Hindawi Publishing Corporation
Journal of Pregnancy
Volume 2013, Article ID 780180, 6 pages
Prepregnancy Physical Activity in relation to
Offspring Birth Weight: A Prospective Population-Based Study
in Norway—The HUNT Study
Silje Krogsgaard, Sigridur L. Gudmundsdottir, and Tom I. L. Nilsen
Department of Human Movement Science, Norwegian University of Science and Technology, 7491 Trondheim, Norway
Correspondence should be addressed to Tom I. L. Nilsen; email@example.com
Received 12 October 2012; Revised 7 December 2012; Accepted 10 January 2013
Academic Editor: Michelle F. Mottola
Copyright © 2013 Silje Krogsgaard et al.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background. The objective was to examine the association between prepregnancy physical exercise and offspring birth weight and
to assess the combined association of pre-pregnancy body mass index (BMI) and physical exercise on birth weight. Methods. The
study included 2,026 women aged 20–39 years participating in the Norwegian HUNT study and linked with the Medical Birth
birth weight. Women who reported no exercise had reduced risk of a macrosomic infant (OR, 0.6; 95% confidence interval (CI),
0.4–0.9)comparedtowomenwithahighexerciselevel.Overweight(BMI ≥ 25.0kg/m2)wasassociatedwithanORof1.9(95%CI,
and risk of macrosomia but only among the least active women.
1.2–2.9) for a macrosomic infant among women who reported low exercise levels, whereas the OR was 1.2 (95% CI, 0.8–1.8) among
women with higher exercise levels. Conclusion. There was some evidence that women who reported no exercise before pregnancy
The proportion of women giving birth to large infants has
increased around the world [1, 2], most likely because of
the rising rates of maternal overweight and obesity [3–7].
Whereas consequences of low birth weight may include
related to increased risk for caesarean section, chorioam-
nionitis, fourth degree perinatal lacerations, postpartum
haemorrhage, shoulder dystocia [9–11], and low Apgar score
. Additionally, high birth weight has been positively
associated with obesity  and type 2 diabetes  in
Previous studies have reported that physical activity in
pregnancy is related to foetal growth rate and birth weight
[15, 16], and that physically active women have a reduced
risk of delivering a large infant [17, 18], possibly by increased
insulin sensitivity . However, not all studies have reported
consistent inverse associations between physical activity in
pregnancy and birth weight [19–22]. Although women who
exercise regularly before pregnancy are more likely to con-
tinue to exercise during pregnancy [23–25], few studies have
examined the associations between prepregnancy physical
activity and birthweight, and the results have been inconsis-
tent [18, 20, 26, 27].
on women participating in a population-based health study
linked with information from the Medical Birth Registry
to study the association between maternal pre-pregnancy
physical exercise and offspring birth weight. Additionally, we
explored the combined association of pre-pregnancy body
mass index and physical exercise on birth weight.
2. Materials and Methods
HUNT study) is a large population-based health study con-
2 Journal of Pregnancy
Table 1: Characteristics of the study population (N = 2,026) according to total leisure time physical exercise.
Total physical exercise levela
No participants (% of total)
Mean age at baseline, y
Mean body mass index, kg/m2
Parity (% primiparous)
Smoking (% current smoking)
Education (% college/university)
Alcohol (% not drinking last 2weeks)
Marital status (% married)
aBased on a summary score of frequency, duration, and intensity of exercise.
study is a collaboration between HUNT Research Centre
(Faculty of Medicine, Norwegian University of Science and
Technology NTNU), Nord-Trøndelag County Council, Cen-
waves; the first was conducted in 1984–1986 (HUNT 1), the
second in 1995–1997 (HUNT 2), and the third in 2006–
2008 (HUNT 3). For the purpose of the present study, we
have used information from the first wave (HUNT 1). In
HUNT 1, 87,285 persons aged ≥ 20 years were invited to
men and 39,390 women). A more detailed description of
participation, method, and procedures of the HUNT study
can be found elsewhere .
For the purpose of the current study, we first selected all
3,739 women aged 20–39 years at participation in HUNT 1
(1984–1986) who gave birth to at least one child during a five
year period after participation. Of these 3,739 women, we
participate, and 77,216 (88.5%) accepted the invitation, filled
excluded 232 women due to pre- (<37 weeks) or postterm
the time of participation).This left 2,026 women available for
(>44 weeks) delivery, 32 women with multiple births, 10
2.2. Study Variables. Information on the offspring was
obtained by a linkage to the Medical Birth Registry of
Norway. These data were obtained for the first child born
during five years after participation in HUNT. The main
outcome variable was newborn birth weight, measured in
grams (g), first analyzed as a continuous variable and then as
a dichotomized variable using 4,000g as cutoff. Macrosomia
was defined as birth weight at or above 4,000g .
Leisure time physical exercise was assessed using three
questions. In the first question, the participants were asked
to report how many exercise sessions (e.g., walking, skiing,
swimming, or other sports) they usually had during a week,
with five response options (0, <1, 1, 2-3, and ≥4 times;
coded 1–5). If the participants reported exercising at least
once a week, they were also asked to report the average
least once a week, a summary score of frequency, duration,
and intensitywas calculated accordingto the followingequa-
intensity (light, moderate, and to exhaustion; coded 1–3) of
the activity. Among participants who reported exercising at
tion:1/5 ∗ frequency+1/4 ∗ duration+1/3 ∗ intensity.This
for each of the three components of the summary score. The
median score value of 1.97 (range, 1.2–3.0) was then used
as a cut-off to classify women into two categories of score
(1) no activity, (2) low activity (<1 session per week), (3)
measures as weight divided by the squared value of height
and BMI ≥ 25.0kg/m2(i.e., overweight or obese) .
2.3. Statistical Analysis. We used linear regression to analyze
the association between measures of leisure time physical
exercise and mean birth weight. We also calculated odds
procedure intended to give equal weight to each component
of physical activity and resulted in a maximum score of 1.0
values (±median). This information was used to construct a
(≥median score value) .
medium activity (<median score value), and (4) high activity
(kg/m2) and categorized into two groups: BMI < 25.0kg/m2
ratios (OR) for having a macrosomic infant (≥4,000g) in
was assessed by a 95% confidence interval (CI). Women
who reported the highest activity level were used as the
considered as potential confounders in the analysis; age (20–
24, 25–29, 30–34, and 35–39 years), smoking (never, former,
and current), frequency of alcohol consumption during the
different categories of leisure time physical exercise using
logistic regression. Precision of the estimated associations
past 2 weeks (none, 1–4 times, ≥5 times, abstainer, and
and parity (primiparous, 1-2 children, and 3–6 children).
Covariates were removed from the model if there was no
meaningful difference between adjusted and unadjusted esti-
and parity. Tests for trend across categories of leisure time
physical exercise were conducted by treating the categories
as an ordinal variable in the regression model.
unknown), education (<10, 10–12, >12 years, and unknown),
Journal of Pregnancy3
Table 2: Maternal pre-pregnancy leisure time physical exercise and mean offspring birth weight from linear regression analyses.
Sessions per week
No. of persons Mean birth weight (g)Crude difference
Adjustedadifference (95% CI)
−26.9 (−137.9 to 84.2)
21.7 (−64.8 to 108.3)
−53.0 (−141.8 to 35.8)
CI: confidence interval.
aAdjusted for maternal age (20–24, 25–29, 30–34, and 35–39 years), smoking (never, former, current, and unknown), and parity (primiparous, 1-2 children,
and 3–6 children).
bP value from trend test when categories were entered as an ordinal variable in the regression model.
cBased on a summary score of frequency, duration, and intensity of exercise.
31.0 (−56.1 to 118.1)
−24.4 (−113.2 to 64.4)
−3.5 (−58.3 to 51.4)
−36.9 (−91.5 to 17.8)
Since maternal BMI could be both an effect modifier and
on the causal pathway between exercise and birth weight,
BMI was not included as a confounder in the primary
analyses. However, additional analysis was conducted for the
time physical exercise in relation to birth weight, using linear
and logistic regression as described earlier. We also included
a product term of BMI (<25 versus ≥25kg/m2) and exercise
well as stratified the analyses of physical exercise on the two
All statistical analyses were performed using the statisti-
cal software SPSS for Windows, version 17.0.
The study was approved by the Regional Committee for
Ethics in Medical Research. All eligible participants received
a written invitation with information about the study, and
all participants gave their consent by filling in and returning
the first questionnaire that was mailed together with the
level (no or low activity versus medium or high activity) to
assess possible interaction between the two variables and as
Descriptive characteristics of the study population are pre-
sented in Table 1. Mean baseline maternal age among the
weight of their offspring was 3,620g (SD, 502), and a total
of 416 (20.5%) newborns weighed 4,000g or more (i.e., had
Table 2 presents results from linear regression show-
ing that there was no clear association between maternal
leisure time physical exercise and mean birth weight of
their offspring, neither in relation to number of exercise
sessions per week (푃 trend, 0.49) nor in relation to total
consistent evidence for an association between maternal
amount of exercise (푃 trend, 0.56). Correspondingly, results
from logistic regression presented in Table 3 provide no
a marcosomic infant (i.e., birth weight>4000g) in association with
Sessions per week
No activity147 20
CI: confidence interval.
(never, former, current, and unknown), and parity (primiparous, 1-2 chil-
dren, and 3–6 children).
bP value from trend test when categories were entered as an ordinal variable
in the regression model.
cBased on a summary score of frequency, duration, and intensity of exercise.
maternal pre-pregnancy leisure time physical exercise.
physical exercise and risk of a macrosomic infant. However,
women who reported being inactive before pregnancy had
a lower risk of giving birth to an infant with excessive birth
weight (OR, 0.6; 95% CI, 0.4–0.9) compared to women with
a high total exercise level (Table 3).
The combined association of pre-pregnancy BMI and
leisure time physical exercise in relation to birth weight is
25.0kg/m2) before pregnancy and reported no or low leisure
shown in Table 4. Women who were overweight (BMI ≥
time physical exercise gave birth to infants with significantly
linear regression and odds ratio (OR) from logistic regression for a macrosomic infant (i.e., birth weight>4000g).
BMI and exercisea
<25kg/m2and no/low level
CI: confidence interval; BW: birth weight.
aBased on a summary score of frequency, duration, and intensity of exercise.
bAdjusted for maternal age (20–24, 25–29, 30–34, and 35–39 years), smoking (never, former, current, and unknown), and parity (primiparous, 1-2 children,
and 3–6 children).
Journal of Pregnancy
Table 4: Maternal pre-pregnancy body mass index (BMI) and leisure time physical exercise related to mean offspring birth weight from
Combined categories of
≥25kg/m2and no/low level
No. of Crude meanAdjustedbmean
difference (95% CI)
−8.4 (−55.6 to 38.8)
134.1 (41.0 to 227.3)
No. of cases
1,046 0.00.0 (reference)
33.1 (−42.5 to 108.7)
−8.6 1.01.0 (0.8–1.2)
186 29.842 1.2 1.2 (0.8–1.8)
37 1.91.9 (1.2–2.9)
higher mean birth weight (134g; 95% CI, 41.0–227.3) and had
a higher risk for a macrosomic infant (OR, 1.9; 95% CI, 1.2–
by body mass index, overweight women (BMI ≥ 25.0kg/m2)
higher odds ratio for a macrosomic infant (OR, 2.0; 95% CI,
level (data not shown). However, there was no statistically
significant interaction between BMI and total exercise level
2.9), compared to women with BMI < 25.0kg/m2who had a
who reported no or low exercise levels had significantly
higher offspring birth weight (132g; 95% CI, 20.4–243.7) and
(푃 = 0.08).
In this large prospective study of Norwegian women, we
found no clear association between reported leisure time
physical exercise level before pregnancy and offspring birth
weight. There was some evidence that inactive women had
a slightly lower likelihood of giving birth to a child with
excessive birth weight than more physically active women,
but the small numbers of inactive women call for a cautious
interpretation. Analysis of the combined association of BMI
and exercise on birth weight showed that women with a
higher birth weight than women with a BMI < 25.0kg/m2,
extent reduces the effect of maternal adiposity on offspring
The suggestive evidence that women who were inactive
before pregnancy had lower risk for delivering a macrosomic
infant is contradictory to some previous studies. Voldner
macrosomia than physically active women (>1h per week).
BMI ≥ 25.0kg/m2gave birth to infants with significantly
but only if they also reported no or low levels of physical
exercise. This could suggest that physical exercise to some
et al.  reported that inactive women (defined as <1h
However, another Norwegian study found no association
between frequency of regular exercise before pregnancy and
per week) had almost a threefold higher odds ratio for fetal
offspring with excessive birth weight (≥90th percentile) .
women could be more extremely sedentary than inactive
In the present study, those who were classified as inactive
reported never engaging in physical exercise, and these
women in other studies. Studies have shown that a sedentary
lifestyle in pregnancy is associated with lower birth weight
 and an increased risk of a very low birth weight infant
. It has been observed that mothers of very low birth
weight infants were less likely to be physically active during
and insufficient physical activities in pregnancy are related
to an inadequate fetal growth , although some women
might be advised to be inactive and at rest to reduce the risk
of adverse pregnancy outcomes.
Unlike the present study, some previous studies have
shown inverse associations between maternal pre-pregnancy
excess weight [17, 18], although the results are not entirely
present study. Hegaard et al.  found no association with
of physical activity, in addition to the possibility for chance
findings in the smaller studies.
There is growing evidence that overweight or obesity
The results from the present study suggest that the effects
of maternal pre-pregnancy overweight were associated with
higher birth weight only among women who reported no or
low level of activity. This is contradictory to the findings by
L¨ of et al.  who showed that a high pre-pregnancy activity
level did not reduce the risk of high birth weight infants
in women who were overweight or gained much weight
in pregnancy. Nevertheless, physical activity may improve
maternal weight control, both before and during pregnancy
design, the large sample size of women reporting physical
activity before pregnancy, and the standardized measures of
size at birth obtained from the Medical Birth Registry of
Norway. However, some of the categories of physical exercise
(e.g., inactive) suffered from small samples size, and this
could result in chance findings. Moreover, as in any obser-
vational study, residual confounding due to unmeasured
or high (≥4,500g) birth weight infants. The inconsistent
results in these studies could be due to different measures
Journal of Pregnancy5
(e.g., smoking during pregnancy and gestational weight
gain) or unknown factors cannot be ruled out. Since the
subject to misclassification, although a validation study has
shown acceptable agreement with objective measures .
with physical activity level during pregnancy [23–25], it has
also been shown that exercise levels decline in pregnancy
[24, 25, 40, 41]. Unfortunately, our data did not allow us to
examine if pre-pregnancy physical exercise was related to the
activity level during pregnancy.
There was no clear association between maternal prepreg-
nancy leisure time physical exercise and offspring birth
gave birth to infants with lower birth weight and had
lower risk for delivering an infant with macrosomia. The
results also show that high maternal pre-pregnancy BMI was
associated with higher mean birth weight and increased risk
of macrosomia in offspring of physically inactive women,
whereas pre-pregnancy BMI was not associated with birth
weight and risk of macrosomia among more active women.
Conflict of Interests
The authors declare that they have no conflict of interests.
 M. S. Kramer, I. Morin, H. Yang et al., “Why are babies getting
bigger? Temporal trends in fetal growth and its determinants,”
Journal of Pediatrics, vol. 141, no. 4, pp. 538–542, 2002.
 J. Ørskou, U. Kesmodel, T. B. Henriksen, and N. J. Secher, “An
increasing proportion of infants weigh more than 4000grams
at birth,” Acta Obstetricia et Gynecologica Scandinavica, vol. 80,
no. 10, pp. 931–936, 2001.
 S. Bhattacharya, D. M. Campbell, W. A. Liston, and S. Bhat-
tacharya, “Effect of body mass index on pregnancy outcomes
in nulliparous women delivering singleton babies,” BMC Public
Health, vol. 7, article 168, 2007.
 H. M. Ehrenberg, L. Dierker, C. Milluzzi, and B. M. Mercer,
“Prevalence of maternal obesity in an urban center,” American
Journal of Obstetrics and Gynecology, vol. 187, no. 5, pp. 1189–
 S. Phelan, C. Hart, M. Phipps et al., “Maternal behaviors during
Research, vol. 2011, Article ID 985139, 9 pages, 2011.
and obesity: modifiable determinants of pregnancy outcome,”
Human Reproduction Update, vol. 16, no. 3, pp. 255–275, 2010.
 M. S. Kramer, “The epidemiology of adverse pregnancy out-
comes: an overview,” Journal of Nutrition, vol. 133, no. 5,
supplement 2, pp. 1592S–1596S, 2003.
 T. Henriksen, “The macrosomic fetus: a challenge in current
obstetrics,” Acta Obstetricia et Gynecologica Scandinavica, vol.
87, no. 2, pp. 134–145, 2008.
 N. E. Stotland, A. B. Caughey, E. M. Breed, and G. J. Esco-
bar, “Risk factors and obstetric complications associated with
macrosomia,” International Journal of Gynecology and Obstet-
rics, vol. 87, no. 3, pp. 220–226, 2004.
 N. Voldner, K. F. Frøslie, L. A. H. Haakstad, K. Bø, and
T. Henriksen, “Birth complications, overweight, and physical
inactivity,” Acta Obstetricia et Gynecologica Scandinavica, vol.
88, no. 5, pp. 550–555, 2009.
 M. C. Jolly, N. J. Sebire, J. P. Harris, L. Regan, and S. Robinson,
“Risk factors for macrosomia and its clinical consequences: a
study of 350,311 pregnancies,” European Journal of Obstetrics
Gynecology and Reproductive Biology, vol. 111, no. 1, pp. 9–14,
 W. McGuire, L. Dyson, and M. Renfrew, “Maternal obesity:
Seminars in Fetal and Neonatal Medicine, vol. 15, no. 2, pp. 108–
 T. Harder, E. Rodekamp, K. Schellong, J. W. Dudenhausen, and
A. Plagemann, “Birth weight and subsequent risk of type 2
diabetes: a meta-analysis,” American Journal of Epidemiology,
vol. 165, no. 8, pp. 849–857, 2007.
 J. F. Clapp III, H. Kim, B. Burciu, and B. Lopez, “Beginning
regular exercise in early pregnancy: effect on fetoplacental
growth,” American Journal of Obstetrics and Gynecology, vol.
183, no. 6, pp. 1484–1488, 2000.
 J. A. Leiferman and K. R. Evenson, “The effect of regular leisure
Journal, vol. 7, no. 1, pp. 59–64, 2003.
 B. W. Alderman, H. Zhao, V. L. Holt, D. H. Watts, and S. A. A.
Beresford, “Maternal physical activity in pregnancy and infant
exercise and excessive newborn birth weight,” Obstetrics and
Gynecology, vol. 114, no. 4, pp. 770–776, 2009.
 A. M. Z. Jukic, K. R. Evenson, J. L. Daniels, A. H. Herring,
A. J. Wilcox, and K. E. Hartmann, “A prospective study
of the association between vigorous physical activity during
pregnancy and length of gestation and birthweight,” Maternal
and Child Health Journal, vol. 16, no. 5, pp. 1031–1044, 2012.
 N. Voldner, K. F. Frøslie, K. Bø et al., “Modifiable determinants
of fetal macrosomia: role of lifestyle-related factors,” Acta
Obstetricia et Gynecologica Scandinavica, vol. 87, no. 4, pp.423–
 M. Juhl, J. Olsen, P. K. Andersen, E. A. Nøhr, and A. M. N.
measures: a study within the Danish National Birth Cohort,”
American Journal of Obstetrics and Gynecology, vol. 202, no. 1,
pp. 63.e1–63.e8, 2010.
 H. K. Hegaard, K. Petersson, M. Hedegaard et al., “Sports and
leisure-time physical activity in pregnancy and birth weight: a
population-based study,” Scandinavian Journal of Medicine and
Science in Sports, vol. 20, no. 1, pp. e96–e102, 2010.
 L. A. H. Haakstad, N. Voldner, T. Henriksen, and K. Bø, “Why
do pregnant women stop exercising in the third trimester?”
during pregnancy: the Norwegian Mother and Child Cohort
Study,” Scandinavian Journal of Medicine and Science in Sports,
vol. 19, no. 5, pp. 637–645, 2009.
6 Journal of Pregnancy Download full-text
 H. K. Hegaard, P. Damm, M. Hedegaard et al., “Sports and
leisure time physical activity during pregnancy in nulliparous
women,” Maternal and Child Health Journal, vol. 15, no. 6, pp.
 C. Fleten, H. Stigum, P. Magnus, and W. Nystad, “Exercise
during pregnancy, maternal prepregnancy body mass index,
1, pp. 331–337, 2010.
 M. L¨ of, L. Hilakivi-Clarke, S. Sandin, and E. Weiderpass,
“Effects of pre-pregnancy physical activity and maternal BMI
on gestational weight gain and birth weight,” Acta Obstetricia et
Gynecologica Scandinavica, vol. 87, no. 5, pp. 524–530, 2008.
 S. Krokstad, A. Langhammer, K. Hveem et al., “Cohort profile:
the HUNTStudy, Norway,” International Journal of Epidemiol-
ogy. In press.
somic births in the United States: determinants, outcomes, and
proposed grades of risk,” American Journal of Obstetrics and
Gynecology, vol. 188, no. 5, pp. 1372–1378, 2003.
 T. I. L. Nilsen, P. R. Romundstad, and L. J. Vatten, “Recreational
physical activity and risk of prostate cancer: a prospective
population-based study in Norway (the HUNT study),” Inter-
national Journal of Cancer, vol. 119, no. 12, pp. 2943–2947, 2006.
 M. I. Both, M. A. Overvest, M. F. Wildhagen, J. Golding, and
H. I. J. Wildschut, “The association of daily physical activity
Journal of Epidemiology, vol. 25, no. 6, pp. 421–429, 2010.
 W. F. Schramm, J. W. Stockbauer, and H. J. Hoffman, “Exercise,
employment, other daily activities, and adverse pregnancy
 M. Y. Takito, M. H. Ben´ ıcio, and L. C. Neri, “Physical activity
by pregnant women and outcomes for newborns: a systematic
review,” Revista de sa´ ude p´ ublica, vol. 43, no. 6, pp. 1059–1069,
 I. O. Frederick, M. A. Williams, A. E. Sales, D. P. Martin, and
gain, and other maternal characteristics in relation to infant
birth weight,” Maternal and Child Health Journal, vol. 12, no. 5,
pp. 557–567, 2008.
 H. M. Ehrenberg, B. M. Mercer, and P. M. Catalano, “The influ-
ence of obesity and diabetes on the prevalence of macrosomia,”
American Journal of Obstetrics and Gynecology, vol. 191, no. 3,
pp. 964–968, 2004.
 K. Melzer, Y. Schutz, M. Boulvain, and B. Kayser, “Physical
dations and pregnancy outcomes,” Sports Medicine, vol. 40, no.
6, pp. 493–507, 2010.
 Z. Sui, R. M. Grivell, and J. M. Dodd, “Antenatal exercise to
improve outcomes in overweight or obese women: a systematic
review,” Acta Obstetricia et Gynecologica Scandinavica, vol. 91,
no. 5, pp. 538–545, 2012.
 N. Kurtze, V. Rangul, B. E. Hustvedt, and W. D. Flanders,
“Reliability and validity of self-reported physical activity in the
Nord-Trøndelag Health Study (HUNT 2),” European Journal of
Epidemiology, vol. 22, no. 6, pp. 379–387, 2007.
 L. A. H. Haakstad, N. Voldner, T. Henriksen, and K. Bø,
“Physical activity level and weight gain in a cohort of pregnant
Norwegian women,” Acta Obstetricia et Gynecologica Scandi-
navica, vol. 86, no. 5, pp. 559–564, 2007.
 D. B. Fell, K. S. Joseph, B. A. Armson, and L. Dodds, “The
impact of pregnancy on physical activity level,” Maternal and
Child Health Journal, vol. 13, no. 5, pp. 597–603, 2009.