information processing
among children were developed, and data on important potential confounders such as maternal intelligence quotient, vision,
and hearing abilities were collected. Efforts were made to standardise procedures and obtain high inter-rater reliability.
Conclusions: We expect that the study will illuminate the significance or lack of significance of maternal lifestyle
during pregnancy and contribute to better understanding the effects of alcohol drinking during pregnancy at low
to moderate consumption levels.
Key Words: Alcohol, binge drinking, cognition, intelligence, lifestyle during pregnancy, neurobehavioural development,
It is well established that exposure to high levels
of certain toxic agents during pregnancy can affect
the risk of malformations and other life long dis-
abilities among offspring. Alcohol, heavy metals, and
certain medications are some of the most widely
recognised teratogens. Some modifiable exposures
are often referred to as lifestyle factors. Common
modifiable lifestyle risk factors include alcohol and
caffeine consumption, tobacco smoking, and diet.
Fetal alcohol exposure is the leading known pre-
ventable cause of birth defects and developmental
disabilities among Western countries . Maternal
Correspondence: Ulrik Schiøler Kesmodel, School of Public Health, Department of Epidemiology, University of Aarhus, Bartholins Alle 2 8000 Aarhus C,
Denmark. Tel: þ45 89 42 60 70. Fax: þ45 86 13 15 80. E-mail: email@example.com
(Accepted 12 November 2009)
? 2010 the Nordic Societies of Public Health
intake of large daily volumes of alcohol during
pregnancy has been shown to be associated with
malformations , mental retardation, behavioural
and psychosocial problems in childhood and adoles-
cence , and, in the most severe cases, fetal alcohol
syndrome (FAS) [2,4].
Maternal intake of about one drink per day or
more has been associated with reduced birthweight
and intrauterine growth restriction , spontaneous
abortion , preterm delivery , and stillbirth .
Furthermore, varying degrees of prenatal alcohol
exposure have been found to be associated with
dysfunctions in the speed and efficiency of informa-
tion processing  and in deficits in specific cognitive
functions such as attention [10–16], learning and
memory [17–19], visual perceptual and visual motor
skills [16–18], language [16–18], and executive
functions [20–22]. However, the results have been
inconsistent with numerous studies showing little or
no effect of more moderate alcohol intake during
pregnancy [23,24], while studies on birth weight ,
preterm delivery , and long term neurodevelop-
mental effects [25,26] have even suggested apparent
beneficial effects of low weekly doses of alcohol.
Tobacco smoking during pregnancy has been
associated with many short-term adverse pregnancy
between smoking during pregnancy and adverse effects
on cognitive and behavioural development .
Experimental studies among animals have sug-
gested that high doses of caffeine can cause beha-
vioural problems among the offspring , although
it is unclear whether these high levels of exposure
are comparable with the ordinary consumption levels
of pregnant women . A single study among
humans showed no association between caffeine
intake during pregnancy and developmental out-
comes at the age of 7 .
Prenatal diet may influence fetal development. It is
well documented that periconceptional supplemen-
tation with folic acid can prevent neural tube defects
. Morover, several studies support that long-
chain n-3 fatty acids, abundant in seafood, may
prolong gestation and reduce the risk of preterm
delivery [34,35]. It is possible that suboptimal nutri-
tion at vulnerable stages in pregnancy might even
affect the risk of diseases that may first appear in
Several dietary factors have been associated with
offspring development. Poor nutrition during brain
development may impair cognitive development 
and supplementation studies with long-chain poly-
unsaturated fatty acids to very preterm infants have
shown significantly enhanced visual acuity, visual
recognition, memory, and performance during early
infancy . Food-borne environmental toxins have
also received great attention in relation to neurode-
velopment. Consumption of seafood contaminated
with methyl-mercury has been associated with subtle
neurodevelopmental deficits [39,40], although some
Many significant limitations are, however, encoun-
tered by most studies of long-term developmental
outcomes, especially studies of prenatal alcohol
exposure. Differences in findings among studies
might be due to a variety of methodological pro-
blems, including: failure to obtain valid and reliable
prospective assessments of drinking patterns during
pregnancy [3,43]; a systematic lack of information on
low-to-moderate-level alcohol exposure [3,43,44]
and pattern (chronic vs. binge) and timing of binges
[3,43–46]; failure to adequately control for con-
founding by parental intellectual abilities [45,47–
50], socioeconomic position [3,45,50,51], and other
environmental factors [3,45,51]; failure to evaluate a
range of developmental outcomes at different ages
[47,50] to assess multiple effects that could emerge
over time; and generally small sample sizes in studies
on neurodevelopment [3,44,52,53], most likely
because of the costs of conducting such studies.
In addition, the task of examining young children
involves special methodological problems: Although
many measures for young children are available for
clinical use, there are few theoretically founded tests
suitable for pre-school-aged children and few tests
that successfully distinguish between different cog-
nitive functions and different aspects of behaviour.
Furthermore, the results of tests for young children
might be affected substantially by response modus,
(i.e. whether motor or verbal skills are required to
perform the tasks). Finally, young children are very
susceptible to factors associated with test adminis-
tration and tester effects [45,47,48,50].
The current study was designed as outlined below
to address and overcome the challenges and limita-
tions of previous studies of the relation between
lifestyle factors during pregnancy and later neurode-
velopment of the child. The study is based on one of
the largest cohorts of pregnant women available in
the world. In this paper we describe the study design,
our implementation of important methodological
decisions and preliminary results regarding partici-
pation as well as inter-rater reliability.
We wanted to create a study population that would
make it possible to examine the relation between
maternal lifestyle during pregnancy and offspring
Alcohol in pregnancy
neurodevelopment at the age of 5 years. The primary
exposure of interest in The
Pregnancy Study is alcohol. Specifically, we wanted
to be able to study the association between average
alcohol intake (drinks per week) before and during
pregnancy and binge drinking (intake of five or more
alcohol containing drinks on a single occasion) at
different points in time during pregnancy. In addi-
tion, the design of the study and data available
through the Danish National Birth Cohort (DNBC)
allow for investigation of several other exposures or
intake, and diet – particularly seafood diet.
This study was designed to address and overcome
the limitations of previous studies by using prospec-
tive data with regard to exposure to risk factors,
obtained at two points in pregnancy, including
information on different drinking patterns and
timing of binge drinking. Thus, we specifically
focused our study on the potential effects of very
low, weekly (rather than daily) alcohol intake and
binge drinking. Furthermore, we used data from a
homogenous, generally middle-class population with
access to comprehensive healthcare free of charge for
the individual (thereby reducing the potential for
confounding) and without the stigma associated with
alcohol that might exist in some countries ,
possibly reducing the magnitude of information bias
associated with self-reported alcohol intake. We also
sampled across different alcohol consumption levels
with sufficient numbers in each stratum to allow for
relevant analyses of the effects of different drinking
We included theoretically founded tests suitable for
young children and, when no relevant tests were
available, we developed and validated new tests
for use in this study and for subsequent clinical use.
In choosing and developing new tests, we wanted
to include tests that tapped fundamental cognitive
processes as well as distinguished between specific
cognitive functions and different aspects of beha-
viour. Finally, we standardised all possible proce-
dures and performed regular inter-rater comparisons
for both groupsof examiners
and physiotherapists) to minimise any potential
effects of examiner, examination site, or other sys-
The study is a prospective follow-up study, based on
a sample from the DNBC  with oversampling of
certain alcohol exposure categories.
The Danish National Birth Cohort
During the period 1997–2003, pregnant women in
Denmark were invited to participate in the DNBC
study at the first antenatal visit by the general
practitioner. Enrolment was completed in the
autumn of 2003 and the database includes 101,042
pregnancies, corresponding to approximately 60% of
those invited and approximately 30% of all pregnant
women in Denmark during the enrolment period
. For this study, data points of interest from the
DNBC study included: two prenatal maternal tele-
phone interviews conducted at week 12 and 30 of
gestation concerning maternal health, use of medi-
cines, socioeconomic status, obstetric history, and
lifestyle factors (e.g. alcohol, smoking, caffeine, and
diet); two postnatal maternal interviews conducted at
6 and 18 months postpartum focusing on family
conditions and offspring health and development.
Additional medical and socioeconomic informa-
tion on DNBC participants can be obtained by
linking with Danish computerised registries, using
the unique Danish personal identification number.
Information on exposures in the DNBC
Both prenatal interviews provide information on the
number of drinks per week of beer, wine, fortified
wine, and spirits that each participating pregnant
woman consumed at the time of the interview; in
addition the first prenatal interview provides infor-
mation on the number of drinks per week consumed
before pregnancy. Such interviews have been shown
to yield valid (relative to other methods) and reliable
information among pregnant Danish women .
The interviews also provide information on binge
drinking (intake of five or more drinks on a single
occasion) and when during pregnancy each binge
episode occurred. This approach has been shown to
yield valid (relative to other methods) and reliable
information on binge drinking, and it is the only
validated method for the collection of data on timing
of binge drinking [57,58]. Furthermore, the prenatal
interviews provide information on smoking habits
before and during pregnancy, intake of caffeine-
containing beverages (coffee, tea, cola, and hot
chocolate), seafood and iron consumption during
pregnancy, and other detailed information on diet.
Outcomes in the lifestyle study
The neurodevelopmental outcomes included mea-
sures of cognitive, behavioural, emotional, and social
functions and measures of growth at 5 years of age
(age span: 60–64 months). The outcomes were
U. S. Kesmodel et al.
assessed by administration of a neuropsychological
test battery and questionnaires on the child’s
social functioning, behaviour, and health to be
answered by parents and staff in the child’s day-care
centre (Table I).
For all outcomes, we hypothesised that for alcohol
consumption at low weekly levels to have any poten-
tial long-term adverse effects, a documented effect at
higher levels should be present. The decision to
include outcomes in this study was based on thor-
ough literature reviews showing negative associations
between high daily intake of alcohol and intelligence
quotient (IQ) , attention [10–16], speed of infor-
[21,22,59], and motor development [16–18].
The age of 5 years was chosen based on considera-
tions of stability of cognitive performance: While the
reliability of IQ measures increases with age into
middle age, IQ from the age of about 5 years is a
reliable measure with reliability about 0.8 . IQ in
childhood and adolescence predicts adult health and
mortality , as well as social socioeconomic posi-
tion [62,63] and social pathology , with high IQ
The prenatal and postnatal telephone interviews
included detailed information on most potential
confounders, e.g. smoking, diet, socioeconomic posi-
tion, and medical and obstetric histories. A parent-
administered questionnaire included information on
the parent’s education, current socioeconomic status,
Table I. Final tests and examinations used in The Lifestyle During Pregnancy Study.
Psychological domain Test instrument used
General intellectual ability (IQ)
WPPSI-R subtests 
– object assembly
– geometric design
– block design
Animal house (WPPSI-R) 
TEACh-5 (test of everyday attention for 5-year-old children)
– balloon hunt
– draw a line
– hide and seek-V
– hide and seek-A
Sternberg task – KVC (Kilburn’s version for children)
BRIEF (behaviour rating inventory of executive functions) 
(parent and teacher part)
SDQ (strengths and difficulties questionnaire)
(parent and teacher part) – modified version 
Draw a person 
Behaviour Checklist (clinical rating)
Attention (visual and auditory)
(sustained, focused, spatial, control/inhibition)
Reaction time and speed of information processing
Prediction of executive function
Social skills, behaviour, including adaptive behaviour
eight subscales/aspects of executive functions
General developmental status
Behaviour during test session
Maternal general intellectual ability (IQ)
Height, weight, head circumference
Motor development (fine and gross)
Background information on SES, upbringing,
disease and medication during childhood, etc.
information (from WAIS) [71,72]
vocabulary (from WAIS) [71,72]
Raven (Raven’s standard progressive matrices) 
Østerberg vision board
Movement ABC 
static and dynamic balance
Photography (full face and profile of child and mother)
WAIS, Wechsler adult intelligence scale.
Alcohol in pregnancy
and lifestyle, etc. In addition, the mother was
administered a test of intelligence.
As opposed to smoking and caffeine intake, high
average alcohol intake and binge drinking after the
first 4 weeks of pregnancy are relatively rare events
. A primary consideration in the development of
the sampling design, therefore, was to include women
with different alcohol drinking patterns (Table II).
The primary analyses were planned to be based on
two distinct drinking patterns: average intake among
four groups (0, 1–4, 5–8, and 9 or more drinks per
week) and binge drinking (yes or no as well as the
number and timing of binge episodes). To study the
interaction between average alcohol intake and binge
drinking, we subsequently defined seven main cate-
gories according to average drinking levels during
pregnancy (categories 1–5) or prepregnancy drinking
levels (categories 6–7). Each of these seven categories
was further stratified on the basis of binge drinking
patterns: whether or not binge drinking occurred
and, if so, during which weeks during the pregnancy
the binging episodes occurred. Category 1 is
the ‘‘unexposed’’ reference group for the study
(no drinking during pregnancy and no binge epi-
sodes). Category 7 comprised women who had
the highest drinking levels before pregnancy (15
or more drinks per week, i.e. above the maximum
intake level recommended by the Danish National
Board of Health). This category was established as a
separate risk category because of the risk of con-
tinued high drinking levels before pregnancy is
A total of 150 women were selected for the
reference group (category 1), while 75 women were
selected for each of the other categories. For a few
categories, however, fewer than 75 women had
reported this drinking pattern in the DNBC, and
hence fewer participants were expected (Table II).
A total of 1450 women were expected to participate
in the initial sample (Table II) based on power
number of pregnancies in each alcohol category
(Table II), as only approximately 65,000 pregnancies
were available from the DNBC when decisions were
made as to the number of participants in this study.
Table II. Primary selection procedure in The Lifestyle During Pregnancy Study.
Weeks of pregnancya
pregnancy1–2 3–45–8 9 or more
1–23–45 or more
NA, not applicable.
aMeasured from the last menstrual period.
bBased on 50% participation in all categories.
cAny combination of two or more ‘‘yes’’.
dRandom sample of women from all subgroups ‘‘in pregnancy’’.
U. S. Kesmodel et al.
During 2005, an additional group of 300 women
who completed a nutritional survey component of
the DNBC and who had different intake patterns of
seafood and iron and different reported durations
of breast feeding were included to ensure sufficient
variance with respect to these factors in the final
sample for supplemental analyses (Table III).
Mothers and children were excluded if they did not
speak Danish, if they had impaired hearing or vision
loss that inhibited the performance of the psycholog-
ical tests or if a child was affected by congenital
disorders that might lead to mental retardation (e.g.
Downs syndrome, autism, and cerebral palsy).
Test battery and pilot testing
The overall purpose of the test battery was to enable a
thorough and theoretically relevant assessment of
each child’s neurodevelopmental status, including
general intelligence, fundamental specific cognitive
functions, and functioning in specific behavioural
domains. Specifically, we wanted measures of verbal
and nonverbalintelligence (IQ), measuresof
sustained and focused attention, and measures of
speed of information processing, executive functions,
learning skills, fine and gross motor development,
social skills and general behavioural development.
Furthermore, height, weight, and head circumfer-
ence were measured, photographs were taken to
allow for subsequent measurement of (dysmorphic)
facial features, and on-site screening of vision and
hearing abilities was conducted. The final test battery
was selected after an extensive pilot phase. Three of
the key issues were the relatively limited time that a 5-
year-old child is able to collaborate on mentally
demanding tasks, the limited number of available
tests and questionnaires, suitable for young Danish
children, and theoretical relevance in relation to
prenatal alcohol exposure and other lifestyle factors.
Table I presents an overview of the final test battery.
With regard to attention and speed of information
processing, we were unable to locate theoretically
founded tests for young children. Therefore, a new
test of attention (TEACh-5) and a children’s version
of Sternberg’s speed of information processing par-
adigm were developed. In developing both tests, it
Table III. Additional selection procedure in The Lifestyle During Pregnancy Study.
Iron from foodsc
Iron from supplements
aNumbers do not add up to 1450 because of missing information for each variable.
bGroup 1 (never eating fish), Group 2 (eating fish each month or less), Group 3 (hot meal each month, sandwich each week),
Group 4 (hot meal and sandwich each week, lowfrequency), and Group 5 (hot meal and sandwich each week, high frequency).
cThe estimated intake in mg/day (energy-adjusted).
Alcohol in pregnancy
was paramount that the tests require as little verbal
skill as possible and making it possible to control for
potential confounding of test results by individual
differences in motor function.
A total of six pilot studies were performed of which
four were mainly related to the development of the
TEACh-5 and the Sternberg method. The main pilot
involved 25 mother–child pairs and was used to test
enrolment procedures, to estimate participation rate,
to ensure comprehensibility of questionnaires, and to
decide on the final content and order of the test
Selected mothers were invited to participate in the
study by letter approximately 8–12 weeks before their
child’s fifth birthday (2003–08). The mother/parents
could return a slip (a) agreeing to participate, (b)
asking for more information by telephone, or (c)
declining to participate. Those agreeing to partici-
pate were contacted by telephone and provided more
detailed project information, and an appointment for
testing the child was made. The parent-administered
questionnaires (Table I) were mailed to the partici-
pants before the day of the testing, and, if permission
was obtained from the parents, questionnaires were
subsequently mailed to the child’s day-care centre.
Further oral information was given on the day of
testing before the consent form was signed. A max-
imum of two reminders were mailed to the parents
and/or day-care institutions if they did not respond to
the initial letter.
The assessments were conducted in a controlled
setting at four regional sites (Copenhagen, Odense,
Aarhus, and Aalborg). The test session lasted
approximately 3 hours, which, based on the pilot
studies, was the maximum length of time that 5-year
olds can be tested. All psychological tests were
administered by one of 10 psychologists with special
training in neuropsychological testing and study
procedures. Assessment of motor development was
conducted by 30 trained physiotherapists (this part of
the study was stopped in February 2006 due to lack
of funding). At the time of the test session, psychol-
ogists and physiotherapists were blinded to exposure
status and to any information regarding the child’s
development in general. The mothers completed
adult IQ tests, while the child was being tested by
the physiotherapist or, from February 2006, at the
end of the test session.
The total cost of the data collection was estimated
at US$ 4,723,537.
Standardisation of procedures and inter-rater reliability
Given the study size and the inherent need to involve
several psychologists and physiotherapists in the data
collection, special attention was paid to methodolog-
ical issues, such as inter-rater reliability and standar-
dised test procedures.
During the pilot projects, the psychologists observed
and supervised each other to ensure a uniform and
standardised administration of all test procedures
and a consistent scoring of all tests. This at-site
supervision procedure was repeated every 6 months
and whenever a new psychologist began working with
the project. In addition, test manuals were continu-
ously supplemented with additional guidelines for
test procedures and scoring criteria (copies available
from the authors).
Approximately every 6 months, each psychologist
blindly rescored a number of subtests administered
by other psychologists. This procedure included the
vocabulary and the geometric design subtests of the
Wechsler preschool and primary scale of intelligence
– revised (WPPSI-R), and the vocabulary subtest
from the Wechsler adult intelligence scale (WAIS).
Subsequently, identified discrepancies were dis-
cussed, the correct scoring was agreed upon, and
scoring criteria were specified. Typically, there was
approximately 97–97.5% agreement before discus-
sion, and about two-thirds of the discrepancies were
deemed to be unavoidable because of ambiguous
scoring criteria in published manuals.
A successful attempt was made to allocate testers
evenly across alcohol exposure categories (Table IV).
Preliminary analyses of the distribution of test scores
among psychologists demonstrated small but detect-
able differences on IQ scores and on some TEACh-5
subtests of attention, but no tester differences with
regard to the Sternberg speed of information pro-
cessing test. However, no systematic pattern of tester
differences was observed and, because tester status
was not associated with exposure categories, tester
effects per se should not lead to confounding.
Before the beginning of the project, the physiothera-
pists participated in an introductory whole daylong
meeting, allowing for in depth discussions of the
project procedures and training in administration
and scoring of the movement ABC test. Every 6
months, all the procedures of the movement ABC
were demonstrated and discussed, using volunteers
U. S. Kesmodel et al.
and video recordings. Where the manual appeared
unclear, additional guidelines for the different pro-
cedures were outlined. There were no substantial or
significant differences in the distribution of mother–
child pairs tested by each physiotherapist across the
alcohol exposure categories.
Initially, we expected a participation rate of about
50% and consequently sampled 3000 mother–child
pairs from the DNBC. In 2005, an additional 600
pairs were sampled to include 300 mother-child pairs
with different intake of seafood diet and iron and
different duration of breast feeding. A total of 3292
mother–child pairs have been invited to participate,
with a consent rate of 57.0%. For various reasons
to do so (e.g. because of illness or lack of motivation).
Consent and participation rates have been almost
equally distributed across alcohol intake categories.
Tests, examinations, and questionnaires
Of all the mother–child pairs invited to participate,
50.8% returned the questionnaire and participated in
the test session, 0.2% only participated in the test
session, whereas 4.3% only returned the question-
naires. Of the participating mothers, 95.3% gave
permission for the questionnaires to be sent to their
child’s day-care centre: 93.4% of the centres returned
the questionnaires after 1–2 reminders. Of the chil-
dren participating in the test session, 83.5% com-
pleted all 13 psychological subtests (subtests from
WPPSI-R, TEACH-5 and draw-a-person), except
the speed of information processing test (completed
by 76.6% of the children). For the remaining
children, single subtests were missing due to e.g.
motivational factors, lack of understanding test
premises, or inability to complete the subtests. Of
the mothers to the participating children, 99.0%
completed all three adult IQ test elements.
The overall participation in the DNBC has been
estimated at approximately 30% . While a strong
selection is likely in the DNBC , and further
selection has taken place in this study, the differential
participation in the DNBC seems to cause only little
if any bias in studies within the DNBC .
Each individual outcome will be analysed in appro-
priate regression models taking into account the
sample design (including timing of binge drinking
within the predefined week groups). The power
calculations shown below relate to these subanalyses.
The information derived through the subanalyses is
usedtobuildalargemixed-model analysis ofvariance
Table IV. Distribution of testers across alcohol sampling categories, based on initial sampling.
Psychologist, numbered according to time of entering the project
p-value (chi-squared test) for the distribution of testers across alcohol sampling categories¼0.242.
Alcohol in pregnancy
model, which takes care of multiple testing issues by
analysing all outcomes simultaneously. Conducting a
power calculation on the combined model is imprac-
tical. However, it is not unreasonable to assume that
the combined power of analysing all outcomes in one
subanalyses, and possibly more powerful.
Based on the groups proposed for follow-up
(Table II), we operated with two levels of analyses:
Primary analyses included analyses of overall average
intake among four groups: 0, 1–4, 5–8, and 9 or more
drinks per week, and binge drinking (yes or no)
(Table V). Assuming a standard deviation of 15, we
calculated the smallest detectable differences for
continuous outcomes (e.g. IQ) at a power of 80%
and an a-level of 0.05 and, with the basic assumption
of disease proportion of 10% among the reference
group, we derived the smallest detectable relative risk
for dichotomous outcomes (Table V).
Subsequent secondary analyses were based on the
stratified sample. Assuming scenarios in which the
analyses would have to be adjusted for differences
over strata, the power calculations were repeated
(data not shown). These calculations indicated only a
very modest loss of power for the moderate strata
differences we expect to see. All power calculations
were done by simulation.
The study was approved by the DNBC Board of
Directors, the DNBC Steering Committee, the
relevant regional ethics committees in Denmark,
the Danish Data Protection Agency, and the
Centers for Disease Control and Prevention (CDC)
Institutional Review Board.
Following the tests, each mother/parents and her/
their child were offered an oral presentation of the
psychologists ‘and the physiotherapists’ overall impres-
sion, if they wished. Naturally, most children were
expected to perform within the normal range on the
neuropsychological tests. But, in the few cases in which
the psychologist felt the child warranted special assis-
tance, the mother/parents were informed of this and
advised to seek professional assistance as provided by
cases, the parents were already aware of the problem.
Although a 3-hour test session may be considered
long for 5-year old children, few participants
expressed that it was too long or strenuous and only
in a few cases were tests abandoned due to exhaus-
tion of the child.
The parents received DKK 400–500 or a book
worth approximately that amount for participating
and the children received avoucher of DKK 100 for a
toy shop. Travelling expenses were covered with
appropriate documentation. These incentives were
chosen based on the feedback from the pilot studies.
We hope that our study of the potential long-term
effects of low-dose weekly alcohol intake and binge
Table V. The smallest detectable differences for continuous outcomes (IQ) and smallest detectable risk ratio for categorised outcomes with
a-level¼0.05, power¼80%. Both calculations based on expected numbers at study initiation and calculations based on the actual
participation are shown. The Lifestyle During Pregnancy Study.
Average alcohol intakeBinge drinking
aBased on smallest detectable difference in IQ from reference IQ of 100 (SD¼15).
bBased on an expected proportion of cognitive and behavioural deficits among the unexposed of 10%.
cAlcohol intake categories included from Table II: 0 (1?1dþ3a); 1–4 (2?2dþ½X3b); 5–8 (4?4cþ½X3b); 9þ (5a?5b).
dExpected numbers at study initiation.
eAlcohol intake categories included from Table II: no (1þ2þ4þ5a); yes (1a?1dþ2a?2dþ3a?3bþ4a?4cþ5b).
fActual numbers at the end of the study.
IQ, intelligence quotient; Ref, reference category, RR, risk ratio.
U. S. Kesmodel et al.
drinking during pregnancy on offspring cognitive
and behavioural development of offspring will move
this field of research forward. Most importantly,
the following methodological issues have been
Prospective datawere obtained for alcohol and other
Reliable information was obtained for low levels
of exposure to alcohol.
For alcohol use, pattern and timing of intake were
Diet, environmental factors and maternal intelli-
A range of outcome measures were administered.
Theoretically appropriate measures of fundamental
cognitive processes were included in addition to
standardised clinical measures.
Several modalities were used for assessments.
Extensive pilot testing ensured that the test battery
was appropriate for 5 year-old children in terms of
the length of the test session and comprehensibility,
and ensured their ability to give a good performance
on all measures.
The overall sample size was large enough to allow
sufficient power in the analyses.
account as potential
Little is known about the potential long-term
effects of low-dose weekly alcohol intake and binge
drinking during pregnancy on the cognitive and
behavioural development of offspring. This also
applies to the potential long-term effects on neuro-
development of prenatal exposure to tobacco smok-
ing, caffeine, and diet. Together with similar studies
of maternal behaviour and lifestyle during pregnancy,
the results of this project will contribute to the
evaluation of long-term developmental effects of
prenatal exposure to alcohol and other lifestyle
In the development of the test battery, we devel-
oped new tests of attention and information process-
ing speed. These tests have the potential to improve
studies of neurocognitive development and could be
a basis for earlier and more precise diagnosis of
children with cognitive dysfunction, including defi-
cits of attention.
It will also be possible to use the study as a basis for
research into relevant biomarkers. The action of
alcohol (ethanol) on neurodevelopment might be
moderated by the host’s (mostly the mother’s)
capacity to detoxify alcohol, which is primarily
determined by the level of enzyme efficiency (e.g.
alcohol dehydrogenase [ADH], and aldehyde dehy-
drogenase [ALDH]). The efficiency of ADH and
ALDH most likely depends on the genetic composi-
tion coding for the ADH and ALDH genes [65,66],
and it is possible to evaluate the potential influence of
genetic components such as single nucleotide poly-
morphisms by measuring these factors in blood
samples taken from all mothers and children in the
DNBC study , including those enrolled in the
The DNBC cohort offers a unique possibility to
evaluate associations between multiple prenatal and
perinatal factors and the long-term development of
offspring. The DNBC is administered by the Danish
National Board of Health and is open to researchers
on application. Also, the lifestyle study is likely to
have global public health implications. Thus, the
results will add important and unique information
regarding issues about maternal alcohol consump-
tion during pregnancy and any later effects on
children who have experienced prenatal alcohol
exposure at low and moderate levels, which is an
issue of worldwide public health interest.
Conflicts of interest
The authors have no conflicts of interest to declare.
The findings and conclusions in this report are those
of the author(s) and do not necessarily represent the
official position of the Centers for Disease Control
This study was supported financially by the Danish
National Board of Health, Centers for Disease
Control and Prevention (CDC), Atlanta, Georgia,
USA, and Lundbeck Foundation, Denmark. We
gratefully acknowledge the participation and enthu-
siasm of the children and their parents in this study as
well as the Danish National Birth Cohort.
The Danish National Research Foundation has
Centre that initiated and created the Danish
National Birth Cohort. The cohort is furthermore a
result of a major grant from this Foundation.
Additional support for the Danish National Birth
Cohort is obtained from the Pharmacy Foundation,
the Egmont Foundation, the March of Dimes Birth
Defects Foundation, theAugustinus Foundation, and
the Health Foundation
Alcohol in pregnancy
 Abel EL, Sokol RJ. Incidence of fetal alcohol syndrome and
economic impact of FAS- related anomalies. Drug Alcohol
 American Academy of Pediatrics Committee on Substance
Abuse. American Academy of Pediatrics Committee on
Disabilities: fetal alcohol syndrome and fetal alcohol effects
[see comments]. Pediatrics 1993;91(5):1004–6.
 Mattson SN, Riley EP, Gramling L, Delis DC, Jone KL.
Heavy prenatal alcohol exposure with or without physical
features of fetal alcohol syndrome leads to IQ deficits. J
 Spohr HL, Willms J, Steinhausen HC. Prenatal alcohol
 Passaro KT, Little RE, Savitz DA, Noss J. The effect of
maternal drinking before conception and in early pregnancy
on infant birthweight. The ALSPAC Study Team. Avon
Longitudinal Study of Pregnancy and Childhood. Epide-
 Windham GC, Fenster L, Swan SH. Moderate maternal and
paternal alcohol consumption and the risk of spontaneous
abortion. Epidemiology 1992;3(4):364–70.
 Kesmodel U, Olsen SF, Secher NJ. Does alcohol increase the
risk of preterm delivery? Epidemiology 2000;11(5):512–18.
 Kesmodel U, Wisborg K, Olsen SF, Hensriksen TB,
Secher NJ. Moderate alcohol intake during pregnancy and
the risk of stillbirth and death in the first year of life. Am J
 Burden MJ, Jacobson SW, Jacobson JL. Relation of prenatal
alcohol exposure to cognitive processing speed and efficiency
in childhood. Alcohol Clin Exp Res 2005;29(8):1473–83.
 Streissguth AP, Barr HM, Sampson PD, Parrish Johnson JC,
Kirchner GL, Martin DC. Attention, distraction and reac-
tion time at age 7 years and prenatal alcohol exposure.
Neurobehav Toxicol Teratol 1986;8(6):717–25.
 Streissguth AP, Martin DC, Barr HM, Sandman BM,
Kirchner GL, Darby BL. Intrauterine alcohol and nicotine
exposure: attention and reaction time in 4-year-old children.
Develop Psychol 1984;20(4):533–41.
 Streissguth AP, Sampson PD, Olson HC, Bookstein FL,
Barr HM, Scott M. Maternal drinking during pregnancy:
attention and short-term memory in 14-year-old offspring –
a longitudinal prospective study. Alcohol Clin Exp Res
 ShaywitzSE, Cohen DJ, ShaywitzBA. Behavior and learning
difficulties in children of normal intelligence born to alco-
holic mothers. J Pediatr 1980;96(6):978–82.
 Brown RT, Coles CD, Smith IE, Platzman KA, Silverstein J,
Erickson S, Falek A. Effects of prenatal alcohol exposure at
school age. II. Attention and behavior. Neurotoxicol Teratol
 Coles CD, Platzman KA, Lynch ME, Freides D. Auditory
and visual sustained attention in adolescents prenatally
 Korkman M, Autti-Ra ¨mo ¨ I, Koivulehto H, Granstro ¨m ML.
Neuropsychological effects at early school age of fetal alcohol
exposure of varying duration. Child Neuropsychol 1998;
 Mattson SN, Riley EP. A review of the neurobehavioral
deficits in children with fetal alcohol syndrome or prenatal
exposure to alcohol.Alcohol
 Mattson SN, Schoenfeld AM, Riley EP. Teratogenic effects
of alcohol on brain and behavior. Alcohol Res Health
 Olson HC, Streissguth AP, Sampson PD, Barr HM,
Bookstein FL, Thiede K. Association of prenatal alcohol
exposure with behavioral and learning problems in early
adolescence. J Am Acad Child and Adolesc Psychiatry
 Kodituwakku PW, Kalberg W, May PA. The effects of
Alcohol Res Health 2001;25(3):192–8.
 Connor PD, Sampson PD, Bookstein FL, Barr HM,
Streissguth AP. Direct and indirect effects of prenatal alcohol
 Noland JS, Singer LT, Arendt RE, Minnes S, Short EJ,
Bearer CF. Executive functioning in preschool-age children
prenatally exposed to alcohol, cocaine, and marijuana.
Alcohol Clin Exp Res 2003;27(4):647–56.
 Kesmodel U. Alkohol og graviditet. [Alcohol in pregnancy].
Ugeskr Laeger 1999;161(36):4989–94.
 Gray R, Henderson J. Report to the Department of Health –
review of the fetal effects of prenatal alcohol exposure.
Oxford: National Perinatal Epidemiology Unit; 2006.
 Kelly Y, Sacker A, Gray R, Kelly J, Wolke D, Quigley MA.
Light drinking in pregnancy, a risk for behavioural problems
and cognitive deficits at 3 years of age? Int J Epidemiol
 Testa M, Quigley BM, Eiden RD. The effects of prenatal
alcohol exposure on infant mental development: a meta-
analytical review. Alcohol Alcohol 2008;38(4):295–304.
 Secker-Walker RH, Vacek PM, Flynn BS, Mead PB.
Smoking in pregnancy, exhaled carbon monoxide, and
birth weight. Obstet Gynecol 1997;89:648–53.
 Wisborg K, Henriksen TB, Hedegaard M, Secher NJ.
Smoking during pregnancy and preterm birth. Br J Obstet
 Wisborg K, Kesmodel U, Henriksen TB, Olsen SF,
Secher NJ. Exposure to tobacco smoke in utero and the
risk of stillbirth and death in the first year of life. Am J
 Weitzman M, Gortmaker S, Sobol A. Maternal smoking
and behavior problems of children. Pediatrics 1992;
 Christian MS, Brent RL. Teratogen update: evaluation of the
reproductive and developmental risks of caffeine. Teratology
 Barr HM, Streissguth AP. Caffeine use during pregnancy
and child outcome: a 7-year prospective study. Neurotoxicol
 Lumley J, Watson L, Watson M, Bower C, et al.
Periconceptional supplementation with folate and/or multi-
vitamins for preventing neural tube defects. Cochrane
Database Syst Rev 2001;(3):CD001056.
 Olsen SF, Secher NJ. Low consumption of seafood in early
pregnancy as a risk factor for preterm delivery: prospective
cohort study. BMJ 2002;324(7335):447.
 Olsen SF, Secher NJ, Tabor A, Weber T, Walker JJ, Gluud C.
Randomised clinical trials of fish oil supplementation in high
risk pregnancies. Fish Oil Trials In Pregnancy (FOTIP)
team. BJOG 2000;107((3):382–95.
 Ravelli AC, van der Meulen JH, Michels RP, Osmond C,
Barker DJ, Hales CN, Bleker OP. Glucose tolerance in adults
U. S. Kesmodel et al.
 Lucas A. Programming by early nutrition: an experimental Download full-text
approach. J Nutr 1998;128(2 Suppl):401S–6S.
 Kretchmer N, Beard JL, Carlson S. The role of nutrition in
the development of normal cognition. Am J Clin Nutr
 Grandjean P, Weihe P, White RF, Debes F, Araki S,
Yokoyama K, et al. Cognitive deficit in 7-year-old children
with prenatal exposure to methylmercury. Neurotoxicol
 Steuerwald U, Weihe P, Jorgensen PJ, Bjerve K, Brock J,
Heinzow B, et al. Maternal seafood diet, methylmercury
exposure, and neonatal neurologic function. J Pediatr
 Davidson PW, Myers GJ, Cox C, Axtell C, Shamlaye C,
Sloane-Reeves J, et al. Effects of prenatal and postnatal
methylmercury exposure from fish consumption on neuro-
development: outcomes at 66 months of age in the Seychelles
Child Development Study. JAMA 1998;280(8):701–7.
 Davidson PW, Myers GJ, Cox C, Wilding GE, Shamlaye CF,
Huang LS. Methylmercury and neurodevelopment: longitu-
dinal analysis of the Seychelles child development cohort.
Neurotoxicol Teratol 2006;28(5):529–35.
 Nordberg L, Rydelius PA, Zetterstrom R. Parental alcohol-
ism and early child development. Acta Paediatr Suppl
 Nulman I, Rovet J, Kennedy D, Wasson C, Gladstone J,
Fried S, Koren G. Binge alcohol consumption by non-
alcohol-dependent women during pregnancy affects child
behaviour, but not general intellectual functioning; a pro-
spective controlled study. Arch Women Mental Health
 Streissguth AP, Barr HM, Martin DC. Maternal alcohol use
and neonatal habituation assessed with the Brazelton scale.
Child Dev 1983;54(5):1109–18.
 Bailey BN, Delaney-Black V, Covington CY, Ager J,
Janisse J, Hannigan JH, Sokol RJ. Prenatal exposure to
binge drinking and cognitive behavioral outcomes at age 7
years. Am J Obstet Gynecol 2004;191(3):1037–43.
 Parry GJ, Ogston SA. EUROMAC. A European concerted
action: maternal alcohol consumption and its relation to the
outcome of pregnancy and child development at 18 months.
Results – child development at age 18 months. Int J
Epidemiol 1992;21(Suppl 1):S72–8.
 Olsen J. Effects of moderate alcohol consumption during
pregnancy on child development at 18 and 42 months.
Alcohol Clin Exp Res 1994;18(5):1109–13.
 Forrest F, Florey CD, Taylor D, McPherson F, Young JA.
Reported social alcohol consumption during pregnancy and
infants’ development at 18 months [see comments]. BMJ
 Sood B, Delaney-Black V, Covington C, Nordstrom-Klee B,
Ager J, Templin T. Prenatal alcohol exposure and childhood
behavior at age 6 to 7 years: I. dose-response effect.
 Goldschmidt L, Richardson GA, Cornelius MD, Day NL.
Prenatal marijuana and alcohol exposure and academic
achievement at age 10. Neurotoxicol Teratol 2004;26(4):
 Aronson M, Hagberg B. Neuropsychological disorders in
children exposed to alcohol during pregnancy: a follow-up
study of 24 children to alcoholic mothers in Go ¨teborg,
Sweden. Alcohol Clin Exp Res 1998;22(2):321–4.
 Korkman M, Kettunen S, Autti-Ramo I. Neurocognitive
impairment in early adolescence following prenatal alcohol
exposure of varying duration. Neuropsychol Dev Cogn Sect
C Child Neuropsychol 2003;9(2):117–28.
 Kesmodel U, Kesmodel PS. Drinking during pregnancy:
attitudes and knowledge among pregnant Danish women,
1998. Alcohol Clin Exp Res 2002;26(10):1553–60.
 Olsen J, Melbye M, Olsen SF, Sorensen TI, Aaby P,
Andersen AM. The Danish National Birth Cohort – its
background, structure and aim. Scand J Public Health
 Kesmodel U, Olsen SF. Self-reported alcohol intake in
pregnancy: comparison between four methods. J Epidemiol
Community Health 2001;55(10):738–45.
 Kesmodel U. Binge drinking in pregnancy – frequency and
methodology. Am J Epidemiol 2001;154(8):777–82.
 Kesmodel U, Frydenberg M. Binge drinking during preg-
nancy – is it possible to obtain valid information on a weekly
basis? Am J Epidemiol 2004;159(8):803–8.
 Roemer FJ, Rowland DY. Long-term developmental out-
comes of method of delivery. Early Hum Dev 1994;
 Schuerger JM, Witt AC. The temporal stability of individ-
ually tested intelligence. J Clin Psychol 1989;45(2):294–302.
 Batty GD, Deary IJ, Gottfredson LS. Premorbid (early life)
IQ and later mortality risk: systematic review. Ann Epidemiol
 McCall RB. Childhood IQ’s as predictors of adult educa-
tional and occupational
 Gottfredson LS. g: highly general and highly practical.
In: Sternberg RJ, Grigorenko EL, editors. The general
factor of intelligence: how general is it? Mahwah, New
Jersey: Erlbaum; 2002. pp. 331–80.
 Nohr EA, Frydenberg M, Henriksen TB, Olsen J. Does low
participation in cohort studies induce bias? Epidemiology
 Chen HJ, Tian H, Edenberg HJ. Natual haplotypes in
the regulatory sequences of affect human alcohol dehydro-
genase 1C (ADH1C)gene
 Choi IG, Kee BS, Son HG, Ham BJ, Yang BH, Kim SH.
Genetic polymorphisms of alcohol and aldehyde dehydro-
genase, dopamine, and serotonin transporters in familial and
non-familial alcoholism. Eur Neuropsychopharmacol 2005.
 Wechsler D. Manual for the Wechsler preschool and primary
scale of intelligence – Revised UK edition. Sidcup, Kent:
The Psychological Corporation; 1990.
 Gioia GA, Isquith PK, Guy SC, Kenworthy L. Behavior
rating inventory of executive function: professional manual.
Odessa, Florida: Psychological Assessment Resources Inc;
 Goodman R. The strengths and difficulties questionnaire:
a research note. J Child Psychol Psychiatry 1997;38(5):
 Naglieri J. Manual for draw a person: a quantitative scoring
system. New York: The Psychological Corporation; 1988.
 Wechsler D.Manual for the Wechsler adult intelligence scale.
New York: The Psychological Corporation; 1955.
 Kasper J. WAIS: Retningslinjer for testprocedure og scoring.
København: Dansk Psykologisk Forlag; 1990.
 Raven J, Raven J, Court J. Manual for Raven’s progressive
Psychologists Press Ltd; 1998.
 Henderson S, Sugden D. Movement assessment battery for
children, 2nd edn. København: Dansk Psykologisk Forlag;
status. Science 1977;
Alcohol in pregnancy