American Journal of Epidemiology
Copyright C 1999 by The Johns Hopkins University School of Hygiene and Public Health
All rights reserved
vol. 149, No. 8
Printed in U.S.A.
Birth Weight and Length as Predictors for Adult Height
HenrikToft S0rensen, u Svend Sabroe,1 Kenneth J. Rottiman,3 Matthew Gillman,* Flemming Hald Steffonsen,1*
Peer Fischer,6 and Thorkild I. A. Serensen7
Adult height has been found to be inversely associated with mortality. Recently, it has been suggested that
growth in utero is linked with adutt risk of several chronic diseases. The authors examined possible associations
between birth weight, birth length, and adult height in young Danish men. They conducted the study in the fifth
conscription district of Denmark including all the men born after January 1,1973 who were residents in the study
area during the period August 1, 1993 to July 31, 1994. The Danish Medical Birth Register contains information
on all births in Denmark since January 1, 1973. Data on height from the Conscription Register were linked to
the Danish Medical Birth Register in 4,300 conscripts examined. Nearly all Danish men have to register with the
draft board around age 18 years of age where they undergo a physical examination. There was a strong positive
association between birth weight and adutt height; for subjects with birth weight £2,500 g, mean height was
175.7 cm, while for those with birth weight £4,501 g, mean height was 184.1 cm. A positive association was also
found between birth length and adult height. For subjects with birth length <47 cm, mean adult height was 175.2
cm, increasing to 184.3 cm at birth length >56 cm. The associations between birth length and adult height
persisted after adjustment for birth weight, gestational age, and other confounders, while the associations
between birth weight and adult height almost disappeared when adjusting for birth length and the same
confounders. Genetic and/or environmental factors operating both during the pre- and postnatal period may be
responsible for the association between birth length and adult height. Am J Epidemiol 1999;149:726-9.
birth weight; body height; body weight; fetal development
Adult height has been found to be inversely associated with mortality in several cohort studies in different countries, with the strongest association found for
mortality from respiratory and coronary heart disease
(1-3). Attained adult height is thought to be partly
determined by factors in utero, in infancy, and in childhood (4-9), but the direct link between prenatal
growth and adult height has not been established. Birth
weight has been associated with adult risk of coronary
heart diseases (10). Several studies including our own
have found that birth weight may be associated with a
later risk of obesity (11, 12), which, as a risk factor for
coronary heart disease, suggests a relation in the opposite direction of that between fetal growth and coronary heart disease. On the other hand, the association
between height and mortality is consistent with the
programming hypothesis that early life factors influence mortality in adulthood (13, 14). Even though this
programming hypothesis is controversial (15), the
associations between early growth, final height, and
morbidity and mortality may indicate that important
early determinants for adult diseases exist.
We have examined possible associations between
birth weight, birth length, and adult height in a population of Danish young men evaluated for military service.
Received for publication March 13, 1998, and accepted for publication August 12, 1998.
1
Danish Epidemiology Science Centre at the Department of
Epidemiology and Social Medicine, University of Aarhus, Aartius,
Denmark.
2
Department of Internal Medicine V, Aartius University Hospital,
Aarhus, Denmark
3
Section of Preventive Medicine and Epidemiology, Boston
University, School of Medicine, Boston, MA.
4
Department of Ambulatory Care and Prevention, Harvard
Medical School and Harvard Pilgrim Health Care, Boston, MA.
5
Department of Internal Medicine M, Aalborg Hospital, Aalborg,
Denmark.
8
Fifth Conscription District, Aalborg, Denmark.
7
Danish Epidemiology Science Centre at the Institute of
Preventive
Medicine,
Copenhagen
University
Hospital,
Copenhagen, Denmark.
Reprint requests to Dr. Henrik Toft Sarensen, The Danish
Epidemiology Science Centre, University of Aartius, HoeghGuldbergsgade 10, DK-8000 Aarhus C, Denmark.
MATERIALS AND METHODS
We conducted the study in the fifth conscription district of Denmark, which mainly includes the counties
of North Jutland and Viborg. The population size of
this region is approximately 700,000, about 14 percent
of the total population in Denmark. Nearly all Danish
men have to report to the draft board around 18 years
of age and rarely beyond age 26 years; at the draft
726
Fetal Growth and Adult Height
board, they undergo physical examination and intelligence testing. A few were exempted from examination.
We studied all the men examined who were born after
January 1, 1973 and residents in the study area during
the period August 1, 1993 to July 31, 1994.
We linked data on height from the examination to
the Danish Medical Birth Register by means of the
central personal registration number (CPR number),
which has been given to all Danes at birth since 1968.
The CPR number is used in all Danish data sources,
thus making linkage between data bases simple and
valid. The Birth Register contains information on all
births in Denmark since January 1, 1973 and data are
collected for ongoing statistics. The data are obtained
from the notifications of birth, which are recorded by
the midwife attending the delivery. In the Birth
Register, information on birth weight is divided into
categories of 250 g, and we used midpoints of the categories in the analyses. Birth length was reported to
the register in centimeters, and was included in the
regression analysis as a continuous variable.
We first examined adult height in relation to the
study variables. To control for confounding, we fitted
linear regression models to the data that included
weight and length at time of birth, birth order, gestational age, marital status, mother's age, and employment status. To obtain a clearer picture of the shape of
the relation between birth length/birth weight and adult
height, we used confounder-adjusted quadratic splines
to smooth the curves (16).
The study was approved by the Regional Ethics
Committee and the Danish Data Protection Agency.
RESULTS
We identified 4,805 draftees born after January 1,
1973 during the study period. Of these, 4,300 draftees
underwent a medical examination. The majority of the
remaining 505 men were exempted because of asthma,
juvenile osteochondrosis, or epilepsy.
Table 1 shows the association between birth weight
and adult height at the draft board examination. There
was a strong positive association between birth weight
and adult height; at birth weight <2,500 g, the mean
height was 175.7 cm, while at birth weight >4,501 g, it
was 184.1 cm. Table 1 also shows a positive association between length at the time of birth and adult
height. At birth length <47 cm, mean adult height was
175.2 cm, increasing to 184.3 cm at birth length >56
cm.
To improve our estimate of the shape of the doseresponse curve relating the birth length to height, we
fitted an unrestricted quadratic spline regression model
with all covariates including birth weight. Figure 1
shows a strong association between birth length and
Am J Epidemiol
Vol. 149, No. 8, 1999
727
TABLE 1. Birth weight, birth length, and gestational age relative to adult height at age 18-20 years among young Danish
men, August 1,1993 to July 31,1994
Mean
adutt
height (cm)
95% Cl*
lor
mean height
175.7
105
177.2
178.0
179.0
180.3
181.1
182.5
184.1
174.7-176.6
176.7-177.8
177.5-178.5
178.6-179.5
179.8-180.7
180.6-181.5
182.0-183.0
183.0-185.2
79
172
661
1,333
1,342
613
124
175.2
176.7
176.8
178.6
180.8
182.4
184.3
173.7-176.8
175.6-177.8
176.4-177.3
178.3-178.9
180.5-181.1
181.9-182.8
183.3-185.3
3,900
399
179.7
178.2
179.5-179.9
177.6-178.9
No.
Birth weight (g)
52,500
2,501-3,000
3,001-3,250
3,251-3,500
3,501-3,750
3,751-4,000
4,001^,500
£4,501
Length at birth (cm)
5*46
47-48
49-50
51-52
53-54
55-56
£57
Gestational age
Full-term
Pre-term
171
604
607
845
776
err
515
* Cl, confidence interval.
adult height. The smoothed curve indicates a peak
around 58 cm after which height declines slightly.
Figure 2 presents the corresponding regression with
birth weight as predictive variable and birth length
included among the covariates. The curve indicates a
positive adjusted association between birth weight and
height, but much less pronounced than the association
with birth length. There was a strong association
between birth weight and length, but the spline curves
indicate that birth length was a much stronger predictor for adult height than birth weight.
DISCUSSION
We found a clear association between birth length,
birth weight, and adult height; the association was
most pronounced for birth length, which outweighed
the effect of birth weight in a common analysis.
However, because birth weight is measured in categories and length as a continuous variable, some residual confounding of birth weight on the length and adult
height association cannot be excluded. The association
persisted when adjusted for possible effects of gestational age, birth order, mother's age, marital status, and
employment status.
Fetal growth is generally considered in terms of
birth weight and not birth length, and it is consequently difficult to make comparisons with postnatal
growth, which is most often expressed in length. A
recent randomized trial in Java (17) showed that
728
Sorensen et al.
Height cm
Height cm
190
190,
185
185
180
180
175
175
170
170
165
165
160
1601 i | i i | | | i i i | | | i |—]—i
\ i i t \ i i i i i i i i i i i i i i i i i i r
n
40 42 44 46 48 50 52 54 56 58 60 62
Birth length cm
Birth weight g
FIGURE 1. Estimated mean adult height according to birth length,
Viborg, North Jutland County, 1973-1994. Smoothed curve was fitted by unrestricted quadratic spline regression with adjustment for
gestational age, birth weight, maternal age, birth order, marital status, and employment.
FIGURE 2. Estimated mean adult height according to birth weight,
Viborg, North Jutland County, 1973-1994. Smoothed curve was fitted by unrestricted quadratic spline regression with adjustment for
gestational age, birth length, birth weight, maternal age, birth order,
marital status, and employment.
among women in the reproductive age, energy supplementation of the last 90 days of pregnancy increased
postnatal growth and reduced malnutrition in
preschool children. The most striking result was the
greater effect on height than on weight. Adult height
may thus be an indicator of early nutrition.
Our findings are in agreement with other studies of
birth weight and height associations using somewhat
different types of data and design. A recent populationbased study that followed small for gestational age
infants to adulthood (18) showed that they had a sevenfold higher risk of being short than subjects who
were not born small for gestational age. Leger et al.
(19) studied 236 full term singleton babies small for
gestational age and 281 with normal birth weight. The
target height was 174 cm versus 175 cm in men and
160 versus 162 in women. A study from Israel (4)
found a strong association between birth weight and
adult height regardless of ethnic and socioeconomic
factors. Another Israeli study with 17 years follow-up
of small for gestational age infants (9) found an
increased risk of short stature at late adolescence.
It seems likely that implicit growth factors as well as
genetic and environmental factors are operating. The
biologic factors determining the growth of the fetus are
the genetic makeup of the fetus combined with the
utero-placental function and thus the substrate availability (17). The association between birth length and
adult height may be explained by several factors that
do not invoke in utero determinants of adult height: I)
tall stature has genetic determinants that may be
expressed in infants as well as adults; 2) although we
adjusted for some social factors, there may be residual
confounding between stature and social and other
environmental factors that affect both birth length and
adult height. Secular trend studies within populations
definitely demonstrate that common general environmental factors influence height of the members of the
Am J Epidemiol Vol. 149, No. 8, 1999
Fetal Growth and Adult Height
population (20). On the other hand, several twin studies have shown that within the population variation in
height at a given point in time is mainly attributable to
genetic differences (5, 21). It has been hypothesised
(15) that "metabolic programming" occurring at critical stages of early development causally determines
the risk of adult diseases such as coronary heart disease, stroke, diabetes mellitus, and hypertension.
However, several methodological problems, such as
loss to follow-up and confounding by socioeconomic
factors, suggest the advisability of cautious interpretations about the possible causal relations between fetal
growth and risk of later disease (15).
ACKNOWLEDGMENTS
Supported by grants from Sygekassernes Helsefond
(11/064-94). The activities of the Danish Epidemiology
Science Centre are financed by a grant from the Danish
National Research Foundation.
The authors thank Hans J0rgen Pedersen for helping with
the data collection.
REFERENCES
1. Waaler HT. Height, weight and mortality. Acta Med Scand
1984;679(suppl):l-56.
2. Peck ANM, VSgero D. Adult body height, self perceived
health and mortality in the Swedish population. J Epidemiol
Community Health 1989;43:380-4.
3. AHbeck P, Bergh C. Height, body mass and mortality; do
social factors explain the association? Public Health 1992;106:
375-82.
4. Seidman DS, Gale R, Stevenson DK, et al. Is the association
between birth weight and height attainment independent of the
confounding effect of ethnic and socioeconomic factors? Isr J
MedSci 1993;29:772-6.
Am J Epidemiol Vol. 149, No. 8, 1999
729
5. Rona RJ. Genetic and environmental factors in the control of
growth in childhood. Br Med Bull 1981;37:265-72.
6. Kariberg J, Albertsson-Wikland K, Baber FM, et al. Bom
small for gestational age: consequences for growth. Acta
Paediatr 1996;417(suppl):8-13.
7. Strauss RS, Dietz WH. Effects of intrauterine growth retardation in premature infants on early childhood growth. J Pediatr
1997; 130:95-120.
8. Bacallao J, Amador M, Hermelo M. The relationship of birth
weight with height at 14 and with the growing process.
Nutrition 1996; 12:250-4.
9. Paz I, Seidman DS, Danon YL, et al. Are children born small
for gestational age at increased risk of short stature? Am J Dis
Child 1993; 147:337-9.
10. Barker DJP. Mothers, babies, and disease in later life. London:
BMJ Publishing Group, 1994.
11. S0rensen HT, Sabroe S, Rothman KJ, et al. Relation between
weight and length at birth and body mass index in young adulthood: cohort study. BMJ 1997;315:1137.
12. Seidman DS, Laor A, Gale R, et al. A longitudinal study of
birth weight and being overweight in late adolescence. Am J
Dis Child 1991; 145:782-5.
13. Leon DA, Smith GD, Shipley M, et al. Adult height and mortality in London: early life, socioeconomic confounding, or
shrinkage. J Epidemiol Community Health 1995;49:10-15.
14. Barker DJP, Osmond C, Golding J. Height and mortality in the
counties of England and Wales. Ann Hum Biol 1990;17:l-6.
15. Joseph KS, Kramer MS. Review of the evidence on fetal and
early childhood antecedents of adult chronic disease.
Epidemiol Rev 1996;18:158-74.
16. Greenland S. Dose-response and trend analysis in epidemiology: alternatives to categorical analysis. Epidemiology 1995;6:
356-65.
17. Kusin JA, Kardjati S, Houdcooper JM, et al. Energy supplemention during pregnancy and postnatal growth. Lancet 1992;
340:623-6.
18. Kariberg J, Albertsson-Wikland K. Growth in full-term small
for gestational age infants: from birth to final height. Pediatr
Res 1995;38:733-9.
19. Leger J, Leyy-Marchal C, Bloch J, et al. Reduced fetal height
and indications for insulin resistance in 20-year olds bom
small for gestational age: regional cohort study. BMJ 1997;
315:341-7.
20. Hasle H, Boldsen JL. Childhood conditions and adult height. J
BiosocSci 1991;23:107-12.
21. Tambs K, Moun T, Eaves LJ, et al. Genetic and environmental
contributions to the variance of body height in a sample of first
and second degree relatives. Am J Phys Anthropol 1992;88:
285-94.