GROWTH IN SWEDEN
Surveillance of growth patterns and
epidemiological monitoring of secular changes in
height and weight among children
and adolescents
Bo Werner
Stockholm 2007
From Karolinska Institutet, Department of Public Health Sciences
Division of Social Medicine, SE-171 76 Stockholm, Sweden
GROWTH IN SWEDEN
Surveillance of growth patterns and epidemiological monitoring
of secular changes in height and weight among children and
adolescents
Bo Werner
ÖREBRO LÄNS LANDSTING
Samhällsmedicinska enheten
ÖREBRO LÄNS LANDSTING
Samhällsmedicinska enheten
Barnhälsovården
ÖREBRO LÄNS LANDSTING
Barnhälsovården
ÖREBRO LÄNS LANDSTING
Stockholm 2007
Growth in Sweden. Surveillance of growth patterns and epidemiological monitoring of secular
changes in height and weight among children and adolescents.
Copyright © Bo Werner
ISBN: 978-91-7357-148-7
Cover: Conscripts’ examination
By Axel Petersson, ”Döderhultaren”, 1868–1925
Photograph of wooden sculpture,
By permission of Döderhultarmuseet, Oskarshamn, Sweden
Karolinska Institutet, Department of Public Health Sciences
Division of Social Medicine
SE-171 76 Stockholm, Sweden
Printed in Sweden by Tryckverkstan, Örebro, 2007
ABSTRACT
This thesis investigates the growth patterns of infants, children and adolescents in all
of Sweden and the secular change for height and weight for schoolchildren. This
thesis is based on the study of two cohorts: First, schoolchildren born 1973 from age
7 y to 18 y, and second children born in 1981 from birth to age 19 y. Cohort 1973
was all children born on the 15th of any month in 1973, 3 579 boys and girls.
Longitudinal data were collected from school health records. Missing cases were
4.5%. Cohort 1981 was all children born on the 15th of any month of 1981, 3 158
boys and girls. Longitudinal data were collected from child health records and
school health records. Missing cases were 1.6%.
Longitudinal data for somatic growth, cohort 1973 from 7 y to 18 y (height and
weight), cohort 1981 from birth to 19 y (height and weight), from birth to age 48 mo
(head circumference) from two nationally representative sample of children in
Sweden, collected from child health records and school health records, can be used for
epidemiological monitoring of growth with fewer missing individuals and at lower
costs compared with other dedicated studies. Data quality is comparable to similar
national surveys.
These studies represent, without selection bias, the current growth situation
among children and adolescents, enabling both epidemiological comparisons over time
and comparisons with other national surveys.
For the first time in Sweden, and without selection bias, means and distribution
of head circumference measurements are documented longitudinally for a nationally
representative sample of infants.
The time trend analyses revealed: An increased rate for those born in 1981
compared with those born in 1973 of relative weight reduction episodes was found for
both boys and girls. The increase for girls were most pronounced, started from a higher
rate and was seen in nearly all body weight categories and in all ages. For boys the
reductions increased for all body weight categories in the age interval 7–9 y; otherwise
the pattern was much more heterogeneous. Body weight and reduction of BMI were
highly correlated in both cohorts, as more of the overweight than the thinner children
reduced their BMI. For girls the increase in reduction rate between 1981-born and
1973-born was highest among the thinnest individuals.
From age 7 y to 18 y a strong positive secular change for BMI exists in all ages,
and the rate of overweight and obesity is increasing for both boys and girls.
Furthermore, obesity was growing more severe.
Keywords: adolescents, birthweight, body mass index, children, epidemiological
monitoring, final height, growth, head circumference, height, infants, length,
longitudinal study, national study, overweight, obesity, Sweden, secular change,
schoolchildren, time trend, weight, weight loss
LIST OF PUBLICATIONS
I
Werner B, Bodin L, Bremberg S. Data of height and weight from school health
records as a national public health surveillance tool – the case of Sweden.
Scand J Public Health 2006;34:406–13
II
Werner B, Bodin L. Growth from birth to age 19 years for children in Sweden
born in 1981 – descriptive values. Acta Paediatr 2006;95:600–13
III
Werner B, Bodin L. Head circumference from birth to age 48 months for
infants in Sweden. Acta Paediatr 2006;95:1601–7
IV
Werner B, Magnuson A, Bodin L. An increasing rate of weight loss among
schoolchildren, especially girls, in Sweden. J Adolesc Health 2007;40:238–44
V
Werner B, Bodin L. Obesity in Swedish schoolchildren is increasing in both
prevalence and severity. 2007 submitted.
All papers are reprinted with the permission of the respective copyright holders.
To my children, from birth to final height. Moa-Lisa -76: 50–167 cm,
Tove -77: 49.5–165 cm and Joel -82: 52–184 cm.
Margit, Signe, Miriam, Elisif, Dagmar.
From Henrik Berg: ”Om barnavård”, Stockholm, 1898
”Pygmies placed on the shoulders of giants see more than the giants
themselves.”
Robert Burton 1621
CONTENTS
1.
BACKGROUND.................................................................................................................. 7
1.1
1.2
1.3
1.4
1.5
2.
3.
Introduction............................................................................................................... 7
1.1.1
Growth – nature and nurture...................................................................... 7
1.1.2
The description of growth........................................................................ 10
History of human growth........................................................................................ 12
1.2.1
The height of man in pre-historical times................................................ 13
1.2.2
The stature of Neolithic and medieval populations................................. 14
1.2.3
What we learn from soldiers.................................................................... 15
1.2.4
The first longitudinal studies ................................................................... 18
Secular changes....................................................................................................... 19
1.3.1
Birthweight............................................................................................... 22
1.3.2
Final height............................................................................................... 23
1.3.3
Age at menarche....................................................................................... 24
1.3.4
Body proportions...................................................................................... 25
1.3.5
Time trend of overweight and obesity..................................................... 27
1.3.6
Dietary changes in Sweden...................................................................... 31
Growth and depending factors ............................................................................... 32
1.4.1
Growth during pregnancy ........................................................................ 33
1.4.2
Stature from birth to final height ............................................................. 39
1.4.3
Weight from birth to adulthood ............................................................... 50
1.4.4
Longitudinal aspects of growth and later health ..................................... 60
Growth surveys, geographical coverage and design.............................................. 62
1.5.1
Growth surveys in Sweden ...................................................................... 63
1.5.2
Growth surveys – international examples ............................................... 68
1.5.3
Internationally used reference values ...................................................... 70
AIMS .................................................................................................................................. 71
2.1
General aim............................................................................................................. 71
2.2
Specific aims........................................................................................................... 71
MATERIAL AND METHODS......................................................................................... 72
3.1
3.2
Target populations and study populations.............................................................. 72
3.1.1
Target populations.................................................................................... 72
3.1.2
Study populations..................................................................................... 72
Data collection ........................................................................................................ 75
3.2.1
Cohort 1973.............................................................................................. 76
3.2.2
Cohort 1981.............................................................................................. 76
1
3.3
3.4
4.
5.
Analysis................................................................................................................... 77
3.3.1
Paper I, II and III...................................................................................... 77
3.3.2
Paper IV ................................................................................................... 78
3.3.3
Paper V..................................................................................................... 79
Ethics....................................................................................................................... 79
Results................................................................................................................................. 80
4.1
Paper I ..................................................................................................................... 80
4.2
Paper II.................................................................................................................... 80
4.3
Paper III .................................................................................................................. 80
4.4
Paper IV .................................................................................................................. 80
4.5
Paper V.................................................................................................................... 81
General discussion.............................................................................................................. 82
5.1
5.2
Epidemiological monitoring................................................................................... 82
5.1.1
The representativity of the samples......................................................... 82
5.1.2
Selection bias ........................................................................................... 83
5.1.3
Data quality .............................................................................................. 84
5.1.4
Comparability........................................................................................... 84
Comparison to other surveys.................................................................................. 85
5.2.1
In Sweden................................................................................................. 85
6.
General conclusions ........................................................................................................... 90
7.
Considerations for further investigations........................................................................... 91
8.
Sammanfattning på svenska............................................................................................... 92
9.
Acknowledgements ............................................................................................................ 93
10. References .......................................................................................................................... 95
11. Papers I – V……………………………………………………………………….…....133
12. Appendix…………………………………………………………………………..…...189
2
ABBREVIATIONS, CONCEPTS AND
GLOSSARY
Adipositas rebound – from a rapid increase of body mass index, (BMI), the first years
of life the BMI subsequently declines and reaches a minimum by the age of about 6
years, before beginning a gradual increase up to the end of growth. The point of
minimum BMI has been called the adiposity rebound
Adolescent growth spurt – the rapid and intense increase in the rate of growth in
height and weight that occurs during the adolescent stage of the human cycle
Anthropometry – the science of measuring the human body (height, weight and
proportions)
Anorexia nervosa – a disorder that is characterized by low body weight, intense fear
of weight gain, and an inaccurate perception of body weight or shape
Anthropometric measures – body measures, for example height, weight, relative
weight and head circumference
Auxology – the science of physical and physiological growth of man (auxein=to grow,
to increase)
Benn index – an index that in a statistical sense is highly correlated with weight and
uncorrelated with height. Benn index is generally formulated weight/heightn, weight
divided by heightn, where n is a number calculated from the data to ensure zero
correlation between the index and height
Binge eating – is a recognized condition featuring episodic uncontrolled consumption,
without compensatory activities, such as vomiting or laxative abuse, to avert weight
gain
Body mass index, BMI – weight (kg) divided by the square of height. Higher BMI
scores indicate that an individual has relatively more weight-for-height than a person
with lower score. In the general population higher BMI scores usually indicate more
body fatness
Bulimia nervosa – a disorder that includes both binge eating, and compensatory
activities
Catch up growth – can be defined as a height velocity above the statistical limits of
normality for age and/or maturity during a defined period of time, following a transient
period of growth inhibition (recovery from disease or re-feeding after short-term
starvation)
Childhood phase – a stage in the human life cycle that occurs between the end of
infancy and the start of the juvenile phase
Cross-sectional study – measurement on a single occasion of individuals grouped by
age, sex, and sometimes other characteristics
Dedicated study – a study performed especially arranged with a controlled situation
when measuring according to method and age
Descriptive – how it (growth) is at a defined point of time or period
Distance curve – a graphic representation of the amount of growth achieved by an
individual over time
Diurnal variation of standing height – the variation that means that an individual is
tallest in the morning after sleep and shortest in the end of the day before sleep
Early maturer – an individual with early puberty and maturing
Eating disorders – includes anorexia nervosa, bulimia nervosa, and binge eating
3
Epidemiology – the study of the causes and transmission of disease
Failure to thrive – low rate of weight gain with no previously known disease
explaining the condition
Final height – when no more height can be achieved
Foetus – stage of prenatal development lasting from the tenth week following
conception to birth
Genotype – the genetic constitution of an individual
Growth – quantitative increase in size or mass
Growth phase – a part of growth of an individual that can be defined specifically
Growth standard see standard values
Head circumference – maximal fronto-occipital circumference or a Frankfurt plane
circumference, but both techniques have largely been replaced by simply measuring
the maximum head circumference
Heritability – an estimate of the relative genetic contribution to the phenotypic
expression of a physical or behavioural characteristic. The value of the heritability
estimate is a function of genetic factors, environmental factors and the interaction of
genetic and environmental factors
Heterosis – the mating of individuals born in different geographic regions, which may
lead to greater genetic variability and ”genetic vitality” in their offspring
Hominid – living human beings and their extinct, fossil ancestors
Infant phase or infancy – a stage in the life cycle of all mammals. For human beings it
lasts from the second month after birth to end of lactation, usually by age 36 months
Juvenile phase – a stage in the life cycle that is defined as the time of life when an
individual is no longer dependent on parents for survival, and prior to that individual’s
sexual maturity
Kurtosis – a statistical measure, indicating whether the distribution around the mean is
thick-tailed (a higher proportion of subjects is found at the extremes of the distribution,
kurtosis >0) or thin-tailed (the opposite case, kurtosis <0)
Late maturers – individuals that have late puberty and maturing. One of the
consequences of this is, for boys for example, sometimes growth up to age 25 years
LMS – a method by Cole to construct growth curves that summarizes the changing
distribution by three curves (L=lambda, M=mu and S=sigma) representing the median,
coefficient of variation and skewness, the latter expressed as a Box-Cox power. Using
penalized likelihood the three curves can be fitted as cubic splines by non-linear
regression, and the extent of smoothing required can be expressed in terms of
smoothing parameters or equivalent degrees of freedom
Longitudinal study – measurements of the same individual or group of individuals,
repeated over time
Low birthweight – a birthweight of 2 500 g or less for a neonate of normal gestation
length
Mean – after adding all variable-values and then dividing the sum with the number of
variable-values
Median value – the value that divides a distribution of data in two similar big parts. Is
the same as P50
Medical Birth Registry, MBR – a registry that since 1973 includes birth data for all
newborns in Sweden
Menarche – the first menstrual period
Migration – the movement of people from place to place
4
Mixed-longitudinal study – a study that combines a cross-sectional method with a
longitudinal. For example 2-y-olds, 4-y-olds and 6-y-olds are followed, longitudinally,
2 years and thus covering age-interval from 2 years to 8 years but it takes only 2 years
to perform
Military Service Conscript Registry, MSCR – a registry that includes, among other
things, growth data collected at military conscription in Sweden
Neolithic period – earlier stone age, follows Palaeolithic period, which ended 10 000
years ago
Neonatal period – from birth to age 28 days
Obesity – BMI >30 for adults or age-related limits suggested, for example, by Cole et
al.
Osteology – the branch of human anatomy that deals with the study of bones and the
skeleton
Osteologist – one who is skilled in osteology, an osteologer
Overweight – BMI >25 for adults or age-related limits suggested, for example, by
Cole et al.
P50 – the 50th percentile; it is the same as median value
Palaeolithic period – later stone age, precedes Neolithic period
Peak height velocity, PHV – the maximal height velocity of an individual achieved
during the adolescent growth spurt
Percentile of growth – a method of ranking growth status for height, weight, etc. of an
individual to other members of a sample or population of people. Example: a child at
the 75th percentile for height is taller than 75 percent of the other children in the group
under consideration
Phenotype – the physical or behavioural appearance of an individual, resulting from
the interaction of the genotype and the environment during growth and development
Ponderal index – weight divided by the cube of height
Prescriptive – how it (growth) ought to be
Prevalence – the total number of all individuals who have an attribute or disease at a
particular time (or during a particular period) divided by the population at risk of
having that attribute or disease at this point in time or midway through the period
Psychosocial short stature – a type of growth retardation produced by a negative
physical and emotional environment for growth
Puberty – an event of short duration (days or a few weeks) that marks the reactivation
of the central nervous system regulation of sexual development. Puberty occurs at the
end of the juvenile stage and is the period that starts with PHV and where menarche of
girls puts a relative upper limit for further growth. Usually a girl grows only 6–8 cm
after menarche, seldom more than 10 cm and just in extreme case 12 cm
Reference values – constructed values as reference to compare individuals with
Register study – a study based on data from a registry
Relative weight – see further: body mass index, ponderal index, Quetelet index, Benn
index
Standard deviation, SD – a measure of distribution
Secular change (secular trend) – the process that results in a change in the mean size
or shape of individuals of a population from one generation to the next
Sedente – a person living in her geographic region of birth; a non-migrating individual
Sexual dimorphism – differences between the sexes in physical appearance,
behavioural performance, and psychological characteristics
5
Skewness – a statistical measure that gives information on the deviation from
normality, within each interval. Should be close to zero in the case of an
approximately normal distribution within the interval
Socio-economic status, SES – an indicator, often defined by measures of occupation
and education of the parents or head of household, used as a proxy for the general
quality of the environment for growth and development of an individual
Standard values (growth standard) – normative values for growth
Quetelet index – see body mass index
6
1. BACKGROUND
1.1
INTRODUCTION
This thesis is aiming to be a part of the description of human somatic growth and to
explore why studies of growth mirrors the social and health situation in populations
or the state of public health in a society. There is a relation between somatic growth
and social conditions both on an individual and a population level.
This thesis is also aiming to be a part of the longitudinal descriptions of health
from birth to adulthood, the life course of the first part of life.
1.1.1 Growth – nature and nurture
To realize the biological potential
Some quotations about nature-nurture:
”Every character is both genetic and environmental in origin. Let us be quite
clear about this. Genotype determines the potentialities of an organism. Environment
determines which or how much of those potentialities shall be realised during
development” [Thoday 1965].
”The height of man is a response to both nature and nurture. If we leave genetic
differences aside changes in living conditions cause dramatic change in average
height. The better those conditions are, the taller the average citizen becomes.
”Although genes are important determinants of individual height, studies of
genetically similar and dissimilar populations under various conditions suggest that
differences in average height across most populations are largely attributable to
environmental factors.” [Steckel 1995].
Human beings are plastic during development, and a single genotype can
produce more than one alternative form of structure or physiological state in response
to environmental conditions [Bateson 2001]. In general, plants and animals have a
predetermined adult genetic size; this is referred to as their genotypic size. However,
observation shows that the population’s mean adult size of any species in the natural
environment always manifests a significant ”growth deficit”, demonstrating that no
living organism has yet been able to achieve its theoretical genetic size. The actual size
is called its phenotypic size. This difference between phenotype and genotype is
commonly referred to as nature-nurture difference: Genotype-Phenotype=Growth
deficit.
In fact, in human beings, the long slow historical reduction of this growth
deficit is scientifically known as the ”secular trend”. Growth deficit on a population
level is the secular trend, or more correctly the secular change, since ”trend” indicates
that it is a positive change, which is not always the case.
The environmental conditions decide
Even though many puzzles connected with human growth remain, the outlines of the
process of growth and its connections with environmental influences are quite clear.
7
Growth is a measure of nutritional status, of the synergistic relationship between
nutritional intake, work effort, and health. Malnourished and unhealthy populations are
short, while well-fed and healthy ones are tall. This fact is of enormous importance to
historians, because it means that height records offer a way of summing up many of
the different influences on human existence that we think of when we speak of the
standard of living. Growth statistics can complement measures of real wages in
discussing the standard of living within individual societies, and they are also a means
of comparing standards of living between different societies within the same genetic
group.
Somatic growth is a mirror of conditions in society [Tanner 1987, Bielicki
1986]. To grow from foetus into man is a trajectory in order to attain the biological
potential and a process of complicated interactions between protective and risk factors.
As Tanner says:
”Provided always that the parental size is known, growth emerges as a prime
measure of a child’s physical and mental health. The study of growth emerges also as a
powerful tool to monitoring the health and nutrition status of populations, especially in
ecological and economical circumstances that are sub-optimal. It can be used for
studying the effect of political organisation upon the relative welfare of various social,
cultural and ethnic groups that make up a modern state. Thus the study of growth has a
very direct bearing upon human welfare. At the same time it provides valuable lessons
on the way in which our biological heritage and our technological culture interact. It
warns us that all too soon we could be technically capable of creating monstrously
specialised or monstrously similar children. It points to our need to be reconciled with
our origins, to see ourselves once again as a part of the natural order; not foetus into
angel, nor foetus onto monster, but foetus into man.” [Tanner 1989 p.239]
Factors of importance
Growth and development result from the integrated work of four groups of factors: (1)
genetic, (2) paragenetic (genetic resonance of the foetus), (3) factors related to lifestyle (work, sleep, etc) and (4) exogenous factors. The latter should be understood as
the modifiers of development both natural and cultural (populational and intra-family)
[Wolanski 1988].
Children’s height is a sensitive indicator of their physical well-being, and has
been recommended as a measure of their nutritional state [WHO 1976]. WHO has in a
series of reports addressed this issue [Jeliffe 1966, WHO 1976, Waterlow 1977, WHO
1983, WHO 1986, WHO 1995]. In 1996 de Onis and Habicht stated that
anthropometry is the single most portable, universally applicable, inexpensive, and
non-invasive method available to assess the proportions, size and, composition of the
human body [de Onis 1996]. Associations between height and environmental
characteristics have been investigated as possible indications of adverse influences on
growth. In such investigations a number of variables including the parental social
class, and the employment, the family size [Bielicki 1986], the number of parents at
home [Garman 1982, Rona 1986], and the presence of overcrowding [Foster 1983,
Rona 1986] have been used to characterize social conditions during childhood [Lissau
1993]. A gradient associating height with social class has been documented [Bielicki
1986], but the existence of a gradient does not in itself indicate a direct influence on
growth.
8
It is necessary to adjust for confounding variables, but this raises a question:
what confounding variables should be allowed for in such analyses? More complicated
to analyse is the pattern of interactions between all the factors (variables) related to
growth as an outcome.
A proposal for health measures
On the subject of measures for health and public health, Sally Mcintyre says:
”There are a couple of problems with measures and measurements when
comparing health of different populations or sub-groups within populations. Mortality,
morbidity, health care utilisation, self-reported health have all their obvious limitations
and problems.” [Macintyre 1988].
She also sets up some criteria for useful measures:
* Dichotomous measures (present/absent) of rare phenomena are less useful than
measures which can be used to characterize every individual and be subdivided into a
number of categories, i.e. those which are continuously (and preferable nearly
normally) distributed.
* We should be seeking for measures that summarize characteristics with substantial
consequences for life-chances, not ones that are trivial.
* The measures should not be overly disease specific or organ specific but should have
general significance for health and longevity.
* The measures should display social patterning rather than randomly distributed
across countries, time and social groups.
* The measures should be acceptable to those on whom they are made, be practicable
in field settings, and be reliable.
She concludes by saying: ”Measures approximate to these criteria are height and body
mass index.” [Macintyre 1988].
Somatic growth indicates health and public health
Other authors, outside auxology, also argue that somatic growth is seen as an
important indicator, both on individual and on population level, for health and public
health [Floud 1990, Steckel 1995].
As Eveleth and Tanner put it: ”At present we know that factors such as
improved nutrition, control of infectious diseases, reduced family size, more
widespread health and medical care, as well as population mobility (both
geographically to urban areas and socially upwards) appear to be responsible for the
increase in growth observed over the past 100 years in industrialized countries and,
recently, in some developing ones. On the contrary, deprivations and poor living
conditions, stress, etc., could strongly affect the sensitive growing years in a negative
way, causing an absence of trend or even a negative tendency.
A child’s growth rate reflects, better than any other single index, his state of
health and nutrition; and often indeed his psychological situation also. Similarly, the
average value of children’s heights and weights reflect accurately the state of a
nation’s public health and the average nutritional status of its citizens, when allowance
is made for differences, if any, in genetic potential.” [Eveleth 1990].
9
Tall is not always best
Studies do not conclude that being tall is in itself a positive benefit. In old-fashioned
warfare being big probably meant being a better target. Neither is longevity a simple
consequence of increased stature. What the studies show is tallness being a product of
good living conditions in a society, meaning good general health, and therefore bodies
that on average are more likely to resist time’s ravages for a longer span of time.
Another statement is important to do, in this context, namely that bigger does not
necessarily mean better. In ”Foetus into Man” James Tanner cites from a study of
Frisancho and his co-workers from the slum outside Cuzco in the Andes in Peru, that it
is the small mothers who have the largest number surviving babies [Tanner 1989,
Frisancho 1980]. Small body size can be an advantage under certain circumstances.
Malcolm in New Guinea [Malcolm 1970] and Stini in Latin America [Stini 1975] have
come to the same conclusion. In an agricultural economy, a small man is more
effective than a tall one and therefore he needs to do less work to earn his living.
Genetics matters
That genetics can explain some of the differences in stature between populations are
illustrated for example by Bogin: The differences between the Netherlands, the US,
Africa Turkana and Japan can be the expression of differences between societies. But
the stature of Indians from Latin America and African pygmies has a more genetical
explanation [Bogin 1999a p.226]. The African Efe pygmies may be, on average, the
shortest people in the world, and their short stature appears to have a strong genetic
component [Rimoin 1968, Merimee 1981, Hattori 1996].
A group with Chinese background living in Kingston, Jamaica was studied. All
the children had from upper-middle to upper socio-economic status and attended
private fee-paying schools. There were no significant differences in height or weight
between the European, African and Afro-European groups. However, the Chinese
sample was significantly shorter and lighter than the other three groups at almost every
age, suggesting a hereditary difference in amount or rate of growth between the
Chinese children and the other samples of children. More recent surveys of the growth
of Chinese, Japanese, and other Asian children, adolescents, and adults find that they
on average remain shorter and lighter than European and African populations living in
the same cities or nations. However the differences in mean size have narrowed over
the past 30 years [Eveleth 1990]. This is too short a time for genetic change, and
indicates that other factors must be influencing the size of members of these
populations.
Differences between different populations with various genetic heredity are
described for example in ”World-wide variation in Human Growth” [Eveleth 1990]
and in ”Foetus into Man” [Tanner 1989].
1.1.2 The description of growth
”Human growth has always fascinated statisticians, judging by the number of
statistics books that use weight or height for examples. The fascination may be due
to not only the ready availability and statistically well-behaved nature of the data,
but also to the fact that growth is universal: everybody has experienced it.” [Cole
10
p.78 in Ulijaszek ed. 1994] The most obvious statistical construct used in the study
of human growth is the fiction that growth is a smooth process. Hermanussen et al.
has shown that on a sufficiently short time scale the process is anything but smooth,
proceeding in fits and starts over periods of weeks [Hermanussen 1988]. Equally, on
a scale of months there are important seasonal influences on growth. This is
particularly so in the developing countries, where growth rates can vary enormously
from one season to another [Maleta 2003], but even in the industrialized world there
is clear evidence of seasonality [Marschall 1971, Gelander 1994].
In the 1960s, Tanner, Whitehouse and Takaishi published several different
charts based on the heights and weights of British children. They distinguished
between charts of attained height and weight, and charts showing the growth in height
or weight over a period of a year or so; these two types of chart they termed distance
and velocity, respectively, by analogy with a moving object [Tanner 1966].
Reference centile curves are used widely in medical practice as a screening
tool. They identify subjects who are unusual, in the sense that their values of some
particular measurement, for example height or weight, lie in one or other tail of the
reference of the distribution. The need for centile curves, rather than a simple reference
range, arises when the measurement is strongly dependent on some covariate, often
age, so that the reference range changes with the covariate. The case for making the
centile curves smooth is to some extent cosmetic – the centiles are more pleasing to
the eye when smoothed appropriately – but there is also the underlying justification
that physiologically small changes in the covariate are likely to lead to continuous
changes in the measurement, so that the centiles ought to change smoothly. In such
cases, fitting discontinuous curves could lead to substantial bias.
Relative weight – a measure with complications
Height and weight is by definition clear. But relative weight is more complicated,
especially concerning growing people. The expression of relative weight ought to be
an index that in a statistical sense is highly correlated with weight and uncorrelated
with height. Benn index is of general form weight/heightn, weight divided by heightn,
where n is a number calculated from the data to ensure zero correlation between the
index and height [Benn 1971]. By testing this relationship you can see that n is about 3
for a newborn and then goes towards 2 at final height. A mean for n 0–18 y is 2.64
[Cole 1996, Bodin 2007].
The relationship between weight-height indices and the triceps skinfold
measure among children age 5–12 y was studied by Michielutte [Michielutte 1984]:
This study examines several weight for height indices – Quetelet’s index W/H2, W/H
and Ponderal index (Rohrer’s index) W/H3 – for their appropriateness in estimating
adiposity among young children. Data were obtained for a sample of 1 668 children
age 5–12 y residing in Forsyth County, North Carolina. Although W/H2 was found to
be the most useful of these indices, the results suggest that no index, including the
triceps skinfold measure, can be considered completely satisfactory in estimating
adiposity among children [Michielutte 1984].
A description is not a prescription
Since there is often not an obvious distinction, in the literature, between how growth is
and how it should be, it is very important to make the difference between these to
11
concepts quite clear. Descriptive values show the up-to-date situation for a population,
i.e. epidemiological monitoring. No exclusions are made and everybody belonging to
the population is included. From a description you can exclude sub-groups with
deviating growth patterns in order to get values that, for different purposes, can be
referred to in order to reveal deviating growth patterns. This you can call reference
values [Sullivan 1991]. If you want to state what is normal you have to have an idea of
what ”normal” is, and also what could count as desirable. If you succeed in this you
have standard values [Preece 1988]. Now you can compare the growth of an individual
or a population with the standard, and conclude if the comparison reveals a normal or
un-normal growth-situation for the individual or population of investigation.
1.2
HISTORY OF HUMAN GROWTH
The beginning of history
The earliest documentation of stature of man is from osteological material from our
ancestors who lived about 2 million years ago. The earliest written records about
human growth date from Mesopotamia, about 3500 BC. Both in Sumerian art and in
life there was a relationship between growth and biosocial conditions.
For the investigation of the historical stature of man there are three main types
of sources:
1. Archaeological, bone and skeleton material from pre-historical times to 18th century.
2. Historical data from registries or other written documentation, from 18th and 19th
century and onwards.
3. Surveys from the middle of 19th century.
Historians and growth
Historians have only recently become familiar with the notion that measurements of
height are relevant to the standards of living [Floud in Komlos 1994]. Early work of
Emmanuel Le Roy Ladurie in this field was sceptically received and was not followed
up [Le Roy Ladurie 1971]. By contrast, human biologists and physical anthropologists
have long believed that human growth is responsive to changes in the environment. An
early pioneer in the measurement of human growth, Adolphe Quetelet (1796–1874), a
Belgian whose work was widely known throughout Europe [Quetelet 1870], analysed
the growth of children and adolescents, as did L.R. Villerme’ in France at the same
time [Villerme’ 1829]. Their research inspired Edwin Chadwick in England to
organize surveys of the height of children in factory districts in the 1830s [Chadwick
1842]. Thereafter, the study of human growth or ”auxology” became a scientific
discipline of its own right, and has given rise to numerous surveys of growth
throughout the world [Eveleth 1990].
Fogel was not the first historian to make use of anthropometry, since Emanuel
le Roy Ladurie had used the records of French conscripts to describe the physical
characteristics of Frenchmen in the early nineteenth century [Fogel 1993]. But Fogel’s
use of anthropometric material, gathered by himself, Engerman and Coopersmith,
represented the first attempt to use such material – in particular the records of the
12
heights of recruits – for comparative purposes and in the study of economic history
and historical demography.
In the light of this knowledge, it is interesting that Karl Marx, presumably
aware of the work of Quetelet, Villerme, and Chadwick, used height data in his
discussion of the impact of industrialization. He referred to J. von Liebig’s description
of the declining health of the French and German populations as shown by the
condition and height of their military recruits, and argued that this decline in England
made factory legislation necessary [Marx 1954].
1.2.1 The height of man in pre-historical times
Stature of our ancestors – from Homo habilis to Homo sapiens sapiens
Height modifications in the course of human evolution can only be made apparent by
means of studying the osseous remains from our ancestors, a difficult task because of
the scarcity of data for earlier periods and the methods used for stature estimation.
In spite of the obvious difficulties represented by the methods for height
estimation and of the problems regarding anthropological remains [Trotter 1958,
Bennike 1985], at present there are enough data about the evolution of this remarkable
characteristic [Garralda 1991].
Homo habilis (2.2–1.8 million years ago) showed a high variability with heights
between 120 and 150 cm.
By Homo erectus (1.8 million–300 000 years ago) for the first time a clear
increase of stature is observed, the variation ranging globally, between 150 and 180
cm. H. erectus show much geographical variation in general appearance and size, since
they were the first humans to occupy and adapt to not only the tropical or sub-tropical
areas but also the coldest climates of Northern Eurasia.
In relation to the data that exist, H. erectus reflects the ”actual human range”
regarding stature. The chronological successors, and probably genetic relatives, the
earlier forms of H. sapiens, furnishing measurements for height estimation, show
similar variation; this is the case for H. sapiens sapiens.
With Homo sapiens sapiens (90 000 years ago) living in the Near East, males
varied from 179 cm to 186 cm and females from 165 cm to 171.5 cm, and for those
living in Europe during the later Palaeolithic age (40 000–10 000 years ago) males
varied from 160.2 cm to 189.0 cm and females from 157.9 cm to 164.6 cm [Garralda
1991].
This high variability (sexual, populational) will remain throughout time; the
general growth pattern of these ancestors are unknown although we can suppose that
long infancy period and pubertal spurt existed; we can suppose that the variability
answers partially to an ecosensitivity that was at least as great as nowadays, and
partially to differences in genetic pool; in fact the higher stature of H. erectus was also
accompanied by brain increase allowing remarkable cultural development, brain sizes
between 850 to 900 cm, better capacity of protection and hunting techniques; for
more recent periods the available information confirms, at the same time, the
ecosensitivity as well as the individual and populational genetic variation.
For details of cranial variation revealed by skeletal collection from Cambridge
see further Lahr [Lahr 1996].
13
1.2.2 The stature of Neolithic and medieval populations
Man settles down
Neolithic age is a prehistorical period that followed the Palaeolithic age, which ended
about 10 000 years ago. The Neolithic age is characterized by people beginning to
settle down, starting to cultivate the land and gathered their livings in villages.
Archaeologists and osteological specialists who have scrutinised skeletons
from those times have been able to show that there existed surprisingly tall individuals
as far back as the hunting-gathering period. They have also argued that these people
lived in a comparatively sound surrounding, enjoying a diet of satisfactory
composition [Åkerman 1988, Bogin 1999].
An overview made by Bennike, osteologist and archaeologist, from
archaeological populations shows findings on mean height for males and females, see
Table 1 [Bennike 1985].
Table 1. An approximation, from osteological material, of stature in historical
times. Means for height in cm.
Period
Male
Female
10000–4000 BC
4000–2300 BC
2300–1800 BC
1800–500 BC
0–400 AD
800–1050 AD
1050–1500 AD
1850 AD
160
165
177
173
176
170
172
165
156
153
166
167
167
159
162
154
If the terracotta warriors of the Qin dynasty army from 2 200 years ago in Xian, China
do represent a selected part of the population and if the heights (169 cm to 196 cm) of
the terracotta warriors reflect the heights of human warriors at that time we have here a
selected sample that mirrors current stature 200 BC.
The medieval situation
”The implication will be that the transition from hunting to agriculture, from mobility
to stable settlements, turned out to be more problematic. This fact has been revealed by
skeleton material. This means that mankind had to pay a rather high price for the
improved security of the agricultural society. As a consequence of a higher density of
population we can register a series of epidemic diseases which have tormented rural as
well as urban populations ever since. Through history peasants have a height 15
centimetres less than hunters-gatherers. In Sweden the height of man seems to have
been rather stable from the Neolithic age to the medieval period” [Åkerman 1988]. But
14
between medieval populations stature can vary because of locally different social and
nutritional conditions [Werdelin 1985].
However, there are exceptions to short height in the Middle Ages in Sweden:
from male peasants living 1361 outside the city of Visby we know from a huge bone
material that the average height among the peasants was almost 169 centimetres
[Hultcrantz 1927]. Around 1900 the average height of men (conscripts) in Sweden just
passed 170 cm.
Out of 3 305 skeletons from medieval Lund (at that time belonging to
Denmark), between 990 AD and 1536 AD, means for height varied from 160 cm to
163 cm for women, and 171 cm to 175 cm for men [Arcini 1999].
From the village of Holje in Blekinge an archaeological example, from around
1700, where the main part of the village inhabitants died during the last epidemic of
plague in Sweden (1710 and 1711) illustrates both height as a measure of living
conditions and the different vulnerability of men and women: ”Both men and women
had an average height that corresponded to the height of the medieval population at
that time, that is 159.8 cm for women and 170.6 cm for men […] the men, age
between 20 y and 40 y when they died, most of them born between 1680 and 1690,
had an average height of 166 cm while the men, age between 40 y and 60+y had an
average height of 171 cm. For the women the corresponding difference was 159 cm,
age 20 y to 40 y, and 162 cm in age interval 40 y to 60+ y […] these data agree with
the periods of famine we know existed in the 1680s and 1690s” [Arcini 2006]. So, in
bad times men seem to be more vulnerable than women, as the average height
decreased 5 cm for men compared to only 3 cm for women.
1.2.3 What we learn from soldiers
The height of soldiers makes history
During the last 100 years, the pattern for human somatic growth has changed in most
countries. The somatic maturing has been achieved at an earlier age, while the adult
somatic size has increased – an example of the secular change. The ideal method to
conclude secular changes is to compare a series of repeated, representative samples
from defined populations. These kinds of samples are only available for young men,
from military conscription, from the middle of 19th century, and to some extent for
slaves in North America [Steckel 1987].
Heights of recruits were recorded in Norway from 1741, and soon afterwards in
England from 1755 and in Sweden from 1761. Later, it was common for conscription
to be based on regular inspections, of very large sections of the young male
population. Thus, records of height, health, and other physical and social
characteristics (such as occupation and place of birth) for entire populations of young
men have survived.
In Sweden, there is height information gathered since 1776 among soldiercrofter-population that allows us to make reconstructions populations born as early as
in the 1740s. Aggregated national statistics about the conscripts can be found from
1840 and onwards, but we have been able to recover local material from certain areas
from 1813 [Åkerman 1988]. The ordinary military personnel were on average taller
than the conscripts, and they were not included among the conscripts in the muster
rolls, but instead measured in a special procedure.
15
In the 20th century the enlistment of conscripts took place already at the age of 20 y,
instead of 21 as was the case in the 19th century (See Table 2); therefore we have to
make special arrangement to combine the older material with the modern. It must also
be said that during the period 1813–60, persons who were rejected because of illness
or physical handicaps are not included in the measurements. After 1860 we have the
same information about them as about the accepted individuals. In the very beginning
there were not many draft dodgers, but their number increased and reached the level of
6 000–8 000 persons out of about 40 000 every year in the middle of the 1870s. We
know that a substantial part of these draft dodgers immigrated to North America and
that this contingent tends to be taller than the average population [Åkerman 1988]. All
this has to be considered when we make comparisons over time.
1840 we have conscription at age 21 y, and the mean height was 165.1 cm, and
1926 when the conscription was at age 20 y the mean height had increased to 172.5
cm. In Anthropologia Suecia the mean heights are described for all the counties around
1800. The national mean height was at this time 170.9 cm for conscripts [Retzius
1902]. Mean height for Swedish conscripts from 1841–45 (21 y) to 1952 (19 y) has
increased with 8.4 cm, see Table 2.
16
Table 2. Mean height, in cm, at time for conscription for men in Sweden from
1841–45 to 1952. (1953 to 2004 see Table 4 page 17)
Age (y) for
conscripts
21
21
21
21
21
21
21
21
21
21
21
21–20
20
20
20
20
20
20
20
20
20
20
20–19
19
19
19
Year of
conscription
1841–45
1846–50
1851–55
1856–60
1861–65
1866–70
1887–90
1891–95
1896–1900
1901–05
1906–10
1911–15
1916–20
1921–25
1926–30
1931–35
1936–40
1941–44
1945
1946
1947
1948
1949
1950
1951
1952
Height
167.4
167.4
167.8
168.1
168.5
169.6
169.2
169.6
170.1
170.8
171.6
172.0
171.7
172.1
173.7
173.2
174.2
174.5
174.7
174.9
174.8
174.9
175.0
175.3
175.6
175.8
Source: MSCR
Height differs between social groups (students, peasants, crofters, workers and farm
labourers) and the differences are more pronounced than geographical differences or
differences between the age cohorts over time. Students were two generations before
the mean population as regards to the bodily height development. The mean height of
students during the years 1873–74 was 173.2 cm. It was not until the 1930s that this
was the mean height of the population in Sweden. Gunnar Sandberg and Herbert
Steckel have used more fragmentary information about soldiers-crofters treating the
whole period between 1750 and 1850 [Sandberg 1980].
Measuring of conscripts on a large scale was done in the beginning of the
1880s for example in Baden, Germany, by Ammon [Ammon 1894] and in Italy by
Livi [Livi 1896].
Comparisons from values at military conscriptions over time have one
complication. Mostly conscripts had not reached their final height at examination. The
age for final height has continuously been getting lower. A hundred years ago final
17
height was reached at age 25 y, and today it will be reached below the age of 20 y
except for late maturers [Hägg 1991]. See further 1.3.2
1.2.4 The first longitudinal studies
A son and some students were the first to be studied longitudinally
The Count du Montbeillard of France measured the stature of his son every sixth
months from the boy’s birth in 1759 to his eighteenth birthday. These data are
usually considered to constitute the first longitudinal study of human growth. This
was thus the first example of a distance curve of growth [Buffon 1777].
Another eighteenth century longitudinal study of growth is that of the
students of the Carlschule, conducted between the years 1772 and 1794. The growth
data showed that the sons of the nobility were, on average, taller than the sons of the
bourgeoisie during the growing years, but that both groups achieved an
approximately equal height at 21 y of age. Thus, the sons of the nobility experienced
an advancement of the rate of growth [Komlos 1992].
In 1835, Lambert Adolphe Quetelet (1796–1874) published the first statistically
complete study of the growth in height and weight of children. Quetelet was the first
researcher to make use of the concept of the ”normal curve” (today commonly called
the normal distribution or ”bell-shaped” curve) to describe the distribution of his
growth measurements, and he also emphasized the importance of measuring samples
of children, rather than individuals, to assess normal variation of growth. The
18
statistical approach of Quetelet was followed in Europe by Luigi Pagliani (1847–
1932). He began his studies on the size and fitness of Italian military personnel. He
later applied his methods to children and demonstrated that the growth status of
orphaned and abandoned boys, ages 10 to 19 y, improved after they were given care at
a state-run agricultural colony. Pagliani also noted that children from the higher social
classes were taller, and heavier, than poverty-stricken children. Finally, he was taking
longitudinal measurements of the same children. From these Pagliani noted that
menarche almost always followed the peak of the rapid increase in growth that takes
place during puberty [Pagliani 1876].
The early, big studies were cross-sectional
Villerme’ [Villerme’ 1829], Chadwick [Chadwick 1842] and Bowditch [Bowditch
1879], and from Scandinavia Wretlind [Wretlind 1878], Vahl [Vahl 1874–83], Key
[Key 1885], Malling-Hansen [Malling-Hansen 1883], and further Schmidt-Monnard
[Schmidt-Monnard 1895], Boas [Boas 1892], Camerer [Camerer 1893], and Dovertie
[Dovertie 1895] all made surveys on the growth of infants, children or adolescents and
were the pioneers in this field. Bowditch was the first to perform mass investigations
of the development of growth in different ages, and in Boston 13 691 boys and 10 904
girls were measured (height and weight) in different schools.
Boas’ scientific discoveries also include his research into the methodology of
growth studies [Boas 1892, Boas 1930]. Among other things Boas provided the
concept of tempo of growth to understand the difference between early and late
maturing individuals.
1.3
SECULAR CHANGES
To follow changes in populations
The word ”secular” has two meanings: worldly, especially pertaining to the material,
non-spiritual world, and just once in an age, indicating a relatively long span of time
[Bogin 1999 p.244]. The process is aptly named because the factors influencing the
secular trend are related to the material conditions of life, and these conditions do act
on human growth over long spans of time.
History reveals that the height of man, over time, both has increased and decreased.
For example, 8 000 years ago the mean height in Latin America was higher than the
relatively recently noted bottom value for mean height in the 1940s. The difference
was about 7 centimetres [Bogin 1999]. In the beginning of the 19th century, body
height was decreasing in many European countries, linked to the effects of the
industrial revolution [Floud 1990].
Secular changes take place not only within one country or region, but also when
people move between places. For example, it is well known that a positive secular
change is associated with migration from a low socio-economic status (SES) to a
higher SES environment or socio-economic improvement [Garn 1987, Bogin 1988]. A
few classic examples of the secular change in the growth of migrant children include
the work of Boas [Boas 1912, Boas 1940] with European immigrants to the United
States, the work of Shapiro [Shapiro 1939] with Japanese immigrants to Hawaii, the
work of Goldstein [Goldstein 1943] and Lasker [Lasker 1952] with Mexican
19
immigrants to the United States, and the studies by Greulich [Greulich 1976] of
American-born children of Japanese descent. Follow-up studies of these same
populations show that, with time, the growth in height of each generation of the
children of migrants continues to increase until it converges on that of the host
population [Roche 1979].
Investigations from Norway [Udjus 1964] and Japan [Tanner 1982] have
shown that the secular trend of body height, which has been ongoing during the 20th
century, has been more pronounced for the legs (length of the legs) than for the trunk
(sitting height). These changes in the proportions of the body ought to have happened
earlier and is a source of error when estimating body height from historical findings of
skeletons.
Secular changes between 1880 and 2000 are reported from Poznan, Poland for
body height and weight in children and adolescents. The results of measurements
obtained in eight subsequent surveys, from 1880 to 2000, were included in the
analysis. In general, in the 20th century, children grew taller and heavier and reached
final body height and weight more rapidly. The biggest differences in body height and
weight in the 20th century, observed at growth spurt, were about 17 cm and 11 kg for
boys, and 13 cm and 13 kg for girls. The magnitude of secular changes in body height
and weight in the 20th century was not stable. There were periods of increased and
decreased intensity of acceleration of physical development (the 1950s and 1970s, and
the 1960s and 1980s, respectively), as well as a period of deceleration (the 1940s). In
the last decade, the tendency has been towards deceleration in most age groups studied
[Krawczynski 2003]. From Zagreb schoolchildren Prebeg reports from secular
changes 1951–91 [Prebeg 1995].
But why does the secular trend proceed in some countries but stops in other?
To what degree is the secular trend inter-generational; do parents transfer their better
health and stature to their children and if so, in what way? Are trends in body-height,
weight and tempo synchronized to each other; do they increase or decrease together?
Overviews of secular trend have been done and broaden the perspective [Roche
1979, van Wieringen 1986, Tanner 1987, Hauspie 1997].
Soldiers tell us about secular changes
Information about adult stature in the past is mainly based on data about slaves or on
data from conscripts. In some countries, at conscription, the shortest were excluded
and thus increased the mean of the examined.
There are data without exclusions in the Netherlands, from 1851 to 1983 [van
Wieringen 1986]. For example, a height of 170 cm, in the 76th centile, in 1863 had
fallen to the 5th centile in 1983. A height of 180 cm in the 98th centile was in the 43rd
centile 120 years later. Besides a minor year-to-year fluctuation and a small decrease
around 1890, it was an increasing tendency during the period. There are still positive
trends in the 1880s and in the1890s [Gerver 1994, Fredriks 2000].
From Italian conscript data a decelerating height is seen. Between 1896 and
1900 mean height decreased with 1.9 cm but this is an exception to a continuously
positive trend during 1854–1963 [Hermanussen 1995].
Recent data for 18-y-old men at conscription show an increasing trend in
mean height between 1960 and 1990 in eleven countries [Schmidt 1995]. The Dutch
are tallest (181 cm 1990) and the Portuguese are shortest (170 cm). The trend seems
to be stronger in the shortest groups since the change has been 2.4 cm/decade in
20
Spain against 0.9 cm in Norway. The tall populations in north Europe show signs of
decelerating trend during this period. Based on surveys of populations, changes of
1.0 cm are typical for Western Europe in last years while Eastern Europe and Japan
have increased with 3.0 cm/decade [Hauspie 1997]. Based in conscription data from
different countries, Larnkjaer states: Secular change in adult stature has come to a
halt in northern Europe and Italy [Larnkjaer 2006].
Except conscript data there are not much data of adult height, especially not
for women from cohorts born before the 1940s. From the 1946 and 1958 cohorts in
the UK, data were collected for all, including their parents. In this way data of adult
height were collected for people born between 1892 and 1958. For men the tendency
was 10.9 mm/decade while for women it was 3.6 mm/decade [Kuh 1991, Wadworth
1997].
Sex-dimorphism, something to consider
The sex-dimorphism in height-tendency, noticed in the study of the 1946 cohort, is
striking and has been noticed historically elsewhere [Arcini 2006]. It was bigger
before 1940 than after. Among parents born before 1905 the fathers were 6.9% taller
than the mothers, and in the 1958-cohort men were 9.3% taller than women. The
authors pointed out that, in general, boys’ growth is more affected by the environment,
it is more plastic than girls’, so in good times boys accelerate more and in bad times
they decelerate [Kuh 1991]. Eveleth and Tanner believe the same [Eveleth 1990].
Anyhow, Kuh et al. do not find the proof for more plasticity for boys convincing and
look for other explanations [Kuh 1991].
What is causing the sexual dimorphism in height tendency puts the light on an
interesting contradiction. In his famous article about ”regression to the mean”, Francis
Galton showed that hereditary height made an 8% sex-difference in height when
adjusting for the sex-difference of the descendants of the family [Galton 1886], and the
same differences are seen in many populations today [Cole 2000]. But Cole like Kuh
found that the percentage difference increased by time. Men have either kept on being
8%, or more, taller than women the last 110 years [Cole 2000, Kuh 1991].
Secular changes at different ages
The size of the secular change of children varies with age. Dutch children were at age
1 y shorter in 1997 than in 1965, while at age 8 y respectively 21 y of age they were
considerably taller. In Japan, data were published from birth to age 17 y for children
between 1940 and 1990, a period of time when dramatic changes in every age took
place [Takaishi 1994, Takaishi 1995]. Changes between 1950 and 1990 show small
effects both for boys and girls before the age of 2 y (about 10 mm/decade at 2 y), just
like the Dutch children. After that the tendency increases with a maximum at 11 y for
girls (about 30 mm/decade) and 14 years for boys (about 35 mm/decade) and return to
values as adult (10 mm/decade girls, 15 mm/decade boys) like those by the age of 2 y
[Cole 2000 p.319]. For boys the curve is still falling at 17 y and probably return (there
are data just to 17 y) to the values at 2 y at the age of 20 y.
Consequently, the secular change in childhood seems to be divided in 3 distinct
periods: before 2 y with a small tendency; from 2 y to puberty where the increasing
tendency is clearly related to the pubertal growth spurt; post-puberty where the
tendency decrease and finish like that of the adult. So, when the pubertal spurt is over,
21
the difference of adult height is the same as was already determined at the age of 2
years. These findings are confirmed by others [Bock 1989, Edlund 1999].
To reveal changes, comparable data are needed
The difficulty to catch secular changes in a valid way increases when using data which
is neither representative nor comparable. Questions: Are measurements taken at the
same age, and are the measures of good quality? Is there selection bias? But, we are
often forced to do comparisons without straight comparability. This is something that
has to be taken into consideration. In Sweden there are some unique possibilities, to
use the Medical Birth Registry and data from conscription, and also to collect
representative samples with few missing cases [Cnattingius 1990].
1.3.1 Birthweight
Birthweight is a central public health measure. The birthweight is the result of the
intra-uterine growth and it is a starting point from which the life course in some
respects is defined by. I developing countries the mean and distribution of birthweight
points out in what direction the development of the country is going.
There is a positive secular change in birthweight in developed countries,
including the UK [Alberman 1991] and Sweden. For Sweden, Meeuwisse suggests
that if increasing size at birth is positively correlated with fatness in adulthood, this
partly underlie the trend of increasing obesity [Meeuwisse 1998].
Looking at the most recent situation in Sweden and the development over the
last 30 years, see Table 3, the picture is like this:
Table 3. Percentage of infants in different birthweight intervals 1973–2004 in
Sweden
Period
1973–
1979
Birthweight
in gram
3 500–4 499 47.0
4 500–5 000 2.4
>5 000
0.3
>4 500
2.7
Source: [MBR]
1974
2.7
1980–
1989
48.3
2.7
0.3
3.0
1984
3.1
1990–
1999
50.0
3.2
0.4
3.6
1994
4.0
2000–
2004
50.4
3.6
0.6
4.2
2004
4.1
The number of infants with birthweight >5000 g have doubled in 20 years.
From 1992 to 2001, mean BMI for newly registered pregnant women, at the prenatal
clinics in Sweden, increased from 23.4 to 24.5. During the same period the rate of
smoking pregnant women decreased and the rate of maternal diabetes increased.
22
1.3.2 Final height
At what age we reach final height has varied through time. Udjus showed that the
average age for completed growth in the Norwegian male population was 29 y in
1780, 25 y in 1860 and 20 y in 1960 [Udjus 1964]. In the Polish male population born
in 1953 the average increment in stature was 2–3 cm between 19–27 years of age
[Hulanicka 1983]. Data from the Solna-study by Taranger et al. show that the mean
height raises up until age 20 y for boys and up till 18 y for girls [Hägg 1991].
For 18-y-old men in Sweden (see Table 4) we can see that from 1953 to 1962
mean height increased with 1.7 cm, and from 1962 to 2004 with 2.8 cm but from 1994
to 2004 just an increase with 0.7 cm, thus a decelerating increase over time. The
development that remains is perhaps that we reach final height at an earlier age. It is
possible to predict a future development for a population when looking at the growth
of very healthy and social privileged groups.
In the Netherlands, data from the repeated national surveys show that the
growing population is extremely tall in all ages, both boys and girls, and the secular
change is developing in another way as in for example Scandinavia [Fredriks 2000].
But the exclusions made in the surveyed Dutch population and a huge number of
missing cases, especially among older teenagers, make comparisons with other
national data hard to evaluate. Exclusion is made for those who is not born in the
country or for those having parents not born in the Netherlands.
From the MSCR we get conscription data from 1953 to 2004 in Table 4:
Table 4. Height (cm) and BMI (kg/m), means for men at conscription from 1953
to 2004 in Sweden (data from 1841 to 1952 see Table 2 page 11)
Age (y) at
conscription
Year of
conscription
Height
BMI
19
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
18
1953
1955
1957
1958
1960
1962
1966
1970/71
1974
1978
1982
1986
1990
1994
1998
2002
2004
175.7
175.7
175.8
175.9
177.0
177.4
177.7
178.3
178.4
179.0
179.1
179.1
179.4
179.5
179.8
180.3
180.2
21.4
20.8
20.9
20.9
20.9
20.9
20.9
21.0
21.3
21.7
21.7
21.9
22.0
22.3
22.5
22.9
23.0
Source: [MSCR]
23
Conscription took place at age 19 y in 1953–1954, but after that at age 18 y. We can
here see that mean height for 18-y-old men has increased by 4.5 cm in 50 years, and
mean BMI has increased with 2.2 units.
Just like increasing positive secular change of height, an earlier age of final
height is a public health measure. Unfortunately, very few investigations of growth
have data that cover the age intervals when final height is achieved: on a population
level, for females somewhere between 16 y and 20 y and for males somewhere
between 18 y and 23 y.
1.3.3 Age at menarche
The decreasing age-trend is decelerating
Age for menarche has decreased sharply during the last 150 years. In for example
Norway [Liestöl 1982] and Denmark [Helm 1987] it has decreased with a speed of 12
months/decade. Menarche around age 13 y was achieved for girls born during World
War II in Belgium [Wellens 1990], around 1946 in England [Roberts 1994], between
1953 and 1970 in Denmark [Helm 1998], and around 1965 in Poland [Hulanicka
1991]. In Belgium between 1919 and 1967 dropped the 90th percentile of age at
menarche faster than the median value while the 10th percentile hardly changed,
showing that late menarche has been less frequent during this period [Hauspie 1997].
Thus, over the last 30 years, the age at menarche has not changed though the positive
secular change for height has kept on. Hulanicka felt that the turning point for age at
menarche depends on a decline in standard of living in Poland after 1978 [Hulanicka
1991], but it might instead be a part of the more general pattern of stabilizing age at
menarche over all of Europe after World War II [Cole 2000].
Relation to social factors
Menarche is achieved at about 12.5 years of age for girls of the middle-class from
many nations, and at 14 years of age or later for girls of the lower socio-economic
classes [Johnston 1974, Eveleth 1990]. The latest median age of menarche on record is
18 years of age for girls from the Bundi tribe of highland New Guinea [Malcolm
1970]. Malnutrition, heavy labour, infectious disease, and living at high altitude are
some of the reasons for delayed maturation of this group. Since the 1970s, the living
conditions of the Bundi have improved, especially in relation to nutrition and medical
care, and the median age of menarche has declined [Worthman 1993].
Physical and physiological stress influences growth and development,
including menarche and menstruation [Ruble 1982]. There are many reports that
highly competitive female track athletes, who enter training before puberty, reach
menarche later than girls in the general population [Malina 1973, Frisch 1980,
Merzenich 1993].
Birthweight has decreased, stayed unchanged or increased, not related to height
and development of menarche. The lack of synchronizing for height, birthweight and
menarche trends tells us that there are different explanations to these phenomena [Cole
2000].
24
Relations to genetics and body-weight
A genetical linkage for menarcheal age is reported by Moisan, who argues that the
most convincing relation that exists is between the menarcheal ages for daughter and
mother [Moisan 1990]. Two other factors are identified to have relevance for age of
menarche: First, the size of a child during childhood. An analysis of 4 000 schoolgirls
in Oslo related menarche to age, height and weight, where weight was shown to have a
strong relation while height failed to have a significant relation [Liestöl 1995]. Many
authors have shown that BMI at age 7 y or BMI before menarche was strongly
predictive for later age at menarche [Stark 1989, Wellens 1992, St. George 1994,
Cooper 1996, Power 1997]. These observations make it likely with a causal relation
between weight and menarche, the so-called ”critical weight-hypothesis” proposed by
Frisch et al. [Frisch 1970]. But Ellison did not succeed in finding any evidence of a
critical weight or obesity at time for menarche. Instead he found that the velocity of
gaining height was a far stronger predictor than weight, showing that the time for
menarche is decided by the degree of maturing, not by body size. [Ellison 1982]. Twin
studies too are showing that the genetic control of sexual maturing is independent of
the genetic system that decides body size [Loesch 1995]. The ”critical weighthypothesis” is now generally called into question, but in spite of that the has been
arisen again and is now linked to the hormone leptin [Clayton 1997]. Furthermore,
lack of linkage was found between height and weight and age at menarche during the
secular shift in growth of Japanese children [Tsuzaki 1989].
Cooper et al. studied the influence of birthweight, and weight and height at age
seven years, on menarcheal age in a national sample of 1 471 girls in England,
Scotland and Wales. They found that girls who were heavier at age seven years had
menarche at an earlier age. The average age of those in the highest fifth of the
distribution of weight at seven years was 7.3 months less than that of those in the
lowest fifth of the distribution. In contrast, girls who were heavier at birth had
menarche at a later age. The extreme fifth’s differed 2.2 months for age at menarche.
They conclude that the observations are consistent with the hypothesis that menarcheal
age is linked to programmed patterns of gonadotrophin release established in utero,
when the foetal hypothalamus is imprinted, and is subsequently modified by weight
gain in childhood [Cooper 1996]. The other factor that affects age of menarche is an
environmental factor linked till development of society is shown by Liestöl by
studying hospital records from 1860 to 1950 in Norway [Liestöl 1982].
Time for menarche is closely related to maturing and tempo of growth. Early
Tanner revealed that menarche appears about 1 year after PHV [Tanner 1962].
1.3.4 Body proportions
Height does not vary as body proportions do
Comparisons of stature and body proportion between blacks and whites in the United
States provide another example of gene-environment interactions and their affect on
growth. Fulwood et al. published data from the first National Health and Nutrition
Examination Survey of the United States, which gathered anthropometric data on a
nationally representative sample of blacks and white aged 18 to 74 years. When the
data are adjusted for differences between the two ethnic groups in income and
25
education, urban or rural residence, and age, there are no significant differences in
average height between black and white men. Nor is there a significant difference in
average height between black and white women [Fulwood 1981]. Despite this, the
body proportions for the two groups are different. Krogman found that for the same
height, blacks living in one American city had shorter trunks and longer extremities
than whites, especially the lower leg and forearm [Krogman 1970]. Hamill et al. found
that this was also true for a national sample of black and white youths 12 to 17 y old
[Hamill 1973].
Examples of differences in body proportion between populations
Differences in body proportion are known from other populations. Eveleth & Tanner
surveyed studies of boys and girls of European (London), African (Ibadan), Asian
(Hong Kong) and Australian Aborigine origin. In proportion to sitting height, the
Australians had the longest legs, followed in order, by Africans, Europeans, and
Asians. Expressed quantitatively: at a sitting height of 60 cm, for example, London
boys have leg lengths averaging 43 cm, Ibadan boys 53 cm and Australian
Aborigine boys 61 cm [Eveleth 1976 p.229]. Though unclear, many researchers
have assumed that the body proportion differences are explainable only in terms of a
genetic model.
Nurture matters
On the other hand, work on the growth of Japanese children suggests that
environmental factors also may be powerful determinants of body proportion. Kondo
& Eto found that between the years 1950 to 1970 the ratio of sitting height to leg
length decreased for Japanese schoolchildren, meaning that the children became
relatively longer-legged with time [Kondo 1975]. Tanner et al. confirmed this finding
by comparing both the rate of growth and the amount of growth for Japanese school
children measured in 1957, 1967 and 1977 [Tanner 1982]. Each successive cohort of
children grew faster, and grew larger, than the previous cohort. Between 1957 and
1977, sitting height showed practically no increase, while increased leg length
accounted for almost all the difference in height (4.3 cm for boys and 2.7 cm for girls).
In 1977, adult Japanese had sitting height to leg length proportions similar to Northern
Europeans, whereas 20 years earlier, the two populations were significantly different.
The major influences on growth that that changed during the past two decades are
improvement in nutrition (especially greater intakes of both protein and energy), health
care, and sanitation [Kimura 1984, Takahashi 1984] and all this mirrors the great
economic and social improvement in Japan the last decades [Marmot 1989].
Similar findings on the plasticity of body proportions are reported by several
researchers who are working in Argentina, Poland and in Mexico [Bolzan 1993,
Wolanski 1979, Dickinson 1990, Wolanski 1993, Siniarska 1995].
After 1977, the average height of Japanese men and women continued to
increase, but at a slower rate [Takaishi 1995], and both leg length and sitting height
seem to have increased at the same rate, at least for young women [Hojo 1981]. Since
1990 there is little evidence for further increase in stature. This means that the body
proportions of Japanese and Northern Europeans remain similar, even though the two
populations differ in mean stature [Bogin 1999 p.243]. It is difficult to establish a
26
genetic difference when comparing populations that live on different continents, since
important environmental variations that can affect growth are likely to exist.
Another approach to the study of population differences in growth is to
compare children and adults of different national or geographic backgrounds living in
the same, or very similar, environments. Ashcroft et al. measured the heights and
weights of 4- to 17-y-old children and adolescents of European, African, AfroEuropean descent [Ashcroft 1964, Ashcroft 1966].
1.3.5 Time trend of overweight and obesity
Despite difficulties with measurements and comparability, the trend is obvious
While measurement inconsistencies make it difficult to provide an overview of global
prevalence of overweight and obesity in children, studies generally report high and
increasing prevalence of overweight and obesity worldwide [WHO 1997, Livingstone
2001, Lobstein 2004]. Current estimates (1998) of childhood obesity prevalence in the
USA suggest that 21.5% of African-Americans, 21.8% of Hispanics and 12.3% of
non-Hispanic whites are overweight, and overweight increased rapidly between 1986
and 1998 [Strauss 2001]. In Australia, most recent estimates, for the ages 7–12 y,
suggest that 16.1–16.9% of boys are overweight and 5.1–6.9% are obese, and for girls
17.4–20.4% are overweight, and 5.7–7.0% are obese [Lazarus 2000]. In the UK,
reports suggest that the prevalence of obesity among children of all ages is increasing
[Reilly 1999, Reilly 2001, Rudolf 2001]. Data from the National Study for Health and
Growth show a dramatic rise in the proportion of overweight primary school children
(aged 4–11 y) in the UK between 1984 and 1994 [Chinn 2001]; overweight increased
from 5.4% to 9.0% in English boys, from 6.4% to 10.0% in Scottish boys, from 9.3%
to 13.5% in English girls, and 10.4% to 15.0% in Scottish girls. Data from a large
survey of young children (age 1 mo to 4 y) in England showed a similar rise in the
prevalence of overweight [Bundred 2001]. In Denmark an increased prevalence of
obesity was identified for birth cohorts from the early 1940s to the mid-1950s and
from late 1960s and onwards [Olsen 2006].
What is ”normal”?
Developing consistent approaches to the measurements of childhood obesity is a
priority issue in this field [Barlow 1998] and international standardized cut-points have
been proposed [Cole 2000]. For defining overweight in children, reference values for
body mass index (BMI) are available from the US Centers for Disease Control and
Prevention (CDC) [Kuczmarski 2000] and the IOTF (International Obesity Task
Force) [Cole 2000]. Current expert opinion supports the use of BMI cut-off points for
children and adolescents, but experts are at variance as to which centiles should be
used for comparison [Bellizzi 1999, Reilly 2002].
Once again – comparability is crucial for epidemiological monitoring
The ongoing worldwide epidemic of overweight and obesity among children and
adolescents is a crisis in public health [Lobstein 2004]. The epidemic needs a reliable
and valid tool for epidemiological surveillance in order to really catch the
27
development. It is both important to describe the depth and speed of the time change
and to discover when a trend is broken. But, what kind of surveys makes it possible to
monitor this epidemic? How is the national situation monitored? It is rare that surveys
are, in a statistical sense, representative for larger populations, and they are very
seldom nationally representative. Data on body weight is either collected by weighing
the individuals in dedicated studies, but more often by simply asking individuals about
their body weight. You have to have a national representative study with minimal
selection bias and it must be possible to repeat later under similar circumstances in
order to study time changes properly. Growth data should be based on measurements.
Sometimes you have a time period covered by register data, with national coverage,
but just for one single age, as is the case with conscription data, for 18-y-old males, in
Sweden [www.pliktverket.se].
The trajectory of obesity is revealed by longitudinal studies
Longitudinal studies of national representative samples make it possible to analyse,
both within the cohorts, and between cohorts. Until now, to the best of our knowledge,
there is only one single example of such an attempt, but that was a mixed longitudinal
study [22]. Examples of growth monitoring by studying large samples do exist in the
UK, the Netherlands, China, Finland, Australia, Canada, Norway and Sweden [Hughes
1997, Fredriks 2000, Luo 2002, Kautiainen 2002, Booth 2001, Tremblay 2002,
Andersen 2005, Rasmussen 1999].
Comparison of prevalence data on obesity in children and adolescents around
the world is difficult because of the lack of standardisation and interpretation of
indicators for being overweight or obese in these age groups. There is actually no clear
opinion on how to express relative weight or how to consider what a ”normal”
development in weight in relation to height is. Usually local or national percentile
distributions of weight for age are used. These distributions may differ between
regions and nations and are also subject to change over time. In addition, different
percentiles are used for the definition of being overweight or obese (e.g. 85th, 90th, 95th,
and 97th percentiles are used in different countries) [Seidell 1999].
Epidemiological monitoring in Sweden
The situation in Sweden is, up until now, best described by data collected at
conscription. Conscript data cover the majority of 18-y-old males, yearly, from year
1944 to 2001, see Tables page 11 and 17. When looking at data from MSCR, Taranger
in 1985 asked: Increasing rate of overweight of young men – a medical risk factor?
[Taranger 1985]. And later, using the same data source, Rasmussen studied all males
born in 1953, 1958, 1963, 1968, and between 1973 and 1977 [Rasmussen 1999]. Of
the 503 689 subjects identified in the RTPs, information on BMI was available for 448
732. The increase between 1971 and 1995 was 4.7 kg for body weight, 0.5 cm for
height, and 1.38 kg/m2 for BMI.
Between 1971 and 1993 muscle power increased by 5.8% among all men. For
all birth cohorts the mean muscle power was higher among the overweight individuals
than among the conscripts with BMI in the normal range. The increase in muscle
power over time was greater among individuals with BMI in the normal range (4.7%)
than among individuals who were overweight (2.4%)
28
Among men born between 1953 and 1963 the prevalence of overweight was
7.2% for those living in large cities, and 9.7% in rural or sparsely populated areas.
Among men born between 1975 and 1977 the prevalence of overweight was 13.3% in
large cities and 17.5% in rural or sparsely populated areas.
Analyses conducted after exclusion of all young men from immigrant families
clearly showed that immigration had no impact on the time trend.
The development of mean BMI among 18-y-old conscripts is illustrated for the
years 1962, 1972, 1982, 1992 and 2000 by values 20.9, 21.3, 21.7, 22.1 and 22.7, see
further Table page 17 [MSCR]. A shift is seen around 1972. Perhaps this indicates that
boys born around 1950 still had a body weight which could be considered as ”normal”.
In Sweden, there are surveys of time trends but they are either local [Berg
2001, Petersen 2003, Mårild 2004, Blomquist 2007] or comparing, not national
representative, cross-sectional material for revealing national trends [Ekbom 2004,
Westerståhl 2003, Eriksson 2005] or just 7-, 10- and 13-y-old children in Stockholm
from 1940–1990 [Cernerud 1991, Cernerud 1993]. Eriksson for example compared a
population-based, selected, material from Gothenburg born 1973–75 with a material
from parts of Stockholm county born 1985–87 [Eriksson 2005], and Blomquist
compared 4-y-olds born in 1998–99 from Västerbotten with the material from
Stockholm born in 1985–87 and with the material from Gothenburg born in 1973–75
in order to reveals national trends [Blomquist 2007].
Explanations for the obesity epidemic
The rising prevalence of obesity in children has resulted in a range of health policy
statements, and halting the rising prevalence of overweight and obesity in children is
nominated as a public health priority [WHO 1997]. The increasing prevalence of
obesity is not under genetic control but rather, related to two major lifestyle factors:
the energy of the diet and an increasingly sedentary lifestyle [Jeffery 1987, Beunen
1992, Brownell 1994, Prentice 1995]. Prevention of overweight and obesity must rely
on the modification of these two factors, and the efficacy of strategies that seek to alter
these factors to cure obesity must be understood.
There are many reports that indicate that the prevalence of obesity and an
increasing average BMI, even in a very short period during the last years, is a fact in
many countries [Troiano 1995, Mei 1997, Hughes 1997].
Interpretation of these increases in childhood and adolescent obesity rates is
difficult. Explanations require unbiased and precise estimates of energy intake and
energy expenditure, and these are often unavailable. Small secular shifts in energy
balance, which are all well within the margin of error of all available methods of
assessment. Such assessments are further complicated by the likelihood that reported
energy intake in children is considerably underestimated [Champagne 1998]. In the
U.S., however, which is one of those countries in which energy intake over the last
decades has, if anything, decreased [Kennedy 1997], there has been a dramatic recent
increase in the prevalence of obesity. Some crude evidence suggests that the reduction
in energy expenditure in children and adults is the most important determinant of this
increase in obesity [Berkowitz 1985]. Several studies have reported low physical
activity in obese children compared with their lean counterparts [Harrell 1997, Maffeis
1997], but this may be either a cause or a consequence of their obesity. Prospective
studies, however, have also linked sedentary behaviour, such as television viewing, to
the development of obesity [Dietz 1985, Robinson 1993, Robinson 1998].
29
What comes first?
Overweight and obesity in childhood are known to have significant impact on both
physical and psychosocial health. For example, hyperlipidemia, hypertension and
abnormal glucose tolerance occur with increased frequency in obese children and
adolescents [Lauer 1975, Steinberger 1995]. In addition, obesity in childhood is
known to be an independent risk factor for adult obesity [Whitaker 1997, Parsons
1999]. Guo and Chumlea report that the risk of developing adult obesity (BMI >28) in
children older than nine years who are obese (defined as BMI above the 95th percentile
for weight), is up to 80% at age 35 y [Guo 1999]. Furthermore, there is evidence of an
association between adolescent obesity and increased risks for health in adult life
[Must 1992, Power 1997, Must 1999]. The relative risks among men were 1.8 (95
percent confidence interval, 1.2 to 2.7; P=0.004), for mortality from all causes and 2.3
(95 percent confidence interval, 1.4 to 4.1; P=0.002), for mortality from coronary heart
disease. It is noteworthy that overweight in adolescence was a more powerful predictor
of risk than overweight in adulthood.
Overviews that make the epidemic situation clear?
By a summary of cross-sectional, nationally representative school-based surveys in
1997–1998 in 15 countries, with self-reported data, for 13- and 15-y-old boys and
girls, the highest prevalence of overweight was found in the United States and the
lowest in Lithuania. The highest rate in Europe was found in Greece [Lissau 2004a].
By presenting prevalence data from 21 surveys in Europe, a tendency was revealed for
higher prevalence of overweight among children in western and especially southern
Europe [Lobstein 2003].
The prevalence of childhood overweight has increased in almost all countries
for which data are available. Exceptions are found among school-age children in
Russia and to some extent Poland during the 1990s. Exceptions are also found among
infant and pre-school children in some lower-income countries. Obesity and
overweight has increased more dramatically in economically developed countries and
in urbanized populations [Wang 2006]. The prevalence of overweight and obesity
among children is rising in the European region, and the annualised rates of increase
are themselves increasing [Jackson-Leach 2006].
These two reports by Lobstein and by Lissau [Lobstein 2003, Lissau 2004b]
were examined by Lissau, and she concluded that the year of data collection, methods
and use of appropriate statistics are of critical importance for the conclusion drawn
from comparative epidemiological surveys on the prevalence of overweight [Lissau
2004b].
In the overview by Jackson-Leach [Jackson-Leach 2006] the figures for
Sweden for instance is based on the surveys by Ekblom [Ekblom 2003] and Petersen
[Petersen 2003]. The first survey have for example decreasing heights (we know that
there is a still ongoing, although weak positive secular change for height), over time,
for the populations compared that points on some problems in comparability and the
second survey just covering the situation in the city of Umeå in the northernmost part
of Sweden. This emphasizes the importance of national representative samples when
making comparisons over time for revealing national trends in overweight and obesity.
30
1.3.6 Dietary changes in Sweden
It is hard to catch changes in eating habits on a national level in a scientific meaning.
Surveys with both a valid content and without considerable number of missing cases
are hard to perform. Anyhow, good surveys exist, though often local, but in 2003 the
National Food Administration performed a survey of 4-, 8- and 1-y-olds from all of
Sweden.
On the other hand we have data for a longer time period regarding the feeding
situation in the first year of life.
Breastfeeding
The rate of breastfeeding in Sweden is possible to follow through official statistics,
from 1964 up until now, except for the years 1976–1985, when official statistics are
missing. Up until the mid-1930s, most children were born at home. As hospital
deliveries increased, the proportion of breastfed children started to decrease. The
decrease continued until the beginning of the 1970s, when a strong attitude shift
occurred towards emphasizing the social and biological advantages of breastfeeding.
This resulted in a strong increase of the breastfeeding frequency up until the mid1980s. After that there seems to have been stagnation, or even a slight decrease.
However, the period of decrease was short. From the beginning of the 1990s we have
again seen a continued increase in the number of breastfed children. Currently,
Sweden is the leading country among the developed nations with respect to the
breastfeeding frequency [Socialstyrelsen Amningsstatistik].
Dietary changes in Västerbotten
The nutrition and health in Swedish children 1930–1980 are described by Persson: In
1930 a nutrition survey was made of 1 675 school children in the county of
Västerbotten in northern Sweden [Odin 1934]. In 1967 a second survey was carried
out in the same area, covering 1 411 children aged 4, 8 and 13 y [Samuelsson 1971]. A
third survey was carried out in 1980 of 572 children in the same age groups.
Underweight and iron deficiency anaemia was prevalent in 1930. In the last study the
average energy intake had decreased from 100 to 87% of the Recommended Dietary
Allowance (the same as Reference Daily Intake). A slight increase in the prevalence of
overweight among 13-y-old children was also noted. The fat intake was lower in 1980
than in 1967. The malnutrition problems of 1930 have been eradicated but new
nutritional problems, linked to the risk of developing obesity and health problems in
adulthood such as coronary heart disease, call for new preventive strategies [Persson
1989]. After these surveys a school based dietary survey in 1989, using seven-day
records, was performed in two cohorts, living in Umeå, Västerbotten, aged 14 and 17 y
[Bergström 1993]. The study comprised 366 boys and 365 girls. When compared to
these previous studies, a striking finding was a decrease in dietary fat intake and an
increase in carbohydrate intake. However, the relative intake of saturated fat had not
changed (15% of total energy). The dietary change was mainly due to an increased
consumption of cereal products.
31
An example from Uppsala and Trollhättan
In a longitudinal study from 1993 to 1999, 208 adolescents were followed from age 15
y to 21 y in Uppsala and Trollhättan. At 17 and 21 y of age, the adolescents consumed
significantly more pasta, vegetables, coffee and tea compared to age 15, while the
frequency consumption of fat spread, milk, bread, potatoes, carrots and buns and
biscuits decreased. The changes between 15 and 17 were smaller than between age 17
and 21. At age 21, the males decreased their intake of fruit, while the females
decreased their intake of meat. No-meat consumers among females increased from 2 to
13%. Milk consumption decreased significantly in both sexes. Breakfast habits did not
change: 90% had breakfast five times/week or more [von Post-Skagegård 2002].
National survey of eating habits for 4-, 8- and 11-y-olds in Sweden
In 2003 a national survey was performed that was completed by 590 out of 924
sampled 4-y-olds, 889 out of 1 209 sampled 8-y-olds and 1 016 out of 1 290 sampled
11-y-olds. No large differences were seen in food choice or nutrient intake between
children from different socio-economic groups. Children to parents with a university
education consumed more fruits and vegetables and had a diet with a slightly higher
nutrient density. Children to parents with a non-Swedish background ate more fruit
and vegetables but drank less milk. The result showed that the most desired changes in
food habits are a lower intake of soft drinks, sweets, crisps, cakes and biscuits and an
increase in the intake of fruits and vegetables [National Food Administration 2003].
1.4
GROWTH AND DEPENDING FACTORS
On the one side we have the whole society that influence the population to grow in
one way, and on the other we have all studies that examine if there is a relation
between growth and one single exposure, trying to take confounding under
consideration. Though we want to know about all the interactions important for
optimal growth and study growth in a ”holistic” way, we almost always have to rely
on studying one factor at a time, and the relation to growth. An ecological or
interactional perspective developed for example by Magnusson or Bronfenbrenner
[Magnusson 1988, Bronfenbrenner 1979] on growth are missing in the literature.
Therefore this will be a review in a variable-oriented way.
A review of growth and depending factors, can be presented like this:
a. Growth during pregnancy-birthweight (1.4.1).
b. Height from birth to final height (1.4.2).
c. Weight from birth to adulthood (1.4.3)
d. Relations between a, b and c and later health (1.4.4)
Poor growth is known to have a relation with poverty and misfortune [Tanner 1992].
Obviously, secular changes in height are affected by socio-economic factors as
social class [Kuh 1991], income and education [Meyer 1999], family-size [Chinn
1989], urban or countryside [Weber 1995] and region [Padez 1999]. Similar factors
affect the tendency in menarcheal age [Laska-Mierzejewska 1982, Bielicki 1986,
32
Rimpelä 1993, Roberts 1994]. It is well-known that growth is affected by bad
housing standard and population density [Foster 1983].
The secular change in final height has its origin during the first two years in
life, probably mainly through increasing growth of the lower extremity. This period
is when the post-natal growth is as fastest and thus most vulnerable to bad
circumstances. In developing countries poor growth during childhood results
sometimes in reduced height, and this we know is concentrated to the first 1–2 years
of life.
Three causes to reduced growth of the child can be identified: nutrition,
infections and mother-child interaction [Waterlow 1994]. Genetical explanations are
not that important while children in developing countries who get their basic needs
met are growing as well as children in the industrialized world. In terms of secular
changes, future generations are coming from the same gene-pool, by definition, and
the time-frame is too narrow for the gene-pool to go through any changes. So,
genetical explanations are not a very likely part in the explanation of secular changes
[Cole 2000].
The pattern of growth in childhood and adolescence is common to all human
populations. Growth is most rapid in infancy, slows down during childhood, and then
accelerates during puberty (the ”adolescent growth spurt”) before slowing again until
final height is achieved sometime between the late teens and mid-twenties. The timing
of particular events, such as the adolescent growth spurt or the achievement of final
height, can vary between different genetic groups, but the overall pattern is common to
all.
1.4.1 Growth during pregnancy
Growth starts at conception. What factors determine birthweight, and do they have any
biological significance for the growth, future health and welfare of the individual? We
know now that there is a relationship between birthweight and future health. In a series
of studies Barker et al. have investigated the relation between intra-uterine growth and
growth early in life and later susceptibility, later health and mortality [Barker 1989,
1992, 1993, 2001, 2002, Forsen 1999, 2000], and other authors have followed [Huxley
2000, Law 2002].
Firstly, there are genetic factors that contribute to birthweight [Johnston 2002],
secondly, maternal nutrition determines size [Stephenson 2002], thirdly, there are
social influences on birthweight [Drever1997, Spencer 2002].
Birthweight has been observed to relate to, among other factors, socioeconomic status [Drever 1997], gestational age [Chamberlain 1975], maternal smoking
[Butler 1972, Fox 1990, Nafstad 1997], gestational diabetes [Metzger 1990], and
famine [Ravelli 1976].
Genetic factors
Epidemiological studies estimate that environmental influences account for about 25%
of the birthweight variance, and genetic influences account for 38–80% of the
birthweight variance [Penrose 1954, Magnus 1984, Magnus 1984, James 1998]. There
is considerable variability in the estimates of the foetal and parental components of
these genetic influences from 18 to 69.4%, and from 3 to 20% variance of birthweight
33
respectively. Overall there is strong evidence that genetic factors play a significant role
in determining birth size.
Birthweight is correlated between half siblings of the same mother but not of
the same father [Gluckman 1994] because of the greater contribution of the maternal
genotype and environment [Walton 1954].
Familial trends in birthweight have also been observed. There is significant
correlation between parental birthweight and birthweight in index cases using multiple
regression analysis (mothers 0.19–0.20; fathers 0.12–0.16) [Langhoff-Roos 1987,
Magnus 1997]. Other authors have reported evidence of strong familial trends in birth
size [Magnus 1984, Carr-Hill 1987, Wang 1995].
Mikulandra reports from a study that the parents (father and/or mother) of male
babies are older than those of girls, that pregnancy for male babies lasts longer and that
male babies are born heavier than girls. With increased weight, and height and BMI in
the father, the birthweight of both male and female babies increases [Mikulandra
2001].
Parental birthweight is a significant and independent predictor of low
birthweight in offspring is reported from a Norwegian study. The estimate of the
hereditability of birthweight in this study is lower than previously estimated from data
within one generation in the Norwegian population [Magnus 2001].
There is growing evidence supporting the roles of certain candidate genes in
influencing size at birth [Liu 1993, Baker 1993, Woods 1996, Hattersley 1998].
Maternal nutrition as a determinant
Maternal nutrition, encompassing maternal dietary intake, circulating concentrations,
utero-placental blood flow, and nutrient transfer across the placenta, influences
birthweight. Summarized by James et al.: Genetic and environmental contributions
(%) to birthweight variation: Genetic factors 38 (maternal genotype 20, foetal
genotype 16 and foetal sex 2), environmental factors 62 (general maternal
environment 18, immediate maternal environment 6, maternal age and parity 8 and
unknown environmental influences 30) [James 1998].
Deficiency and foetal growth
The effects of severe macronutrient deficiency depend on the stage of gestation.
During the Dutch famine in 1944–45, a 50% reduction in energy intake during the first
trimester was associated with increased placental weight but not with change in
birthweight. Maternal under-nutrition in late gestation was associated with reduced
placental and foetal weights [Lumey 1998].
Embryo transfer and litter reduction experiments similarly show that maternal
environment predominantly influences later foetal growth [Snow 1989]. Although
macronutrient deficits in later pregnancy would be expected to exert greatest impact on
birthweight, the human foetus weighs only 20% of term weight at 24 weeks [James
1998], catch up growth often occurs [Coutant 1998, Fewtrell 2001]. In contrast, the
earlier in postnatal life that under-nutrition occurs, the more likely it is to have
permanent – programming, that is – effects [Widdowson 1963]. In normal pregnancies
of malnourished women, dietary supplementation during late pregnancy increases
birthweight [Prentice 1991].
34
In developed countries, dietary macronutrient or micronutrient deficiency is
rarely thought to be responsible for clinically significant impaired foetal growth
[Robinson 2000]. Lower birthweight is associated with lower social class, but although
it is often assumed that this is nutritional, there are many confounders such as smoking
and genetic factors. Recent human pregnancy studies do not confirm the dietary
hypothesis [Godfrey 1997, Mathews 1999], but these studies have been criticised
[Symonds 2000]. Contemporary studies in Australia, however, indicate that nearly
30% of women who deliver babies with a low birthweight (<2 500 g) suffer from
eating disorders [Conti 1998]. Experimentally increasing maternal nutrition in sheep
enhances birthweight [Symonds 2000].
In developing countries, maternal dietary intake can affect birthweight, and
intervention helps. In developed countries, epidemiological studies and experiments
using animals indicate that modest reductions in maternal food intake could affect
survival at birth and longevity, in the absence of pathological changes in birthweight
[Ravelli 1998, Heasman 2000]. It appears to be earlier maternal nutrient restriction
that increases placental size [Heasman 1998] and alters the expression of genes
regulating the glucocorticoid and renin-angiotensin systems [Whorwood 2001].
Improved long-term nutritional situation and living conditions seems to be the
most important prerequisites to counteract low birthweight in developing countries
[Andersson 1997]. Muller et al. come to the same conclusion after studying the
situation in Papua New Guinea (PNG). Besides improving maternal health,
interventions for improving birthweights in PNG should therefore aim at strengthening
the economic base of rural populations and promote the cultivation and consumption
of high quality foods [Muller 2002].
Over-nutrition of mother, and foetal growth
Many researchers have found that heavier women get heavier infants [Langhoff-Roos
1987, Rössner 1990]. For every extra maternal kilogram the birthweight will increase
with 15–20 g [Cresswell 1997]. This cannot be accepted as the only or even as a main
cause to the observed increase of birthweight over some decades [Meeuwisse 1998].
But Forsum et al. concludes that the relation between the body weight of the mother
before and during pregnancy is important for the size of her baby [Forsum 2003]. See
further 1.3.1.
Social influences
Obvious relation though social factors are hard to isolate
Birthweight, like growth, is determined by the complex interplay of genetic and
environmental factors. The proportional contribution of these influences is unclear.
However, birthweight varies within genetically similar populations [Mackenbach
1992, Elmén 1996, Spencer 1999], suggesting that environmental factors play a
significant role. Positive or negative secular changes in birthweight are explained by
an environmental influence [Rush 1983, Power 1994, Chike-Obi 1996, Brieger 1997,
Bonellie 1997, Meeuwisse 1998, Orskou 2001]. Birthweight also shows a reverse
social gradient such as that increasing disadvantage is associated with decreasing
birthweight [Mackenbach 1992, Elmén 1996, Spencer 1999].
35
The effects of social class on birthweight, and its interaction with other
maternal factors, were examined in groups of women bearing small-for-dates (SFD),
average-for-dates (AFD) and large-for-dates (LFD) babies. The relative risk of a lower
social class women having an SFD baby steadily decreased from 1.75 to 1.20 as
adjustment was cumulatively made for smoking, hypertension, maternal age and
height [Ounsted 1982].
The relationship between social position and birthweight is being studied. A
total population of 102 638 single born, first-born infants was included in a Norwegian
study. Children of parents with a high education (more than 15 years of schooling) had
the highest birthweight. The same tendency appeared for paternal socio-economic
status, but the differences were comparatively small. When examining income the
pattern was different. The highest maternal income had the highest proportion of low
birthweight offspring. When examining parental education jointly, it was found that
the mothers’ education had the greatest impact on birthweight. By adjusting for female
smoking, the association between maternal education and birthweight was weakened.
However, assuming that birthweight is decreased by 200 g from smoking the effect
was still significant [Arntzen 1994].
Interventions to increase birthweight in disadvantaged groups have been largely
unsuccessful [Paneth 1995], and, although mean birthweight has increased [Alberman
1991, Meeuwisse 1998, Orskou 2001], the rate of change is slow and the gradient
remains unchanged.
The social gradient in birthweight probably arises as a result of the
accumulation and addition of risk and protective factors over time [Kuh 1997] and
across generations [Emanuel 1992], rather than resulting from risk exposures within
the index pregnancy. Biological processes during the foetal period are systematically
linked to the social circumstances of the mother, but in a different way in the 1920s
and in 1985 in Sweden [Vågerö 1999].
Practically all social differences in birthweight are related to the differences in
maternal age, parity, height, and smoking habits. If a socio-economic indicator is to be
included in the analysis of birthweights (for other reasons like international
comparisons), education is recommended [Nordström 1994].
Environmental factors
Environmental factors with a known association to birthweight are nutrition, smoking,
maternal ill health, and genital infection. Adult height has a known association with
relative nutritional impairment in childhood [Berney 2000], and maternal height is an
important determinant of birthweight [Kramer 1987].
Maternal health and birthweight
The association of other factors such as stress [Hoffman 1996] and exposure to some
types of work during pregnancy [Homer 1989, Ahlborg 1989] remains unproven, but
some authors notice that work during pregnancy appears to reduce the mean foetal
weight in both primigravidas and multigravidas, and rest during the last six weeks of
pregnancy improves foetal weight significantly [Alegre 1984]. Järvelin et al. stated
after analysing a northern Finland cohort that their result taken together with the
spatial clustering of birthweights, suggests that there may be important determinants of
birthweight that have yet to be identified [Järvelin 1997]. Evidence for an independent
36
effect of stress is slight, but one study does show stress exerting an effect through
increased smoking [Sheehan 1998], while another one does not [Hedegaard 1996].
Other risk factors for low birthweight such as maternal age, although not it selves
environmental factors, are strongly influenced by the social environment. Severe
energy restriction during pregnancy, such as occurs in some developing countries
[Achadi 1995] and was noted in the 1945 Dutch Hunger Winter [Hart 1993], reduces
birthweight but, randomised controlled trials of nutritional interventions in the index
pregnancy have failed to show convincing benefit [Kramer 1998]. Nutrition may exert
its effect over a longer period through an effect on maternal growth in childhood
[Baird 1980] and possibly through an intergenerational effect [Emanuel 1992].
Maternal ill health has been associated with reduced birthweight [Baird 1974], and
genital infection exerts its influence through increasing the risk of preterm delivery
[Divers 1993]. It is concluded that hypertensive pregnancies are characterized by
lower birthweight and shorter gestational period. However, intra-uterine growth
retardation is not a general characteristic of hypertension in pregnancy [Himmelmann
1996].
Smoking mothers and their babies
The association of smoking with a reduction in birthweight is well established [Butler
1972, Rush 1974, 1983, Kramer 1987, Ahlborg 1991, Wilcox 1993, Nordström 1994,
Bjerregaard 1996, Ellard 1996, Groff 1997, Secker-Walker 1997, England 2001].
Brooke et al. conclude that smoking was the most important single factor (5%
reduction in corrected birthweight). Passive smoking was not significant (0.5%
reduction). After smoking was controlled for, alcohol had an effect only in smokers
and the effects of caffeine became non-significant. Only four of the socio-economic
and stress factors significantly reduced birthweight and these effects became nonsignificant after smoking was controlled for [Brooke 1989]. The relation between
passive smoke exposure and birthweight is not that convincing [Martin 1986, Rubin
1986].
Recently, Ingvarsson et al. conclude: ”Smoking during pregnancy causes
symmetrical fetal growth impairment, possible due to decreased oxygen transport to
the fetus and decreased concentrations of fetal insulin, insulin-like growth factor I and
IGF binding protein 3.” [Ingvarsson 2007]
In most investigations, the reduction in birthweight has been found to range
between 170 and 235 g, even after adjustment for maternal age, height, weight, weight
gain, alcohol consumption and birthweight of previous children [Rush 1974, Davies
1976, Wainright 1983, Secher 1990].
The first-born babies
In general terms a woman is more likely to deliver a heavier baby in her second
pregnancy than in her first pregnancy [Wilcox 1996]. In Sweden there is an
increasing birthweight from the first-born to the third child, and after that
birthweight decreases [MBR].
37
Coffee, any relation?
A significant reduction in birthweight was found to be associated with an average
caffeine intake of >=71 mg per day, after adjustment for gestational age, infant sex,
and maternal height and weight, but only in infants born to non-smoking mothers
[Vlajinac 1997]. Caffeine intake during pregnancy is a risk factor for intra-uterine
growth retardation but not for preterm delivery or low birthweight concludes Fortier et
al. [Fortier 1993]. As a result, Fenster et al. suggest that heavy caffeine consumption
increases the risk for foetal growth retardation [Fenster 1991].
Alcohol, any relation?
Mills noted an effect of alcohol to reduce birthweight [Mills 1984] and Larroque noted
a significant reduction in birthweight associated with an average daily alcohol
consumption of three drinks or more after gestational age, infant sex, maternal age,
parity, weight, and height, and cigarette smoking had been controlled for [Larroque
1993].
Other drugs
Hatch could not detect any effect of marijuana on foetal growth [Hatch 1986]. After a
meta-analysis English et al. conclude: There is inadequate evidence that cannabis, at
the amount typically consumed by pregnant women, causes low birthweight [English
1997].
Intra-uterine growth retardation
Small-for date babies, many causes
Both the degree and the cause of growth retardation decide the post-natal growth
together with the environment. Intra-uterine growth retardation is one of the most
frequent types of severe shortness of stature persisting during childhood and adulthood
[Job 1991]. By definition, 3% of newborns, either at term or before, are below the
statistical limits of normality for length, regardless of the term [Usher 1969], and only
part of them will have a postnatal catch-up growth. Intra-uterine growth retardation
(IUGR) is a heterogeneous group of disorders, including many malformative or
dysmorphic syndromes with or without detectable abnormality of chromosomes and/or
genes, a variety of foetal diseases due to maternal conditions (hypertension, drugs,
alcohol and other toxins, etc.), to materno-foetal transmission of viruses or other
infectious agents, or to anatomical and functional abnormalities of the placenta, uterus,
or cord. Such causes may in fact be lacking or not detected, so that many cases of
IUGR remain unexplained. The heterogeneity of small-for-date babies followed in
neonatology clinics makes a comprehensive appreciation of postnatal events difficult.
However, longitudinal studies show that approximately two-thirds of IUGR newborns
have some catch-up growing during their first months of life [Tanner 1975, Nilsen
1984, Thierot-Prevost 1988, Fitzhardinge 1989]. Those whose height remains
subnormal after 1 year will probably not have a catch-up growth in the following
years. And part of those whose height had improved during their first postnatal year
will have an insufficient growth velocity during their second to fourth year, so that
38
they might thereafter return at the third percentile of height or below. No fixed
correlation in this evolution of postnatal growth has been demonstrated.
1.4.2 Stature from birth to final height
Gender differences
Gender differences, and different susceptibility for bad times
From the first longitudinal study, the study of Montbeillard’s son, we can see the four
distinct phases of growth velocity: 1. From birth to 3 years of age, the infant phase,
growth decelerates rapidly from its maximum value of 22 cm/year to 6 –7 cm/year. 2.
From 3 to about 7 years of age, the childhood phase, growth remains fairly constant.
After age 7 years until about 11 years, the juvenile phase, growth decelerates again, at
first rather slowly but then much faster. 4. From about 11 to 18 years of age, the
adolescence phase, there is a classic adolescent growth spurt, with an acceleration
period from about 11 to 14.5 years followed by a deceleration period that continues
until 18 years and beyond
In Europe today men are on average 12–13 cm taller than women. Secular
change in this gender difference has been shown [Hall 1978]. Many data suggest that it
was a small difference in height between the genders during periods with bad life
circumstances. These observations support the opinion that growth of male individuals
more easily depends of environmental factors like under-nutrition and diseases than do
the growth of female individuals [Eveleth 1990, Kuh 1991, Arcini 2006].
Longitudinal studies, that is, studies of growth of the same individuals from
birth to adult age, has shown that gender differences in height are built up by four
components of growth [Prader 1984], see Table 4.
Table 4. Differences between boys and girls in growth
1. Differences in pre-pubertal growth-velocity, which is faster
for boys before birth and during the first year of life.
2. Longer pre-pubertal period of growth for boys, who get
their peak velocity in height about 2 years earlier.
3. More intensive peak height velocity during puberty for
boys.
4. More growth for girls after the pubertal peak height
velocity period
Difference
+1.5 cm
+6.5 cm
+6.0 cm
-1.5 cm
+12.5 cm
Thus, the result is that in general men are 8% taller than women.
After menarche a girl normally gains 6–8 cm, seldom more than 10 cm [Tanner
1989]. Late maturers among boys can grow up to 25 years of age [Hägg 1991].
Whether you are an early or a late maturer, it is only the legs that show a smaller
adult size for early maturers. The shorter growth period is compensated by a higher
39
prepubertal growth velocity and a higher level in pubertal years. The pubertal peak
is a little larger for early maturing boys but not for girls [Gasser 2001].
Genetic factors
The old debate between the relative importances of genes versus the environment in
human development is largely ignored by most researchers today. In reality, the
biological development of the human being is always due to the interaction of both
genes and the environment [Bogin 1999]. It is erroneous to consider whether one or
the other is more important; genes are inherited and ”… everything else is developed”
[Tanner 1978].
The human phenotype is the outcome of this interaction. All of the measurable
characteristics of the human body, of human behaviour, and of the human mind are
phenotypic traits. Although it is possible to describe a human genotype with great
precision, such knowledge by itself provides very little information about how a
human being will develop without also knowing about the environment with equal
precision [Bogin 1999 p.240]. See also 1.1.1 and 1.4.1 (Genetic factors).
Family-related factors
Comparative growth of primary schoolchildren from one and two parent families has
been studied by Garman: Children from one parent families were shorter than children
from two parent families; however, once heights had been adjusted for birthweight,
number of siblings, mother’s height, father’s height, and mother’s education this was
no longer the case. The higher prevalence of low birthweights and shorter parents that
account for the shorter stature of one parent children are factors that cannot be ignored
in a consideration of the health and growth of this group of children, and obesity may
be a potential health problem among the one parent family children [Garman 1982].
Socio-economic status (SES)
SES and social class have a strong impact on growth
It is often argued that socio-economic status, SES, is, in reality, only a proxy for better
health care, which reduces childhood mortality and morbidity, and results in increased
growth. However, the SES effect is more subtle than this. Bielicki & Welon listed four
primary factors relating SES and growth: (1) higher SES allows for better nutrition, (2)
better health care, (3) reduced physical labour for children, and (4), greater growthpromoting psychological from parents, schools, and peers [Bielicki 1982].
In all societies, social class and SES are powerful influences on human physical
and psychological growth and development. Conversely, stature, body composition
(fatness and muscularity), and rate of development influence the social, emotional, and
economic status of children, youth, and adults. An appreciation of these biosocial
interactions between growth and socio-economic status (SES) has been realized only
in the past 200 years. Following the studies of children in the Carlschule in the
eighteenth and the work of Pagliani in Italy and Bowditch in the United States and
Key in Sweden in the nineteenth century, it has been known that children of lower
SES are generally smaller and mature less rapidly than children of higher SES.
40
The National Child Development Study of the UK is a longitudinal study of the
influence of SES on growth in height and weight. Data from this study are based on
the population of all infants born in England, Scotland, and Wales between March 3
and 9, 1958. Lasker and Mascie-Taylor published the mean height, weight, and BMI
of these boys and girls at ages 7, 11, 16, and 23 y, stratified by the social class of the
male head of the household (grouped from I to V with I as the upper class). They
found that mean body size is significantly related to social class, and declines,
generally, from social class I to V at each age. The data for males are paralleled
compared to data for females. The statistical impact of the social class effect is
achieved by age 7 y, and is then maintained to age 23 y. The notable exception to the
stability of size by social class over time is the change in BMI between ages 11 and 16
y. The three lower classes, III, IV, and V, have a significant increase in BMI compared
with the social class I, and social class II. The change is due to absolutely faster rate of
growth in weight of the lower three classes [Lasker 1989, Lasker 1996]. Lasker and
Mascie-Taylor also analysed the association between body size of the boys and girls in
the study with the social mobility of their father. They found that the study subjects
with fathers who moved up one social class were large for their old social class but
small for their new social class. The effect of the size of the child and social mobility
of the father was found for the 7-y-olds, but was not statistically significant for 11-or
16-y-olds. Changing of occupation improves the family situation and, at an age most
susceptible, improvement leads to better growth [Lasker 1989]?
The same situation, as for growth and the relation to SES, is between socioeconomic status and maturation [Johnston 1974, Bielicki 1982, Low 1982].
Small or no effect?
In recent years, only three nations, The Netherlands, Norway, and Sweden, as far as it
is documented, seem to have eliminated or powerful decreased the differences in
growth between children from higher and lower SES levels [Brundtland 1975,
Lindgren 1976, Brundtland 1980, Roede 1985, Lindgren 1995]. In Sweden for
example, in 1979 Lindgren reported that differences decreased during the era after
World War II, but in 1980 she reported that significant differences in height and head
circumference were found between 5.5-y-old children of the highest and lowest SES
groups. The reason for the reappearance of an SES effect is not completely clear, but it
is perhaps an effect of the fact that the gap between rich and poor in Sweden has been
widening during the last decades [Lindgren 1994].
Data were analysed for 8 491 representative sample children measured in
England and Scotland in 1987 and 1988, and 3 203 inner city children were measured
in England in 1987. Height was negatively associated with social class but the
association was not significant after allowing for biological variables. A negative
gradient of height with size of sib-ship was evident in white children but less so in
Afro-Caribbean and Asian children. The individual associations of 11 different
environmental characteristics were examined after allowing for biological factors and
size of sib-ship. Consistent associations with height included a negative gradient of
height with increasing latitude and an association of taller stature with increasing
maternal age. A social class gradient in height is accounted for by association with
biological factors, particularly the parental heights; environmental attributes are
weakly associated with height after allowing for biological factors [Gulliford 1991].
41
The contents of the social variable and the relation to growth is demonstrated
by Söderfeldt and Werner by comparing the relation between the height and weight at
age 4 y with three different measures for social class [Söderfeldt 1988].
There can be no doubt that physical abuse and neglect of children leads to short
stature and failure to thrive. The correlation between speech and developmental delay
and shortness leads to the conclusion that both are manifestations of long-term
unhappiness and disruption within the family. The short, abused children will show a
recovery of growth with the more radical changes in the home circumstances, and this
can be used to gauge the ”success” of any intervention [Wales 1991].
Social mobility and growth
Empirical data have been gathered on 6 162 19- and 20-y-old Flemish males. The
mobility categories have been set up on the basis of the social professional status of
the father of the tested subjects and the social professional aspiration level aimed at
by the tested subjects themselves.
Statistically significant relations of varying importance have been found
between the nature and the degree of social mobility and the vast majority or the
anthropological characteristics considered in the present investigation. According to
the social professional origin, the aspiration level being kept constant, the former
relations disappear almost completely, except for groups originating from
agriculturalists [Cliquet 1968].
The present findings may at least partially be considered as being the result of
social assortment. Up to a certain level, the social assortment of genotypes should be
taken into consideration.
The hen or the egg – what comes first?
Scott et al. found that taller women from Aberdeen, Scotland worked at more skilled
jobs than shorter women. Furthermore, when taller women married, their husband
worked at more skilled jobs than the husbands of shorter women [Scott 1956]. These
findings was confirmed by Schreider in a study of British women showing that,
regardless of the occupation of the father, taller women tended to marry men with
more highly skilled jobs and shorter women tended to marry men performing semiskilled or unskilled labour [Schreider 1964].
Taller people achieve higher educational status than shorter people, as was
shown by Parnell for university students in England and other European countries
[Parnell 1954] and by Kimura for Japanese university students [Kimura 1984].
Bielicki and Charzewski found that within families the taller siblings were better
educated than shorter siblings. The average difference between the better educated and
the less educated brothers was 1.26 cm, a statistically significant difference. The result
indicates that tallness leads to a higher level of education [Bielicki 1983]. These
findings were confirmed by Susanne [Susanne 1980]. By studying stature, upward
social mobility and the nature of statural differences between social classes, Bielicki
could confirm the positive effect of social mobility upwards on growth [Bielicki
1992].
Numerous studies, from a diversity of disciplines, find that the taller, non-obese
man or woman is given preference in many arenas linked with SES, such as the
perception of intelligence, academic performance, and social skills, as well as initial
42
job hiring and perception of both current and future job performance [Hensley 1987,
Loh 1993, Lindeman 1994].
Psychosocial factors
The psychosocial short stature is a well-studied condition that shows how the
endocrine system mediates the relationship between psychological factors and physical
growth. Psychosocial short stature is a clinical condition of retarded growth stature that
cannot be ascribed to an organic problem with the child, but rather to behavioural
disturbance and emotional stress in the environment in which the child lives. A
diagnosis of psychosocial short stature is confirmed when the child is removed from
that environment, and growth is spontaneously restored. Spitz carefully investigated
the causes of poor growth experienced by emotionally disturbed infants and children
[Spitz 1945].
Growth failure caused by emotional deprivation in children was first reported
more than 50 years ago [Talbot 1947] and was subsequently the subject of further
reports [Patton 1962, Powell 1967, Powell 1967]. Powell et al. suggested that
emotional deprivation of children with psychosocial short stature resulted in growth
hormone deficiency secondary to hypopituarism [Powell 1967, Powell 1967], and that
this was reversible many months after the change in environment [Powell 1967,
Powell 1973]. It was proposed that the diagnosis of psychogenic growth failure should
be based on the child’s response to being separated from the parents, as expressed by
an increase in growth velocity [Job 1987, Golkhe 1998]. 18 children diagnosed as
having psychosocial short stature, and who had a change in their environment,
increased their mean height velocity and attained a stature towards the lower limit of
the range of their biological expected final height [Golkhe 2002].
Life-style factors
Sleep habits and height at age 5 to 11 y were investigated by Gulliford. Shorter
duration of slow wave sleep and lower growth hormone responses have been reported
in children with short stature caused by psychosocial deprivation. An investigation was
carried out to determine whether lower total sleep duration was associated with shorter
stature, and it was investigated in a sample of children taking part in the National
Study of Health and Growth. Parental responses to a self-administered questionnaire
were used to estimate usual times for going to sleep at night and usual time for waking
in the morning for 5 145 children aged 5 to 11 y. After adjusting for the effects of
other variables known to be associated with height, it was shown that there was a weak
negative association between sleep duration and height. It is concluded that variation
in sleep duration between children is unlikely to have an important influence on
growth [Gulliford 1990].
43
Nutritional factors
Breastfeeding and growth in infancy (See 1.4.3, page 45)
Dietary factors and growth
Growth and nutrition are closely correlated. Nutritional biochemists have determined
that there are 50 essential nutrients required for growth, maintenance, and repair of the
body [Guthrie 1995]. The essence to grow and gain weight is the focus for a healthy
situation, but since a few decades another focus emerged, namely the growing problem
with overweight and obesity.
Nutritional deficiency as a cause of growth failure. Golden lines up what is the
requirement of the chondrocyte, in particular, which will be crucial to longitudinal
growth. They are likely to require following nutrients for their metabolic machinery:
water, nitrogen, sulphur, phosphorus, threonine, lysine, potassium, sodium,
magnesium, zinc, oxidisable substrate (energy), carbon skeletons of essential amino
acid [Golden 1992].
Under-nutrition is the single most important cause of growth retardation worldwide, resulting from an adaptive response to decreased food availability [Nikens 1976,
Torun 1988], however, nutritional dwarfing found in a paediatric endocrine practice in
the United States usually does not involve the chronic growth problem of povertyrelated malnutrition frequently observed throughout the world. Poor growth and
delayed sexual development among suburban middle class adolescents is usually due
to inadequate nutrient intake due to psychological causes [Lifshitz 1987, Lifshitz
1992]. Also, poor growth and inadequate nutrition have been seen in relation to
organic problems such as chronic inflammatory bowel disease [Kelts 1979, Kirschner
1983, Hildebrand 1994]. Nutritional dwarfing patients usually show a deteriorating
linear growth and delayed sexual development that is accompanied by inadequate
weight gain. This pattern of growth is similar in patients with any of the different types
of nutritional dwarfing syndromes including those of organic nature, such as seen in
chronic inflammatory bowel disease [Kelts 1979, Kirschner 1983]. Catch up growth is
usually achieved with nutritional rehabilitation [Kelts 1979, Kirschner 1983, Pugliese
1983].
Nutrition is a critical factor. The contents of nutrition are more important than
the amount. Animal protein, fat, micronutrients, vitamins, essential fatty acids and
amino-acids everything is important [Allen 1994], and to support this Tanner and
Takahashi have connected the strong secular trend in height in Japan to the strongly
increasing intake of milk since the end of World War II [Tanner 1982, Takahashi
1984, Takaishi 1994].
The milk hypothesis
To what extent can a specific food make an impact on human growth and
development? There has been considerable debate in this question, and of all the foods
studied so far, milk has generated the most important results.
Orr and Leighton & Clark gave schoolchildren, in several cities in Scotland, an
extra pint of milk per day for seven months. Some children received whole milk and
some skimmed milk. Both groups increased faster in height and weight than two
44
control groups of children of the same ages, one group given no supplement and the
other given a supplement of biscuits, equalling the milk in total calories. Since the
biscuits supplement had no effect on growth, it was concluded that some factor in
milk, either whole or skimmed, accelerated growth [Orr 1928, Leighton 1929]. Several
other studies afterwards show similar results [Spies 1959, Lampl 1978, Little 1983].
Takahashi made a strong case in favour of what may be called the ”milk hypothesis”,
that posits that increased consumption of milk by infants, children and adolescents is
directly related to greater average height of a population [Takahashi 1984]. He found
an association between changes in dietary practices, especially milk consumption, and
growth in Japan. From 1966 to 1976 the milk consumption rose from 55 to 100 grams
per person and day. The height of schoolboys, aged 6 to 17 y, rose by an average of
4.1 cm between 1930 and 1960, a period of relatively great social and economic
change, and by an average of 5.3 cm between 1960 and 1975. Although other factors
(low rate of disease, reduced family size, society under development) may also have
contributed to height increase, the milk was one of the most important reasons for the
increase.
Migration
The relation between migration and growth was recognized early
The extent to which migrants differ from non-migrants in their physical characteristics,
and the reasons for the differences found, have been the subject of numerous studies,
beginning with Livi [Livi 1896] and Ammon [Ammon 1894], both cited by Boas
[Boas 1922] and followed by the classical investigations of Boas [Boas 1912] and
Shapiro [Shapiro 1939]. Livi and Ammon found, separately, that the children of urban
migrants in Italy were taller than rural sedentes (the non-migrating rural population).
Livi believed the cause was heterosis, and Ammon believed that the action of natural
selection was the explanation. Boas found that neither natural selection nor heterosis
could adequately account for changes in growth. Rather these were due to biological
plasticity in the face of the new urban environments.
Japanese in Japan and USA
Shapiro’s Japanese migrant study compared the growth of Hawaiian-born of Japanese
immigrant parentage, Japan-born Japanese who migrated to Hawaii, and Japanese
sedentes living in the same villages from which the migrants originated. The sedentes
and the recent immigrants differed in a few anthropometric measurements, some
increasing and some decreasing with migration. The largest differences were between
the immigrants and the Hawaiian-born. The latter were taller and more linear in body
build than their parents or the sedentes. Shapiro argued that, with migration, there were
improvements in diet, health care, and socio-economic status and that these conditions,
associated with an urban lifestyle, were responsible for the growth changes. For
example, Kobliansky and Arensburg have shown that migrants are non-random
samples of the populations from which they depart. These authors, in their study of
internal migration, conclude: ”… an attempt was made to assess the relationship
between population mobility and morphology. Samples of stationary urban and village
national groups from Europe were compared to similar mobile populations.
Statistically significant differences were observed between migrants and the original
45
stationary population, proving the process of selection. The higher values of length,
width, weight stature, etc. are found generally in the mobile population.” [Kobliansky
1977]
Migration and growth are related – but the pattern can change
In fact, a relationship between migration and larger body size, especially greater
stature, has been shown frequently [Shapiro 1939, Goldstein 1943, Lasker 1952,
Lasker 1961] but also the reverse [Macbeth 1987]. The urban advantage in growth was
not always present. Meredith [Meredith 1979], Malina et al. [Malina 1981], and
Wolanski [Wolanski 1988] reviewed studies from the period 1870–1920 that showed
that rural-living children in the United States and Europe were taller than their urban
peers.
By 1930, this pattern was reversed and urban children were consistently taller
and heavier than rural children. Other sources of data provide some ideas about
changes in the rural and urban environment at the turn of the century [Eveleth 1990,
Fogel 1986, Bogin 1988]. For example, prior to 1850, mortality rates for infants,
children and adults were higher in the cities than in the rural areas of Europe and the
United States. Around the year 1900 the trend reversed, due to improvements in
sanitation, water treatment, and food preservation in the city. Urban children benefited
from these improvements, as reflected in their greater stature and weight, and more
rapid maturation, compared with rural children. All this shows that it is not the shape
of the ecological exposure that is important, but the content. In many instances, the
motivation for migration is either to improve the life circumstances of an individual or
a family, or to maintain standards perceived as deteriorating in the original place of
residence [White 1980]. Therefore, much migration is associated with expectations of
improved health and nutritional status. Where these expectations are realized, and
where migrants are studied at some time after their arrival in the new environment,
plasticity rather than initial selectivity may well explain the larger dimensions of
migrants [Susanne 1984].
There is certainly evidence in some studies of selectivity of migrants at point of
departure with regard to socio-economic and/or demographic variables [Clarke 1984].
Biological selectivity of migrants can therefore be divided into that which is the direct
result of some intrinsic propensity for mobility, and that which is the indirect result of
an association between biological characteristics and other characteristics related to the
decision to move [Macbeth 1984]. In practice these two components cannot easily be
disentangled.
Biological selection?
Several studies provide tentative support for biological selection. The study of
Steegmann of eighteenth century British military recruits, found that conscripts who
migrated from their county of birth were significantly taller (169.1 cm) than recruits
living in the county of their birth (167.6). One must remember, though, that he also
found that urban-born recruits were shorter than rural-born men. Steegmann pointed
out that eighteenth century Britain was a developing country and that urban areas were
characterized by food shortages and unhygienic conditions. Thus migration to the city
was probably not responsible for an increase in stature; rather taller men living in rural
counties were more likely to migrate than shorter men. Whether these were genetically
46
taller men, or individuals who had experienced better living conditions prior to
migration, was not known [Steegmann 1985].
Kaplan found that migration was selective for physical type. Growth
differences between migrants and sedentes were found too soon after migration to be
due to an environmental change [Kaplan 1954]. Illsley et al. studied those to migrated
to and from Aberdeen, Scotland. They found that migrants were, on average, taller
than rural or urban sedentes. Migrants had generally better health than sedentes.
Migrant women had lower rates of low birthweight and perinatal death for their
children. Finally, migrants were generally of higher socio-economic status than
sedentes. This study suggests what the true meaning of migrant selection may be that it
is more likely to be selection for socio-economic status than biological selection per se
[Illsley 1963]. The study by Kobliansky and Arensberg [Kobliansky 1974] confirms
this, and so does Mascie-Taylor [Mascie-Taylor 1984]. Boas’ thoughts on human
growth response to migration were correct [Boas 1912].
Greek children in Sweden
Immigrant children from Greece was studied by Neiderud: They were all second
generation immigrants, consisting of all healthy children born in Sweden by Greek
parents in the catchment area of a single child welfare centre in Sweden in the
southern Helsingborg. The children represented 36 families with 52 children who were
born between 1974 and 1980. The parents had emigrated from a rural area in northern
Greece in the late 1960s or early 1970s for economic reasons [Neiderud 1992].
Children born between 1974 and 1980 in the two villages of Terpillos and Dipotamos
in the northern part of Greece were 66 from 48 families. The mean age for the
immigration group was 5.6 years and for the Greek group 5.3 years. The immigrated
group was significantly taller than the Greek group but had the identical height-for-age
as the Swedish group.
Refugee children, in Sweden, in the 1980s
The health of newly resettled refugee children from Chile and the Middle East was
studied by Hjern: 62 children from Chile and 43 children from Middle East were
studied. They arrived in Sweden during a period of 13 months, from July 1986 until
July 1987. They were measured at a first point as soon as possible and then subsequent
measurements 5–7 months and 17–19 months after arrival in Sweden. Both refugee
groups had a significant catch-up growth in height/age values from 6 to 18 months
after arrival in Sweden. Chileans had a significant rise in weight/height values, a rise
that was not seen in the Middle East group. After 18 months in Sweden 16 of the
Chilean were obese. No child was obese in the Middle East group [Hjern 1990].
New Americans
A semi-longitudinal study on growth and development was initiated on immigrant and
refugee school-aged children in San Francisco. Anthropometric values (height, weight
among others) were collected soon after their arrival in the United States and repeated
at 3-month intervals for 1 year. Comparisons for median growth rate indicated that
most cohorts exhibited a median growth velocity that was close to, or exceeded, the
median for US white children. There was also significant improvement in weight-for-
47
age. The results indicated that these immigrant and refugee children accelerated their
growth markedly in an optimum nutritional environment and were in a period of catchup growth [Schumacher 1987].
Geographical, climatical and ecological factors
Population density and growth
The relation of the height of primary schoolchildren to population density was studied
by Foster. Shorter stature was found with increasing population density, and this
relation remained after allowing for child’s birthweight, mother’s height, father’s
height, social class, number of siblings, and additionally a measure of home
overcrowding. The results are in contrast to those found elsewhere, and are unlikely to
indicate a cause-and-effect relationship between population density and height. The
findings cannot be explained either by known differences between urban and rural
areas in social characteristics or by plausible hypothesis of genetic heterogeneity
[Foster 1983].
High altitude compared to low altitude
More than 25 million people live in high altitude regions of the world; that is at
altitude of 3 000 meters above sea level or higher. High altitude environments impose
a number of stress factors on people, including hypoxia, high solar radiation, cold, low
humidity, high winds, and rough terrain with severe limitations of agricultural
productivity of the land. Of these, cold temperatures and hypoxia, the lack of sufficient
oxygen delivery to the tissues of the body, has been considered to be the most
important determinants of growth at high altitude. When statistically adjusted lowaltitude infants had a birthweight of 3 415 g compared to high-altitude infants with
birthweight of 3 133 g. Birthlength was 49.6 cm compared to 49.0 cm [Baker 1977].
This study and more recent research indicate that problems with the delivery of
oxygen to the foetus are part of the reason for the growth delay. One classic study by
Frisancho & Baker found that Peruvian children living above 3 000 meters are shorter
and lighter, on average, than lowland Peruvian children of the same age [Frisancho
1970]. In contrast, Clegg et al. found that high-altitude-living Ethiopians were taller
and heavier than ethnically similar people living at low altitude in Ethiopia. In this
case, the lowland population suffered from malaria and intestinal parasites that may
have affected their growth. Altitude, meaning hypoxia, was shown to be of secondary
importance in this early study [Clegg 1972]. Similar findings are reported by other
authors [Frisancho 1975, Mueller 1980]. How arctic populations grow is reported by
Jamison [Jamison 1990].
The sunlight in Stockholm and elsewhere
Nylin published what was probably the first study of the sunlight effect on growth. He
exposed one group of Stockholm boys (n=45) to ”sunlamp” treatments (using a lamp
that produced both visible and ultraviolet light) during the winter months and
compared their growth to a group of boys (n=292) not receiving treatment. During the
three-month period of treatment, the experimental group averaged 1.5 cm more growth
in height than the control group. During the summer, the control group grew at a faster
48
rate than the experimental group, so that over the entire year there was no difference
between the two groups in total height gain [Nylin 1929].
Two other studies lend support to an association between variation in sunlight
and growth rate. Vincent and Dierickx found that healthy children living near the
equator in Kinshasa (the Leopoldville), Zaire, grew more rapidly in height in the dry
season than in the rainy season. Diet, temperature, humidity, and sunlight variation
were considered as possible influences. No evidence in favour of the first three was
found [Vincent 1960]. Seasonal variation in growth was also studied by Bogin. 246
healthy children, juveniles, and adolescents of high socio-economic status living in
Guatemala City were followed during a 14-month period. About 75 percent of the
children grew at a significantly faster rate during the dry season than during the rainy
season. Conversely, about 25% grew at their fastest rate during the rainy season. This
latter is perhaps an example of that no single exposure works isolated [Bogin 1978].
Seasonal variation and month of birth
The cause of seasonal variation in height growth rate is not completely understood;
however investigators suggest that seasonal periodicity in sunlight may act on the
human endocrine system. A relation between light and growth has been known since
1919, when it was shown that ultraviolet light could cure rickets, a disease of bone
growth. Buffon could by scrutinizing the data from the son of Montbeillard conclude
among other thing a seasonal variation in rate of growth; the boy grew faster in the
summer than in the winter [Buffon 1777].
Seasonal variation in weight is explained by two categories of exposure. The
first category includes population suffering from seasonal food shortages or disease as
shown from Gambian studies [Billewicz 1967, Billewicz 1982]. The second category
includes healthy, well-nourished populations, and in this case the causes are not so
easily discerned. By scrutinizing 29 studies, Bogin [Bogin 1977] found that 22 showed
that maximum increments in height and weight do not occur at the same time of the
year. The authors of some of these studies speculate that children have a natural,
endogenous rhythm for growth in weight that is independent of both the seasonal
rhythm of growth in height and of seasonal variation in climate. That height is
increasing more during the summer season is reported by many authors [Dahlberg
1931, Broman 1942, Bogin 1978, Gelander 1994], and so do weight [Wretlind 1878,
Marschall 1971, Bogin 1978].
A month-of-birth effect on the height of both young people and adults has
been reported in a few studies. The validity of this phenomenon is less well
established and its causes are not clear. Henneberg and Louw made a cross-sectional
study of 1 165 boys and girls, 6 to 18 y old, attending schools in Cape Town, South
Africa. The participants in the study were all judged to be healthy and wellnourished. Those individuals born during the months of August to January were
found to average about 0.7 cm taller and 800 g heavier than individuals born from
February to July. The differences were statistically significant, though small
[Henneberg 1990]. Henneberg and Louw made a later study, where they measured 1
522 boys and girls, 6 to 18 y old, living in the poorest region of the Cape Province of
South Africa. Most of these participants showed evidence of chronic undernutrition.
Even so, a significant month-of-birth effect was evident in the data but the timing
was shifted. Individuals born during the months of November to April were taller
and heavier than those born during May to October [Henneberg 1993].
49
Weber et al. discovered a seasonal variation in body height of 507 000 18-y-old
Austrian conscripts related to their month of birth with a difference between the
highest and lowest mean value of 0.6 cm with the highest value for the month of April
[Weber 1998].
Infections
What all trained paediatricians recognize is the relation between weak growth and
infections. Cole has studied the phenomenon for children younger than 2 years of age.
Diarrhoea, infections in lower respiratory tract and measles all have a strong effect on
weight gain and reduced weight between 17 and 67 g/day [Cole 1989].
Infections are important in the way they affect and interact with nutrition.
Infection can result in gastro-intestinal influence and malabsorbation as a result. An
infection can also result in more infections by affecting the immune defence
[Waterlow 1994].
The impact of diseases
Disease acts in the range from stopping growth or severely decreasing growth velocity
to having no impact at all on growth. Some examples are given here.
Final height was reduced only in children with Crohn’s disease. Height velocity
was significantly decreased, between 6.0 years of age and the onset of puberty in
children with ulcerative colitis. In children with Crohn’s disease height velocity was
significantly decreased from 3 years to the end of pubertal period. Peak height velocity
was assessable in 112 children and it was on average delayed, especially for children
with CD [Hildebrand 1994].
Coeliacs with gastrointestinal symptoms treated before adulthood reach a mean
height similar to that of the normal population, and their final height is slightly better
(3 cm) than that of coeliacs who have not received treatment. This difference appears
to occur only in males and might be due to a more rapid evolution of puberty in girls
on a gluten free diet [Caacciari 1991]. Growth retardation is a major complication of
chronic renal insufficiency in children [Rodriguez-Soriano 1991]. Respiratory
conditions, including asthma, and height in primary school are studied in the UK and
the impact is presented [Somerville 1993].
Diurnal variation
Diurnal variation of standing height results in higher values in the morning and lower
values in the evening [Ashizawa 1990, Lampl 1992]. This is important to recognize
when measuring children in order to survey them.
1.4.3 Weight from birth to adulthood
Trying to conclude the cause of the epidemic of obesity Skelton says: When people
ask: ”What is causing this epidemic – genes, diet, exercise, TV, fast food?” the answer
is ”Yes – all of them!” The complex interplay of genetic predispositions in an
environment geared to making us to eat more and exercise less has led to our recent
crisis [Skelton 2004]. Long time ago, before this secular trend started Börjeson told us:
50
”Under normal conditions, obesity arises almost only when there is a genetic
disposition.” [Börjeson 1976]
Nature
It is widely acknowledged that genetic factors play a role in obesity, and that fatness is,
to some extent ’heritable’. This topic has formed the subject of several reviews in the
last decade [Maes 1997]. The influence of heredity on obesity has been demonstrated
in family studies [Court 1977, Merritt 1982]. Especially those by Mossberg [Mossberg
1989, DiPietro 1994] by Shields [Shields 1986] and by Börjeson [Börjeson 1976] and
Stunkard et al. [Stunkard 1990] on twins and by Stunkard and Sörensen [Stunkard
1986] by adopted children. From a study in 1999 Stunkard concluded: The result
suggest that genetic influences on the body weight of infants may be independent of
those that influence BMI in adults, a circumstance that could complicate the search for
genetic determinants of obesity [Stunkard 1999].
Data from Kaplowitz et al. suggested that this relationship may be stronger
between mothers and their offspring than fathers and offspring, and that the motheroffspring relationship strengthens over time [Kaplowitz 1988]. Whitaker et al. found
that parental obesity was a more important predictor of offspring obesity earlier in
childhood (<6 y), becoming less important with increasing age of the child [Whitaker
1997]. Data from Lake et al. showed that parental obesity influences tracking if both
parents are obese [Lake 1997].
Other types of studies: all other studies identified, family and sibling
[Wilkinson 1977, Vuille 1979, Garn 1981, Weinberg 1982, Khoury 1983, Rosenbaum
1987, Burns 1989, Grinker 1989, Wada 1990, Moll 1991, Ayatollahi 1992, Tienboon
1992, Maffeis 1994, Rice 1995, Klesges 1995, Guillaume 1995, Hashimoto 1995,
O’Callaghan 1997, Goran 1998, Maffeis 1998, Prescott-Clarke 1998] twins [Brock
1975, Allison 1994] and adoption studies [Stunkard 1986, Sörensen 1992, Vogler
1995] support the view that genetic factors contribute to fatness level, although the
estimated effects vary between studies.
There would seem to be little doubt that the development of fatness has a
genetic component. What is far less clear, and is important from a public health
perspective, is how much of the variation in fatness is due to the genetic component,
and how much to the environment. Also important is the influence of the environment
on the genetic component, i.e. gene-environment interactions. But if an individual is
susceptible to being fat, the type of exposition factors and the power of them will
decide what the result will be.
Nutritional factors
Breastfeeding and growth
More than 25 years ago, Waterlow and Thomson speculated that ”growth faltering”
around the age of 3 months of breastfed infants was reflecting inadequate energy
supply [Waterlow 1979] and so did Duncan et al. [Duncan 1984]. Now we know that
this problem, which seemed to be obvious, was not real, and mostly dependent on that
an improper norm was used (the old NCHS standard from the 1920s). Among the
earliest to address this issue was the Cambridge group that found that though breastfed
51
Gambian infants were smaller than Cambridge infants, the weight gain patterns by the
two groups were remarkable identical expressed as percent of the reference value for
weight-for-age. This obvious ”faltering” did reflect a tendency of breastfed infants to
show an accelerating growth in early months of life. In other words: the pattern
described in the international reference and in many other curves did not equal reflect
the growth of breastfed infants in industrialized and developing countries [Whitehead
1984]. This discovery has been made by successors [Dewey 1995], who pointed out
the necessity of a new reference for infants [de Onis 1997, de Onis 2003]. 1999 a
proposal of a new reference was made from Garza [Garza 1999] and from Williams
[Williams 2002].
An example is given by Cole. Out of 252 infants were 63 bottle-fed, 69 breastfed/bottle-fed and 120 breast-fed only for more than 24 weeks. 120 infants that were
breast-fed more than 24 weeks, with solid food introduced at an average age of 15
weeks, were followed from birth, being weighed every 4 weeks up to 1 year of age.
Those with long breastfeeding were somewhat heavier than the British reference
[Freeman 1995, Cole 1998] and crossed the centiles upwards in order to reach +0.3
SDS at age 2 mo but afterwards crossed the centiles downwards -0.2 SDS at age 12
mo [Cole 2002].
The literature on the relationship between early infant feeding and growth
shows that after the first 3 or 4 months, breast-fed infants in the developed world are
lighter than formula-fed infants with markedly lower adiposity. There is some
evidence of a slightly lower rate of linear growth over the first year or so. These
differences in weight and length do not appear to persist beyond the first years of life.
In the developing world the situation is very different. The growth curves of breast-fed
infants of malnourished mothers may falter between the third and sixth month of life.
However, the generally poor quality of the supplementary foods offered in the
developing world, and the increased risk of diarrhoeal infections, mean that
supplementary feeding before the age of 6 months is unlikely to lead to a growth
advantage and may well instead lead to growth faltering [Rogers 1997].
Scholtens recently stated that compared with non-breast-fed children, breastfed children tend to have a lower BMI at about 1 y of age. The association between
breastfeeding and BMI between 1 and 7 y of age was negligible, while a high BMI at 1
y of age was strongly associated with high BMI between 1 and 7 y of age.
Physical activity and inactivity
”Normal” physical activity is by definition desirable. But an amount of physical
activity can be counteracted by an amount of physical inactivity. Physical inactivity
could be described as the time spent watching television or the time spent in front of a
computer at home [Dietz 1985, Robinson 1993, Gortmaker 1996, Robinson 1996].
For 8- and 13-y-olds in the early 1980s in Sweden Sunnegårdh concluded: In
spite of increased sports involvement in Swedish children over the years, a reduction
of physical activity was indicated by a tendency towards a higher body fat content
despite a lower mean energy intake as compared with such values obtained 10–15
years ago in Swedish children of equal ages [Sunnegårdh 1986]. By studying 2 200 9y-old boys and girls, Harrell tells us: Children from a higher socio-economic status
(SES), especially boys, reported a greater proportion of sedentary activities than lower
SES children. Significantly more nonobese than obese children were reported a
vigorous (high-intensity) activity as one of their top three activities [Harrell 1997].
52
There is a relationship between physical inactivity and adiposity in pre-pubertal boys
Maffeis has found [Maffeis 1997].
The activity for 52 children, aged 4 to 8 y, was measured during the first 3 days
of life, and in this survey, neonatal adiposity was not significantly correlated with
parental adiposity, neonatal physical activity, or gender, nor was neonatal activity
significantly correlated with adiposity in childhood. Neonatal adiposity did not predict
adiposity in childhood. However, in a stepwise multiple regressions, parental adiposity
and the children daytime high activity levels were significantly associated with
childhood adiposity. The age or gender of the child did not significantly correlate with
childhood adiposity. As parental adiposity increased or daytime high activity of a child
decreased, the adiposity in a 4- to 8-y-old child was likely to increase [Berkowitz
1985].
Westerståhl et al. compared 16-y-olds in 1974 and 1995 and noticed that
although more adolescents participated in leisure-time sports activity in 1995 than in
1974 the BMI increased from 1974 to 1995. One explanation could be that the lifestyle
of adolescents between sports training sessions may have become more sedentary
[Westerståhl 2003, Westerståhl 2003]. This could fit the finding among conscripts by
Rasmussen et al., that among 18-year-old men the muscle-strength has increased at the
same time that BMI increased over time [Rasmussen 1999].
Psychosocial factors
The classical overview of body size, of infants and children, around the world, in
relation to socio-economic status, was made by Meredith [Meredith 1984]. The
relation was convincingly shown. This relation is obvious both in developing countries
[Arteaga 1982] as in developed [Rolland-Cachera 1986]. In Sweden Mellbin has
shown that psychosocial stress and psychosocial problems are related to an increase in
relative weight in schoolchildren [Mellbin 1989, Mellbin 1989].
Socio-economic status has a direct influence on weight status in developing
countries. Low-weight subjects are encountered in low-income families, and obesity is
more frequent in high-income families [Arteaga 1982]. By contrast, in countries with
abundant food supplies, weight status and economic status is inversely related
[Arteaga 1982], and a lower prevalence of obesity is recorded in higher social classes.
The same inverse association is recorded in the study of 18-y-old males in Sweden
[Oldenburg 1999]. A large number of lifestyle characteristics, such as eating habits
and physical activity, may account for differences between socio-economic groups.
SES and obesity – not a simple relation
In a review, Parson showed that the relation between obesity and SES was not clear
[Parson 1999].
Sobal and Stunkard have published a comprehensive review about SES and
obesity, which includes 144 studies from all over the world, of men, women and/or
children [Sobal 1989]. Despite the fact that these studies used a range measure of SES
and fatness, and spanned over a period up to 45 y, a strong negative relationship
between SES and fatness was obvious in women across the developed world, such as
that fatness increased from higher social groups (professional and management) to
lower social groups (unskilled and manual). Of 54 studies, 85% demonstrated a
negative relationship, 13% showed no relationship, and only 2% a positive relationship
53
(greater fatness among higher social groups). Among men and children, however, the
relationship was far less consistent. In stark contrast, in developing countries, not a
single study found a negative relationship between SES and fatness in women. Indeed,
the majority reported among men and women, and in children.
Since 1995, children have been included in the Health Survey. Combined
data from 1995 show no pattern in social class and prevalence of overweight in
males aged 2–15 y or 16–24 y or females aged 2–15 y. In females aged 16–24 y,
prevalence of obesity was higher in manual and non-manual social classes.
The current systematic review examines evidence that low SES in childhood
promotes the development of fatness later in life.
Nine studies (12 papers) spanned over a period within childhood [Wilkinson
1977, Kramer 1981, Garn 1981, Kramer 1986, Power 1988, Lasker 1989, Agras 1990,
Power 1997, Power 2000], girls [Wilkinson 1977, Zack 1979, Berkowitz 1985,
Weststrate 1986, Kramer 1986, Agras 1990] and three observed a negative relationship
between SES and fatness at age 17 y [Wilkinson 1977, Weststrate 1986, Agras 1990].
Two studies found no associations between SES and fatness [Berkowitz 1985, Agras
1990], and in one study the relationship was not explicit [Zack 1979]. One study
observed a negative relationship in boys and girls separately and a further study found
a negative relationship for girls but not boys [de Spiegelaere 1998].
The influence of social factors other than social class, such as family size,
number of parents at home and child-care, has not been well researched.
Breastfeeding as a protective factor against obesity and a promoting factor for
stature
Compared to bottle-fed infants, breast-fed infants in the 1920s and 1930s were taller in
childhood and adulthood. As stature is associated with health and life expectancy, the
possible long term impact of infant feeding on adult mortality patterns merits further
investigation [Martin 2002].
Agras found that breastfeeding only protects against early adiposity to the age
of 6 months [Agras 1987] and later he found that predictive factors at 6 years of age
were adiposity at birth, with greater adiposity at birth predicting greater fatness at 6
years, parental education, with less education associated with fatness, and a prolonged
period of breastfeeding with delayed introduction of solid food [Agras 1990].
Breastfeeding is protecting for obesity at age 6 y [Gunnarsdottir 2003] and up
through adolescence [Elliot 1997, Gillman 2001, von Kries 1999] and even up to
adulthood that is studied within the Boyd-Orr cohort in the US [Bray 2006, Martin
2002].
The protective effect of breastfeeding on future overweight seems to be
explained only partially by decreased maternal feeding restriction [Taveras 2006].
Parsons summarizes the literature
Parsons summarize a review of the literature 1999 as follows:
* Risk factors for obesity included parental fatness, social factors, birthweight, timing
or rate of maturation, physical activity, dietary factors and other behavioural or
psychological factors.
54
* Offspring of obese parent(s) were consistently seen to be risk of fatness, although
few studies have looked at this relationship over longer periods of childhood and into
adulthood. The relative contributions of genes and inherited lifestyle factors to the
parent–child fatness remain largely unknown.
* No clear relationship is reported between socio-economic status (SES) in early life
and childhood fatness. However, a strong consistent relationship is observed between
low SES in early life and increased fatness in adulthood. Studies investigating SES
were generally large but very few considered confounding by parental fatness. Women
who change social class (social mobility) show the prevalence of obesity of the class
they join, an association that is not present in men. The influence of other social
factors such as family size, number of parents at home and child care has been little
researched.
* There is good evidence from large and reasonable long-term studies for an
apparently clear relationship for increased fatness with higher birthweight, but in
studies that attempted to address potential confounding by gestational age, parental
fatness, or social group, the relationship was less consistent.
* The relationship between earlier maturation and greater subsequent fatness was
investigated in predominantly smaller, but also a few large studies. Again, this
relationship appeared to be consistent, but in general, the studies had not investigated
whether there was confounding by other factors, including parental fatness, SES,
earlier fatness in childhood, or dietary or activity behaviours.
* Studies investigating the role of diet or activity were generally small, and included
diverse methods of risk factor measurement. There was almost no evidence of an
influence of activity in childhood on later fatness. No clear evidence for an effect of
infant feeding on later fatness emerged, but follow-up to adulthood was rare, with only
one study measuring fatness after 7 y. Studies investigating diet in childhood were
limited and inconclusive. Again, confounding variables were seldom accounted for.
* A few, diverse studies investigated associations between behaviour or psychological
factors and fatness, but mechanisms through which energy balance might be
influenced were rarely addressed.
The authors, after scrutinizing all the relevant studies, concluded: Many of the risk
factors investigated are related, and may operate on the same causal pathways.
Inherent problems in defining and measuring these risk factors make controlling for
confounding, and attempts to disentangle relationships more difficult. A given risk
factor may modify the effect of another, and cumulative effects on the development of
obesity are likely, both over time for specific risks, or at any particular time over a
range of risk factors. An additional approach to addressing these issues may be to use
large samples on which more basic measures of risk factors have been collected.
[Parsons 1999].
55
Prognosis of obesity in infancy and childhood
Studies examining individual fatness development also suggest an early origin of adult
obesity [Rolland-Cachera 1984, Prokopec 1993, Whitaker 1997]. Children have a
rapid increase in BMI during the first year of life. The BMI subsequently declines and
reaches a minimum by the age of 6 y, before beginning of a gradual increase up to the
end of growth. The point of minimal BMI has been called the adiposity rebound
[Rolland-Cachera 1984, Dietz 1994]. The age at adiposity rebound is associated with
BMI development. An early adiposity rebound is associated with an increased risk of
adult obesity [Rolland-Cachera 1984, Prokopec 1993], and this was found to be
independent of both parental obesity and the child’s BMI at adiposity rebound
[Whitaker 1997].
A longitudinal study of nutrition and growth has investigated the early
determinants of the age at adiposity [Rolland-Cachera 1995]. The only significant
association between nutritional intake at the age of 2 y and age at adiposity rebound
was a high percentage of energy provided by protein in the diet, i.e. the higher the
percentage of protein, the earlier the adiposity rebound. By the age of 1 y, the infant
diet is characterized by a high intake of proteins and a low intake of fat [RolandCachera 1999]. To compare with the content of human milk where 7% of the energy is
protein and 50% is fat. Childhood obesity is characterized by an increased stature and
muscle mass [Valoski 1990].
450 children at age 6 months were followed to the age of 9 years. Weight gain
in infancy is also a poor predictor of 9-y-old obesity. Changes from obese to nonobese or lean are often not linear. There is evidence that impending or actual obesity
begins at ages 6 to 9 y with some predictability provided as early as age 2 y for girls,
age 3 y for boys [Shapiro 1984].
226 children were followed from the first year of life to age 4 y. 23 children
were considered obese at some point during the first year of life but only 4 at age 4 y.
Only 3 of 23 of the obese children during the first year were still obese at age 4 y. The
weight and length of the children obese at 0–1 y of age were significantly increased at
age 4 y. Over-nutrition occurred during the first year in 26 infants and the number of
obese infants in this group was significantly increased at age 7–12 mo and of
overweight children during the first two years of life. At age 4 y, however, none of
them were either obese or overweight [Sveger 1978].
27 899 infants were followed from birth to age 7 y and a pattern of rapid
weight gain during the first 4 months of life was associated with an increased risk of
overweight status at age 7 years, independent of birthweight and weight attained at age
1 year [Stettler 2002].
In a 40-year follow up of 500 overweight children Mossberg found that the
degree of obesity in the family and the degree of overweight in puberty were the most
important predictors of adult body overweight. Factors of minor importance were
birthweight of more than 4 500 g and the age of onset of obesity. Whether an obese
infant will stay obese depends very much on the degree of obesity in the family,
especially in the mother [Mossberg 1989].
A review of the literature about whether obese children become obese adults
you find in Serdula et al. [Serdula 1993, Serdula 1993] and also by Sörensen
[Sörensen 1988]. The proportion remaining obese in adulthood is reported to be
between 40% and 85% [Lloyd 1961, Court 1977, Taitz 1983].
56
Dealing with the question whether chubby infants become obese adults and
coming to somewhat different results, see further [Charney 1976, Börjeson 1976,
Rolland-Cachera 1989, Rolland-Cachera 1990, Valdez 1996].
A large national cohort of children studied from birth to 36 y was used to test
the predictive value of childhood obesity for obesity in adult life. Only 21% of obese
36-y-olds had been obese at age 11 y, and even when associated social factors were
taken into account the correctly predicted percentage was much lower than the
prediction rate achieved using body mass data from age 26 y. At the age of 36 y, 3 754
of the 5 362 members of the study population were contacted, and out of these, 3 322
were successfully interviewed in their homes by specially trained nurses [Braddon
1986].
Early growth and abdominal fatness in adult life was studied by Law. 845 men
born between 1920 and 1930, and 239 men born between 1935 and 1943 were
followed-up. The tendency to store fat abdominally, which is known to increase the
risk of cardiovascular disease and diabetes independently of obesity, may be a
persisting response to adverse conditions and growth failure in fetal life and infancy
[Law 1992].
Overweight children tend to be taller, have advanced bone age, and mature
earlier than non-overweight children, because height gain accelerates or follows
shortly upon excessive weight gain [Dietz 1998], but this do not result in higher final
height but just final height at an earlier age.
Weight loss
Is the rate of eating disorders increasing, or not?
Obesity and overweight among children and adolescents are becoming increasingly
prevalent around the world [Lobstein 2004], and increased rates of eating disorders are
reported for example in the United States and Sweden [Lucas 1994, Edlund 1994,
Edlund 1999]. However, in a review of the existing literature, Fombonne found no
epidemiologic evidence for an increasing frequency of eating disorders or anorexia
[Fombonne 1995, Fombonne 1996]. Fombonne concluded: ”Although based on a
limited number of studies, the empirical evidence does not support secular changes in
the incidence of bulimic disorders. In keeping with this conservative conclusion, it is
noted that high rates of dieting and body dissatisfaction were already reported 30 years
ago among adolescent females.” [Fombonne 1996] On the other hand, more recent
studies conclude that the rate of eating disorders really is increasing for example in
Japan, the United States and in Europe [Nakayama 2001, Kohn 2001, Morande 1999,
Martin 1999, Bulik 2006, Steiner 1998].
In Sweden, the risk for anorexia is four times higher among women than men
for individuals born between 1935 and 1958 [Bulik 2006]. The prevalence of anorexia
defined by DSM-IV [First 1997] criteria is 1.2% for females and 0.3% for males in
this population. The rate is higher among those born after 1945 [Bulik 2006].
Weight concerns and weight loss practices among schoolchildren
When asking schoolchildren about weight concerns, Edlund et al. [Edlund 1994,
Edlund 1999] found that the rate of weight concerns is increasing and that these
57
concerns are beginning at an earlier age among Swedish children today. Among
adolescent girls, 60–70% reported having been on a diet during the previous year [Hill
1993]. Also, dieting behaviour has been reported among preadolescent females
[Edlund 1996, Gustavsson-Larsson 1992] and is increasing with age [Edlund 1999].
To estimate the rate of various weight loss practices in U.S. adolescents Serdula
et al. [Serdula 1993] surveyed 11 467 high school students in 1989. They found that
44% of the females and 15% of the males reported that they were trying to lose
weight. From the same investigation Felts et al. [Felts 1996] concluded that onequarter of U.S. high school students perceived themselves to be overweight; threequarters of these students are trying to lose weight. French et al. [French 1995]
reported from the U.S. that the frequency of dieting, weight change history, and
specific weight loss behaviours in a population-based sample of 1 015 female 9th–12th
graders was caused by healthy weight loss behaviours more often than unhealthy
weight loss behaviours (healthy behaviours such as exercise, 32.4%; decrease in fat
intake, 26.0%; reduction in snack intake, 25.0%; and reduction in kilocalorie intake,
22.4%; and unhealthy behaviours such as fasting, 8.1%; diet pills, 5.4%; and vomiting,
4.4%). Obesity status and restrained eating scores were positively related to more
weight loss episodes, pounds lost, and weight fluctuations and to increased use of
healthy weight loss methods and weight loss programs.
In Australia Nowak et al. [Nowak 1996] surveyed 791 8th graders from private
schools in north Queensland. Only 41% of the girls and 54% of the boys were satisfied
with their weight; 52% of the girls and 27% of the boys wanted to lose weight. When
surveyed, 35% of the girls and 22% of the boys were trying to lose weight. One of the
conclusions from this study is that the drive to lose weight is already apparent among
12–14-y-old adolescents. In another Australian study, McCabe et al. [McCabe 2001]
investigated 1 266 adolescents (622 boys, 644 girls) with respect to their body image
and body change strategies, as well as the socio-cultural influences on these variables.
Girls were found to be less satisfied with their bodies and were more likely to adopt
strategies to lose weight, whereas boys were more likely to adopt strategies to increase
weight and muscle tone. Respondents with higher body mass index (BMI) had greater
body dissatisfaction and more weight loss strategies, but there were no differences
between BMI groups in weight gain or strategies to increase muscles. The influences
of the media on the desire to alter weight, as well as feedback from parents and both
male and female peers, were stronger for girls.
All these surveys deal with questionnaires to adolescents about weight
concerns or weight loss practices. They all show that these phenomena are frequent,
especially among girls. But, to the best of our knowledge, there are no surveys that
longitudinally have collected data of recorded weight loss. Furthermore, we have not
found comparable surveys that could catch the factual weight loss situation over time.
Body image among adolescents
Bergström et al. [Bergström 2000] report that in Sweden today, many adolescents,
especially girls, over-evaluate their body size. Another point of view comes from
Kaltiala-Heino et al. [Kaltiala-Heino 2003]. On the basis of surveys that cover a 20year period in Finland they state: ”Even if the adolescent population has gained
weight, they are less concerned at being overweight than earlier. It seems that
adolescents compare themselves rather to the peers close to them than to ideal models
58
provided by culture at large”. However, most studies report a stronger desire to be
thinner among those with a high BMI [Patton 1990, Hill 1995, Striegel-Moore 1995].
Anthropometry, body composition and body image in dieting and non-dieting
8–16-y-old Swedish girls was studied by Edlund et al.: The subjects participated in a 3
y prospective longitudinal study and were selected randomly after stratification for
grades from those scoring in the upper vs. the lower thirds of the Children’s Eating
Attitudes Test (ChEAT) score distribution. The ChEAT was completed 6 months
before the present study, together with a demographic and dieting questionnaire and a
questionnaire for the estimation of body size. In total 43% (n=52) admitted ever
dieting (”Dieters”) and 25% (n=30) admitted that they were currently trying to lose
weight. The anthropometric and body composition data indicated that ChEAT Highscores and Dieters were somewhat fatter than Low-scorers and Non-dieters, although
this pattern was not shown among the 8-y-olds or the 14-y-olds (High-scorers). The
mothers of the ChEAT High-scorers were found to be somewhat fatter than the other
mothers. A current vs ideal body shape discrepancy was shown for both High-scorers
and Dieters, with a larger discrepancy for the Dieters. All groups believed that their
parents were aspiring for a leaner body [Edlund 1999].
81 infants were investigated to evaluate the clinical significance of low rate of
weight gain. An organic aetiology was found in 28 infants. The remaining infants were
analysed with special respect to their psychosocial environment. Growth was assessed
by rate of weight gain. Thirty-four infants had subnormal rates. Comparisons were
made with 18 infants with low but normal rates and a control group of 72 infants.
Several psychosocial risk factors were over-represented in infants with subnormal rate
of weight gain such as employment, bad health in fathers, dependence on social
welfare etc. No significant differences were found in perinatal factors except for
birthweight. The magnitude of the load of adverse factors within the family was
measured as a score. A significant negative correlation was found between the score
obtained and the rate of weight gain [Kristiansson 1981].
Weights were retrieved from child health records for an annual cohort of 3 418
children, aged 18–30 months, to explore the relationship between deprivation and
weight gain. Children from deprived areas were smaller at all ages with a widening
gap: by 1 year of age, they were three times as likely as affluent children to be below
the third centile for weight [Wright 1994].
Catch-up growth
Catch-up growth was introduced as a term by Prader [Prader 1963]. Catch-up growth
may be defined as a height velocity above the statistical limits of normality for age
and/or maturity during a defined period of time, following a transient period of growth
inhibition. One of the first papers in paediatric literature on catch-up growth was
published by Bauer [Bauer 1954] where he described the growth pattern of 34 children
with nephritic syndrome.
The variability of catch-up growth, not only among suffering from diseases but
also among individuals with the same disorder, suggests that influences on catch-up
growth are multifactorial [Boersma 1997].
59
1.4.4 Longitudinal aspects of growth and later health
Every point on the curve of development of an individual contains the possibilities for
different development depending on what exposure that exists. If there is, on one
point, a decelerated situation, and not too strong impact, then a positive exposure
facilitates a catch-up. If the development has been optimal, a positive exposure just
maintains that positive development. A point or a period with retardation may lead to
an increased susceptibility and with unfavourable exposure in the future lead to disease
and premature mortality. More knowledge about this longitudinally relations has been
reported the last 15 years.
Poor socio-economic circumstances in early life may lead to biological
vulnerability in later life [Wadworth 1997], and adult health behaviours seem to have
socio-economic roots early in life [Lynch 1997].
During the last 15 years, however, a considerable body of evidence has
accumulated, suggesting that height at maturity is also an important predictor of the
probability of dying and of developing chronic diseases at middle and late ages
[Waaler 1984, Marmot 1984, John 1988, Forsen 1999]. BMI has similar predictive
properties [Heywood 1983, Waaler 1984, Martorell 1985, Payne 1992, Osmani 1992,
Srinivasan 1992].
Birthweight and childhood growth and their relations to later
health
Given the importance of birthweight for infant, childhood, and adult health [Barker
1992], 150–200 g social gradient in mean birthweight and 30% of births less than
2 500 g attributable to social inequalities [Spencer 1999] is a key public health issue.
Reductions in inequalities in infant mortality and many childhood and adult health
inequalities, key government health targets are unlikely to be achieved without a
narrowing of the social gradient in birthweight.
Paediatricians have long been familiar with the increased risk of mortality and
early morbidity of babies born very small or very early. These babies have a greater
risk of dying throughout the first year of life [Murphy 1992, Bernstein 2000].
In the last 20 years or so, there has been increasing evidence that size at birth is
also associated with later health, particularly with chronic degenerative diseases that
are major causes of death in middle and later life. The best documented relations are
those between smaller size at birth and higher death rates from coronary heart disease
and stroke [Rich-Edwards 1997, Osmond 1993, Stein 1996, Leon 1998, Eriksson
2001]. Smaller size at birth is also related to increased levels of cardiovascular risk
factors such as hypertension [Huxley 2000, Curhan 1996 women, Curhan 1996, men,
Pharoah 1998], type II diabetes mellitus [Phillips 1998], and hyperlipidemia [Barker
1993]. High birthweight is normally associated with long-term health. However,
people with high birthweight have higher death rates from prostate cancer [Ekbom
1996] and possibly breast cancer [Vatten 1996]. They may also be at risk of obesity
[Barker 1997, Parsons 1999] but some doubt about this [Allison 1995] and type II
diabetes mellitus [Dabelea 1999, Rich-Edwards 1999].
By reviewing the literature since 1956 and 40 years forwards, Law et al.
concluded that blood pressure is inversely related to birthweight in children and adults.
60
The positive results in neonates and the inconsistency in adolescence may be related to
the unusual growth dynamics during these phases of growth [Law 1996].
Other reports also show that growth and living conditions in childhood and
hypertension and cardio-vascular disease in adult life are related [Arnesen 1985, Blane
1996, Forsen 1999, Barker 2002].
By following 5 914 infants born in 1982 from birth to age 18 y, Victora et al.
concluded that height was primarily determined by fetal and infant growth. Weightrelated indices, including the fat/lean mass ratio, were more strongly influenced by
later growth. No clear critical windows of growth during which absolute tissue masses
are programmed could be identified [Victora 2006].
By examining all children born from 1977 to 2004 in Denmark (N=1.7 million)
Hviid and Melbye found that birth weight was inversely associated with a risk of
infectious disease hospitalization; among children aged 0–14 y, the risk of
hospitalization increased 9% for each 500 g-reduction in birth weight (increase rate
ratio = 1.09, 95% confidence interval: 1.09, 1.11). The effect was found to peak in
infancy and to persist until 10 years of age. It was present also in children born at term
(37–41 weeks of gestation) [Hviid 2006].
However, Barker’s conclusion that foetal origin of adult disease is a clear fact
has been questioned by Lucas [Lucas 1999].
Final height and its relation to mortality and morbidity
1993 Robert Fogel stated: ”Extensive clinical and epidemiological studies over the
past two decades have shown that height at given ages, weight at given ages, and
weight-for-height (BMI) are effective predictors of the risk of morbidity and mortality.
Until recently, most of the studies have focused on children under age 5 y, using one
or more of the anthropometric indicators at these ages to assess risks of morbidity and
mortality in early childhood, and it was at these ages that the relevance of
anthropometric measures was established most firmly.” [Fogel 1993] Many examples
of this exist [Sommer 1975, Chen 1980, Billewicz 1982, Kielmann 1983, Martorell
1985].
13 146 persons whose weights and heights were measured between the ages of
15 and 18 y in Hagerstown, Maryland, during the period 1933–1945 were studied. The
associations between growth velocities or attained height with mortality tended to be
inversed, although not statistically significant. These results are compatible with the
existence of positive associations of overweight in school-age children with long-term
morbidity and seem to allay fears that harm could come from increased growth rates in
childhood. Without jeopardizing growth, the avoidance of overweight in childhood
might reduce mortality in the middle age [Nieto 1992].
A study of 3 332 men and women born in 1946 investigated the relation
between blood pressure at age 36, and birthweight and BMI in childhood,
adolescence and adulthood. Low birthweight and high BMI at age 36 y were
independently related to high blood pressure. A reduction in the percentage of low
birth-weight babies born in the fourth decade of this century would only have a
negligible effect on the incidence of adult hypertension 30–40 years later [Holland
1993]. Cardiovascular disease and diabetes have been shown to be associated with
retarded growth during fetal life and infancy [Sveger 1978, Barker 1992].
Within the Nurses’ Health Study, where a prospective cohort of 121 700 US
female nurses aged 30–55 in 1976 was examined after a 14 years follow-up and the
61
conclusion was: The result support the hypothesis that height is inversely related to
the risk of coronary heart disease in women [Rich-Edwards 1995].
Observational evidence points towards positive associations of height with risk
of some cancers (e.g. prostate, breast, and colon) [Gunnell 2001], and inverse
associations with insulin resistance and coronary heart disease [Williams 1997, Davey
Smith 2001].
The Norwegian experience about height and mortality
It must be stressed that we are talking about averages, not individuals, but there are
clear links between height and health. In Norway, a survey by Hans Waaler showed
that tall men and women live significantly longer. For example, women aged 40–44 y,
those standing between 145 cm and 149 cm, had a mortality rate double that of women
165 cm to 169 cm tall. Norwegian men aged 55 to 59, who stood 150 cm to 155 cm,
had mortality rate twice that of men whose height was 185 cm to 189 cm [Waaler
1984]. Se Jousilathi [Jousilathi 2000].
Fogel reproduces a diagram by Waaler and shows that short Norwegian men
between 1963 and 1979, aged 40–59 y (by establishing the predictive power of height
and BMI with respect to morbidity and mortality) were much more likely to die than
tall men. Indeed, the risk of mortality for men with heights of 165 cm was on average
71% greater than that of men who measure 182.5 cm. The Norwegian curve has
relative risk that reach a minimum of between 0.6 and 0.7 at a height of about 187.5
cm and a relative risk of about 2 at about 152.5 cm. By comparing this to another
result of a survey of US soldiers Fogel concludes: the relative risk of morbidity and
mortality depends not on the deviation of height from the current mean but from an
ideal mean: the mean associated with full genetic potential [Fogel 1993].
Waaler has also studied the relationship in Norway between BMI and the risk
of dying in a sample of 1.7 million individuals. Although the observed values of the
BMI ranged between 17 and 39, over 80% of the males at age 40 and older had BMIs
within the range 21–29. Within the range 22–28, the curve is fairly flat, with the
relative risk of mortality hovering close to 1.0. However, at BMIs of less than 22 and
over 28, the risk of dying increases quite sharply as the BMI moves away from its
mean value. The BMI curves are much more symmetrical than the height curves
indicating that high BMIs are as risky as low ones. The Norwegian survey shows that
an adult male with a BMI of 25 who is 164 cm tall runs about 55 percent greater risk
of dying than a male who is 183 cm tall and also has a BMI of 25. A BMI of 25 is
”ideal” for men in the neighbourhood of 175 cm, but for tall men (greater than 183
cm) the ideal BMI is between 22 and 24, while for short men (under 168 cm) the
”ideal” BMI is about 26.
1.5
GROWTH SURVEYS, GEOGRAPHICAL COVERAGE
AND DESIGN
It is often unclear if surveys of growth in different countries have a national
representative purpose or not. ”Reference values” are created and the question of
whether it is a description of a local population, and to the best ambition, in spite of
that, representative for a country, is seldom declared. It is with this distinction between
62
surveys based on local populations and based on some type of national representative
sample that the following overview shall be read. Longitudinal surveys are very rare.
1.5.1 Growth surveys in Sweden
From the first surveys of children’s growth up until today, different attempts have been
made to describe the national growth-situation of children and adolescents.
Surprisingly, the ambition was highest in the early studies. Key, for example,
performed in 1883 an impressing data collection, from a national point of view, with a
discussion about the survey’s weakness and what is desirable in a future of
geographical and social coverage in order to get a national representative material
[Key 1885]. But later, in 1942, Broman et al. are doing something in the lines af whet
Key wanted [Broman 1942]. Anyhow, it is not until a study by Lindgren et al. that a
national representative sample of data was collected, for children born in 1967, with
height and weight data from school health records, longitudinally, from age 7 y to age
16 y [Lindgren 1986].
Table 5. Surveys of growth in Sweden
Author
N
Ages
Birth years
Measures
Wretlind 1878
Petterson 1882
Key 1885
Dovertie 1895
Hultcrantz 1896
Steenhoff 1900
Norinder 1907
Dovertie 1914
Sundell 1917
Törnell 1920
Ljunggren 1925
Kjerrulf 1927
Dahlberg 1931
Ljunggren 1933
Broman 1942
Samuelsson 1971
Ljung 1974
Karlberg 1976
Persson 1985
Sunnegårdh 1986
Lindgren 1986
Mellbin 1989
Mellbin 1989
Cernerud 1991
Karlberg 1994
Aurelius 1994
Werner 2006
Werner 2006
38
1 675
18 019
485
1 896
1 331
422
2 313
18 000
523
2 703
?
962
63 497
8 441
1 401
740
212
572
682
2 907
944
5 399
10 127
3 650
2 471
3 579
3 107
7–16y
0–1y
7–21y
8–15y
21y
7–15y
8–15y
9–16y
7–14y
9–16y
7–13y
7–13y
8–13y
7–14y
0–18y
4,8,13y
10–16y
0–25y
4,8,13y
8,13y
7–16y
7–15y
4–18y
7,10,13y
0–18y
0–6y
7–18y
0–19y
1863–67
1860s
1862–76
1880–87
1875
1885–92
1891–98
1897–1904
1902–9
1903–10
1911–17
1906–12
1916–21
1913–20
1920–38
1954,-59,-63
1955
1954–55
1967,-72,-76
1972, -67
1967
1963
1965–67
-33,-43,-53,-63
1973–75
1980
1973
1981
W
W
H,W
H,W
H
H,W
H,W
H,W
H,W
H,W
H,W
H,W
H
H,W
H,W
H,W
H,W
H,W,h
H,W
H,W
H,W
H,W
H,W
H,W
H,W,h
H,W,h
H,W
H,W,h
Location and remark
Göteborg, girls
Uppsala, born at hospital
Stockholm + other places
Kristianstad
All of Sweden, conscripts
Sundsvall
Hjortkvarn
Jönköping, boys
Stockholm
Norrtälje
Skåne
Stockholm, girls
Uppsala, one year longitudinal
All of Sweden
Stockholm, Malmö, Östergötland
Västerbotten
All of Sweden, longitudinal
Solna
Västerbotten,Uppland,Västergötland
Västerbotten, Uppsala, Älvsborg
All of Sweden, longitudinal
Uppsala, longitudinal
Uppsala, longitudinal
Stockholm
Göteborg, longitudinal
Stockholm, longitudinal
All of Sweden, longitudinal
All of Sweden, longitudinal
Measures: H=height, W=weight. h=head circumference
Surveys with locations on different places are marked in bold
63
Local surveys
The pioneer in Sweden is Wretlind (1878) – the first person in Sweden who weighed
schoolchildren regularly. He investigated girls in Gothenburg. Without taking height
under consideration, he registered mean values for body weight for age 7 y and 8 y in
spring and autumn for the three schools, finding a seasonal variation in their weight
[Wretlind 1878]:
”För att få en säkrare grund för mitt omdöme om skolgångens
inflytande på flickornas helsa och utveckling, än som kan erhållas
genom speciella iakttagelser af ofvan antydda art, har jag vid de
omnämnda mönstringarna äfven tagit flickornas kroppsvigt medels en
vanlig decimalvåg. Jag har nämligen antagit, att stillastående i
kroppsvigt eller ännu mer minskningen af densamma vore det bästa
beviset för en sjuklig eller åtminstone ganska svag kropp”.
E. W. Wretlind 1878.
Weight, during the first year of life, was investigated by Pettersson for children in
Uppsala. 1 675 children born in hospital were investigated, thus a strongly selected
sample [Pettersson 1882]. Schoolchildren in Kristianstad were surveyed by Dovertie
[Dovertie 1895] and in another survey by him in Jönköping, 2 313 boys aged 9–16 y
[Dovertie 1914]. In a survey in Sundsvall, with 654 boys and 677 girls, age 8 to 15 y,
height was measured without shoes on and weight with clothes on, and because of that
body weight was counted minus 1/17 of the weight of body and clothes together.
Means for height and weight were presented for schoolchildren, inspired by Hertel in
Copenhagen who had done this as a routine already in 1880 [Steenhoff 1900]. In the
spring of 1906, 422 schoolchildren were surveyed in Hjortkvarn in the middle of
Sweden. These were children from the countryside in Närke [Norinder 1907]. In 1917
Sundell reported from a huge survey of 7–14-y-olds (for example, it included 2 366
10-y-olds) [Sundell 1917]. In Norrtälje, Törnell surveyed 523 9–16-y-olds in 1919
[Törnell 1920]. Another early investigator of weight of infants was Höjer, 0 to 1 y,
[Höjer 1926].
Within a study of child health and nutrition in a northern Swedish county
medical and anthropometrical examinations were carried out [Samuelsson 1971]. 1
401 children, 4-, 8- and 13-y-olds, were selected from three areas of Västerbotten
county that differed from one another geographically and in part socio-economically;
namely, one urban area and two rural areas. All the examinations were performed
during the autumn of 1967 and thus the children were born in 1954, 1959 and 1963.
Within the so called Solna-study, a prospective longitudinal study within a
multi-centre European study, the somatic development of children in a Swedish urban
community was studied. At first reported from birth to age 16 y [Karlberg 1976], but
then up to age 25 y by Taranger et al. [Hägg 1991]. Data on somatic development
were collected for 212 children (122 boys and 90 girls) born between April 1955 and
March 1958. 183 of 212 newborns were recruited antenatally by asking every fourth
pregnant woman at the antenatal clinic of the community chosen for the study. The
other 29 children constituted a pilot group and were the first children selected for the
study. At age 16 y, 170 children (84.4%) were still being followed in the study.
They were examined at the following ages: 1 month (4 weeks), 3 months (13
weeks), 6 months (26 weeks), 9 months (39 weeks), 12 months, 18 months, 2 years
64
and then subsequently once a year. In order that the investigation may be considered to
represent true ages the following tolerances regarding time have been aimed at: for the
age of 4 weeks±2 days; for 3–18 months ± 1 week; for 2–3 years ±2 weeks and
subsequently - 2 weeks and + 3 weeks. It is the material from this study that construct
the national growth charts used in Child Health care and School Health care all over
Sweden between 1973 and 2001.
In the city of Uppsala, Mellbin et al. conducted a study in 1989 [Mellbin 1989,
Mellbin 1989]. The study comprises two materials: First, 971 children born in 1963 in
the city of Uppsala and for whom weight data were available from the first year of life,
and after the first school health examination at the age of 7 y they were followed up at
10, 13 and 15 y. At the last examination 944 children (97%) could still be included in
the study. Second, a sample of 5 399 Swedish schoolchildren (all schoolchildren in the
municipality of Uppsala born in 1965–67 who were residents in this municipality both
in the first and fourth grade were included (2 738 boys and 2 661 girls). The children
were then followed up to age 18 y.
By collecting data from school health records for 7-, 10- and 13-y-old children
in Stockholm born in 1933, 1943, 1953 and 1963, Cernerud described the
development for height and weight over time by using samples of about 25% from
these four Stockholm birth cohorts [Cernerud 1991].
The population-based reference values from Göteborg, from birth to 18 y of
age, is based on a study population of 5 111 final grade school children who were born
in Göteborg, or in the surrounding areas. Data were collected from child health and
school health records. Out of this population was 4 488 born in 1974. Information on
319 girls and 304 boys was not collected, since some of them were not willing to or
failed to attend the last investigation at school. Exclusion criteria, such as lack of
information at birth, multiple births, prematurity and growth disorders were applied in
the selection of subjects. 1 849 boys and 1 801 girls remained [Albertsson-Wikland
2002]. Body mass index reference values were reported in 2001 [ Karlberg 2001]. This
material has been the reference values within Child Health Care and School Health
Care since 2001.
Surveys that cover different geographical populations in Sweden
First performed was the classical investigation by Key [Key 1885]. The aim was to
shape some reference-values for children’s growth in Sweden, from age 7 y to 21 y.
The population who was investigated contained both pupils from State secondary
schools, 14 590 boys, (in Stockholm) and girls’ high schools, 2 971 girls, (from
different places in Sweden), and from preparatory schools and elementary schools, 458
boys and girls, (in Stockholm). Key pleads the necessity of such investigations in
every country, with regard to the influence of race, climate, and social and hygienic
conditions on bodily development. An interesting point is that Key was already
maintaining the need of investigations in different social classes of the population with
different living conditions since the development of the child at different ages can
differ quite a lot from one social level to another.
65
”Längdmåttet har såsom komitén i sitt frågeformulär begärde, tagits,
sedan skodonen blifvit afdragna. Vid vägningarna hafva ej kläderna
frånräknats. I allmänhet, eller åtminstone ofta, torde de väl å gossarna
hafva utförts, sedan rockarna eller jackorna blifvit aftagna; men
bestämda uppgifter härom saknar jag”.
A. Key 1885
Heights for conscripts, at age 21 y, are presented by the professor in anatomy who was
very active examining conscript data from late 19th century and forwards [Hultcrantz
1896, Hultcrantz 1927].
Ljunggren made a huge survey of 63 497 schoolchildren aged 7–14 y from
different places in Sweden [Ljunggren 1933].
”Barnen vägas och mätas fullt nakna med undantag av de större
flickorna, som få vara iklädda byxor, ev. linnen. Vikten av dessa plagg
fråndrages sedan de funna viktsiffrorna. Förrättningen skall alltid äga
rum vid samma tid på dygnet, ty såväl längd som vikt varierar under
dagens lopp. Helst bör förättningen äga rum samma månadsdag, så att
man lätt kan beräkna ökningen per månad. Vid mätning av
kroppslängden användes ett vid väggen fastsatt, justerat måttband av
trä eller metall. Härvid tillses att barnet står i full sträckställning med
lätt utåtriktade fötter samt med de slutna hälarna, sätet, ryggen och
huvudet intill måttbandet. Huvudet hålles så, att den normala
blickriktningen är vågrät. Avläsning bör ske med hjälp av vinkelhake
och vattenpass”.
C.A. Ljunggren 1933
Broman et al. had the ambition to establish new Swedish norms for height and weight
as a basis for practical hygienic and medical work. Their material, in total 8 441
individuals, was collected during 1938 and 1939: In Stockholm 4 379 individuals (age
0 to 18 y), in Malmö 2 662 individuals (age 1 to 18 y) and in the county of
Östergötland 1 400 individuals (age 7 to 14 y). The material was cross-sectional and
all the children were measured once standardised. A comparison between the figures
for the children from different regions shows small differences. Stockholm comes first
both in regards to height and weight and is followed by Malmö. The figures for
Östergötland are the lowest. Figures were found to indicate that children born in the
autumn have somewhat greater height and weight at birth than children born in the
spring [Broman 1942]
Ljung et al. made a study with the ambition to create new standards that should
replace the standards made up from Broman’s work for schoolchildreny. The material
used is the same as for the study of Lindgren below. The sample includes two main
groups: One group consists of pairs of twins, the other of one or two classmates to the
twins – of the same sex, born in the same month and year (mostly in 1955, a few in
1954) as the twins they were to match. The samples were drawn from urban
elementary schools where the possibilities of making uniform arrangements for the
different tests and observations were supposed to be better than in rural schools.
[Ljung 1974].
66
In 1985 Persson presented a material with heights and weights of 572 children
in ages 4, 8 and 13 y from three communities in Västerbotten, Uppland and
Västergötland [Persson 1985].
The aims of a study by Sunnegårdh was to study the height, weight and
skinfold thickness in randomly selected 8- and 13-y-old children in Sweden and to
compare the results with earlier figures for Swedish children. The study was
performed in 1980–81. The investigation comprised 682 children born in 1967 and
1972 in four different regions of Sweden, namely Vilhelmina and Dorotea, Umeå,
Uppsala and Tierp, and Vänersborg and Trollhättan. Altogether 789 children were
selected at random from a total population of 5 342 children, 8 and 13 y old, by use of
the Swedish central population register, and from these 15 children were excluded
because of chronic diseases. In addition, 18 children moved from the regions before
the investigation was completed. 70 refused to participate for various reasons and one
child was inaccessible, whereas 338 8y-olds, and 347 13-y-olds participated. The
anthropometric measurements were performed by four paediatricians. Although the
children in this study were randomly selected they cannot be regarded as representing
all children in Sweden of comparable ages, as the most heavily urbanised areas and the
southern part of the country were not included [Sunnegårdh 1986].
The only study in Sweden, before the present studies within this thesis, of a
representative national sample was carried out by Lindgren et al. by collecting data for
height and weight from school health records for children born in 1967, age 7 to 16 y
[W-Lindgren 1986]. By a sampling in many steps the group of children was picked
out: First they chose communities all over Sweden out from four factors: number of
inhabitants, number of socialist respective rightwing voters, number of occupied in
public sector and number of immigrant-pupils. Thus were 29 communities selected
and within the communities a number of different classes, argued to represent a miniSweden. In every participating school, the school nurse copied the health records for
every pupil born the 5th, 10th, 15th, 20th and 25th in each month in 1967. The study was
based on records from 1 390 girls and 1 517 boys. Missing number was about 2%. The
description was made for mean values, height and weight, for different ages boys and
girls, and for social class based on the occupation of the father.
The two materials from Ljung and Lindgren (Ljung 740 children born in 1955
and Lindgren 2 907 children born in 1967) were together presented as the basis for
Swedish population reference standards for height, weight, and body mass index
attained at 6 to 16 y (girls) or 6 to 19 y (boys) in 1995 [Lindgren 1995]. The proportion
of children whose parents were immigrants in the 1967 sample was approximately
10%, in the 1955 sample the proportion was about 2%. By comparing the two samples
in the overlapping ages there were no consistent differences between the heights of the
two samples, but the earlier sample averaged 3% lower in weight for both girls and
boys, with no apparent age trend. All weights from this 1955-born sample were
increased by 3% in the subsequent calculations.
They compared this with age-overlapping data from a material of children born
in 1980 living in a suburb of Stockholm at age 6 y [Lindgren 1994]. They saw very
small differences between the samples in height and weight at age 6 y. The authors
argue that there are no differences between national samples over time (55 to 67-born)
and no differences between regions.
67
1.5.2
Growth surveys – international examples
National surveys
Norway
The study group consisted of 3 068 children, 1 479 girls and 1 589 boys, born 1956–
68. The children lived in the parishes of Sletteland and Landås where the population
may be regarded as representative for an average town in Norway. They were
measured at intervals of about one year during the period 1971–74. Height, weight and
10 other anthropometric variables were measured the first year and, for most children,
also the second year. In the two subsequent years, height and weight measurements
were followed up in many of the children. On 966 of the 3 068 children only height
and weight were registered. No children were excluded from the study except those
with diseases or malformations affecting their general condition severely [Waaler
1983]. Values from birth to age 4 y are based on data of infants born 1982–84
[Knudtzon1988]. Norwegian percentile curves are presented by Knudtzon and Waaler
[Knudtzon 1988].
Iceland
Height and weight were measured in 5 554 randomly selected Icelandic schoolchildren
(2 779 boys and 2 775 girls) at the age of 6 to 16 y. The total number of Icelandic
children in the selected groups are 47 583 [Dagbjartsson 1989].
United Kingdom
Cameron [Cameron 2002] examines the base for different growth curves in United
Kingdom and discusses in what circumstances they will be adequate.
In 2001, a group reviewed all the growth reference charts in the United
Kingdom [Wright 2002] and concluded that the main recommendation is to use the
United Kingdom 1990, UK90 [Cole 1995, Cole 1995, Freeman 1995, Cole 1998] for
all ages. The Buckler references [Buckler 1997] are suitable for assessing cross
sectional height in isolation, but the UK90 should be used where both weight and
height are being evaluated. A simple method for assessing centile change based on
UK90 is described by Cole [Cole 1998].
The Netherlands
The Netherlands have a strong tradition with their four cross-sectional surveys in
1955, 1965, 1980 and 1996 [de Wijn 1960, van Wieringen 1971, Roede 1985, Fredriks
2000] with nationally representative samples. Roede et al. investigated, crosssectionally, in 1979–80, a material collected within the Netherlands’ third nation-wide
survey of growth [Roede 1985], and in 1997 a fourth nation-wide survey in the
Netherlands was carried out [Fredriks 2000]. Within these two Dutch studies the
stratified samples were nationally representative of the demographic, geographic, and
socio-economic distribution of the population.
To document the distribution of BMI in the Netherlands when obesity was less
common, a nationally representative growth survey was made. 41 000 boys and girls
age 0–20 y were the subjects [Cole 1999].
A nation-wide growth study performed in the Netherlands in 1979–1980
[Roede 1985]. Length, height and weight were measured in 42 000 healthy 0–19-y-old
68
individuals of Dutch origin. The stratified sample was nationally representative of the
demographic, geographic and socio-economic distribution of the population. The
measuring was performed during routine medical examination programmes by 200
teams of medical officers and nurses in the child health sector. The survey was crosssectional, with a restricted period of data collection to avoid the effect of secular
changes within the survey period. Nearly 46 000 recording forms were received. After
exclusions and the rejection of outliers (children with fewer than two Dutch parents,
infants with birthweights below 2 500 g, and children with pathological conditions),
data on 21 521 males and 20 245 females were available for analysis [Fredriks 2000].
Furthermore, special surveys have been done for children with Moroccan or Turkish
origin [Fredriks 2003, Fredriks 2004, Fredriks 2005].
Ireland
A cross-sectional study from Ireland of 3 509 children was carried out on
approximately 100 boys and 100 girls of each year of age from 5 to 19 y. A sample of
schools was selected to include pupils of different socio-economic class, and we
measured 50 children of each sex and year of age, in both rural and urban areas. 227
were excluded because of illness, non-Irish parentage, or inadequate information
[Hoey 1987].
Europe
Fully longitudinal data from the Euro-Growth Study is the base for Euro-Growth
references for length and BMI. Birth dates were between August 1990 and October
1993 for 2 145 children collected in 21 study centres in 11 European countries
[Haschke 2000, van’t Hof 2000].
Cuba
In 1970 the population of Cuba was 8 575 00 of whom 3 900 000 were between 0–20
y of age. To distribute the sample size within the population consideration was given
to:
First, the number of children under 20 y by province.
Second, the average number of children under 20 y per household. This average
was different for urban and rural zones. For example, in metropolitan Havana it was
1.6; in urban Oriente province it was 1.8, while Oriente province had 3.1 per family.
Third, the sample was initially stratified by province according to these figures.
The final sample size was 56 000; 28 000 children of each sex. The sample should
enable the investigators to achieve two main objectives: precision (particularly for the
outlying centiles) and representativity. In addition, it should allow comparisons
between different regions of the country and different ethnic and social groups. One
measuring team was assigned to each province, with one in reserve nationally. There
were 9 teams because Havana and Oriente provinces, being more populated, had two
each [Jordan 1975, Jordan 1979].
United States
Several large, expansive long-term studies were initiated in the US. These include Fels
study and this is the only one of these studies still active, which began in 1929 [Roche
1992]. The sample of the Fels study is healthy, well-nourished boys and girls, living in
small urban communities and rural areas of south-western Ohio [Roche 1992]. This
data material was the foundation for the former NCHSC growth charts and was also
69
used as an international standard in spite of the sample being neither socially nor
geographically representative [Hamill 1977, Hamill 1979].
In the US there are Centers for Disease Control and Prevention 2000 Growth
Charts for the United States [Kuczmarski 2000, Ogden 2002]. These charts are based
on national data collected in a series of 5 surveys between 1963 and 1994.
Australia
During 1997, 2 828 females and 2 871 males (age range 5–7 y) were measured when
entering primary school by school nurses using a Minimeter, in North Brisbane,
Australia (93% of entrants) [Jack 1999].
China
Nationwide growth surveys have been performed once every 10 years in China since
1975 (1975, 1985, 1995 and 2005?). In the second survey in 1985, data for almost half
a million children and adolescents, aged 7–18 y, were collected [Zhang 1988]. In the
second survey the physical growth of 175 290 Chinese children less than 7 y in age
was undertaken in the urban and rural areas of ten provinces [Ching 1989]. The third
survey was performed in 1995, and from this survey, data for the Beijing population
were retrieved to construct the height for age, and weight for age percentile curves [Li
1999].
A survey on the physical growth of Chinese children under 7 y in the urban and
suburban rural areas of nine cities of China 1995 was compared with the survey ten
years earlier (1985) [Anonymous 1998].
Body mass index reference curves for Chinese children are reported by Leung.
[Leung 1998]
Turkey
In 1973 growth data were published for the first time in Turkey (Neyzi 1973), but a
more recent survey showed body mass index references for Turkish children aged 6 to
18 y. It was constructed by first collecting height and weight measurements biannually
for 1 100 boys and 1 019 girls. The children were born in 1975 or later and all children
belonged to the highest socio-economic group [Bundak 2006, Neyzi 2006].
1.5.3 Internationally used reference values
For establishing a standard definition for child overweight and obesity worldwide, six
large nationally representative cross sectional growth studies (Brazil, Great Britain,
Hong Kong, the Netherlands, Singapore and the United States) were pooled together
in order to get reference values for BMI, IOTF [Cole 2000]
The WHO Multicentre Growth Reference Study (MGRS) (1997–2003)
collected primary growth data and related information from 8 440 children from
widely differing ethnic backgrounds and cultural settings (Brazil, Ghana, India,
Norway, Oman and USA). Eligibility criteria included breastfeeding, no maternal
smoking and environments supportive of unconstrained growth. The study combined
longitudinal (birth to 24 mo) and cross-sectional (18–71 mo) components [WHO Child
Growth Standards 2006]. The construction of the child growth standards are described
by Borghi et al. [Borghi 2006].
70
2. AIMS
2.1
GENERAL AIM
The general aim of this thesis was to explore if it was possible, on a national level, to
monitor central measures of public health among infants, children and adolescents,
height and weight, by collecting data from health records from all of Sweden.
All children should be included and missing cases should be very few.
A good description of growth could reveal how the situation is, and in what
direction the development goes, and be a base for creating references values for
different purposes.
2.2
SPECIFIC AIMS
The specific aims, each related to one of the five papers, were:
Paper I: To investigate the suitability of using routine height and weight data from
records within a school health service system, for population monitoring of child and
adolescent growth on a national level, in order to produce a base for both descriptive
and prescriptive values.
Paper II: To describe Swedish children’s growth (i.e. height, weight and body mass
index) from birth to age 19 years.
Paper III: To describe the development of head circumference, from birth to age 48
months, of infants in Sweden.
Paper IV: To determine changing patterns over time by comparing the rates of relative
weight loss (body mass index, BMI) among boys and girls, from age 7 to 18 years, in
two birth cohorts.
Paper V: To monitor and describe, on a national level, the development of body mass
index (BMI), overweight and obesity from age 7 to 18 years of schoolchildren in
Sweden during a period of 8 years.
71
3. MATERIAL AND METHODS
3.1
TARGET POPULATIONS AND STUDY POPULATIONS
3.1.1 Target populations
Birth cohort 1973
The target population is the birth cohort 1973
Living born: 109 663
Average age of the mothers: 26.5 y
Breastfeeding at the age of 2 mo 32%, at the age of 4 mo 14%, at the age of 6 mo 7%
Migration during 1973
Immigration: 29 443 individuals
Emigration: 40 342 individuals
Birth cohort 1981
The target population is the birth cohort 1981
Living born: 94 064
Average age of the mothers: 28.0 y
Breastfeeding data not available (see study population)
Migration during 1981
Immigration: 32 272 individuals
Emigration: 29 440 individuals
3.1.2 Study populations
Cohort 1973
Every born in 1973 on the 15th of every month and living in Sweden 31.12 1989: 3
749 [Statistics Sweden 1999]
Number with collected data for the sample: 3 579 individuals
Observation period: age 7 y to age 18 y.
72
Table 6. Individuals born the 15th of every month. Countywise, number of
individuals 31.12 1989 and collected
County
Stockholm
Uppsala
Södermanland
Östergötland
Jönköping
Kronoberg
Kalmar
Gotland
Blekinge
Kristianstad
Malmöhus
Halland
Göteborg och Bohuslän
Älvsborg
Skaraborg
Värmland
Örebro
Västmanland
Dalarna
Gävleborg
Västernorrland
Jämtland
Västerbotten
Norrbotten
All
31.12 1989
668
136
114
183
141
81
107
31
76
137
328
131
275
211
124
116
113
108
125
133
113
40
126
132
Collected
649
127
98
177
129
77
101
30
74
120
311
113
297
197
121
113
106
104
121
126
106
35
123
119
3 749
3 579
Breastfeeding data are not available (see target population).
Children born outside Sweden: 225
Diagnosis (from health records) of 32 children with disorders of major impairment for
growth:
Status post polio, Cong. dislocation of the hip, Coeliaci 2 children, Right leg
shortened, Juvenile chronic arthritis, Diabetes mellitus 2 children, Heart failure
op.,Growth error 8 children, Growth error and anorexia, Mental retardation and inborn
error of metabolism, Myotonicity of m. sternocleidomastoideus, Polyneuropati: Mb
Charcot Marie Tooth, Hypothyreosis, Mb Turner, Mb Down, Hereditary angioneurotic
syndrome, Antihidotic ectodermal dysplasia, Hashimoto struma, Cerebral palsy
(Wheel chair), Pubertas tarda, Scoliosis, Nephrectomy, Dwarfism.
Cohort 1981
Every born in 1981 on the 15th of every month and living in Sweden 31.12 1989: 3 158
[Statistics Sweden 1999]
Number with collected data for the sample: 3 107 (1 548 boys/1 559 girls)
73
Observation period: Birth to age 19 y (height and weight)
Birth to age 48 mo (head circumference)
Table 7. Breastfeeding in percent.
Age, in
months
<2
>2
2–4
>4
4–6
>6
Breast-fed
Exclusively
42.5
57.5
20.8
36.7
22.6
14.1
Breast-fed and
formula-fed
30
70
22
47.8
22.2
25.6
Table 8. Individuals born the 15th of every month. Countywise, number of
individuals 31.12 1989 and collected
(5th- and 10th and 20th-born in some counties with low population density, not
studied within this thesis)
County
Stockholm
Uppsala
Södermanland
Östergötland
Jönköping
Kronoberg
Kalmar
Gotland
Blekinge
Kristianstad
Malmöhus
Halland
Göteborg och Bohuslän
Älvsborg
Skaraborg
Värmland
Örebro
Västmanland
Dalarna
Gävleborg
Västernorrland
Jämtland
Västerbotten
Norrbotten
All
74
5th-born
10th-and 20thborn
31
48
45
31.12 1989
Collected
587
114
83
169
116
71
97
24
54
122
263
95
247
176
110
104
94
90
114
97
98
39
106
88
589
116
84
166
130
75
104
25
51
117
289
94
238
179
112
101
101
89
107
103
94
40
112
85
51
103
101
3 158
3 107
334
45
Added sample from counties with low population density (not studied within this
thesis)
Born on the 5th of every month
Available 31.12 1989
Collected
Västerbotten
113
103
Norrbotten
105
101
Jämtland
54
51
Blekinge
49
48
Born on the 5th, the 10th, and the 20th of every month
Gotland
77
76
Immigrated children after 31.12 1989 (not studied within this thesis): 94
Infants with birthweight <2 500 g: 118
Children born outside Sweden: 239
Children with a disorder with major impairment for growth: 24
Diagnosis (from health records) of 24 children with chronic disease of major
impairment for growth:
Many and long stays in hospital, Growth hormone deficiency 3 children, Pulmonal
hemosiderosis (lungtransplantation), Multiple intolerance for food, Aneurysm op.
Crippled, Cerebellar dyspleoi, Intestinal resection, Pubertas praecox ?, Mb Down and
Heart failure, Coeliaci, Born with structural scoliosis, Growth error 4 children,
Pulmonalis stenosis, Cerebral palsy, Spinal atrophy of the muscles, Spinal atrophy,
Disease of long duration, Juvenile chronic arthritis, Hypofyseal short in stature.
About head circumference 0–48 mo
Of the 3 107 children, 123 individuals had immigrated between age 48 mo and 8 y and
are thus excluded since they are not represented by any measurements for the age
interval investigated. Thus 2 984 individuals could be studied from birth to age 48 mo
(see Table 1 in Paper III).
The instructions for measuring head circumference during the period in
question are as follows: ”Measuring is done with a measuring-tape, horizontally at
the place where the largest head circumference is, read to the closest millimetre.”
[Anvisningar och kommentarer]
3.2
DATA COLLECTION
The first wave of data collection for both cohort 1973 and cohort 1981 took place
during the autumn in 1989.
75
3.2.1 Cohort 1973
Almost every Swedish child attends school from age 7 to 19 years and during this time
information on growth is recorded on a fairly regular basis by school nurses. Thus, we
chose to study a national random sample of records from all children born on the 15th
of any month in 1973 and living in Sweden on December 31st 1989. The study design
was longitudinal, and for each child measurements of height and weight were recorded
with the date that the measurement was obtained. The data collection from school
health records, including both public and private schools, was conducted in three
waves: the first wave when the children were 16 years old (in grade 9 in compulsory
school), and the second when they were 18 years old (one year before Swedish
adolescents usually leave secondary school). Finally, in the third wave we looked for
records after the adolescents had left school in local community archives.
The nurses also recorded if the child was born outside Sweden or had a chronic
disease. A chronic disease involving major growth impairment (as judged by an
experienced paediatrician, the author, BW) was present in 32 children. Records were
collected for 3 579 of the 3 749 adolescents, bringing the number of missing cases to
4.5%. An example of a page in a school health record is shown in appendix.
A unique study identification code number was appended to school records by
school nurses, so that linkage between data collection waves was possible without
using the official registration number assigned to all Swedish residents. We anticipated
that avoidance of the official registration code was expected to help us to keep the
number of non-responders as low as possible. If the adolescent signed a written
consent form after age 18 years (completed by approximately 80%), the official 10digit identification number was used for record linkage. Where school records were
incomplete for boys at age 18 years, data from the MSCR were used instead.
In the records, weight was noted in kilograms and height in centimetres with a
numerical accuracy of at least ±0.5 (kg/cm). After data entry we plotted height and
weight by age for each child to identify registration errors.
3.2.2 Cohort 1981
Several times during their preschool years, almost 100% of Swedish children visit
childhood health care centres where they are weighed and measured. These records are
then sent to the school nurses when the children enter school. Almost all children stay
in school from age 7 to 19 y, and at school they are fairly regularly weighed and
measured by school nurses. As all this information is collected in their records, the
prerequisites for monitoring children’s growth and selecting suitable samples are met.
Collection of data from the records was done on three occasions in school: in
grade 2, in grade 9 and finally in grade 12. Most children are 8, 16 and 19 y old
respectively in those grades. For approximately 10% of the children, in particular those
who moved or changed schools, records could not be obtained from the school nurses.
In many such cases it was possible to find the records, as the records were deposited in
the local community archives after the children had left school. Through this combined
effort, childhood health records and school health records were collected for 3 107 of
the 3 158 children, meaning that missing cases accounted for only 1.6% of the
76
sampling frame. An example of a page with growth data in a child health record is
shown in appendix.
In this investigation, school nurses were asked to mark the record of each child
with a unique study identification code, to make it possible to combine the data from
the three data collection occasions without using the official 10-digit identification
number. If the adolescent signed an informed consent form after age 18 y, then the
official 10-digit number could be used. In cases where the school health records were
missing values for boys at around age 18 y, and if there was informed consent, data
from the MSCR were added.
In the records, all data regarding height and weight were registered by date.
Weight was noted in kilograms with a numerical accuracy of at least ± 0.05 from 0 to
6 y and ± 0.5 from age 7 to 19 y, height was noted in centimetres with a numerical
accuracy of at least ± 0.5. After data entry, we plotted height and weight by age for
each individual in order to identify registration errors. A summary of the data sources
is found in Table 1.
3.3
ANALYSIS
3.3.1 Paper I, II and III
Height and weight
For cohort 1973 and 1981 data were analysed using a cross-sectional approach. To do
this we had to take into account that height and weight were not measured at the same
ages in different children. The adjustment was performed using piecewise linear
regression as follows.
For both cohort 1973 and cohort 1981 we centred on the whole numbers from
age 7 to 18 years (for cohort 1981 up to 19 years), and defined equal intervals, e.g. 7
years ±180 days. Within each one of these intervals, one randomly chosen observation
from each individual (when available) was used to form a linear regression of the
outcome (height and weight) on age, measured in days. Thus, observations within each
interval were statistically independent in the sense that each individual contributed
only one observation. The midpoints of the intervals were from 7 up to 19 years with
consecutive steps of one year. A suitable length of the intervals, from the midpoint to
the upper or lower limit, was tested with values ranging from 45, 90, and up to 180
days. The final calculation used an interval of 180 days. Linearity within each interval
was examined for each chosen length using local regression smoothing [Cleveland
1979], and comparing the smoothed curve with the straight line. Based on this
comparison, we accepted the linear function as a satisfactory approach within each
interval. Calculations were performed for males and females separately, both with and
without exclusions.
The linear regression was used to estimate the mean of the outcome for the
centre of each interval. The variability within each interval was derived as the square
root of the residual variance of the regression line, which gives a standard deviation.
SD, Skewness and kurtosis were calculated using the residuals produced by linear
regression. Kurtosis indicates whether the distribution around the mean is thick-tailed
77
(a higher proportion of subjects are found at the extremes of the distribution, kurtosis
>0) or thin-tailed (the opposite case, kurtosis <0).
For cohort 1981 we defined intervals with midpoints at months 1 to 12, and
then at every half-year from 1.5 y up to and including 6.0 y. A suitable interval length,
from the midpoint to the upper or lower limit, was tested, with values ranging from 7
days (in the first months of life) to 180 days (for age 7 y and older). Within each one
of these intervals, one randomly chosen observation for each individual (when
available) was used to run a linear regression of the outcome (height and relative
weight) on age measured in days.
Head circumference in cohort 1981
We defined intervals with midpoints at various ages from 15 d to 48 mo. In each
interval, one randomly chosen observation for each individual (when available) was
used to form a linear regression of the outcome on age measured in days. Thus, the
observations within each interval were statistically independent, in the sense that each
individual contributed only one observation. A suitable interval length, from the
midpoint to the upper or lower limit, was tested with values ranging from 7 d at ages
15 and 30 d, 10 d at ages 45 and 60 d, 14 d at ages 75, 90 and 120 d, 21 d at ages 6– 8
mo, 30 d at ages 9–17 mo, 60 d at age 18 mo, and then 90 d at ages 24–48 mo. The
linear equation assumed to adequately describe the relationship between head
circumference and age within each interval was examined using local regression
smoothing [Cleveland 1979]; then the smoothed curve was compared with the straight
line.
3.3.2 Paper IV
Change in BMI was calculated as the difference in BMI between all adjacent pairs of
measurements within each subject. The relative change was obtained by dividing the
difference in BMI by the first of the two BMI measurements. BMI reduction was
defined as a negative relative change in BMI.
Consecutive multiple BMI reductions within the same subject, with no
intervening BMI increase, was defined as one episode of BMI reduction from the first
measurement taken before the reduction to the last measurement before BMI
increased. The BMI reduction episodes were divided into five categories: no reduction
(including increases), <5%, 5–<10%, 10–<20%, and ≥ 20%. A further classification
was to dichotomize the reduction as above or below the 5% cut off value.
We categorized the age of the children when the measurement before the
reduction was taken into four intervals: 7–9 y, 10–12 y, 13–15 y, and 16–18 y. If an
individual had more than one BMI reduction episode during one age interval, only the
episode with the largest BMI reduction was considered. We also categorized the BMI
at the start of the reduction according to the BMI percentiles calculated from the 1973
cohort, given sex and age: <40th, 40th to <60th, 60th to <70th, 70th to <90th, and ≥90th
percentile. This was done because means for BMI increased for cohort 1981 compared
to cohort 1973 in all ages for both boys and girls with exception for girls age 17 and
age 18 y. The increase in age 7 and 8 y was around 1.5%, and for older ages around
2.5%. If a child was measured more than once during the same age interval but had no
78
BMI reduction we randomly sampled one measurement to assign this child to one of
these BMI categories.
Logistic regression was used to analyse the presence (yes/no) of a BMI
reduction with the cut off at 5% defined previously. As explanatory variables cohort
(born in 1973 or in 1981), sex, age, and BMI before the reduction were used. Age and
BMI before the reduction were categorized in the intervals described above. We also
tested for statistical interactions between the explanatory variables. The effect
parameter is expressed as the odds ratio (OR) and its 95% confidence interval (CI).
Because each individual could have more than one BMI reduction episode the CI was
adjusted for the intraindividual correlations caused by repeated measurements. This
was done by using the cluster option in the logistic regression procedure in the
program package STATA [release 9, StataCorp, College Station, Texas, USA].
3.3.3 Paper V
Data were analysed with a cross-sectional as well as a longitudinal approach. The
cross-sectional approach is more thoroughly described in paper I and in paper II.
In the longitudinal approach the two cohorts were analysed together. We
formed a regression model with outcome (height or BMI) as a function of two sets of
categorical variables, one for cohort (1973 or 1981) and the other for age in whole
years (7, 8, 9 and up to 18 y). Interaction between these two variables was also
introduced and tested. The parameters in this model correspond to means for each age
category and for each cohort estimated simultaneously using all selected data from the
age groups in both cohorts.
The longitudinal model was estimated with a mixed model approach; using the
statistical package SAS (Version 8.2, SAS Institute, Cary, NC, US) with covariance
structure according to an autoregressive structure of the first order, degrees of freedom
according to the Satterthwaite approximation, and restricted maximum likelihood
estimation method (REML). We calculated the marginal means at each age and
contrasted these means for the two cohorts. Adjustment of p-values in case of several
tests was performed. The cross-sectional analysis was done with the statistical package
SPSS (Version 13, SPSS Inc, Chicago, Il, US), supplemented with user-defined
commands.
3.4
ETHICS
These studies were approved by the Ethics Committee of the Örebro County Council,
Sweden.
79
4. RESULTS
4.1
PAPER I
Descriptive statistics for height and weight, from age 7 to 18 y, calculated from data
obtained from school health records are summarized and compared with previous
Swedish studies. The data include measurements of 3 579 of 3 749 sampled children;
4.5% of individuals are missing, so non-response bias is minimal. The effect of
exclusion of children with chronic diseases involving major growth impairment and/or
exclusion of children born outside Sweden have a minor impact on the results due to
the relatively small number of excluded children.
4.2
PAPER II
The data, from birth to age 19 y, include measurements of 3 107 of 3 158 sampled
children; 1.6% of individuals are missing, so non-response bias is minimal. Thus,
statistical descriptions and comparisons with similar data sets can be based on data
with unusually high national representativeness. Selected subgroups (i.e. individuals
born outside Sweden, suffering from chronic disease causing major growth
impairment, or with birthweight <2 500 g) deviate in growth pattern, and exclusion of
these subgroups increases means and decreases SD for height but only slightly
influences summary statistics for weight and body mass index.
4.3
PAPER III
From the sample of 3 158, data for 3 107 infants were collected, making the
percentage of missing individuals 1.6%, thus non-response bias is minimized and high
national representativeness is achieved. Summary statistics for head circumference are
presented. Comparisons are made to international reference materials, and to previous
and current Swedish reference curves.
4.4
PAPER IV
An increasing rate, by comparing individuals born in 1973 and in 1981, of relative
weight reduction episodes was found for both boys and girls. The increase for girls
was most pronounced and started from a higher rate and it was seen in nearly all body
weight categories and in all ages. For boys the increase in reductions occurred for all
body weight categories only in the age interval 7–9 y, otherwise there is a much more
heterogeneous pattern. There is a high correlation between body weight and reduction
of BMI, as more individuals with overweight reduces their BMI compared to thinner
individuals. For girls, the highest increase in rate is seen among the thinnest
individuals.
80
4.5
PAPER V
From age 7 y to 18 y a strong positive secular change for BMI exists in all ages, and
the rate of overweight and obesity is increasing for both boys and girls. Furthermore,
obesity is growing more severe.
81
5. GENERAL DISCUSSION
5.1
EPIDEMIOLOGICAL MONITORING
There are at least four requirements for adequate national epidemiological monitoring.
Firstly, the sample ought to be representative of the population. Secondly, there should
be few missing subjects and they should not be systematically different from the nonmissing subjects. Thirdly, the data ought to be valid and measurements should have
acceptable precision. Fourthly, it should be possible to use the material for
comparisons over time and therefore collected in a way that can be repeated.
The results indicate that our national samples better assess height, weight and
head circumference among infants, children and adolescents than previously employed
methods. This conclusion is based on four findings.
5.1.1 The representativity of the samples
Firstly, the sampling frame is by definition nationally representative and furthermore
we succeeded in collecting data for almost the whole sample. In practice, the data were
tested among 18-years old males through comparison with measurements collected at
conscription for all 18-year-old males: the school record data were found to be
representative for both cohort 1973 and cohort 1981.
The external validation was made of the cohort 1973 sample for 17–19 yearold males by a comparison with the national MSCR (see Table 4 in Paper I). We
performed the calculations using data from MSCR with the same methodology used
for the child health records, i.e. age-corrected by piece-wise linear regression of height
and weight on age at measurement. For this we required the date of measurement, thus
47 739 measurements were available. For age 18 years, where the majority of the data
in MSCR are found, there are only very small differences in mean value and SD
between the two data sets. For weight there are practically no differences in skewness
and kurtosis, with values of 1.2 and 3.1, respectively, from the school record data and
1.3 and 3.6, respectively, from the MSCR material. For the 19-year olds the difference
in height between the two datasets (180.9 cm and 180.1 cm) is just outside the 95%
confidence limits, 0.04 to 1.56, just outside zero.
To test the representativity of the cohort 1981 sample, we compared summary
statistics of the sample with statistics from two nationwide registries containing height
and weight data: 1. The MSCR, which includes measurements for approximately 80%
of all males born in 1981 and holding Swedish citizenship at age approximately 18 y.
2. MBR covers all newborns in Sweden.
1. Comparison between data in cohort 1981 and conscript data is shown in Table 5
Paper II. The most measurements are available for individuals around age 18.0 and
18.5 y, both in the present study and in the MSCR. For height in the present study at
age 18.0 y after exclusions, the mean is almost the same as in the MSCR (0.6 cm
lower in present study without exclusions). At age 18.5 y the present study and the
82
MSCR have the same means, within 0.1 cm (the present study with exclusions has a
0.7 cm higher mean). For weight, the means in the present study without exclusions
and means in the MSCR correspond well at both age 18.0 and 18.5 y (exclusions
increase means by 0.6 kg). For both height and weight the registry has less variability
as measured by the SD.
2. Comparison with MBR data. See Table 4 Paper II. The comparison was made
without exclusions and revealed no differences. In this sense, the present data seem
valid or representative of all boys and girls.
5.1.2 Selection bias
Secondly, few subjects were lost from the sample and they can presumably be
accounted for, as some children are never measured at school if they only live in
Sweden for a short period of time. A sparse pattern of measurements for one
individual is sometimes explained by migration, by chronic disease, or disability being
treated within the general health care system. This might explain why these individuals
never utilise the school health system. Since the data from the subjects with few
measurements are available to us, they can either be included or excluded when
creating reference data, depending on the purpose.
For boys at age 16.0 y there were no statistically significant differences in
height or weight between those for whom measurements from age 19.0 y were
available, and those for whom measurements at this age were unavailable. Therefore,
we have no evidence of internal selection at age 19.0 y, despite the low number of
measurements, much lower than for age 18.0 y and 18.5 y.
Analogous results were obtained for girls in comparing the height at age 14.5 y
of those who were and those who were not measured at age 18.5 y. For BMI there
was, however, a significant difference at age 16.0 y between those girls who were and
those who were not measured for BMI at age 18.5 y. The mean BMI at age 16.0 y for
those who, at a higher age, were not measured is 0.5 units higher than for those who
were measured at age 18.5 y. One interpretation of this finding is that internally
missing cases are selective and probably more likely for overweight or obese subjects.
Therefore the mean BMI for girls of age 18.5 y should be considered to be
underestimated.
Tests of internal selection bias concerning head circumference was performed.
Fewer measurements were made at age 12 mo and above for both boys and girls. To
determine whether this fact could influence the analysis, we compared the 10 mo
measurements for those who did not have measurements at 12 mo, 18 mo, 24 mo, 36
mo and 48 mo respectively with those who did have measurements at these ages. The
differences in the 10 mo means were +0.11 cm, -0.05 cm, +0.03 cm, +0.13 cm and
+0.05 cm. We also analysed a possible correlation between head circumference at
birth and age at final measurement. The correlation was close to 0, and there was no
evidence that those with very low or very high values at birth were observed for a
longer period.
83
5.1.3 Data quality
Thirdly, the quality of data in the present study is acceptable for growth monitoring
among school children. Dealing with data not primarily produced for a scientific
purpose raises some important questions about data quality. What is the reliability and
validity of measurements carried out in schools? Even though school nurses are well
trained for measuring height and weight, their actual measuring practice may not be
well controlled and the records imprecise [Strandberg 2001]. In their daily work,
nurses are most concerned with identifying abnormal growth patterns. Many pupils,
especially the older ones, do not want to undress when they are examined, although
nurses are instructed to weigh the pupils in form grade 11 in their underwear.
However, we believe that measurements are taken with the subject barefoot. At
military conscription all males are measured barefoot, wearing only underpants. It is
therefore very important to make comparisons at age 18 years between the means in
this study and the data from the conscription material. For weight, there is a difference
showing that on average, the males in the study set are heavier. One explanation could
be that the MSCR has excluded more obese individuals (they do not attend military
conscription). Another explanation could be that at military conscription boys are
always measured in underpants, but in school there are some individuals who are
measured clothed.
In conclusion this representative sample with few missing individuals with
well-defined exclusion criteria can be used to create accurate growth charts, both for
descriptive and prescriptive purposes. An advantage of using this material is that this is
possible at a fraction of the cost required for large dedicated cross-sectional
longitudinal studies of growth.
5.1.4 Comparability
Fourthly it is possible to replicate the sampling procedure, facilitating future
comparative studies. The collection of this information is cost-efficient compared with
material collected for a specific research project. We have shown, in this thesis, the
possibility to repeat surveys with high national coverage and minimal selection bias.
This makes it possible to do comparisons that have been done in Paper IV and Paper
V.
In order to survey secular changes, in the best of all worlds, the contents of time
change should be the only exposure. But what impact has the following known
circumstances on health, on growth? Cohort 73 versus cohort 81, the size of the birth
cohort (109 663 versus 94 445), the average age of the mother (26.5 y versus 28.0 y)
and rate of breastfeeding (>6 mo 7% versus 25.6 %).
84
5.2
COMPARISON TO OTHER SURVEYS
5.2.1 In Sweden
Height and weight
We compared data of cohort 1973 with the prospective study by Karlberg (K)
[Karlberg 1976], used as national reference material from 1973 to 2000, and with the
second reference dataset reported by Albertsson-Wikland (A-W) [Albertsson-Wikland
2002], that was used as national reference material from 2001. Table 5 Paper I shows
differences in summary statistics between the school record material and the studies by
(K) and (A-W).
Height
The study by (K) showed lower means for height at all ages for both sexes compared
with our data. The means for height in (A-W) are higher at every age than those
produced using our school records (the amount of increase 0.7 to 1.7 cm). Mean
heights for boys in (A-W) at age 18 years (180.4 cm) are higher than means at age 18
years at any time in the MSCR. It has been reported that all cohorts born between 1953
and 1981 represented in the MSCR have means less than or equal to 179.5 cm
[Rasmussen 1995 and 2000]. Comparisons for SD, skewness as well as kurtosis in the
three studies reveal no substantial differences between them all.
Weight
The study by (K) shows lower means for weight at all ages compared with the present
school record material, with a difference for 7-year old males of 1.4 kg, with a
difference of 5.2 kg at age 16 years. The differences between the school record
material and (A-W) are in both directions and no particular pattern can be found, for
both sexes. For variability within age groups, measured by SD, the school record data
have, in general, somewhat higher values than the other two data sets. In particular
boys show higher values.
The present school record data show skewness similar to the study of (K) at
young ages for both sexes, but for adolescents the values in the present school record
study are higher. For males of all ages skewness is lower in the present study than in
the study of (A-W). For girls there are no differences or only somewhat higher values
in our data compared with (A-W). The values that differ most between the three
studies are those for kurtosis. For males aged 7 to 12 years, kurtosis is much higher in
both our study and (K) compared with (A-W). The differences at ages 13 y to 16 y
between (K) and (A-W) are smaller and in both directions. Values in our data for this
age group are higher compared with both (K) and (A-W). For females, the values tend
to be higher in our study compared with both (K) and (A-W).
However, mean height for males at age 18 years is higher in the reference study
from the same time period presented by (A-W). This could be due to exclusion of
some individuals in their study, thus introducing potential selection bias. Also, their
sampling frame was restricted to the urban area of a big city and urban populations are
reported to deviate both in height and weight compared with other areas (urban
individuals are taller and less heavy) [Rasmussen 1995]. Moreover, skewness and
85
kurtosis for weight were slightly higher in the present study compared with (A-W).
Our experience is that exclusion of more obese individuals has a profound effect on
skewness, and in particular on kurtosis, and the individuals who failed to attend the
last investigation at school (319 females and 304 males) in the (A-W) study might
have been more over-weight or obese than the investigated group.
Head circumference, comparison between cohort 1981 and Swedish
surveys
Figure 2 in paper III presents comparisons, for girls, between findings of the present
study and three Swedish reference materials [Karlberg 1976, Lindgren 1994,
Albertsson-Wikland 2002]
The means at all ages from the present study are markedly higher than those
from the study of Karlberg [Karlberg 1976], higher from age 24 mo to 48 mo
compared to the study of Lindgren [Lindgren 1994], and lower at birth but somewhat
higher from age 3 mo to 18 mo but similar from age 18 mo to 36 mo compared to the
study of Albertsson-Wikland (A-W) [Albertsson-Wikland 2002]. Notably, the
reference currently in use in Sweden, based on the study by A-W, states the mean
circumference at birth for girls to be 34.9 cm. This is higher than the mean values for
the ”healthy” population studied by Niklasson [Niklasson 1991], i.e. mean head
circumferences for girls of 34.7 cm, and higher than the means in MBR for those born
before 1974, i.e. circumference for girls of 34.2. As mentioned earlier, the values from
the present study correspond well with data from MBR. The most pronounced
differences between our study and the study by (A-W) are skewness and kurtosis for
birth. We report skewness of -0.3 for boys and -0.7 for girls, while their study says
2.08 and 2.74, and we report kurtosis of 0.7 and 3.1, while their study states 14.52 and
22.56. These figures from (A-W) are remarkably high and indicate presence of
extremes or outliers.
The differences between the former Swedish reference values and those of this
study are large, most likely due to the equipment used to measure the head
circumference. In the study of Karlberg [Karlberg 1976] a narrow (0.6 cm) metal tape
was used with firm tension applied when measuring to minimize the influence of hair
and subcutaneous tissue; this produced lower values than those measured with a wider,
plastic–linen tape, commonly rather loosely applied. Comparison with the current
Swedish reference is interesting, because its data and those of the present study are
comparable in one important respect: both datasets are based on measurements
collected from existing records, so the measurement routines and techniques and the
documentation methods are likely to be similar. Thus the differences could probably
be explained as either due to different sampling frames (national versus local),
different selection criteria (none versus some specified criteria), or a mixture of both
these factors. In the current Swedish reference material just 37.7% of the population (1
381 out of 3 650) is represented with values for head circumference at birth. Reasons
for this are not presented which makes it hard to evaluate the result of a comparison.
Comparison of head circumference in the present study with other reference
materials
Figure 1 in paper III presents comparisons, for boys, between findings of the present
study and reference materials from other parts of the world, including both pooled and
86
selected materials [Fredriks 2000, Haschke 2000, Ogden 2002] and national
representative materials [Jordan 1975, Zhang 1988].
The Dutch study [Fredriks 2000] has approximately 0.5 lower values at all ages
than the present study. Euro-Growth [Haschke 2000] values are lower at all ages, a
small difference at low ages and more markedly from age 12 mo and up. The values
from the CDC charts are higher from birth to age 9 mo, but from age 12 mo to 48 mo
the values are more than 1.0 cm lower. The CDC charts for age 10 mo and older have
lower values than those from all other materials, except the Chinese [Zhang 1988] and
Cuban data [Jordan 1975]. The values from both Cuba and China are much lower from
age 3 mo and up than those from the present study. Comparison not found in the figure
show that the Norwegian reference [Knudtzon 1988] has higher values at all ages
except at age 48 mo than the present study. The agreement between values at ages 36
and 48 mo from the present study and those from UK90 [Cole 1998] is good, but not
in the case of birth head circumference, which is 1.0 cm higher in UK90.
Height and weight, comparison between cohort 1981 and the current
Swedish reference values
For the present study the percentage differences from the reference values are shown
in Figure 3 Paper II both without and with exclusions, i.e. for children with
birthweight <2 500 g, those suffering from chronic disease of major impairment for
growth and those born outside Sweden. These subgroups, among others, are
excluded in the current Swedish reference material. For both boys and girls, height
in the present study is somewhat lower before age 4 y, and then again somewhat
lower from age 7 y up to age 18 y for values calculated without exclusions.
Exclusions increase the differences for height, so from age 5 y to age 13 y the values
in the present study are somewhat higher but from age 14 y to 18 y they are almost
the same as in the current Swedish reference.
The major differences appear when comparing BMI values. The big difference
at age 3 mo might depend on a miscalculation in the current Swedish reference
material [He 2000]. We reached this conclusion by recalculating the BMI for 3 mo by
using the means for height and weight at age 3 mo (0.25 y) from the same material
[Albertsson-Wikland 2002], which could explain the difference of 6–7 percent. For
both boys and girls the BMI values follow the same pattern. From about age 6 y and
up to age 18 y for boys and from age 6 y up to 17 y for girls the differences show 1 to
5% higher values in the present study.
Comparison between cohort 1981 and other nationwide descriptive studies
We have compared the present study with some of the descriptive studies previously
mentioned, although these studies do not always cover the same birth years. The
studies in question are from Sweden [Lindgren 1986], UK [Hughes 1997], the
Netherlands [Roede 2000, Fredriks 2000], China [Zhang 1988], and Cuba [Jordan
1975]. The differences in findings between these studies and the present study with
regard to males are presented in Figure 1 Paper II (height) and Figure 2 Paper II
(BMI). Means for height are presented, but not for BMI, with its more skewed
distribution; median values, when available, are presented for some studies. Heights
and BMI values found in the present study were higher than those found in the
earlier Swedish study, and also higher than those found in the survey from UK;
87
heights found in the present study were lower and BMI was higher than in the Dutch
survey, heights from age 1.5 y were higher than in the Chinese study, and heights
were much higher than in the Cuban study.
A comparison between present study, cohort 1981, and WHO Child Growth
Standards
The MGRS (1997–2003) was a population-based study covering the cities of Davis,
California, USA; Muscat, Oman; Oslo, Norway; and Pelotas, Brazil; and selected
affluent neighbourhoods of Accra, Ghana, and South Delhi, India.
Longitudinal data from birth to 24 mo and cross-sectional study of infants aged 18 to
71 mo. Individual inclusion criteria were: the absence of health or environmental
constraints on growth, mothers wiling to follow MGRS feeding recommendations (i.e.
exclusive or predominant breastfeeding for at least 4 mo; introduction of
complementary foods by the age of 6 mo; partial breastfeeding continued for at last 12
mo), no maternal smoking before and after delivery, single term birth, and absence of
significant morbidity [WHO 2006 p.56–65].
Table 9. Comparison between cohort 1981, boys+girls, with exclusions and WHO
Child Growth Standards, boys+girls, pooled values and highest mean length for a
single country within the pool.
Length and height from birth to age 60 months.
n
WHO
Age Cohort
Pooled
(mo) 1981
.
0
50.6 2711
49.5
6
67.1 2023
66.7
12
75.5 1106
75.0
18
82.4 2150
81.8
24
87.4
849
87.4
48 104.1 2009
103.5
60 111.7 376
110.3
No=Norway, Bra=Brazil, Gha=Ghana
n
WHO
Max.mean
n
Country
1742
1648
1594
1535
1524
478
465
50.4
67.9
75.5
82.4
88.3
104.9
112.5
300
286
272
285
280
71
76
No
No
No
Bra
Bra
Bra
Gha
Unfortunately, MGRS data are not presented for boys and girls separately. The
comparison reveals that means for cohort 1981 after exclusions are very close to those
of the country with highest means for ages 0 to 18 mo. The same value for cohort 81
as the pooled value at age 24 mo (Brazil is 0.9 to 1.1 cm higher). At age 48 mo cohort
81 has 0.9 cm higher value than pooled value and 0.7 cm lower than Brazil. At age 60
it is hard to compare because of low numbers. For cohort l981 there is possible to
make comparison with those who were breast-fed according to the inclusion criteria
made by MGRS. The result of these analyses is not published yet but shows a positive
impact for gaining length for those who are breast-fed.
88
General remarks
It is hard to compare apples with pears. When there is a national representative study
(Cuba, China) to compare with, it is either from another time or a cross-sectional
survey (The Netherlands, China), or within Sweden, selective samples or local
samples. Anyhow, the comparisons with socially selective and pooled material
(WHO), you can get an idea of the differences and to what extent they can be
explained. But the question about comparability remains.
89
6. GENERAL CONCLUSIONS
Paper I: Longitudinal data for somatic growth (height and weight) from age 7 to 18
years from a nationally representative sample of children in Sweden, collected from
school health records, can be used for epidemiological monitoring of growth with
fewer missing individuals and at lower costs compared with other dedicated studies.
Data quality is comparable to similar national surveys. The data are suitable for
descriptive analysis of growth and other forms of observational study.
Paper II: This study represents, without selection bias, the current growth situation
among children and adolescents, enabling both epidemiological comparisons over time
and comparisons with other national surveys.
Paper III: For the first time in Sweden, and without selection bias, means and
distribution of head circumference measurements are documented longitudinally for a
nationally representative sample of infants.
Paper IV: No previous studies have shown, to the best of our knowledge, an
increasing rate of weight loss, among schoolchildren, but some have shown an
increasing rate of weight concerns. However, in this study, a significant increase was
found, for both males and females, most marked for females. In the lowest age interval
the increase was strongest. The increasing rate is the result of an addition of weight
loss both from those who are overweight, and those who are normal, or underweight
individuals.
Paper V: National representative longitudinal data for BMI from two cohorts, where
non-response bias are minimized, made it possible to monitor and reveal a nation-wide
strong positive secular change for BMI in Sweden, over a period of 8 year for
individuals aged 7 y to 18 y.
Summary conclusion
In order to monitor changes in growth patterns, among children and adolescents in a
huge population, it is important to have data that are representative at several main
points. With this study we have achieved representativity at several main points, and
we can conclude that there still is a visible positive secular change for height both for
boys and girls in Sweden today, as well as a marked positive change in BMI for both
boys and girls from age 7 to 18 y. At the same time it is an increasing rate of weight
loss especially among girls.
Multiple comparisons with other surveys reveal that it is hard to achieve
comparability and those rare surveys that are national representative are old and
therefore not easy to compare with.
It is possible to monitor a national situation by collected data from child
health records and school health records and more cost-effective than to perform a
dedicated study with the aim to present a national situation for growth.
90
7. CONSIDERATIONS FOR FURTHER
INVESTIGATIONS
The materials within this thesis are further analysed, ongoing or planned as follows:
2. From description to prescription. In a suitable way depending on purposes transform
the description of cohort 1981 to growth standards (height, weight and head
circumference).
3. Analyse regional differences within cohort 1973 and cohort 1981.
4. Analyse the growth pattern of the exclusively breast-fed children.
5. Analyse the impact of breastfeeding on both height and weight longitudinally.
6. Analyse growth and migration within Sweden and to Sweden.
7. Analyse healthy or unhealthy behaviour among those who lose weight.
8. Look further into aspects of parametric analysis of longitudinal growth data with
irregular spacing in time [Lundholm 2007].
9. Plan for new data collections in order to epidemiologically monitor growth in
Sweden with cohort 1973 and cohort 1981 as bases.
91
8. SAMMANFATTNING PÅ SVENSKA
Denna avhandling undersöker tillväxtmönstret hos barn och ungdomar i hela Sverige
och de sekulära förändringarna i längd och vikt för skolbarn. Avhandlingen bygger
på studier av två populationer: skolbarn födda 1973 från 7 års ålder till 18 år, och
barn födda 1981 från födelsen till 19 års ålder.
Gruppen med barn födda 1973 utgörs av alla födda den 15:e i varje månad
(15-födda), 3 579 pojkar och flickor. Uppgifter om längd och vikt hämtades från
skolhälsovårdsjournaler. Uppgifter saknas för 4.5%.
Gruppen men barn födda 1981 utgörs av alla 15-födda, 3 158 pojkar och
flickor. Uppgifter om längd, vikt och huvudomfång hämtades ur barnhälsovårdsjournaler samt dessutom längd- och viktuppgifter ur skolhälsovårdsjournaler.
Uppgifter saknas för 1.6%.
Longitudinella data för längd och vikt från födelsen till 19 års ålder samt
huvudomfångsuppgifter från födelsen till 4 års ålder från nationellt representativa
urval av barn insamlade på detta sätt kan användas till epidemiologisk bevakning av
tillväxtförhållanden med marginella bortfallsproblem, jämförbar kvalitet och till en
lägre kostnad jämfört med studier baserade på speciellt vägda och mätta barn och
ungdomar.
De nu aktuella studierna representerar med ett minimalt bortfall den aktuella
tillväxtsituationen och möjliggör jämförelser över tid och jämförelser med andra
nationella tillväxtundersökningar.
Analyser av förändringar över tid (8 år) visar att en ökad frekvens av relativa
viktnedgångar kan konstateras både för pojkar och för flickor. Ökningen är mest
uttalad för flickor, en ökning som sker från en högre frekvens än för pojkar. För flickor
ses ökningen i nästan alla kroppsviktskategorier och i alla åldrar, 7 till 18 år. För
pojkar ökar viktnedgångsfrekvensen i alla kroppsviktskategorier i åldern 7 till 9 år,
men det är en i övrigt mer heterogen bild.
Kroppsvikt och relativ viktnedgång var starkt korrelerad i bägge kohorterna då
fler av de överviktiga än av de smala barnen och ungdomarna minskade sin relativa
vikt. För flickor var ökningen av viktminskning starkast bland de smalaste individerna.
Från 7 års ålder och upp till 18 år existerar en stark positiv sekulär förändring i
body mass index (BMI) i alla åldrar, och frekvensen övervikt och obesitas ökar både
för pojkar och för flickor. Dessutom blir obesitas mer allvarlig.
92
9. ACKNOWLEDGEMENTS
From the very first steps taken in this thesis until the end 19 years later, there are many
to whom I am very grateful. Some have been important since they were there at the
very beginning, some have been important during the whole trajectory and some have
been important in the process of reaching a final solution. Some have always been
important. I thank you all.
From the national ambition in 1988, started and inspired by Gösta
Samuelsson, I was included in the reference and project group for realizing this
project: To create a data base for somatic growth for preschool ages (cohort born in
1981) and in school ages (cohort 1973) with the double aim to make a description of
growth in Sweden and to construct a base for creating ”normal” values for growth.
Since the project group came to an end in 1990, I have been aware of ”the loneliness
of the long distance runner”. I have been surfing on the work of Annika Strandell and
Karin Grundberg and also the work of Gunilla Westin-Lindgren.
Bengt-Erik Ginsburg from the National Board of Health and Welfare was a
strong support. When the main institution for the project, the Skolöverstyrelse, was
reconstructed, it was a major setback. But the show had to go on. Through a very kind
support by Örebro County Council, by supplying resources within the Department of
Public Health and Community Medicine (Samhällsmedicinska enheten, SME) and
within the Division of Child Health Care (Barnhälsovårdsenheten, BHVE), I did
manage to keep on collecting data and register it and finally succeeded in fulfilling this
project. Many thanks to Sven Larsson, Lena Kördel, Margareta Zander, Peggie
Lönngren and most of all to Lars Ekholm (SME) and Leif Ekholm (BHVE). By
giving our group both moral and financial support EpC Socialstyrelsen, through
Anders Ericson and Petra Otterblad-Olausson, acted as door openers in the long
process of data collection. My bosses, during the main part of this period within
Örebro County Council, were Ulf Marcusson and Peter Baeckström. I am also very
grateful to two very important persons within the County council, Raul Björk and
Mats Lindström, in order to maintain the perspective of monitoring public health and
make scientific knowledge a foundation for wise political decisions.
Lennart Bodin, my friend on the 19 years’ journey and my co-author. He is not just a
friend but a giant of statistics. Which is fortunate, since I am but a pygmee of the
above mentioned, possessing only rudimentary skills in data-processing and statistical
analysis. Without the co-operation of Lennart this thesis would never have been
finished.
Lars Ekholm, my close friend who has been involved in the logistics of all my data
collections throughout the years, and finally I succeeded in bringing one project to an
end. With a never-ending patience he has kept in order all the quarter of a million datapoints that are collected in this project.
Leif Svanström, the head of Division of Social Medicine at Karolinska Institutet.
From 1982, when I first became a doctoral student, he has supported and encouraged
me.
93
Gösta Samuelsson, one of the initiators of a national data collection of height and
weight. A member of the society of auxologists and a strong supporter of our project. I
thank him with all my heart.
Bengt Erik Ginsburg, one of the initiators of the project, at that time representing the
National Board of Health and Welfare. He has been supporting our group far beyond
his responsibility.
Annika Strandell, former head of the Skolöverstyrelse and the one who was mainly
responsible for the first data collection wave. Together with Bengt Erik she was
responsible for the start of this project. I miss the Skolöverstyrelse, and I miss working
with Annika in a longitudinal perspective.
John Taranger, friend and an auxological guru. Left the active auxiological science
after reporting from the Solna-study. Fortunately he kept on being a very skilled
auxological speaking partner for me.
Anders Ericson, the late head of EpC at Socialstyrelsen. In the turbulent world of
national epidemiological surveillance of growth in Sweden, he was a strong support,
as long as he was among us, to our project though the years. We have received
financial support by EpC.
Sven Bremberg, my tutor. He came to my rescue when my former tutor moved to
Denmark three years ago.
Anders Magnuson, statistician and co-author in paper IV. We have tested models and
analysis of this paper during one year and re-written the paper hundreds of times.
Thank you for your patience with me.
Finn Diderichsen, an old friend, and though he was not in this field of paediatric
auxological he helped me as a tutor until he left for Denmark.
Sven Larsson, at that time when data collection started in 1989 as the head of
Department of Public Health and Community Medicine in Örebro, let resources within
the department work with the project. I am deeply thankful.
Peggie Lönngren, registered the first half of the amount of data with a remarkable
precision, during 1989 and 1992.
Tove Werner, registered data in the end of data collection.
Moa-Lisa Björk, translating and helpful in the processing from a premature
manuscript to a readable book.
Joel Björk-Werner, registered data in the end of the project.
Tove, Moa and Joel all belong to my loving growing people.
Leif Ekholm, a friend and great support in the final process.
All school nurses, the heroes in every day life helping us to collect data from school
health records all over the country.
All responsible working in the archives in the communities and County councils.
Petra Otterblad Olausson, for helping with MBR data.
Pliktverket in Karlstad, for help with conscript data.
Karin Grundberg, together with Annika Starndell responsible for the first data
collection wave
Staffan Mjönes, member of the first planning group.
Gunilla Westin-Lindgren, member of the first planning group.
Örebro County Council, financial support and helping me to work with this project.
94
10. REFERENCES
Aarts C, Kylberg E, Hofvander Y, Gebre-Medhin M. Growth under privileged
conditions of healthy Swedish infants exclusively breast-fed from birth to 4–6
months: a longitudinal prospective study based on daily records of feeding.
Acta Paediatr 2003;92:145–51
Achadi EL, Hansell MJ, Sloan NL, et al. Women’s nutritional status, iron consumption
and weight gain during pregnancy in relation to neonatal weight and length in
West Java, Indonesia. Int J Gyn Obstet 1995;48 Suppl:S103–19
Agostini C, et al. Growth patterns of breast-fed and formula fed infants in the first 12
months of life: an Italian study. Arch Dis Child 1999;81(5):395–9 Agras WS,
Kraemer HC, Berkowitz RI, Hammer LD. Influence of early feeding style on
adiposity at 6 years of age. J Pediatr 1990;116:805-9
Agras WS, Kraemer HC, Berkowitz RI, Korner AF, Hammer LD. Does vigorous
feeding style influence early development of adiposity? J Pediatr
1987;110:799-804
Agras WS, Kraemer HC, Berkowitz RI, Hammer LD. Influence of early feeding style
on adiposity at 6 years of age. J Pediatr 1990;116:805-9
Ahlborg G Jr, Hogstedt C, Bodin L, Barany S. Pregnancy outcome among working
women. Scand J Work Environ Health 1989;15(3):227-33
Ahlborg G Jr, Bodin L. Tobacco smoke exposure and pregnancy outcome among
working women. A prospective study at prenatal care centers in Orebro
County, Sweden. Am J Epidemiol 1991;133(4):338-47
Alberman E. Are our babies becoming bigger? J R Soc Med 1991;84:257-60
Albertsson-Wikland K, Luo ZC, Niklasson A, Karlberg, J. Swedish population-based
longitudinal reference values from birth to 18 years of age for height, weight
and head circumference. Acta Paediatr 2002;91(7):739–54
Alegre A, Rodriguez-Escudero FJ, Cruz E, Prada M. Influences of work during
pregnancy on fetal weight. J Reprod Med 1984;29(5):334-6
Allen LH, Uauy R. Guidelines for the study of mechanisms involved in the prevention
or reversal of linear growth retardation in developing countries. Eur J Clin Nutr
1994;48 Suppl 1:S212-6
Allison DB, Heshka S, Neale MC, Heymsfield SB. Race effects in the genetics of
adolescents’ Body Mass Index. Int J Obes 1994;18:363-8
Allison DB, Paultre F, Heymsfield SB, Pi-Sunyer FX. Is the intra-uterine period really
a critical period for the development of adiposity? Int J Obes Relat Metab
Disord 1995;19(6):397-402
Ammon O. Die Körpergrosse der Wehrpflichtigen in Grossherzogtum Baden in den
Jahren 1840-1864. Beiträge zur Statistik der Grossherzogtum Baden. Part 5
1894
Andersen L, Lillegaard ITL, Overby N, Lytle L, Klepp K-I, Johansson L. Overweight
and obesity among Norwegian schoolchildren: Changes from 1993 to 2000.
Scand J Public Health 2005;33:99-106
Anonymous. A survey on the physical growth of children under 7 years in the urban
and suburban rural areas of nine cities of China in 1995. Chung-Hua I Hsueh
Tsa Chih (Chinese Medical Journal) 1998;78:187-91
95
Anvisningar och kommentarer till journal inom barnhälsovården. Socialstyrelsen
1981. (Instructions and comments to child health care records. National Board
of Health and Welfare, Sweden) (In Swedish)
Arcini C. Health and disease in early Lund. Archaeologica Lundensia VIII, Lund,1999
Arcini C. Offer för pest, svält och tobak (Victims of plague, famine and tobacco). In
Pestbacken. Arcini C, Jakobsson B, Persson B. Riksantikvarieämbetet 2006 I:I
Stockholm (In Swedish)
Arnesen E, Forsdahl A. The Tromso heart study: cornary risk factors and their
association with living conditions during childhood. J Epidemiol Community
Health 1985;39(3):210-4
Arntzen A, Samuelsen SO, Magnus P, Bakketeig LS. Birth weight related to social
indicators in Norway. Eur J Public Health 1994;4:92-7
Arteaga H, Dos Santos JE, Dutra de Oliveira JE. Obesity among schoolchildren of
different socio-economic levels in a developing country. Int J Obes
1982;6:291-7
Ashcroft MT, Lovell HA. Heights and weights of Jamaican children of various racial
origins. Trop Geogr Med 1964;4:346-53
Ashcroft MT, Heneage P, Lovell HA. Heights and weights of Jamaican schoolchildren
of various ethnic groups. Am J Phys Anthropol 1966;24:35-44
Ashizawa K, Kawabata M. Daily measurements of the heights of two children from
June 1984 to May 1985. Ann Hum Biol 1990;17(5):437-43
Ayatollahi SM. Obesity in school children and their parents in southern Iran. Int J
Obes 1992;16:845-50
Baird D. The epidemiology of low birth weight: changes in incidence in Aberdeen,
1948-72. J Biosoc Sci 1974;6:623-41
Baird D,. Environment and reproduction. Br J Obstet Gynaecol 1980;87:1057-67
Baker J, Liu JP, Robertson EJ, et al. Role of insulin-like growth factors in embryonic
and postnatal growth. Cell 1993;75:73-82
Baker P. ed. The Biology of High Altitude Peoples. Cambridge: Cambridge University
Press,1977
Barker DJP, Winter PD, Osmond C, Margetts B, Simmonds SJ. Weight in infancy and
death from ischaemic heart disease. Lancet 1989;ii:577-80
Barker DJP, ed. Fetal and infant origins of adult disease. London: BMJ Publications
Group,1992
Barker DJP, Godfrey KM, Osmond C, Bull AR. The relation of foetal length, ponderal
index and head circumference to blood pressure and the risk of hypertension in
adult life. Paediatr Perinat Epidemiol 1992;6:35-44
Barker DJP, Martyn CN, Osmond C, Hales CN, Fall CHD. Growth in utero and serum
cholesterol concentrations in adult life. BMJ 1993;307:1524-7
Barker DJP, Osmond C, Simmonds SJ, Wield GA The relation of small head
circumference and thinness at birth to death from cardiovascular disease in
adult life BMJ 1993;306:422-5
Barker M, Robinson S, Osmond C, Barker DJ. Birth weight and body fat distribution
in adolescent girls. Arch Dis Child 1997;77(5):381-3
Barker DJP, Forsen T, Uutela A, Osmond C, Eriksson JG. Size at birth and resilience
to effects of poor living conditions in adult life: a longitudinal study. BMJ
2001;323:1273-6
96
Barker DJP, Forsen T, Eriksson JG, Osmond C. Growth and living conditions in
childhood and hypertension in adult life:a longitudinal study. J Hypertension
2002;20:1951-6
Barlow SE, Dietz WB. Obesity evaluation and treatment: Expert Committee
recommendations. The Maternal and Child Health Bureau, Health Resources
and Services Administration and the Department of Health and Human
Services. Pediatrics 1998;102(3): URL:
www.pediatrics.org/cgi/content/full/102/3/e29.98393765
Bateson P. Fetal experience and good adult design. Int J Epidemiol 2001;30:928-34
Bellizzi MC, Dietz WH. Workshop on childhood obesity: summary of the discussion.
Am J Clin Nutr 1999;70(1):S173-5
Benn RT. Some mathematical properties of weight-for-height indices used as
measures of adiposity. Br J Prev Med 1971;25:42
Bennike P. Paleopathology of Danish skeletons. Copenhagen:Akademisk Forlag,1985
Berg I-M, Simonsson B, Brantefors B, Ringquist T. Prevalence of overweight and
obesity in children and adolescents in a county in Sweden. Acta Paediatr
2001;90:671-6
Bergström E, Hernell O, Persson LÅ. Dietary changes in Swedish adolescents. Acta
Paediatr 1993;82:472-80
Bergström E, Stenlund H, Svedjehäll B. Assessment of Body Perception among
Swedish Adolescents and Young Adults. J Adolesc Health 2000;26:70-5
Berkowitz RI, Agras WS, Korner AF, Kraemer HC, Zeanah CH. Physical activity and
adiposity: A longitudinal study from birth to childhood. J Pediatr
1985;106:734-8
Berney L, Biane D, Davey Smith G, et al. Lifecourse influence on health in early old
age. In: Graham H, ed. Understanding health inequalities. Buckingham and
Philadelphia: Open University Press, 2000
Bernstein IM, Horbar JD, Badger GJ, et al. Morbidity and mortality among very-lowbirth-weight neonates with intra-uterine growth restriction. Am J Obstet
Gynecol 2000;182:198-206
Beunen GP, Malina RM, Renson R, Simons J, Ostyn M, Lefevre J. Physical activity
and growth, maturation and performance: a longitudinal study. Med Sci Sports
Exerc 1992;24(5):576-85
Bielicki T, Welon Z. Growth data as indicators of social inequalities: the case of
Poland. Yearbook of Physical Anthropology 1982;25:153-67
Bielicki T, Charzewski J. Body height and upward social mobility. Ann Hum Biol
1983;10:403-8
Bielicki T, Waliszko Q, Hulanicka B, Kotlarz K. Social-class gradients in menarcheal
age in Poland. Ann Hum Biol 1986;13:1-11
Bielicki T. Physical growth as a measure of the economic well-being of populations:
the 20th century. In: Falkner F, Tanner JM, eds. Humans growth: a
comprehensive treatise. Vol.3 2nd Ed. New York:Plenum Press,1986. p 283-305
Bielicki T, Waliszko H. Stature, upward social mobility and the nature of statural
differences between social classe. Ann Hum Biol 1992;19(6):589-93
Billewicz WZ. A note on body weight and seasonal variation. Hum Biol 1967;39:24150
Billewicz WZ, MacGregor IA. A birth to maturity longitudinal study of heights and
weights in two West African (Gambian) villages, 1951-1975. Ann Hum Biol
1982;9:309-20
97
Bjerregaard P, Hansen JC. Effects of smoking and marine diet on birthweight in
Greenland. Arctic Med Res 1996;55(4):156-64
Blane D, Hart CL, Smith GD, Gillis CR, Hole DJ, Hawthorne VM. Association of
cardiovascular disease risk factors with socio-economic position during
childhood and during adulthood. BMJ 1996;313(7070):1434-8
Blomquist HK, Bergström E. Obesity in 4-year-old children more prevalent in girls
and in municipalities with a low socioeconomic level. Acta Paediatr
2007;96:113-6
Boas F. The growth of children, II. Science 1892;19:281-2
Boas F. Changes in the bodily form of descendants of immigrants. Am Anthropol
1912;14:530-63
Boas F. Report on an anthropometric investigation of the population of the United
States. J Am Stat Assoc 1922;18:181-209
Boas F. Observations on the growth of children. Science 1930;72:4-48
Boas F. Race, Language, & Culture. New York: Free Press,1940
Bock RD, Sykes RC. Evidence for continuing secular increase in height within
families in the United States. Am J Hum Biol 1989;1:143-8
Bodin L. Personnel communication
Boersma B, Wit JM. Catch-up growth. Endocr Rev 1997;18:646-61
Bogin BA. Periodic Rhythm in the Rates of Growth in Height and Weight of Children
and its Relation to Season of the Year. Ph.D. Dissertation. Ann Arbor:
University Microfilms,1977
Bogin BA. Seasonal pattern in the rate of growth in height of children living in
Guatemala. Am J Phys Anthropol 1978;49:205-10
Bogin B. Rural-to-urban migration. In Biological Aspects of Human Migration, ed.
CGN Mascie-Taylor & GW Lasker, Cambridge:Cambridge University
Press,1988. p 90-129
Bogin B. Patterns of Human Growth. 2nd edition. Cambridge University Press,1999
Bogin B, Keep R. Eight thousand years of economic and political history in Latin
America revealed by anthropometry. Ann Hum Biol 1999;26(4):333-51
Bolzan AG, Guimarey LM, Pucciarelli HM. Crecimento y dismorfismo sexual de
escolares segun la ocupation laboral paterna. Archivos Latinoamericanos de
Nutricion 1993;43:132-8
Bonellie SR, Raab GM. Why are babies getting heavier? Comparison of Scottish
births from 1980 to 1992. BMJ 1997;315(7150):1205
Booth M, Wake M, Armstrong T, Chey R, Hesketh K, Mathur S. The epidemiology of
overweight and obesity among Australian children and adolescents, 1995-1997.
Aust N Z J Public Health 2001;25(2):162-9
Borghi E, de Onis M, Garza C, Van den Broeck J, Frongillo A, Grummer-Strawn L,
Van Buuren S, Pan H, Molinari L, Martorell R, Onyango AW, Martines JC.
Construction of the World Health Organization child growth standards:
selection of methods for attained growth curves. Statist Med 2006;25:247-65
Bowditch HP. The growth of children: A supplementary investigation. Tenth Annual
Report. Massachusetts State Board of Health,1879. p 35-62 Also :Eighth
Annual report of the State Board of Health of Mass. Boston,1877
Braddon FEM, Rodgers B, Wadsworth MEJ, Davies JM. Onset of obesity in a 36 year
birth cohort study. BMJ 1986;293:299-303
Brieger GM, Rogers MS, Rushton AW, Mongeli M. Are Hong Kong babies getting
bigger? Int J Gynaecol Obstet 1997;57(3):267-71
98
Brock CG, Huntley RM, Slack J. Influence of heredity and environment in
determination of skinfold thickness in children. BMJ 1975;2:719-21
Broman B, Dahlberg G, Lichtenstein A. Height and weight during growth. Acta
Paediatr 1942;30:1-66
Bronfenbrenner U. The Ecology of Human Development: Experiments by Nature and
Design. Cambridge, MA: Harvard University Press, 1979
Brooke OG, Anderson HR, Bland JM, Peacock JL, Stewart CM. Effects on birth
weight of smoking, alcohol, caffeine, socioeconomic, and psychosocial stress.
BMJ 1989;298:795-801
Brownell KD, Rodin J. The dieting maelstrom. Is it possible and advisable to lose
weight? American Psychologist 1994;49(9):162-9
Brundtland GH, Liestöl K, Wallöe L. Height and weight of schoolchildren and
adolescent girls and boys in Oslo 1970. Acta Paediatr Scand 1975;65:565-73
Brundtland GH, Liestöl K, Wallöe L. Height, weight and menarcheal age of Oslo
schoolchildren during the last 60 years. Ann Hum Biol 1980;7:307-22
Buckler J, Tanner J. Revision and update of Tanner-Whitehouse clinical longitudinal
charts for height and weight. Eur J Pediatr 1997;156:248-9
Buffon G. Histoire Naturelle, generale et particuliere avec la description du cabenet de
Roi. 44 Vols. Fourth Supplement. Paris:Imprimerie Royal,1777
Bulik CM, Sullivan PF, Tozzi F, et al. Prevalence, heretability, and prospective risk
factors for anorexia nervosa. Arch Gen Psychiatry 2006;63:305-12
Bundak R, Furman A, Gunoz H, Darendeliler F, Bas F, Neyzi. Body mass index
references for Turkish children. Acta Paediatr 2006;95:194-8
Bundred P, Kitchiner D, Buchan I. Prevalence of overweight and obese children
between 1989 and 1998: population based series of cross-sectional studies.
BMJ 2001;322:1-4
Burns TL, Moll PP, Lauer RM. The relation between ponderosity and coronary risk
factors in children and their relatives. The Muscatine Ponderosity Family
Study. Am J Epidemiol 1989;129:973-87
Butler NR, Goldstein H, Ross EM: Cigarette smoking in pregnancy: its influence on
birth weight and perinatal mortality. BMJ 1972;2:127-30
Börjeson M. Overweight children. Acta Paedriatr Scand 1962;51 Suppl 132:23-4
Börjeson M. The aetiology of obesity in children. Acta Paediatr Scand 1976;65:279-87
Cacciari E, Corazza GR, Salardi S, Pascucci MG, Tacconi M, Cicognani A, Tassinari
D, Biasco G, Volta U, Lazzari R, Gasbarrani G. What will be the adult height
of coeliac patients? Eur J Pediatr 1991;150:407-9
Camerer W. Untersuchungen uber Massenwachstum und Längenwachstum der
Kinder. Jahrbuch fur Kinderheilkunde 1893;18:254-64
Cameron N. British growth charts for height and weight with recommendations
concerning their use in auxological assessment. Ann Hum Biol 2002;29(1):110
Carr-Hill R, Campbell DM, Hall MH, Meredith A. Is birth weight determined
genetically? BMJ 1987;295:687-9
Cernerud L. Growth and social conditions. Height and weight of Stockholm
schoolchildren in a public health context. Doctoral thesis, The Nordic School
of Public Health, Göteborg, Sweden,1991
Cernerud L. Height and body mass index of seven-year-old Stockholm schoolchildren
from 1940 to 1990. Acta Paediatr 1993;82:304-5
99
Chadwick E. Report on the Sanitary Condition of the Labouring Population of Great
Britain.1842. Reprint. Edinburgh:Edinburgh University Press,1965
Chamberlain R, Chamberlain G, Howlett B, Claireaux A. British Births 1970: a survey
under the joint auspices of the National Birthday Trust Fund and the Royal
College of Obstetricians and Gynaecologists. Vol.1:The First Week of Life.
Heineman Medical: London,1975
Champagne CM, Baker NB, Delany JP, Harsha DW, Bray GA. Assessment of energy
underreporting by doubtly labelled water and observations on reported nutrient
intakes in children. J Am Diet Assoc 1998;98:426-33
Charney E, Goddman HC, McBride M, Lyon B, Pratt R. Childhood antecedents of
adult obesity: Do chubby infants become obese adults? N Engl J Med
1976;295:6-9
Chen L, Chowdhury AKM, Huffman SL. Anthropometric assessment of energyprotein malnutrition and subsequent risk of mortality among pre-school aged
children. Am J Clin Nutr 1980;33:1836-45
Chike-Obi U, David RJ, Coutinho R, Wu SY. Birth weight has increased over a
generation. Am J Epidemiol 1996;144(6):563-9
Ching RJ, Chen JQ, Han HT, Ma LH, Gu YF, Cui KX, Song X. Physical growth of
Chinese children from birth to 6 years old in urban and rural areas in 1985. J
Trop Pediatr 1989;35(5):255-60
Chinn S. Mixed longitudinal studies: their efficiency for the estimation of trends over
time. Ann Hum Biol 1988;15(6):443-54
Chinn S, Rona RJ, Price CE. The secular trend of height in primary schoolchildren in
England and Scotland 1972-9 and 1979-86. Ann Hum Biol 1989;16:387-95
Chinn S, Rona RT. Prevalence and trends in overweight and obesity in three cross
sectional studies of British children, 1974-94. BMJ 2001;322:24-6
Clarke JI. Mobility, location and society. In Migration and Mobility: Biosocial Aspects
of Human Movement, edited by AJ Boyce, Symposia of the Society for the
Study of Human Biology, Vol 23 London:Travor & Francis,1984
Clayton PE, Gill MS, Hall CM, Tillman V, Whatmore AJ, Price DA. Serum leptin
through childhood and adolescence. Clin Endocrinol 1997;46:727-33
Clegg EJ, Pawson IG, Ashton EH, Flinn RM. The growth of children at different
altitudes in Ethiopia. Philosophical Transactions of the Royal Society of
London 1972;264B:403-37
Cleveland WS. Robust locally weighted regression and smoothing scatterplots. J Am
Stat Assoc 1979;74:829–36
Cliquet RL. Social mobility and the anthropological structure of populations. Hum
Biol 1968;40:17-43
Cnattingius S, Ericson A, Gunnarskog J, Kallen B. A quality study of a medical
registry. Scand J Soc Med 1990;18(2):143-8
Cole TJ. Relating growth rate to environmental factors – methological problems in the
study of growth-infection interaction. Acta Paediatr 1989;Suppl 350:14-20
Cole TJ, Green PJ. Smoothing references centile curves: the LMS method and
penalized likelihood. Stat Med 1992;11:1305-19
Cole TJ. Conditional reference charts to assess weight gain in British infants. Arch Dis
Child 1995;73:8-16
Cole TJ, Freeman JV, Preece MA. Body mass index references curves for the UK,
1990. Arch Dis Child 1995;73:25-9
100
Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight,
height, body mass index and head circumference fitted by maximum penalized
likelihood. Stat Med 1998;17:407-29
Cole TJ, Roede MJ. Centiles of body mass index for Dutch children aged 0 - 20 years
in 1980 - a baseline to assess recent trends in obesity. Ann Hum Biol
1999;26(4):303-8
Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for
child overweight and obesity worldwide: international survey. BMJ
2000;320:1240–53
Cole TJ. Secular trends in growth. Proceedings of the Nutrition Society 2000;59:31724
Cole TJ. Galton’s midparent height revisited. Ann Hum Biol 2000;27(4):401-5
Cole TJ, Paul AA, Whitehead RG. Weight reference charts for British long-term
breastfed infants. Acta Paediatr 2002;91:1296-1300
Conti J, Abraham S, Taylor A. Eating behaviour and pregnancy outcome. J
Psychosom Res 1998 ;44:465-77
Cooper C, Kuh D, Egger P, Wadworth M, Barker D. Childhood growth and age at
menarche. Br J Obstet Gynaecol 1996;103(8):814-7
Court J. Obesity in childhood. Med J Aust 1977;1:888-91
Coutant R, Carel JC, Letrat M, et al. Short stature associated with intrauterine growth
retardation: final height of untreated and growth hormone treated children. J
Clin Endocrinol Metab 1998;83:1070-4
Cresswell JL, Barker DJP, Osmond C, Egger P, Philips DIW, Fraser RB. Fetal growth,
length of gestation, and polycystic ovaries in adult life. Lancet 1997;350:11315
Curhan GC, Chertow GM, Willett WC, Spiegelman D, Colditz GA, Manson JE,
Speizer FE, Stampfer MJ. Birth weight and adult hypertension and obesity in
women. Circulation 1996;94(6):1310-5
Curhan GC, WillettWC, Rimm EB, Spiegelman D, Ascherio AL, Stampfer MJ. Birth
weight and adult hypertension, diabetes mellitus, and obesity in US men.
Circulation 1996;94(12):3246-50
Dabelea D, Pettitt DJ, Hansson RL, Imperatore G, Bennett PH, Knowler WC. Birth
weight, type II diabetes, and insulin resistance in Pima Indian children and
young adults. Diabetes Care 1999;22:944-50
Dagbjartsson A, Palsson G, Arnorsson VH, Thorsson AV. Height and weight of
Icelandic children aged 6 to 16 years. A cross-sectional study. Acta Paediatr
Scand 1989;Suppl 359:21
Dahlberg G. Korrelationserscheinungen bei nicht erwachsenen Individuen und eine
Theorie über den Wachstumsmechanismus im Hinblick auf intermittierende
Umweltsfaktoren. Z Morphol Anthropol 1931;29:288-306
Davies DP, Gray OP, Ellwood PC, et al. Cigarettte smoking in pregnancy:
Associations with maternal weight gain and fetal growth. Lancet 1976;385-7
Davey Smith G, Greenwood R, Gunnell DJ, Sweetnam P, Yarnell J, Elwood P. Leg
length, insulin resistance, and coronary heart disease risk: the Caerphilly Study.
J Epidemiol Community Health 2001;55:867-72
Department of Health. Health inequalities: Health Secretary announces new plans to
improve health in poorest areas. 28th February, 2001 URL:
http://www.doh.gov.uk/ healthinequalities/press.htm
101
Dewey KG, Heinig MJ, Nommsen LA, Peerson JM, Lönnerdal B. Growth of breastfed and formula-fed infants from 0 to 18 months: The DARLING Study.
Pediatrics 1992;89(6):1035–41
Dewey KG, Peerson JM, Brown KH, Krebs NF, Michaelsen KF, Persson LA, et al.
World Health Organisation Working Group on Infant Growth. Growth of
breast-fed infants deviates from current reference data: a pooled analysis of US,
Canadian and European data sets. Pediatrics 1995;96:495-503
Dickinson F, Cervera M, Murguia R, Uc L. Growth, nutritional status and
environmental change in Yucatan, Mexico. Stud Hum Ecol 1990;9:135-49
Dietz WH, Jr., Hartung R. Changes in height velocity of obese preadolescents during
weight reduction. AJDC 1985;139:705-7
Dietz WH, Gortmaker SL. Do we fatten our children at the TV set? Obesity and
television viewing in children and adolescents. Pediatrics 1985;75:807-12
Dietz WH. Critical periods in childhood for the development of obesity. Am J Clin
Nutr 1994;59:955-9
Dietz WH. Health consequences of obesity in youth: childhood predictors of adult
disease. Pediatrics 1998;101(Suppl):518-25
DiPietro L, Mossberg H-O, Stunkard AJ. A 40-years history of overweight children in
Stockholm: life-time, overweight, morbidity, and mortality. Int J Obes
1994;18:585-90
Divers MJ, Lilford RJ. Infection and preterm labour: a meta-analysis. Contemp Rev
Obstet Gynaecol 1993;5:71-84
Dovertie Hygiae 1895;57:254
Dovertie 1914 (Cited in Törnell 1920)
Duncan B, Schaefer C, Sibley B, Fonseca NM. Reduced growth velocity in
exclusively breast-fed infants. AJDC 1984;138:309-13
Dössing J. Gennemsnitsvaedier for vaegt-höjde-alder forhold for drenga og piger i
skole-åldern. (Means for weight-height-age relations for boys and girls of
school age). Ugeskr Laeger 1950;112:1171-81 (In Danish)
Dössing J. Determination of Individual Normal Weights of Schoolchildren.
Copenhagen: Ejnar Munksgaard,1952
Edlund B, Hallqvist G, Sjödén P-O. Attitudes to food, eating and dieting behaviour in
11 and 14-year-old Swedish children. Acta Paediatr 1994;83(6):572-7
Edlund B, Sjödén P-O, Gebre-Medhin M. Anthropometry, body composition and body
image in dieting and non-dieting 8-16-year-old Swedish girls. Acta Paediatr
1999;88:537-44
Edlund B, Halvarsson K, Gebre-Medhin M, Sjödén P-O. Psychological correlates of
dieting in Swedish adolescents: A cross-sectional study. Eur Eat Disord Rev
1999;7:47-61
Ekblom Ö, Oddsson K, Ekblom B. Health-related fitness in Swedish adolescents
between 1987 and 2001. Acta Paediatr 2004;93:681-6
Ekbom A, Hisieh C, Lipworth L, Wolk A, Ponte’n J, Adami H-O, Trichopoulus D.
Perinatal characteristics in relation to incidence of and mortality of prostate
cancer. BMJ 1996;313:337-41
Ellard GA, Johnstone FD, Prescott RJ, Ji-Xian W, Jian-Hua M. Smoking during
pregnancy: the dose dependence of birthweight deficits. Br J Obstet Gynaecol
1996;103(8):806-13
Elliot KG, Kjolhede CL, Gournis E, Rasmussen KM. Duration of breastfeeding
associated with obesity during adolescence. Obes Res 1997;5(6):538-41
102
Ellison PT. Skeletal growth, fatness and menarcheal age: a comparison of two
hypothesis. Hum Biol 1982;54:269-81
Elmén H, Höglund D, Karlberg P, et al. Birth weight for gestational age as a health
indicator: birth weight and mortality measures at a local level. Eur J Public
Health 1996;6:137-41
Emanuel I, Filakti H, Alberman E, et al. Intergenerational studies of human birth
weight from the 1958 birth cohort. 1. Evidence for a multigenerational effect.
Br J Obstet Gynaecol 1992;99:67-74
England LJ, Kendrick JS, Wilson HG, Merritt RK, Gargiullo PM, Zahniser SC.
Effects of smoking reduction during pregnancy on the birth weight of term
infants. Am J Epidemiol 2001;154(8):694-701
English DR, Hulse GK, Milne E, Holman CD, Bower CI. Maternal cannabis use and
birth weight: a meta-analysis. Addiction 1997;92(11):1553-60
Eriksson J, Forsen T, Tuomiletho J, Osmond C, Barker D. Size at birth, childhood
growth and obesity in adult life. Int J Obes 2001;25(5):735-40
Eriksson M, Rasmussen F, Nordqvist T. Changes in shape and location of BMI
distribution of Swedish children. Acta Paediatr 2005;94:1558-65
Eveleth PB, Tanner JM. Worldwide Variation in Human Growth. 2. ed. Cambridge:
Cambridge University Press,1990
Felts WM, Parrillo AV, Chenier T, et al. Adolescents’ perceptions of relative weight
and self-reported weight-loss activities: Analysis of 1990 YRBS (Youth Risk
Behavior Survey) national data. J Adolesc Health 1996;18:20-6
Fenster L, Eskenazi B, Windham GC, Swan SH. Caffeine consumption during
pregnancy and fetal growth. Am J Public Health 1991;81:458-61
Fewtrell MS, Morley R, Abbott RA, et al. Catch-up growth in small for gestational age
term infanst: a randomised trial. Am J Clin Nutr 2001;74:516-23
First M, Spitzer R, Gibbon M, et al. Structured Clinical Interview for DSM-IV Axis I
Disorders. Research Version, Patient Edition. New York, Biometrics Research,
New York State Psychiatric Institute,1997
Fitzhardinge PM, Inwood S. Long-term growth in small-for-date children. Acta
Peadiatr Scand 1989;Suppl 349:27-33
Floud R, Wachter K, Gregory A. Height, health and history. Nutritional status in the
United Kingdom, 1750 - 1980. Cambridge studies in population economy and
society in past time. Cambridge University Press,1990
Fogel RW. Physical growth as a measure of the economic wellbeing of populations:
The eighteenth and nineteenth century. In Human Growth, 2nd edn, Vol. 3 ed.
F. Falkner & JM Tanner. New York: Plenum,1986. p 263-81
Fogel RW. New sources and new techniques for the study of secular trends in
nutritional status, health, mortality, and the process of aging. Historical
Methods 1993;26:5-43
Fombonne E. Anorexia nervosa. No evidence of an increase. Br J Psychiat
1995;166:462-71
Fombonne E. Is bulimia nervosa increasing in frequency? Int J Eat Disord
1996;19:287-96.
Forsen T, Eriksson JG, Tuomiletho J, Osmond C, Barker DJ. Growth in utero and
during childhood among women who develop coronary heart disease:
longitudinal study. BMJ 1999;319(7222):1403-7
103
Forsen T, Eriksson J, Tuomiletho J, Reunanen A, Osmond C, Barker D. The fetal and
child growth of persons who develop type 2 diabetes. Ann Intern Med
2000;133(3):176-82
Forsum E, Boström K, Eriksson B, Olin-Skoglund S. Kvinnans vikt före och under
graviditet har betydelse för barnet.(A woman’s weight before and during
pregnancy is of importance to her infant). Läkartidningen 2003;100:3954-8 (In
Swedish)
Fortier I, Marcoux S, Beaulac-Baillargeon L. Relation of caffeine intake during
pregnancy to intrauterine growth retardation and preterm birth. Am J Epidemiol
1993;137:931-40
Foster JM, Chinn S, Rona RJ. The relation of the height of primary schoolchildren to
population density. Int J Epidemiol 1983;12:199-204
Fredriks AM, van Buuren S, Wit JM, Verloove-Vanhorick SP. Body index
measurements in 1996-7 compared with 1980. Arch Dis Child 2000;82:107-12
Fredriks AM, van Buuren S, Burgmeijer RJ, Meulmeester JF, Beuker RJ, Brugman E,
Roede MJ, Verloove-Vanhorick SP, Wit JM. Continuing positive secular
growth change in the Netherlands 1955-1997. Pediatr Res 2000;47(3):316-23
Fredriks AM, van Buuren S, Jeurissen SER, Dekker FW, Verloove-Vanhorick SP, Wit
JM. Height, weight, body mass index and pubertal development references for
children of Turkish origin in the Netherlands. Eur J Ped 2003;162:788-93
Fredriks AM, van Buuren S, Jeurissen SER, Dekker FW, Verloove-Vanhorick SP, Wit
JM. Height, weight, body mass index and pubertal development references for
children of Moroccan origin in The Netherlands. Acta Paediatr 2004;93:817-24
Fredriks AM, van Buuren S, Hira Sing RA, Wit JM, Verloove-Vanhorick SP.
Alarming prevalences of overweight and obesity for children of Turkish,
Moroccan and Dutch origin in the Netherlands according to international
standards. Acta Paediatr 2005;94:496-8
Freeman JV, Power C, Rodgers B. Weight-for-height indices of adiposity:
relationships with height in childhood and early adult life. Int J Epidemiol
1995;24(5):970-6
Freeman JV, Cole TJ, Chinn S, Jones PRM, White EM, Preece MA. Cross-sectional
stature and weight reference curves for the UK, 1990. Arch Dis Child
1995;73:17–24
French SA, Perry CL, Leon GR, et al. Dieting behaviors and weight change history in
female adolescents. Health Psychol 1995;14:548-55
Frisancho AR, Baker PT. Altitude and growth: a study of the patterns of physical
growth of a high altitude Peruvian Quechua population. Am J Phys Anthropol
1970;32:279-92
Frisancho AR, Borkan GA, Klayman JF. Pattern of growth of Lowland and Highland
Peruvian Quechua of similar genetic composition. Hum Biol 1975;47:233-43
Frisancho AR, Guire K, Babler W, Borkan G, Way A. Nutritional influence on
childhood development and genetic control of adolescent growth of Quechuas
and Mestizos from the Peruvian lowlands. Am J Phys Anthropol 1980;52:36775
Frisch RE, Revelle R. Height and weight at menarche and a hypothesis of critical body
weights and adolescent events. Science 1970;169:397-9
Frisch RE, Wyshak G, Vincent L. Delayed menarche and amenorrhea in ballet
dancers. N Engl J Med 1980;303:17-9
104
Fulwood R, Abraham S, Johnson C. Height and Weight of Adults Ages 18-74 Years
by Socioeconomic and Geographical Variables. Vital and Health Statistics,
Series 11, No. 224, DHEW Pub. No. (PHS) 81-1674. Washington DC: US
Government Printing Office, 1981
Galton F. Regression towards mediocrity in hereditary stature. Journal of the
Anthropological Institute of Great Britain and Ireland 1886;15:246-63
Garman AR, Chinn S, Rona RJ. Comparative growth of primary schoolchildren from
one and two parent families. Arch Dis Child 1982;57:453-8
Garn SM, Bailey SM, Solomon MA, Hopkins PJ. Effect of remaining family members
on fatness prediction. Am J Clin Nutr 1981;34:148-53
Garn SM, Hopkins PJ, Ryan AS. Differential fatness gain of low income boys and
girls. Am J Clin Nutr 1981;34(8):1465-8
Garn SM. The secular trend in size and maturation timing and its implication for
nutritional assessment. J Nutr 1987;117:817-23
Garner P, Panpanich R, Logan S. Is routine growth monitoring effective? Arch Dis
Child 2000;82(3):197-201
Garralda MD. Evolution of human height. In: Hernandez M, Argente J.(eds) Human
growth: Basic and clinical aspects. Proceedings of the Sixth Congress of
Auxology, Madrid, Spain, Excerpta Medica: Amsterdam 1992. p 135-41
Garza C, Frongillo E, Dewey KG. Implications of growth patterns of breast-fed infants
for growth references. Acta Paediatr 1994;Suppl 402:4-10
Garza C, de Onis M. A new international growth reference for young children. Am J
Clin Nutr 1999; 70:S169-72
Gasser T, Ziegler P, Seifert B, Prader A, Molinari L, Largo R. Measures of body mass
and of obesity from infancy to adulthood and their appropriate transformation.
Ann Hum Biol 1994;21(2):111-25
Gasser T, Sheehy A, Molinari L, Largo RH. Growth of early and late maturers. Ann
Hum Biol 2001;28(3):328-36
Gelander L, Karlberg J, Albertsson-Wikland K. Seasonality in lower length and height
velocity in prepubertal children. Acta Paediatr 1994;83:1249-54
Gerver WJ, De Bruin R, Drayer NM. A persisting secular trend for body measurement
in Dutch children. The Oosterwolde II Study. Acta Paediatr 1994;83:812-4
Gillman MW, Rifas-Shiman SL, Camargo CA, Jr., Berkey CS, Frazier AL, Rockett
HR, et al. Risk of overweight among adolescents who were breastfed as infants.
JAMA 2001;285(19):2461-7
Gluckman P, Harding JE. Nutritional and hormonal regulation of fetal growth;
evolving concepts. Acta Paediatr 1994;Suppl 399:60-3
Godfrey K, Robinson S. Maternal nutrition, placental growth and fetal programming.
Proc Nutr Soc 1997;57:105-11
Golden MHN. Nutritional deficiency as a cause of growth failure. In: Hernandez M,
Argente J.(eds) Human growth: basic and clinical aspects. Proceedings of the
Sixth International Congress of Auxology, Madrid, Spain,1991. Excerpta
Medica:Amsterdam, 1992. p 175-82
Goldstein MS. Demographic and Bodily Changes in Descendants of Mexican
Immigrants. Austin: Institute of Latin-American Studies,1943
Gohlke BC, Khadilkar VV, Skuse D, Stanhope R. Recognition of children with
psychosocial short stature: a spectrum of presentation. J Pediatr Endocrinol
Metab 1998;11:509-17
105
Golkhe BC, Stanhope R. Final height in psychosocial short stature: is there complete
catch-up? Acta Paediatr 2002;91:961-5
Goran MI, Shewchuk R, Gower BA, Nagy TR, Carpenter WH, Johnson RK.
Longitudinal changes in fatness in white children: no effect of childhood
energy expenditure. Am J Clin Nutr 1998;67:309-16
Gortmaker SL, Must A, Perrin JM, Sobol AM, Dietz WH. Social and economic
consequences of overweight in adolescence and young adulthood. New Engl J
Med 1993;329:1008-12
Gortmaker S, Must A, Sobol A, Peterson K, Coldiz GA, Dietz WH. Television
viewing as a cause of increasing obesity among children in the United States.
Arch Pediatr Adolesc Med 1996;150:356-62
Greulich WW. Some secular changes in the growth of American-born and native
Japanese children. Am J Phys Anthropol 1976;45:553-68
Grinker JA, Garn SM, Rosenberg KR. Effects of parental weight and fatness on
relative weight and fatness of young children. Comparison of cross-sectional
and prospective clinical data. Ann NY Acad Sci 1989;575:521-4
Groff JY, Mullen PD, Mongoven M, Burau K. Prenatal weight gain patterns and infant
birthweight associated with maternal smoking. Birth 1997;24(4):234-9
Grummer-Strawn LM, Garza C, Johnson CL. Childhood growth charts. Pediatrics
2002;109(1):141-2
Guillaume M, Lapidus L, Beckers F, Lambert A, Björntorp P. Familial trends of
obesity through three generations: the Belgian-Luxembourg child study. Int J
Obes1995;19 Suppl 3:S5-9
Gulliford MC, Price CE, Rona RJ, Chinn S. Sleep habits and height at ages 5 and 11.
Arch Dis Child 1990;65:119-22
Gulliford MC, Chinn S, Rona RJ. Social environment and height: England and
Scotland 1987 and 1988. Arch Dis Child 1991;66:235-40
Gunnarsdottir I, Thorsdottir I. Relationship between growth and feeding in infancy and
body mass index at the age of 6 years. Int J Obes 2003;27:1523-7
Gunnell D, Okasha M, Smith GD, Oliver SE, Sandhu J, Holly JM. Height, leg length,
and cancer risk: a systematic review. Epidemiol Rev 2001;23:296-325
Guo SS, Chumlea WC. Tracking of body mass index in children in relation to
overweight in adulthood. Am J Clin Nutr 1999;70:S145-8
Gustavson-Larson AM, Dale Terry R. Weight-related behaviors and concerns of
fourth-grade children. J Am Diet Assoc 1992;92:818-22
Guthrie H, Picciano MF. Human Nutrition. St. Louis: Mosby,1995
Hall RL. Sexual dimorphism for size in seven nineteenth century Northwest Coast
populations. Hum Biol 1978;50:159-71
Hallman N, Bäckström-Järvinen, Kantero R-L, Tiisala R. Finnish centre for study in
child growth and development. Annales paediatrician Fenniae 1966;12:8-12
Hamill PVV, Johnston FE, Lemshow S. Body weight, stature, and sitting height: white
and Negro youths 12-17 years, United States. DHEW Publication No. (HRA)
74-1608. Washington, DC: US Government Printing Office,1973
Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF. NCHS growth curves for
children birth-18 years. United States. Vital Health Stat 11 1977(165):i-iv:1-74
Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. Physical
growth: National Center for Health Statistics percentiles. Am J Clin Nutr
1979;32(3):607-29
106
Harrell JS, Gansky SA, Bradley CB, McMurray RG. Leisure time activities of
elementary schoolchildren. Nurs Res 1997;46:246-53
Hart N. Famine, maternal nutrition and infant mortality: a re-examination of the Dutch
Hunger Winter. Popul Stud (Camb) 1993;47:27-46
Haschke F, van’t Hof MA, and the Euro-Growth Study Group. Euro-Growth
references for length, weight, and body circumferences. J Pediatr Gastroenterol
Nutr 2000; 31 Suppl 1: S14–38
Hashimoto N, Kawasaki T, Kikuchi T, Takahashi H, Uchiyama. Influence of parental
obesity on the physical constitution of preschool children in Japan. Acta
Paediatr Jpn 1995;37:150-3
Hatch EE, Bracken MB. Effect of marijuana use in pregnancy on fetal growth. Am J
Epidemiol 1986;124:986-93
Hattersley AT, Beards F, Ballantyne E, et al. Mutations in the glucokinase gene of
the fetus result in reduced birth weight. Nat genet 1998;19:268-70
Hattori Y, Vera JC, Rivas CI, Bersch N, Bailey RC, Geffner ME, Golde DW.
Decreased insulin-like growth factor I receptor expression and function in
immortalized African Pygmy T cells. J Clin Endocrinol Metab 1996;81:225763
Hauspie RC, Vercauteren M, Susanne C. Secular changes in growth and maturation:
an update. Acta Paediatr 1997;Suppl 423:20-7
He Q, Albertsson-Wikland, Karlberg J. Population-based body mass index reference
values from Göteborg, Sweden: birth to 18 years of age. Acta Paediatr
2000;89:582-92
Heasman L, Clarke L, Firth K, et al. Influence of restricted maternal nutrition in early
to mid gestation on placental and fetal development at term. Pediatr Res
1998;44:546-51
Heasman L, Clarke L, Stephenson T, Symonds MF. Effect of maternal nutrient
restriction in early to mid gestation and thyrotrophin-releasing hormone on
lamb survival following Caesarean section delivery near term. Can J Physiol
Pharmacol 2000;78(7):571-7
Hedegaard M, Henriksen TB, Sabroe S, Secher NJ. The relationship between
psychological distress during pregnancy and birth weight for gestational age.
Acta Obstet Gynecol Scand 1996;75(1):32-9
Helm P, Helm S. Uncertainties of age at menarche in the nineteenth century, revised
mean for Denmark, 1835. Ann Hum Biol 1987;14:371-4
Helm P, Grönlund L. A halt in the secular trend towards earlier menarche in Denmark.
Acta Obstet Gynecol Scand 1998;77:198-200
Henneberg M, Louw GJ. Height and and weight differences amoung South African
Urban schoolchildren born in various months of the year. Am J Hum Biol
1990;2:227-33
Henneberg M, Louw GJ. Further studies on the month-of-birth effect on body size:
rural schoolchildren and an animal model. Am J Phys Anthropol 1993;91:23544
Hensley W, Cooper R. Height and occupational success: a review and critique.
Psychological Reports 1987;60:843-9
Hermanussen M, Geiger-Benoit K, Burmeister J, Sippell WG. Periodical changes of
short term growth velocity (‘mini growth spurts’) in human growth. Ann Hum
Biol 1988;15:103-9
107
Hermanussen M, Burmeister J, Burkhardt V. Stature and stature distribution in recent
West German and historic samples of Italian and Dutch conscripts. Am J Hum
Biol 1995;7:507-13
Heywood PF. Growth and nutrition in Papua New Guinea. J Hum Evolution
1983;12:131-43
Hildebrand H, Karlberg J, Kristiansson B. Longitudinal growth in children and
adolescents with inflammatory bowel disease. J Pediatr Gastroenterol Nutr
1994;18(2):165-73
Hill AJ. Pre-adolescent dieting: implications for eating disorders. Int Rev Psychiatry
1993;5:87-100
Hill AJ, Silver EK. Fat, friendless and unhealthy: 9-year old children’s perception of
body shape stereotypes. Int J Obes 1995; 19:423-30
Himmelmann K, Himmelmann A, Niklasson A, Svensson A. Hypertension in
pregnancy and size at birth. Blood Pressure 1996;5(5):278-84
Hjern A. The health of newly resettled refugee children from Chile and the Middle
East. Doctoral thesis, Karolinska Institutet, Stockholm, Sweden,1990
Hoey HM, Tanner JM, Cox LA. Clinical growth standards for Irish children. Acta
Paediatr Scand 1987;Suppl 338:1-31
van’t Hof MA, Haschke F, Darvay S, and the Euro-Growth Study Group. EuroGrowth on increments in length, weight, and head and arm circumferences
during the first 3 years of life. J Pediatr Gastroenterol Nutr 2000;31:S39-47
van’t Hof MA, Haschke F, and the Euro-Growth Study Group. Euro-Growth
references for body mass index and weight for length. J Pediatr Gastroenterol
Nutr 2000;31:S48-59
Hoffman S, Hatch MC. Stress, social support and pregnancy outcome: a reassessment
based on recent research. Paediatr Perinat Epidemiol 1996;10:380-405
Hojo T, Takemoto R, Shinoda K. The secular unchangebility in relative sitting height
of female Kyushiutes. Journal of the University of Occupational and
Environmental Health, Japan 1981;3:203-5
Holland FJ, Stark O, Ades AE, Peckham CS. Birth weight and body mass index in
childhood, adolescence, and adulthood as predictors of blood pressure at age
36. J Epidemiol Community Health 1993;47:432-5
Hornell A, Hofvander Y, Kylberg E. Solids and formula: association with pattern and
duration of breastfeeding. Pediatrics 2001;107:e38
Homer CJ, James SA, Siegel E. Work-related psychosocial stress and risk of preterm,
low birth weight delivery. Am J Public Health 1989;80:173-7
Hughes JM, Chinn LL, Rona RJ. Trends in growth in England and Scotland, 1972 to
1994. Arch Dis Child 1997;76:182-9
Hulanicka B, Kotlarz K. The final phase of growth in height. Ann Hum Biol
1983;10:429-34
Hulanicka B. Deceleration of age at menarche in Poland. Ann Hum Biol 1991;18:50713
Hultcrantz JV. Uber die Körperlänge der schwedischen Wehrpflichtigen. Centralblatt
fur Anthropologie 1896 årg.1 nr 4
Hultcrantz JV. Uber die Zunahme der Körpergrösse in Schweden in den Jahren 18401926. Nova Acta Regiae Societatis Scientarium Upsaliensis. Ser. 4 Volumen
extra ordinem editum. Uppsala,1927
108
Huxley RR, Shiell AW, Law CM. The role of size at birth and postnatal catch-up
growth in determining systolic blood pressure: a systematic review of the
literature. J Hypertens 2000;18:815-31
Hviid A, Melbye M. The impact of birth weight on infectious disease hospitalization
in childhood. Am J Epidemiol 2007;165:756-61
Hägg U, Taranger J. Height and weight velocity in early, average, and late maturers
followed to the age of 25: A prospective longitudinal study of Swedish urban
children from birth to adulthood. Ann Hum Biol 1991;18:47-56
Höjer AJ. Körpergewichte von Säuglingen in einigen Vororten Stockholms 19201924. Acta Paediatr 1925;5(1-2):59-78
Illsley R, Finlayson A, Thompson B. The motivation and characteristics of internal
migrants: a socio-medical study of young migrants in Scotland. Milbank Mem
Fund Q 1963;41:115-144 & 41;217-48
Ingvarsson RF, Bjarnason AO, Dagbjartsson A, Hardardottir H, Haraldsson A,
Thorkelsson T. The effects of smoking in pregnacy on factors influencing fetal
growth. Acta Paediatr 2007;96:383-6
Jack M, Pinnington J, El-Saadi O, Latham S, Cotteril A. Secular trend in height in
Australian school children: the need for new height standards? Acta Paediatr
1999;Suppl 428:130 (poster abstract)
James DK, Stephenson TJ. Fetal nutrition and growth. In: Chamberlain G, Broughton
Pipkin F,eds. Clinical physiology in obstetrics. Oxford:Blackwell
Science,1998. p 467-97
James WP. The socio-economy of obesity. In: Chadwick DJ, Cardew G (eds). The
origin and consequences of obesity. J Wiley:Chichester,1996. p 1-16
Jamison P. Secular trends and the pattern of growth in arctic populations. Soc Sci Med
1990;30:751-9
Jebb SA. Aetiology of obesity. Br Med Bull 1997;53:264-85
Jeffery RW, Forster JL. Obesity as a public health problem. Greenwich, Johnson WG,
JAI Press, 1987:253-71
Jelliffe DB. The assessment of the nutritional status of the community. WHO
Monograph Series No.53. Geneva: World Health Organization,1966
Job JC. Growth retardation of intrauterine onset: Postnatal follow-up. In: Hernandez
M, Argente J. (eds) Human growth: basic and clinical aspects. Proceedings of
the Sixth International Congress of Auxology, Madrid, Spain,1991. Excerpta
Medica: Amsterdam,1992. p 275-84
Job JC, Chaussain JL, Golse B. Psychogenic growth retardation. Criterial study of
diagnostic data. Arch Fr Pediatr 1987;44:331-8
John AM. The plantation slaves in Trinidad, 1783.1816: A mathematical and
demographic inquiry. Cambridge: Cambridge University Press,1988
Johnston FE. Control of age at menarche. Hum Biol 1974;46:159-71
Johnston LB, Clark AJL, Savage MO. Genetic contributing to birth weight. Arch Dis
Child Neonatal Ed 2002;86:F2-3
Jordan J, Ruben M, Hernandes A, Bebelagua A, Tanner JM, Goldstein H. The 1972
Cuban National Child Growth Study as an example of population health
monitoring: Design and methods. Ann Hum Biol 1975;2:153-72
Jordan JR. Desarollo Humano en Cuba. Instituto de la Infancia. La Habana
Cuba:Editorial Cientifico-Tecnica; 1979 (In Spanish with translation in
English))
109
Jousilathi P, Tuomiletho J, Vartiain E, Eriksson J, Puska P. Relation of adult height to
cause-specific and total mortality: a prospective follow-up study of 31,199
middle-aged men and women in Finland. Am J Epidemiol 2000;151(11):111220
Järvelin MR, Elliot P, Kleinschmidt I, Martuzzi M, Grundy C, Hartikainen AL,
Rantakallio P. Ecological and individual predictors of birthweight in a northern
Finland cohort 1986. Paediatr Perinatal Epidemiol 1997;11(3):298-312
Kaltiala-Heino R, Kautiainen S, Virtanen SM, et al. Has the adolescents’ weight
concern increased over 20 years? Eur J Public Health 2003;13:4-10.
Kaplan B. Environment and human plasticity. Am Anthropol 1954;56:780-99
Kaplan HI, Kaplan HS. The psychosomatic concept of obesity. J Nerv Ment Dis
1957;125:181-201
Kaplowitz HJ, Wild KA, Mueller WH, Decker M, Tanner JM. Serial and parent- child
changes in components of body fat distribution and fatness in children from the
London Longitudinal Growth Study, ages 2 to 18 years. Hum Biol
1988;60:739-58
Karlberg J, Luo ZC, Albertsson-Wikland K. Body mass index reference values (mean
and SD) for Swedish children. Acta Paediatr 2001;90:1427-34
Karlberg P, Taranger J. The somatic development of children in a Swedish urban
community Acta Paediatr Scand 1976;Suppl 258:1-148
Kautiainen S, Rimpelä A, Vikat A, Virtanen SM. Secular trends in overweight and
obesity among Finnish adolescents in 1977-1999. Int J Obes 2002;26:544-52
Kelts DG, Grand RJ, Shen G, Watkins JB, Werlin SL, Boehme C. Nutritional basis of
growth failure in children and adolescents with Crohn’s disease.
Gastroenterology 1979;76:720-7
Kennedy E, Powell R. Changing eating patterns of American children: a view from
1996. J Am Coll Nutr 1997;16:524-9
Key A. Redogörelse för den hygieniska undersökningen. Bilaga E till
Läroverkskomitens underdåniga utlåtande och förslag angående organisationen
af rikets allmänna läroverk och dermed sammanhängande frågor. (The School
Committee’s humble report and proposals for the organisation of the Secondary
Schools and thereto related questions). Stockholm,1885 (In Swedish)
Khoury P, Morrison JA, Laskarzewski PM, Glueck CJ. Parent-offspring and sibling
body mass index associations during and after sharing of common household
environments: the Princeton School District Family Study. Metabolism
1983;32:82-9
Kielman AA, et al. Child and maternal health services in rural India: The Narangwal
experiment. Vol. 1, Integrated nutrition and health. Baltimore: John Hopkins
University Press,1983
Kimura K. Studies on growth and development in Japan. Yearbook of Physical
Anthropology 1984;27:179-214
Kinra S, Nelder RP, Lewendon GJ. Deprivation and childhood obesity: a crosssectional study of 20 973 children in Plymouth, UK. J Epidemiol Community
Health 2000;54:456-60
Kirschner BS. Growth retardation in Crohn’s disease: the merits of aggressive
nutritional therapy. Clin Nutr 1983;2:26-8
Kjerrulf H. Om s.k. harmonisk vikt och dess praktiska betydelse. Svenska
Läkartidningen 1924;46:1125-33 (In Swedish)
110
Kjerrulf H. Resultat från mätningar och vägningar februari 1927 i Adolf Fredriks
folkskola, Stockholm. Svenska Läkartidningen 1927;24:323-5 (In Swedish)
Klesges LH, Haddock CL, Stein RJ, Klesges LM, Eck LH, Hansson CL. Relationship
between psychosocial functioning and body fat in preschool children: a
longitudinal investigation. J Consult Clin Psychol 1992;60:793-6
Klesges RC, Klesges LM, Eck LH, Shelton ML. A longitudinal analysis of accelerated
weight gain in preschool children. Pediatrics 1995;95:126-30
Knudtzon J, Waaler EW, Solberg LK, Grieg E, Skjaerven R, Steen J, Irgens LM.
Höyde, vekt og hodeomkrets hos 0-4 år gamle barn. (Height, weight and head
circumference of children 0–4 y of age). Tidsskr Nor Laegeforen
1988;108(26):2136-42 (In Norwegian)
Knudtzon J, Waaler PE, Solberg LK, Stehen J. Nye norske percentilkurver för höyde,
vekt og hodeomkrets for aldermen 0-17 år. Tidskr Nor Laegeforen
1988;108(26):2125-35 (In Norwegian)
Kobliansky E, Arensburg B. Changes in morphology of human populations due to
migration and selection. Ann Hum Biol 1977;4:57-71
Kohn M, Golden NH. Eating disorders in children and adolescents: epidemiology,
diagnosis and treatment. Paediatric Drugs 2001;3:91-9
Komlos J, Tanner JM, Davies PS, Cole T. The growth of boys in the Stuttgart
Carlschule, 1771-93. Ann Hum Biol 1992;19(2):883-7
Komlos J. ed. Stature, Living Standards, and Economic Development. Chicago, IL:
University of Chicago Press,1994
Kondo S, Eto M. Physical growth studies on Japanes-American children in
comparison with native Japanese. In Comparative Studies of Human
Adaptability of Japanese, Caucasians, and Japanese-Americans, ed SM
Horvath, S Kondo, H Matsui & H Yoshimena. Tokyo: Japanese International
Biological Program,1975. p 13-45
Kramer MS, et al. Do breast-feeding an delayed introduction of solid foods protect
against subsequent obesity? J Pediatr 1981;98(6):883-7
Kramer MS, Barr RG, Pless IB, Bolisjoyly C, McVey-White L, Leduc DG.
Determinants of weight and adiposity in early childhood. Can J Public Health
1986;77 Supp1: 98-103
Kramer MS. Determinants of low birth weight: methotodological assessment and
meta-analysis. Bull WHO 1987;65:666-737
Kramer MS. Balanced protein/energy supplementation in pregnancy (Cochrane
Review). In: Cochrane Library, Issue 4,1998
Krawczynski M, Walkowiak J, Krzyzaniak. Secular changes in body height and
weight in children and adolescents in Poznan, Poland, between 1880 and 2000.
Acta Paediatr 2003;92:277-82
von Kries R, Koletzko B, Sauerwald T, von Mutius E, Barnert D, Grunert, et al.
Breastfeeding and obesity: cross sectional study. BMJ 1999;319(7203):147-50
Kristiansson B, Karlberg J, Fällström SP. Infants with low rate of weight gain.I. A
study of organic factors and growth patterns. Acta Paediatr Scand 1981;70:65562
Krogman WM. Growth of the head, face, trunk, and limbs in Philadephia white and
Negro children of elementary and high school age. Monographs of the Society
for Research in Child Development 1970;20:1-91
Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, et al.
CDC growth charts: United States. Adv Data 2000(314):1-27
111
Kuh DL, Power C, Rodgers B. Secular trends in social class and sex differences in
adult height. Int J Epidemiol 1991;20:1001-9
Kuh D, Power C, Blane D, et al. Social pathways between childhood and adult health.
In:Kuh and Ben-Schlomo Y, eds. A life course approach to chronic disease.
Oxford: Oxford Medical Publications,1997
Jackson-Leach R, Lobstein T. Estimated burden of paediatric obesity in Europe. Part
1. The increase in the prevalence of child obesity in Europe is itself increasing.
Int J Peadiatr Obes 2006;1:26-32
Lahr MM. The evolution of modern diversity. A study of cranial variation. Cambridge
Studies in Biological Anthropology 18. Cambridge University Press,1996
Lake JK, Power C, Cole TJ. Child to adult body mass index in the 1958 British birth
cohort: Association with parental obesity. Arch Dis Child 1997;77:376-81
Lampl M, Johnston FE, Malcolm LA. The effects of protein supplementation on the
growth and skeletal maturation of New Guinean school children. Ann Hum
Biol 1978;5:219-27
Lampl M. Further observations on diurnal variation in standing height. Ann Hum Biol
1992;19(1):87-90
Langhoff-Roos J, Lindmark G, Gustavsson KH, et al. Relative effect of parental birth
weight on infant birth weight at term. Clin Genet 1987;32:240-8
Larnkjaer A, et al. Secular trend in adult stature has come to a halt in northern Europe
and Italy. Acta Paediatr 2006;95:754-5
Larroque B, Kaminski M, Lelong N, Subtil D, Dehaene P. Effects on Birth Weight of
Alcohol and Caffeine Consumption during Pregnancy. Am J Epidemiol
193;137:941-50
Laska-Mierzejewska T, Milicer H, Piechaczek H. Age at menarche and its secular
trend in urban and rural girls in Poland. Ann Hum Biol 1982;9:227-33
Lasker GW. Environmental growth factors and selective migration. Hum Biol
1952;24:262-89
Lasker GW, Evans FG. Age, environment and migration: further anthropometric
findings on migrant and non-migrant Mexicans. Am J Phys Anthropol
1961;19:203-11
Lasker GW, Mascie-Taylor CG. Effects of social class differences and social mobility
on growth in height, weight and body mass index in a British cohort. Ann Hum
Biol 1989;16:1-8
Lasker GW, Mascie-Taylor CGN. Influence of social class on the correlation of stature
of adult children with that of their mothers and fathers. J Biosoc Sci
1996;28:117-22
Lauer RM. Coronary heart risk factors in school children: The Muscatine study. J
Paediatr 1975;86:697-706
Law CM, Barker DJP, Osmond C, Fall HD, Simmonds SJ. Early growth and
abdominal fatness in adult life. J Epidemiol Community Health 1992;46:184-6
Law CM, Shiell AW. Is blood pressure inversely related to birth weight? The strength
of evidence from a systematic review of the literature. J Hypertens
1996;14(8):935-41
Law CM. Significance of birth weight for the future. Arch Dis Child Fetal Neonatal
Ed 2002;86:F7-8
Lazarus R, Wake M, Hesketh K, Waters E. Change in body mass index in Australian
primary school children, 1985-1997. Int J Obes 2000;24(6):679-84
112
Leighton G, Clark ML. Milk consumption and the growth of schoolchildren. Lancet
1929;1:40-3
Leon DA, Lithell HO, Vågerö D, Koupilova I, Mohsen R, Berglund L, Lithell U-U,
McKeigue PM. Reduced fetal growth rate and increased risk of iscaemic heart
disease mortality in 15 thousand Swedish men and women born 1915-29. BMJ
1998;317:241-5
Le Roy Ladurie E, Bernageau EN. Etude sur un Contingent Militaire(1968). Mobilite’
geographique, delinguance et stature, mises en rapport avec d’autres aspects de
la situation des conscrits. Annales de demographie historique 1971:311-37
van Lenthe FJ, Kemper CG, van Mechelen W. Rapid maturation in adolescence results
in greater obesity in adulthood: the Amsterdam Growth and Health Study. Am
J Clin Nutr 1996;64:18-24
van Lenthe FJ, van Mechelen W, Kemper HC, Post GB. Behavioral variables and
development of a central pattern of body fat from adolescence into adulthood in
normal-weight whites: the Amsterdam Growth and Health Study. Am J Clin
Nutr 1998;67:846-52
Leung SS, Cole TJ, Tse LY, Lau JT. Body mass index reference curves for Chinese
children. Ann Hum Biol 1998;25(2):169-74
Li H, Leung SS, Lam PK, Zhang X, Chen XX, Wang SL. Height and weight
percentile curves of Beijing children and adolescents 0-18 years, 1995. Ann
Hum Biol 1999; 26(5): 457-71
Liestöl K. Social conditions and menarcheal age: the importance of early years of life.
Ann Hum Biol 1982;9:521-37
Liestöl K, Rosenberg M. Height, weight and menarcheal age of schoolgirls in Oslo –
an update. Ann Hum Biol 1995;22(3):199-205
Lifshitz F, Moses N, Cervantes C, Ginsberg L. Nutritional dwarfing in adolescents.
Sem Adolesc Med 1987;3:255-6
Lifshitz F. Nutrition and growth. In: Hernandez M, Argente J.(eds) Human growth:
basic and clinical aspects. Proceedings of the Sixth International Congress of
Auxology, Madrid, Spain,1991. Excerpta Medica:Amsterdam 1992. p 205-13
Lindeman M, Sundvik L. Impact of height assessment on Finnish female job aplicants´
managerial activities. J Social Psychol 1994;134:169-74
Lindgren G. Height, weight and menarche in Swedish urban school children in relation
to socio-economic and regional factors. Ann Hum Biol 1976;3(6):501-28
Lindgren G, Strandell A. Fysisk utveckling och hälsa. En analys av
hälsokortsuppgifter för grundskoleelever födda 1967 [Physical development
and health. An analysis of data in health records of pupils born in 1967],
Rapport 4 . Stockholm: Högskolan för lärarutbildning i Stockholm,
Institutionen för pedagogik,1986 (In Swedish)
Lindgren GW, Hauspie RC. Heights and weights of Swedish school children born in
1955 and 1967. Ann Hum Biol 1989;16:397-406
Lindgren G, Aurelius G, Tanner J, Healy M. Standards for height, weight and head
circumference from one month to six years based on Stockholm children born
in 1980 Acta Paediatr 1994;83:360-6
Lindgren G, Strandell A, Cole T, Healy M, Tanner J. Swedish population reference
standards for height, weight, and body mass index attained at 6 to 16 years
(girls) or 19 years (boys). Acta Paediatr 1995;84:1019-28
113
Lissau I, Sörensen TI. School difficulties in childhood and risk of overweight and
obesity in young adulthood: a ten year prospective population study. Int J Obes
1993;17:169-75
Lissau I. Overweight and obesity epidemic among children. Answer from European
countries. Int J Obes 2004;28:S10-S15 (a)
Lissau I, Overpeck D, Ruan J, Due P, Holstein BE, Hediger ML, and the Health
Beahaviour in School-aged Children Obesity Working Group. Body mass
index and overweight in adolescents in 13 European countries, Israel and the
United States. Arch Pediatr Adolesc Med 2004;158:27-33 (b)
Little MA, Galvin K, Mugambi M. Cross-sectional growth of nomadic Turkana
pastoralists. Hum Biol 1983;55:811-30
Liu JP, Baker J, Perkins AS, et al. Mice carrying null mutations of the genes encoding
insulin-like growth factor I (Igf-I) and type 1 IGF receptor (Igf1r). Cell
1993;75:59-72
Livi R. Anthropometica Militaire, parte1. Dati demografici e etnologici. Rome:
Giornale Medico del Regio Esercito 1896
Livingstone MBE. Childhood obesity in Europe: a growing concern. Public Health
Nutr 2001;4(1A):109-16
Ljung BO, Bergsten-Brucefors A, Lindgren G. The secular trend in physical growth in
Sweden. Ann Hum Biol 1974;1:245-56
Ljunggren CA. Bidrag till kännedom om svenska folkskolebarns standardvikt. Nord
Hyg Tidskr 1925;6:241-56 (In Swedish)
Ljunggren CA. Skolbarnens hälsofostran. Handbok för lärare och föräldrar. (The
schoolchildren’s health education. Handbook for teachers and parents). Natur
& Kultur, Stockholm 1933 (In Swedish)
Lloyd J, Wolff OH, Whelen WS. Childhood obesity. A long-term study of height and
weight. BMJ 1961;ii:145-48
Lobstein T, Baur L, Uany R for the IOTF Childhood Obesity Working Group. Obesity
in children and young people. A crisis in public health. Obesity Reviews
2004;5 (Suppl):4-85
Lobstein T, Frelut M-L. Prevalence of overweight among children in Europe. Obes
Rev 2003;4:195-200
Loesch DZ, Huggins R, Rogucka E, Hoang NH, Hopper JL. Genetic correlates of
menarcheal age: a multivariate twin study. Ann Hum Biol 1995;22:479-90
Loh ES. The economic effects of physical appearance. Social Science Quarterly
1993;74:420-38
Low WD, Kung LS, Leong JCY. Secular trend in the sexual maturation of Chinese
girls. Hum Biol 1982;54:539-51
Lucas A, Fewtrell MS, Cole TJ. Fetal origins of adult disease – the hypothesis
revisited. BMJ 1999;319:245-9
Lumey LH. Compensatory placental growth ocio restricted nutrition in early
pregnancy. Placenta 1998;19:105-12
Lundholm C. Aspects on parametric analysis of longitudinal growth data with
irregular spacing in time. Doctoral thesis in progress, University of Örebro,
Örebro, Sweden,2007
Luo J, Hu FB. Time trends of obesity in pre-school children in China from 1989 to
1997. Int J Obes 2002;26:553-8
114
Lynch JW, Kaplan GA, Salonen JT. Why do poor people behave poorly? Variation in
adult health behaviours and psychosocial characteristics by stages of the socioeconomic lifecourse. Soc Sci Med 1997;44:809-19
Löfgren J, Andersson O, Arvidsson B, Bergman LR, Elo S, Palmer M, Werner B. Låg
födelsevikt en stark riskfaktor för låg bentäthet hos premenopausala kvinnor.
(Low birthweight is a strong risk factor for low bone density of praemenopausal women) Sv. Läkaresällskapets Riksstämma 2000 Abstract (in
Swedish)
Macbeth HM. Biological selectivity in migrants. In Migration and Mobility: Biosocial
Aspects of Human Movement, edited by AJ Boyce, Symposia of the Society
for the Study of Human Biology, Vol.23 London:Travor & Francis,1984. p
197-205
Macbeth HM, Boyce AJ. Anthropometric variation between migrants and nonmigrants: Orkney Islands, Scotland. Ann Hum Biol 1987;14:405-14
McCabe MP, Ricciardelli LA. Parent, peer, and media influences on body image and
strategies to both increase and decrease body size among adolescent boys and
girls. Adolescence 2001;36:225-40
Macintyre S A review of the social patterning and significance of measures of height,
weight, blood pressure and respiratory function Soc Sci Med 1988;27(4):32737
Mackenbach JP,. Socio-economic health differences in the Netherlands: a review of
recent empirical findings. Soc Sci Med 1992;34:213-26
Maes HNM, Neale MC, Eaves LJ. Genetic and environmental factors in relative body
weight and human adiposity. Behav Genet 1997;27:325-51
Maffeis C, Micciolo R, Must A, Zaffanello M, Pinelli L. Parental and perinatal factors
associated with childhood obesity in north-east Italy. Int J Obes 1994;18:301-5
Maffeis C, Zaffanello M, Schutz Y. Relationship between physical inactivity and
adiposity in prepubertal boys. J Pediatr 1997;131:288-92
Maffeis C, Talamini G, Tato L. Influence of diet, physical activity and parents’ obesity
on children’s adiposity: a four-year longitudinal study. Int J Obes 1998;22:75864
Magnus P. Further evidence for a significant effect of fetal genes on variation in birth
weight. Clin genet 1984;26:289-96
Magnus P, Berg K, Bjerkedal T, et al. Parental determinants of birth weight. Clin
Genet 1984;26:397-405
Magnus P. Causes of variation in birth weight: a study of offspring of twins. Clin
genet 1984;25:15-24
Magnus P, Bakketeig LS, Hoffmann. Birth weight of relatives by maternal tendency to
repeat small-for-gestational-age (SGA) births in successive pregnancies. Acta
Obstet Gynecol Scand 1997; Suppl 165:35-8
Magnus P, Gjessing HK, Skrondal A, Skjaerven R. Paternal contribution to birth
weight. J Epidemiol Community Health 2001;55(12):873-7
Magnusson D. Individual Development from an Interactional Perspective. A
Longitudinal Study. Paths through Life. Vol.1. Hillsdale, New Jersey:Lawrence
Erlbaum Associates, 1988
Malcolm LA. Growth and development in the Bundi children of the New Guinea
highlands. Hum Biol 1970;42:293-328
115
Maleta K, Virtanen SM, Espo M, Kulmala T, Ashborn P. Seasonality of growth and
the relationship between weight and height gain in children under three years
of age in rural Malawi. Acta Paediatr 2003;92:491-7
Malina RM, Harper AB, Avent HH, Campbell DE. Age at menarche in athletes and
non-athletes. Med Sci Sports 1973;5:11-3
Malina RM, Himes JH, Stepick CD, Lopez FG, Buschang PH. Growth of rural and
urban children in the Valley of Oaxaca, Mexico. Am J Phys Anthropol
1981;55:269-80
Malling-Hansen R. Ueber Periodicität im Gewicht der kinder. Taubstummen instituts
in Kopenhagen. Fragment I. Kopenhagen,1883
Marmot MG, Shipley MJ, Rose G. Inequalities of death-specific explanations of a
general pattern. Lancet 1984;1:1003-6
Marmot MG, Smith GD. Why are the Japanese living longer? BMJ 1989;299:1547-51
Marx K. Capital:a critique of political economy. Book 1:The process of production of
capital. London: Lawrence & Wishart 1954
Marschall WA. Evaluation of growth rate in height over periods of less than one year.
Arch Dis Child 1971;46:414-20
Martin TR, Bracken MB. Association of low birth weight with passive smoke
exposure in pregnancy. Am J Epidemiol 1986;124:633-42
Martin AR, Nieto JM, Jimenez MA, et al. Unhealthy eating behaviour in
adolescents. Eur J Epidemiol 1999;15:643-8
Martin RM, Davey SG, Mangtani P, Frankel S, Gunnell D. Association between breast
feeding and growth: the Boyd-Orr cohort study. Arch Dis Child Fetal Neonatal
Ed 2002;87:F193-F201
Martorell R. Child growth retardation: A discussion of its causes and its relationship to
health. In Nutritional adaption in man, edited by Blaxter K and Waterlow JC
1985:13-29. London och Paris:John Libby,1985
Mascie-Taylor CGN. The interaction between geographical and social mobility. In
Migration and Mobility, ed. AJ Boyce,London:Taylor and Francis,1984. p 16178
Mathews F, Yudkin P, Neil A. Influence of maternal nutrition on outcome of
pregnancy: prospective cohort study. BMJ 1999;319:339-43
MBR, The Swedish Medical Birth Register, National Board of Health and Welfare,
Sweden (Medicinsk födelseregister EPC Socialstyrelsen) URL:
www.sos.se/epc/fodelse/mfr
Meeuwisse G, Otterblad Olausson P. Allt större andel nyfödda väger över fyra kilo.
(Increasing birthweights in the Nordic countries; a growing proportion of
neonates weight over four kg) Läkartidningen 1998;95(48):5488-92 (In
Swedish)
Mei Z, Scanlon KS, Grummer-Strawn LM, Freedman DS, Yip R, Trowbridge FL.
Increasing prevalence of overweight among US low-income pre-school
children: the CDC Nutrition Surveillance 1983 to 1995. Pediatrics
1997;101(1):E12
Mellbin T, Vuille J-C. Rapidly developing overweight in school children as an
indicator of psychosocial stress. Acta Paediatr Scand 1989;78:568-75
Mellbin T, Vuille J-C. Further evidence of an association between psychosocial
problems and increase in relative weight between 7 and 10 years of age. Acta
Paediatr Scand 1989;78:576-80
116
Meredith HV. Comparative findings on body size of children and youths living at
urban centers and in rural areas. Growth 1979;43:95-104
Meredith HV. Body size of infants and children around the world in relation to socioeconomic status. Advances in child development and behaviour 1984;18:81145
Merimee TJ, Zapf J, Froesch ER. Dwarfism in the pygmy, an isolated defiency
insulin-like growth factor I. N Engl J Med 1981;305:965-8
Merritt R. Obesity. Current Probl Pediatr 1982;12:1-58
Merzenich H, Boeing H, Wahrendorf J. Dietary fat and sports activity as determinants
for age at menarche. Am J Epidemiol 1993;138(4):217-24
Metzger BE, Silverman BL, Freinkel N, Dooley SL, Ogata ES, Green OC. Amniotic
fluid insulin concentration as a predictor of obesity. Arch Dis Child
1990;65:1050-2
Meyer HE, Selmer R. Income, educational level and body height. Ann Hum Biol
1999;26(3):219-23
Michaelsen KF, Larsen PS, Thomsen BL, Samuelson G. The Copenhagen cohort
study of infant nutrition and growth:duration of breast feeding and influencing
factors. Acta Paediatr 1994;83:565-71
Michielutte R, Diseker RA, Corbett WT, Schey HM, Ureda JR. The relationship
between weight-height indices and the triceps skinfold measure among children
age 5 to 12. Am J Public Health 1984;74:604-6
Mikulandra F, Tadin I, Grguric J, Zakanj Z, Perisa M. Influence of father’s weight and
height on weight of male and female newborns. Collegium Antropologicum
2001;25(1):59-63
Mills JL, Graubard BI, Harley EE, Rhoads GG, Berendes HW. Maternal alcohol
consumption and birth weight. JAMA 1984;252:1875-9
Moisan J, Meyer F, Gingras S. A nested case-control study of the correlates of early
menarche. Am J Epidemiol 1990;132:953-61
Moll PP, Burns TL, Lauer RM. The genetic and environmental sources of body mass
index variability: the Muscatine Ponderosity Family Study. Am J Hum Gen
1991;49:1243-55
Monteiro CA, Benicio MH, Iunes R, Gouveia NC, Taddei JA, Cardoso MA.
Nutritional status of Brazilian children: trends from 1975 to 1989. Bull World
Health Organ 1992;70:657-66
Morande G, Celada J, Casas JJ. Prevalence of Eating Disorders in a Spanish Schoolage Population. J Adolesc Health 1999;24:212-9
Mossberg H-O. 40-year follow-up of overweight children. Lancet 1989 August
26:491-3
MSCR, The Swedish Military Conscript Registry, Pliktverket Karlstad, Sweden
(Mönstringsregistret, Pliktverket Karlstad) URL: www.pliktverket.se
Mueller WH, Murillo F, Palamino H, Badzioch M, Chakraborty R, Fuerst P, Schull
WJ. The aymara of Western Bolivia: V. Growth and development in an
hypoxic environment. Hum Biol 1980;52:529-46
Muller I, Betuela I, Hide R. Regional patterns of birthweights in Papua New Guinea in
relation to diet, environment and socio-economic factors. Ann Hum Biol
2002;29(1):74-88
Murphy D, Sellers S, MacKenzie I, et al. Case-control study of antenatal and
intrapartum risk factors for cerebral palsy in very preterm singleton babies.
Lancet 1995;346:1449-54
117
Must A, Jacques PF, Dallal GE, Bajema CJ, Dietz WH. Long-term Morbidity and
Mortality of Over-weight Adolescents - A Follow-up of the Harvard Growth
Study of 1922 to 1935. N Engl J Med 1992;327:1350-5
Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. Int
J Obes 1999;23 Suppl 2:S2-11
Mårild S, Bondestam M, Bergström R, Ehnberg S, Hollsing A, Albertsson-Wikland K.
Prevalence trends of obesity and overweight among 10-year-old children in
western Sweden and relationship with parental body mass index. Acta Paediatr
2004;93:1588-95
Nafstad P, Jaakola JJ, Hagen JA, Pedersen BS, Qvigstad E, Botten G, Kongerud J.
Weightgain during the first year of life in relation to maternal smoking and
breast feeding in Norway. J Epidemiol Community Health 1997;51:261-5
Nakayama T, Nozoe S. Anorexia nervosa (AN) – epidemiology, cause, therapy,
outcome. Nippon Rinsho (Japanese Journal of Clinical Medicine)
2001;59:534-9 (In Japanese).
National Food Administration. Dietary habits and nutrient intake in Swedish
children 4 year old and school children in grade 2 and 5. Riksmaten –
children 2003. Livsmedelsverket, Uppsala, Sweden. Abstract in English
Neiderud J. Greek immigrant children in southern Sweden. Living conditions,
nutrition, growth and dental health in comparison with Greek and Swedish
children. Doctoral thesis, Uppsala University, Uppsala, Sweden,1992
Neyzi O, Yalcindag A, Alp H. Heights and weights of Turkish children.
Environmental Child Health 1973;19:5-13
Neyzi O, Furman A, Bundak R, Gunoz H, Darendeliler F, Bas F. Growth references
for Turkish children aged 6 to 18 years. Acta Paediatr 2006;95:1635-41
Nielsen GA, Thomsen BL, Michaelsen KF. Influence of breastfeeding and
complementary food on growth between 5 and 10 months. Acta Paediatr
1998;87(9):911-7
Nieto FJ, Szklo M, Comstock GW. Childhood weight and growth rate as predictors of
adult mortality. Am J Epidemiol 1992;136:210-3
Nikens PR. Stature reduction as an adaptive response to food production in
Mesoamerica. J Arch Sci 1976;3:21-41
Niklasson A, Ericson A, Fryer JG, Karlberg J, Lawrence C, Karlberg P. An Update of
the Swedish Reference Standards for Weight, Length and Head Circumference
at Birth for Given Gestational Age (1977-1981). Acta Paediatr Scand
1991;80:756-62.
Nilsen ST, Finne PH, Bergsjö P, Stammes O. Males with low birth weight examined at
18 years of age. Acta Paediatr Scand 1984;73:168-75
Nordström ML, Cnattingius S. Smoking habits and birthweights in two successive
births in Sweden. Early Hum Dev 1994;37(3):195-204
Norinder I. Bidrag till kännedomen om folkskolebarns kroppsutveckling och
hälsotillstånd. (Contribution to the knowledge of body development and state
of health of elementary schoolchildren). Hygiae 1907;69:1199-1205 (In
Swedish)
Nowak M, Speare R, Crawford D. Gender differences in adolescent weight and shaperelated beliefs and behaviour. J Paediatr Child Health 1996;32:148-52
Nylin G. Periodical variation in growth, standard metabolism and oxygen capacity in
the blood in children. Acta Med Scand 1929;31:1-207
118
O’Callaghan MJ, Williams GM, Andersen MJ, Bor W, Najman JM. Prediction of
obesity in children at 5 years: a cohort study. J Peadiatr Child Health
1997;33:311-6
Odin M. Sjukdomar och sjukdomsfrekvens i övre Norrland särskilt med hänsyn till
födans sammansättning. Del II i Kungl. Medicinalstyrelsen: En socialhygienisk
undersökning i Västerbotten och Norrbottens län, 1929-31. (Diseases and
frequency of diseases in upper northern Sweden regarding the composition of
food). Lund: Håkan Olssons Boktryckeri, 1934 (In Swedish)
Ogden CL, Kuczmarski RJ, Flegal KM, Mei F, Guo S, Wie R, Grummer-Strawn LM,
Curtin LR, Roche AF, Johnson CL. Centers for Disease Control and Prevention
2000 Growth Charts for the United States: Improvements to the 1977 National
Center for Health Statistics Version. Pediatrics 2002;109:45-60
Oldenburg CEM, Cotton-Oldenburg NU, Rasmussen F. Ethnic differences in being
overweight and obesity among male military conscripts born in Sweden
between 1973 and 1975. Eur J Public Health 1999;9:146-9
Olsen LW, Baker JL, Holst C, Sorensen TIA. Birth cohort effects on the obesity
epidemic in Denmark. Epidemiology 2006;17:292-5
de Onis M, Habicht J-P. Anthropometric reference data for international use:
Recommendations from a WHO Expert Committee. Am J Clin Nutr
1996;64:650-8
de Onis M, Yip R. The WHO growth chart: historical considerations and current
scientific issues. Bibl Nutr Dieta 1996 (53):74-89
de Onis M, Garza C, Habicht J-P. Time for a new growth reference. Pediatrics
1997;100:e8
de Onis M, Onyango AW. The Centers for Disease Control and Prevention 2000
growth charts and the growth of breastfed infants. Acta Paediatr 2003;92:413-9
Orr JB. Milk consumption and the growth of schoolchildren. Lancet 1928;1:202-3
Orskou J, Kesmodel U, Henriksen TB, Secher NJ. An increasing proportion of infants
weigh more than 4000 grams at birth. Acta Obstet Gynecol Scand
2001;80(10):931-6
Osmani SR. On some controversies in the measurement of undernutrition. In Nutrition
and poverty, edited by SR Osmani. Oxford: Clarendon Press,1992
Osmond C, Barker DJP, Winter PD, Fall CHD, Simmonds SJ. Early growth and death
from cardiovascular disease in women BMJ 1993;307:1519-23
Ounsted M, Scott A, Ounsted C. Transmission through the female line of a mechanism
constraining human fetal growth. Ann Hum Biol 1986;13:143-51
Padez C, Johnston F. Secular trends in male adult height 1904-1996 in relation to place
of residence and parent’s education level in Portugal. Ann Hum Biol
1999;26:287-98
Pagliani. Sopra alcuni fattori dello sviluppi uman; richerche anthropometriche. Atti
della Reale Accademia di Scienze di Torino 1876;11:92-120
Paneth NS. The problem of low birth weight. The Future of Children. 1995. Article
accessed through http:// www.futureofchildren.org/LBW/03LBWPANA
Parnell RW. The physique of Oxford undergraduates. J Hygiene 1954;52:396-8
Parsons TJ, Power C, Logan S, Summerbell CD. Childhood predictors of adult
obesity: a systematic review. Int J Obes 1999;23 Suppl 8:S1-107
Patton PG, Gardner LI. Influence of family environment on growth: the syndrome of
maternal deprivation. J Pediatr 1962;30:957-64
119
Patton GC, Johnson-Sabine E, Wood K, Mann AH, Wakeling A. Abnormal eating
attitudes in London schoolgirls – a prospective epidemiological study: outcome
at twelve month follow-up. Psychol Med 1990;20:383-94
Payne P. Undernutrition: Measurement and implications.In Nutrition and Poverty,
edited by SR Osmani. Oxford:Clarendon Press,1992
Penrose LS. Some recent trends in human genetics. Caryologia 1954;Suppl.6:521-30
Persson E-K, Hagberg G, Uvebrant P. Hydrocephalus prevalence and outcome in a
population-based cohort of children born in 1989-1998. Acta Paediatr
2005;94:726-32
Persson LÅ. Infant feeding and growth – a longitudinal study in three Swedish
communities. Ann Hum Biol 1985;12:41-52
Persson LÅ, Samuelsson G, Sjölin S. Nutrition and health in Swedish children 19301980. Three nutrition surveys in a northern Swedish county. Acta Paediatr
Scand 1989;78:865-72
Petersen S, Brulin C, Bergström E. Increasing prevalence of overweight in young
schoolchildren in Umeå, Sweden, from 1986 to 2001. Acta Paediatr
2003;92:848-53
Pettersson OV. Om vigtsförhållanden hos späda barn under första lefnadsåret.
Föredrag på Upsala Läkareförenings högtidsdag den 18 Sept. 1882. Upsala
Läkarefören Förhandl 1882/83;XVIII(1):1-31 (In Swedish)
Pharoah PO, Stevenson CJ, West CR. Association of blood pressure in adolescence
with birtweight. Arch Dis Child Fetal Neonatal Ed. 1998;79:F114-18
Phillips DIW. Birth weight and the future development of diabetes. Diabetes Care
1998;21:8150-5
von Post-Skagegård M, Samuelsson G, Karlström B, Mohsen R, Berglund L, Bratteby
L-E. Changes in food habits in healthy Swedish adolescents during the
transition from adolescence to adulthood. Eur J Clin Nutr 2002;56:532-8
Powell GF, Brasel JS, Blizzard RM. Emotional deprivation and growth retardation
simulating idiopathic hypopituitarism. 1. Clinical evaluation of the syndrome.
N Engl J Med 1967;276:1271-8
Powell GF, Brasel JS, Raiti S, Blizzard RM. Emotional deprivation and growth
retardation simulating idiopathic hypopituitarism. 2. Endocrinologic evaluation
of the syndrome. N Engl J Med 1967;276:1279-83
Powell GF, Hopwood NJ, Barratt ES. Growth hormone studies before and during
catch-up growth in a child with emotional deprivation and short stature. J Clin
Endocrinol Metab 1973;37:674-9
Power C, Moynihan C. Social class changes in weight-for-height between childhood
and early adulthood. Int J Obes 1988;12:445-53
Power C. National trends in birth weight: implications for future adult disease. BMJ
1994;308:1270-1
Power C, Lake JK, Cole TJ. Body mass index and height from childhood to adulthood
in the 1958 British born cohort. Am J Clin Nutr 1997;66:1094-101
Power C, Hertzman C, Matthews S, Manor O. Social differences in health: life-cycle
effects between ages 23 and 33 in the 1958 British birth cohort. Am J Public
Health 1997;87:1499-503
Power C, Lake JK, Cole TJ. Measurement and long term health risks of child and
adolescent fatness. Int J Obes 1997;21:507-26
Power C, Parsons T. Nutritional and other influences in childhood as predictors of
adult obesity. Proc Nutr Soc 2000;59(2):267-72
120
Prader A, Tanner JM, von Harnack GA. Catch-up growth following illness or
starvation . An example of developmental canalization in man. J Pediatr
1963;62:646-69
Prader A Biomedical and endocrinological aspects of normal growth and
development. In:Borms J, Hauspie R, Sand A, Susanne C, Hebbelinck M (Eds)
Human growth and development. New York, Plenum Press,1984. p 1-22
Prebeg Z, Juresa V, Kujundzic M. Secular growth changes in Zagreb schoolchildren
over four decades, 1951-91. Ann Hum Biol 1995;22(2):99-110
Preece MA. Standardisation of growth. Acta Paediatr Scand 1989;Suppl 349:57-64
Prentice AM. Can maternal dietary supplement help in preventing infant malnutrition?
Acta Paediatr 1991;Suppl 374:67-77
Prentice AM, Jebb SA. Obesity in Britain: gluttony or sloth? BMJ 1995;311:437-9
Prentice AM. Obesity – the inevitable penalty of civilization? Br Med Bull
1997;53:229-37
Prescott-Clarke P, Primatesta P (eds) Health survey for England: the health of young
people ´95-´97. London:The Stationary Office,1998
Prokopec M, Bellisle F. Adiposity in Czech children followed from one month of age
to adulthood: analysis of individual BMI patterns. Ann Hum Biol 1993;20:51725
Pugliese MT, Lifshitz F, Grad G, Marks-Katz M. Fear of obesity: a cause of short
stature and delayed puberty. N Engl J Med 1983;309:513-8
Quetelet A. Sur l´homme et Anthropome’trie. Brussel,1870
Rajan U. Obesity among Singapore students. Int J Obes 1994; 18 (Suppl 2):27
Rasmussen F, Johansson M, Hansen HO. Trends in overweight and obesity among 18year-old males in Sweden between 1971 and 1995. Acta Paediatr 1999;88:4317
Rasmussen F, Johansson M. Increase in the prevalence of overweight and obesity from
1995 to 1998 among 18-year-old males. Acta Paediatr 2000;89:888-9
Ravelli GP, Stein ZA, Susser MW. Obesity in young men after famine exposure in
utero andearly infancy. New Engl J Med 1976;295:349-53
Ravelli ACJ. Van der Meulin JHP, Michels RPJ, et al. Glucose tolerance in adults
after in utero exposure to the Dutch famine. Lancet 1998;351:173-7
Reilly JJ, Dorosty AR, Emmett PM. Prevalence of overweight and obesity in British
children: cohort study. BMJ 1999;319:1039
Reilly JJ, Dorosty AR. Epidemic of obesity in UK children. Lancet 1999;354:1874-5
Reilly JJ, Wilson M, Summerbell CD, Wilson D. Obesity in childhood and
adolescence: evidence based answers to common questions. Arch Dis Child.
2002;86:302-5
Retzius G. Anthropologia Suecia. Beitrage zur Anthropologie der Schweden nach den
auf Veranstaltung der Schwedischen Gesellschaft fur Anthropologie und
Geographie in den Jahren 1897 und 1898 ausgefuhrten Erhebungen ausgearb.
Stockholm 1902
Rice T, Bouchard C, Perusse L, Rao DC. Familial clustering of multiple measures of
adiposity and fat distribution in the Quebec Family Study: a trivariate analysis
of percent body fat, body mass index, and trunk-to-extremity skinfold ratio. Int
J Obes 1995;19:902-8
Rich-Edwards JW, Manson JE, Stampfer MJ, Colditz GA, Willett WC, Rosner B,
Speizer FE, Hennekens CH. Height and the risk of cardiovascular disease in
women. Am J Epidemiol 1995;142:909-17
121
Rich-Edwards JW, Stampfer MJ, Manson JE, Rosner B, Hankinson SE, Colditz GA,
Willett WC, Hennekens CH. Birth weight and risk of cardiovascular disease in
a cohort of women followed up since 1976. BMJ 1997;315:396-400
Rich-Edwards JW, Colditz GA, Stampfer MJ, Willett WC, Gillman MW, Hennekens
CH, Speizer FE, Manson JE. Birthweight and the risk for type 2 diabetes
mellitus in adult women. Ann Intern Med 1999;130:278-84 1976
Rimoin DL, Merimee TJ, Rabinowitz D, Cavalli-Sforza LL, McKusick VA. Genetic
aspects of clinical endocrinology. Recent Progress in Hormone Research
1968;24:365-467
Rimpelä AH, Rimpelä MK. Towards an equal distribution of health? Socioeconomic
and regional differences of the secular trend of the age of menarche in Finland
from 1979 to 1989. Acta Paediatr 1993;82:87-90
Roberts DF. Secular trends in human growth and maturation in British girls. Am J
Hum Biol 1994;6:13-8
Robinson JS, Moore V, Owens JA, et al. Origins of fetal growth restriction. Eur J
Obstet Gynecol Reprod Biol 2000;92:13-9
Robinson TN, Hammer LD, Killen JD, Kraemer HC, Wilson DM, Hayward C, Taylor
CB. Does television viewing increase obesity and reduce physical activity?
Cross-sectional and longitudinal analyses among adolescent girls. Pediatrics
1993;91:273-80
Robinson TN. Does television cause childhood obesity? JAMA 1998;279:959-60
Roche AF. Secular trends in human growth, maturation, and development.
Monographs of the Society for Research in Child Development 1979;44:1-120
Roche AF. Growth, maturation, and body composition: the Fels Longitudinal Study
1929-1991. Cambridge: Cambridge University Press,1992
Roede MJ, Van Wieringen JC. Growth diagrams 1980. Netherlands third nation-wide
survey. Tijdschr Soc Gezondheidsz 1985; Suppl 63:1-34
Rodriguez-Soriano J, Vallo A, Martul P. Growth in renal insufficiency. In: Hernandez
M, Argente J.(eds) Human growth: basic and clinical aspects. Proceedings of
the Sixth congress of Auxology, Madrid, Spain, 1991.Excerpta
Medica:Amsterdam,1992. p 111-6
Rogers IS, Emmett PM, Golding J. The growth and nutritional status of the breast-fed
infant. Early Hum Dev 1997;49:S157- 74
Rolland-Cachera MF, Deheeger M, Bellisle F, Sempe´ M, Guilloud-Bataille M, Patois
E. Adiposity rebound in children: a simple indicator for predicting obesity. Am
J Clin Nutr 1984;39:129-35
Rolland-Cachera MF, Bellisle F. No correlation between adiposity and food intake:
why are working class children fatter? Am J Clin Nutr 1986;44:779-87
Rolland-Cachera MF, Bellisle F, Deheeger M, Pequinot F, Sempe M. Influence of
body fat during childhood on body fat distribution in adulthood: A two decade
follow-up study. Int J Obes 1990;14:473-81
Rolland-Cachera MF, Deheeger M, Akrout M, Bellisle F. Influence of macronutrients
on adiposity development: a follow-up study of nutrition and growth from 10
months to 8 years of age. Int J Obes 1995;19:573-8
Rona RJ, Chinn S. National study of health and growth: social and biological factors
associated with height of children from ethnic groups living in England. Ann
Hum Biol 1986;13:453-71
Rona RJ, Chinn S. School meals, school milk and height of primary schoolchildren in
the eighties. J Epidemiol Community Health 1989;43:66-71
122
Rosenbaum PA, Elston RC, Srinivasan SR, Webber LS, Berenson GS. Cardiovascular
risk factors from birth to 7 years of age: the Bogalusa Heart Study. Predictive
value of parental measures in determining cardiovascular risk factor variables
in early life. Pediatrics 1987;80 Suppl:807-16
Ruble DN, Brooks-Gunn J. The experience of menarche. Child development
1982:1557-66
Rubin DH, Krasilnikoff PA, Leventhal JM, Weile B, Berget A. Effect of passive
smoking on birth-weight. Lancet 1986;August 23:415-7
Rudolf MCJ, Sahota P, Barth JH, Walker J. Increasing prevalence of obesity in
primary school children: cohort study. BMJ 2001;322:1094-5
Rush D, Cassano P. Relationship of cigarette smoking and social class to birth weight
and perinatal mortality among all births in Britain, 5-11 April 1970. J
Epidemiol Community Health 1983;37:249-55
Rush D. Examination of the relationship between birth weight and cigarette smoking
during pregnancy and maternal weight gain. J Obstet Gynaecol Br Commonw
1974;81:746-52
Rössner S, Öhlin A. Maternal body weight and relation to birth weight. Acta Obstet
Gynecol Scand 1990;69:475-8
Samuelsson G. An epidemiological study of child health and nutrition in a northern
Swedish county. Medical and anthropometrical examinations. Acta Paediatr
Scand 1971;60:653-65
Sandberg LG, Steckel RH Heights and economic history: the Swedish case. Ann Hum
Biol 1987;14:101-10
Sandberg LG, Steckel RH. Soldier, soldier what made you grow so tall? Economy and
History 1980;23-81
Schmidt-Monnard K. Uber den Einfluss der Jahreszeit und der Schule auf das
Wachstum der Kinder. Jahrbuch fur Kinderheilkunde 1895;40:84-107
Schmidt IM, Jörgensen MH, Michaelsen KF. Height of conscripts in Europe: is
postneonatal mortality a predictor? Ann Hum Biol 1995;22:57-67
Schreider E. Recherches sur la stratification sociale des caracteres biologiques.
Biotypologie 1964;26:105-35
Schumacher LB, Pawson IG, Kretchmer N. Growth of immigrant children in the
newcomer schools of San Francisco. Pediatrics 1987;80:861-8
Scott EM, Illsley LP, Thomson AM. A psychological investigation of primigravidae.
Maternal social class, age, physique and intelligence. J Obstet Gynaecol Br
Emp 1956;63:338-43
Secher NJ, Hjortdal J, Hjortdal V. Smoking affects fetal growth selectively. Acta
Obstet Gynecol Scand 1990;69:469-71
Secker-Walker RH, Vacek PM, Flynn BS, Mead PB. Smoking in pregnancy, exhaled
carbon monoxide, and birth weight. Obstetrics & Gynecology 1997;89(5 Pt
1):648-53
Seidell JC. Obesity: a growing problem. Acta Paediatr 1999;Suppl 428:46-50
Serdula MK, Ivery D, Coates RJ, Freedman DS, Williamsson DF, Byers T. Do obese
children become obese adults? A review of the literature. Prev Med
1993;22:167-77
Serdula MK, Collins ME, Williamsson D, et al. Weight Control Practices of U.S.
Adolescents and Adults. Ann Int Med 1993;119:667-71
123
Shapiro HL. Migration and environment. A study of the physical characteristics of the
Japanese immigrants to Hawaii and the effects of environment on their
descendants. New York: Oxford University Press,1939
Shapiro LR, Crawford PB, Clark MJ, Pearson DL, Raz J, Huenemann RL. Obesity
prognosis: a longitudinal study of children from the age of 6 months to 9 years.
Am J Public Health 1984;74:968-72
Sheehan TJ. Stress and low birth weight: a structural modeling approach using real life
stressors. Soc Sci Med 1998;47:1503-12
Shields A, Sörensen T, Hanis C, et al. An adoption study of human obesity. N Engl J
Med 1986;314:193-8
Scholtens S, Gehring U, Brunekreef B, Smit HA, de Jongste JC, Kerkhof M, Gerritsen
J, Wijga AH. Breastfeeding, weight gain in infancy, and overweight at seven
years of age. Am J Epidemiol 2007;165:919-926
Siniarska A. Family environment and body build in adults of Yucatan (Mexico). Am J
Phys Anthropol 1995; Suppl 20:196
Skelton JA. Childhood Obesity: Overview. Monograph: Part 1. Mead Johnson &
Company 2004
Snow MHL. Effect of genome on size at birth. In: Sharp F, Milner R, Fraser R, eds.
Fetal Growth. London:Royal College of Obstetricians and Gynaecologists,
1989:3-12
de Spiegelaere M, Dramaix M, Hennart P. The influence of socio-economic status on
the incidence and evolution of obesity during early adolescence. Int J Obes
1998;22:268-74
Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of the literature.
Psychol Bull 1989;105:260-75
Socialstyrelsen Amningsstatistik. (Official statistics of breast-feeding in Sweden).
Available at http://www.socialstyrelsen.se (In Swedish with summary in
English)
Sommer A, Lowenstein MS. Nutritional status and mortality: A prospective validation
of the QUAC stick. Am J Clin Nutr 1975;28:287-92
Somerville SM, Rona RJ. Respiratory conditions, including asthma, and height in
primary school. Ann Hum Biol 1993;20(4):369-80
Spencer NJ, Bambang S, Logan S, et al. Socio-economic status and birth weight:
comparison of an area-based measure with the Registrar General’s social class.
J Epidemiol Community Health 1999;53:495-8
Spencer N, Logan S. Social influences on birth weight. Arch Dis Child Fetal Neonatal
Ed 2002;86:F6-7
Spies H, Dreizen S, Snodgrasse RM, Arnett CM, Webb-Peploe H. Effect of dietary
supplement of non fat milk on human growth failure. Am J Dis Childhood
1959;98:187-97
Spitz RA. Hospitalism: an inquiry into the genesis of psychiatric conditions in early
childhood. Psychoanal Study Child 1945;1:53-74
Srinivasan TN. Undernutrition: Concepts, measurement, and policy implication. In
Nutrition and poverty, edited by SR Osmani. Oxford: Clarendon Press,1992
St Georg IM, Williams S, Silva PA. Body size and the menarche: the Dunedin Study. J
Adolesc Health 1994;15(7):573-6
Stark O, Peckham CS, Moynihan C. Weight and age at menarche. Arch Dis Child
1989;64(3):383-7
124
Statistics Sweden. Befolkningsutvecklingen under 250 år. Historisk statistik för
Sverige. Statistiska centralbyrån. (Population development during 250 years.
Historical stastitics for Sweden). Demografiska rapporter 1999:2 (In Swedish)
Steckel RH. Stature and the standing of living. J Econ Lit 1995;23:1903-40
Steckel RH. Growth depression and recovery: the remarkable case of American slaves.
Ann Hum Biol 1987;14:11-132
Steegmann AT, Jr. 18th century British military staure: growth cessation, selective
recruiting, secular trends, nutrition at birth, cold and occupation. Hum Biol
1985;57:77-95
Steenhoff G. Bidrag till kännedomen om vår folkskolehygien. Hygiae1900;62:145-65
Stein CE, Fall CH, Kumaran K, Osmond C, Cox V, Barker DJ.Fetal growth and
coronary heart disease in south India. Lancet 1996;348:1269-73
Steinberger J. Relationship between insulin resistance and abnormal lipid profile in
obese adolescents. J Paediatrics 1995;86:697-706
Steiner H, Lock J. Anorexia nervosa and bulimia nervosa in children and adolescents:
a review of the past 10 years. J Acad Child Adolesc Psychiatry 1998;37:352-9
Stephenson T, Symonds ME. Maternal nutrition as a determinant of birth weight. Arch
Dis Child Fetal Neonatal Ed 2002;86:F4-F6
Stettler N, Zemel BS, Kumanyika S, Stallings VA. Infant Weight Gain and Childhood
Overweight Status in a Multicenter, Cohort Study. Pediatrics 2002;109:194-9
Stini WA. Adaptive strategies of human populations under nutritional stress. In
Biosocial Interactions in Population Adaptation, ed. Watts ES, Johnston FE &
Lasker GW.The Hague: Mouton, 1975. p 19-41
Strandberg U, Bergfors G, Blomquist HK. Även erfarna distriktssköterskor får olika
mätvärden. (Measurements of height, weight and head circumference in child
health services. Even experienced nurses obtain divergent results).
Lakartidningen 2001;98:2330-3 (In Swedish)
Strauss RS, Pollack HA. Epidemic increase in childhood overweight, 1986-1998.
JAMA 2001;286:2845-8
Striegel-Moore RH, Schreiber GB, Pike KM, Wilfrey DE, Rodin J. Drive for thinness
in black and white preadolescent girls. Int J Eating Disord 1995;18:59-69
Stunkard AJ, Sörensen TI, Hanis C, Teasdale TW, Chakraborty R, Schull WJ,
Schulsinger F. An adoption study of human obesity. New Engl J Med
1986;314:193-8
Stunkard AJ, Harris JR, Pedersen NL, McClearn GE. The body mass index of twins
who have been reared apart. N Engl J Med 1990;322:1483-7
Stunkard AJ, Berkowitz RI, Stallings VA, Cater JR. Weights of parents and infants: is
there a relationship? Int J Obes Relat Metab Disord 199;23:159-62
Sullivan K, Trowbridge F, Gorstein J, Pradilla A. Growth references. Lancet
1991;337:1420-1
Sundell C. Mätningar å Stockholms förskolebarn. (Measurements of elementary
schoolchildren in Stockholm). Hyg Tidskr. 1917;10:25-7 (In Swedish)
Sunnegårdh J Physical activity in relation to energy intake, body fat, physical work
capacity and muscle strength in 8-and 13-year-old children in Sweden.
Doctoral thesis. University of Uppsala. Acta Universitatis Upsaliensis 41,1986
Susanne C. Interrelations between some social and familial factors and stature and
weight of young Belgian male adults. Hum Biol 1980;52:701-9
125
Susanne C. Biological differences between migrants and non-migrants. In Migration
and Mobility: Biosocial Aspects of Human Movement, edited by AJ Boyce,
Symposia of the Society for the Study of Human Biology, Trevor&Francis,
London, 1984;23:179-93
Sveger T. Does overnutrition or obesity during the first year affect weight at age four?
Acta Paediatr Scand 1978;67:465-7
Symonds ME, Budge H, Stephenson T. Limitations of models used to examine the
influence of nutrition during pregnancy and adult disease. Arch Dis Child
2000;83:215-9
Söderfeldt B, Werner B. Class inequalities in children’s health indicators in a Swedish
community with a comparison of three measures of social class – a
methodological study. In: Söderfeldt B. Inequality in health: a comparative
methodological analysis of a new way to measure social classes. Doctoral
thesis, Karolinska Institutet, Stockholm, Sweden,1988. p 159-83
Sörensen TI, Sonne-Holm S. Risk in childhood of development of severe adult
obesity: Retrospective, population-based cohort study. Am J Epidemiol
1988;127:104-13
Sörensen TI, Holst C, Stunkard AJ. Childhood body mass index – genetic and familial
environmental influences assessed in a longitudinal adoption study. Int J Obes
1992;16:705-14
Taitz L. The obese child. Oxford: Blackwell,1983. p 142-88
Takahashi E. Secular trend in milk consumption and growth in Japan. Hum Biol
1984;56:427-37
Takaishi M. Secular changes in growth of Japanese children. J Pediatr Endocrinol
1994;7:163-73
Takaishi M. Growth standards for Japanese children – an overview with special
reference to secular change in growth. In Essays on Auxology Presented to
James Mourilyan Tanner by Former Colleagues and Fellows. 1995:302-311
(Hauspie R, Lindgren G and Falkner F editors). Welwyn Garden City, Herts.:
Castlemead Publications,1995
Talbot NB, Sobel EH, Burke BS, Lindenmann E, Kaufmann SB. Dwarfism in healthy
children: its possible relation to emotional, nutritional and endocrine
disturbances. N Engl J Med 1947;236:783-93
Tanner JM. Growth at Adolescence. Oxford: Blackwell,1962
Tanner JM, Whitehouse RH, Takaishi M. Standards from birth to maturity for height,
weight, height velocity, and weight velocity: British children, 1965 Part I. Arch
Dis Child 1965;41:454-71
Tanner JM, Whitehouse RH,, Takaishi M. Standards from birth to maturity for height,
weight, height velocity, and weight velocity: British children, 1965. Part II.
Arch Dis Child 1966;41:613-35
Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height
velocity, weight velocity, and stages of puberty. Arch Dis Child 1975;51:170-9
Tanner JM A history of the study of human growth. Cambridge University Press,1981
Tanner JM, Hayashi T, Preece MA, Cameron N. Increase in length of leg relative to
trunk in Japanese children and adults from 1957 to 1977: comparison with
British and with Japanese Americans. Ann Hum Biol 1982;9:411-23
Tanner JM, Davies PS. Clinical longitudinal standards for height and height velocity
for North American children. J Pediatr 1985;107(3):317-29
126
Tanner JM. Growth as a mirror of the condition of society: secular trends and class
distinctions. Acta Paediatrica Japonica 1987;29:96-103
Tanner JM. Foetus into Man: Physical growth from conception to maturity. Ware:
Castlemead Publications,1989
Tanner JM. Growth as a measure of the nutritional and hygienic status of a population.
Hormone Research 1992;38,Suppl 1:106-15
Taranger J. Ökande övervikt hos unga män - en medicinsk riskfaktor? (Increasing
overweight of young men – a medical risk factor?) Läkartidningen
1985;81:1147 (In Swedish)
Taranger J, Lewin T, Karlberg P. Continuing secular trend of height of Swedish
conscripts. Ann Hum Biol 1978;5:203-5
Taveras EM, Rifas-Shiman SL, Scanlon KS, Grummer-Strawn LM, Shery B, Gillman
MW. To what extent is the protective effect of breastfeeding on future
overweight explained by decreased maternal feeding restriction? Pediatrics
2006;118:2341-8
Thierot-Prevost G, Boccara JF, Francoual C, Badoual J, Job JC. Serum IGF1 and
serum growth-promoting activity during the first postnatal year in infants with
IUGR. Pediat Res 1988;24:380-3
Thissen JP, Ketelsleger JM, Underwood LE. Nutritional regulation of the insulin-like
growth factors. Endocr Rev 1994;15:80-101
Thoday JM. Geneticism and environmentalism, in Meade JE and Parkes AS. 1965.
Tienboon P, Rutihauser IH, Wahlqvist ML. A family study of coronary risk factors in
Geelong. Aus J Public Health 1992;16:20-5
Torun B, Viteri FE. Protein energy malnutrition. In: Shils ME and Young VR(eds)
Modern Nutrition in Health and Disease, 7th ed. Lea and Febigen,
Philadelphia,1988. p 746-73
Tremblay MS, Katzmarzyk PT, Willms JD. Temporal trends in overweight and
obesity in Canada, 1981-1996. Int J Obes 2002;26:538-43
Troiano RP, Flegal KM, Kuczmarski RJ, Campbell SM, Johnson CL. Overweight
prevalence and trends for children and adolescents. Arch Pediatr Adol Med
1995;149:1085-162
Trotter M, Gleser GC.A reevaluation of estimation of stature based on measurements
of stature taken during life and long bones after death. Am J Phys Anthrop
1958;16:79-123
Tsuzaki S, Matsuo N, Ogata T, Osano M. Lack of linkage between height and weight
and age at menarche during the secular shift in growth of Japense children. Ann
Hum Biol 1989;16:429-36
Törnell G. Ett svenskt allmänt läroverk. En skolhygienisk studie. Nordisk hygienisk
tidskrift 1920;1:6-23 (In Swedish)
Udjus LG. Anthropometrical changes in Norwegian men in the twentieth century. The
Anatomical Institute, Anthropological Department, University of Oslo and the
Armed Forces Medical Services. Oslo: Universitetsforlaget,1964
Ulijaszek SJ, Mascie-Taylor CGN, ed. Anthropometry: the individual and the
population. Cambridge Studies in Biological Anthropometry 14, Cambridge
University Press,1994
Usher RH, Mac Lean FH. Intrauterine growth of liveborn Caucasian infants at the sea
level: standards obtained from measurements in 7 dimensions of infants born
between 25 and 44 weeks of gestation. J Pediatr 1969;74(6):901-10
127
Vahl M. Mitteilungen ueber das Gewicht nicht erwachsener Mädchen (1874-1883).
Vorsteher der Erziehungs-Anstalt zu Jägerspis. Copenhagen, 1884 (fördrag vid
läkarkongress i Köpenhamn 1884)
Valdez R, Greenlund KJ, Wattigney WA, Bao W, Berenson GS. Use of weight-for
height indices to predict adult overweight: the Bogalusa Heart Study. Int J Obes
1996;20:715-21
Valoski A, Epstein LH. Nutrient intake of obese children in a family-based
behavioural weight control program. Int J Obes 1990;14:667-77
Victora CG, Barros FC, Lima RC, Horta BL, Wells J. Anthropometry and body
composition of 18 year old men according to duration of breast feeding: birth
cohort study from Brazil. BMJ 2003;327:901-5
Villerme’ L. Memoire sur la taille de l’homme en France. Annales d’hygiene publique
1829;1:551-9
Vincent M, Dierickx J. Etude sur la croissance saisonnaire des escoliers de
Leopoldville. Annales de la Societe Belge de Medecine Tropicale 1960;40:83744
Vlajinac HD, Petrovic RR, Marinkovic JM, Sipetic SB, Adanja BJ. Effect on caffeine
intake during pregnancy on birth weight. Am J Epidemiol 1997;145(4):335-8
Vogler GP, Sörensen TI, Stunkard AJ, Srinivasan MR, Rao DC. Influences of genes
and shared family environment on adult body mass index assessed in an
adoption study by a comprehensive path model. Int J Obes 1995;19:40-5
Vuille JC, Mellbin T. Obesity in 10-years-olds: an epidemiological study. Pediatrics
1979; 64:564-72
Vågerö D, Koupilova I, Leon DA, Lithell U-B. Social determinants of birthweight,
ponderal index and gestational age in Sweden in the 1920s and the 1980s. Acta
Peadiatr 1999;88:445-53
Waaler PE. Anthropometric Studies in Norwegian Children. Acta Paediatr Scand.
1983; Suppl 308
Waaler H. Height, weight and mortality. The Norwegian Experience. Acta Med Scand
1984; Suppl 679
Wachholder A, Hauspie RC, Vercauteren M. Standards for growth in height and
weight and for height velocity in Belgian boys and girls aged 3-18 years. Acta
Paediatr Scand 1989;Suppl 349:147 (poster abstract)
Wadsworth MEJ. Health inequalities in the life course perspective. Soc Sci Med
1997;44:859-69
Wadsworth ME, Kuh DJ. Childhood influences on adult health: a review of recent
work from the British 1946 national birth cohort study, the MRC National
Survey of Health and Development. Paediatr Perinat Epidemiol 1997;11(1):2–
20
Wahlström H. En presentation av ett tillvägagångssätt för att bilda referenskurvor för
barns längd, vikt och huvudomfång.(A presentation of the way to construct
reference curves for height, weight and head circumference of children). Duppsats i statistik. Institutionen för ekonomi, statistik och ADB. Högskolan i
Örebro,1998 (In Swedish)
Wainright RL. Change in observed birth weight association with change in maternal
cigarette smoking. Am J Epidemiol 1983;117:668-75
Wales JKH, Taitz LS. Patterns of growth in abused children. In: Hernandez M,
Argente J. (eds) Human growth: basic and clinical aspects. Proceedings of the
Sixth International Congress of Auxology, Madrid, Spain, 1991,1992. p 151-9
128
Walton A, Hammond J. The maternal effects on growth and conformation in Shire
horse-Shetland pony crosses. Proc R Soc Lond B Biol Sci 1954;125:311-35
Wang X, Zuckerman B, Coffman GA, et al. Familial aggregation of low birth weight
among whites and blacks in the United States. N Engl J Med 1995;333:1744-9
Wang Y, Lobstein T. World trends in childhood overweight and obesity. Int J Pediatr
Obes 2006;1:11-25
Waterlow JC, Buzina R, Keller W, Lane JM, Nichaman MZ, Tanner JM. The
presentation and use of height and weight data for comparing nutritional status
of groups of children under the age of 10 years. Bull WHO 1977;55:489-98
Waterlow JC, Thomson AM. Observation on the adequacy of breastfeeding. Lancet
1979;2:238-42
Waterlow JC. Summary of causes and mechanisms of linear growth retardation. Eur J
Clin Nutr 1994;48 Suppl 1:S210
Weber GW, Prossinger H, Seidler H. Height depends on month of birth. Nature
1998;391:754-5
Weinberg R, Webber LS, Berenson GS. Hereditary and environmental influences on
cardiovascular risk factors for children: the Bogalusa Heart Study. Am J
Epidemiol 1982; 116:385-93
Wellens R, Malina RM, Beunen G, Lefevre J. Age at menarche in Flemish girls:
current status and secular change in the 20th century. Ann Hum Biol
1990;17:145-52
Wellens R, Malina RM, Roche AF, Chumlea WC, Guo S, Siervogel RM. Body size
and fatness in young adults in relation to age at menarche. Am J Hum Biol
1992;4:783-7
Werdelin L. The stature of some medieval Swedish populations. Fornvännen
1985;2:133-41
Werner B, Bodin L, Bremberg S. Data of height and weight from school health
records as a national public health surveillance tool – the case of Sweden.
Scand J Public Health 2006;34:406-13
Werner B, Bodin L. Growth from birth to age 19 for children in Sweden born in 1981
– descriptive values. Acta Paediatr 2006;95:600-13
Werner B, Bodin L. Head circumference from birth to age 48 months for infants in
Sweden. Acta Paediatr 2006;95:1601-7
Werner B, Magnuson A, Bodin L. Increasing rate of weight loss among
schoolchildren, especially girls, in Sweden. J Adolesc Health 2007;40:238-44
Werner B, Bodin L. Obesity in Swedish schoolchildren is increasing in both
prevalence and severity. 2007 submitted
Westerståhl M, Barnekow-Bergkvist M, Hedberg G, Jansson E. Secular trends in body
dimensions and physical fitness among adolescents from 1974 to 1995. Scand J
Med Sports 2003;2:128-37
Weststrate JA, Van Klaveren H, Deurenberg P. Changes in skinfold thicknesses and
body mass index in 171 children initially 1 to 5 years of age: a 5.5 year followup study. Int J Obes 1986;10:313-21
Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting adult obesity
from childhood and parent obesity. N Engl J Med 1997;337:869-73
White P, Woods R. The Geographical Impact of Migration. London:Longman,1980
Whitehead RG, Paul AA. Growth charts and the assessment of infant feeding practices
in the western world and in developing countries. Early Hum Dev 1984;9:187207
129
Whitehead RG, Paul AA. Growth patterns of breastfed infants. Acta Paediatr
2000;89:136-8
Whorwood CB, Firth KM, Budge H, et al. Maternal undernutrition during early-to
mid-gestation programmes tissue-specific alterations in the expression of the
glucocortoid receptor, 11 -hydroxysteroid dehydrogenase isoforms and type 1
angiotensin II receptor in neonatal sheep. Endocrinology 2001;142:2854-64
Widdowson EM, McCance RA. The effect of finite periods of undernutrition at
different ages on the composition and subsequent development of the rat. Proc
R Soc Lond 1963;158:329-42
van Wieringen JC. Secular growth changes. In Human Growth: A Comprehensive
Treatise 1986 vol 3 pp 307-31 (Falkner F and Tanner JM, editors). New york:
Plenum Press,1986
van Wieringen JC, Wafelbakker F, Verbrugge HP, de Haas JH. Growth diagrams 1965
Netherlands. Nederlands Instituut voor Praeventieve Geneeskunde/WolterNoordhoff, Leiden/Groningen,1971. p 1-69
de Wijn JF, Haas JH. Groeidiagrammen van 1:25 jarigen in Nederland (Growth
diagrams for ages 1-25 years in the Netherlands). Nederlands Instituut voor
Praeventieve Geneeskunde, Leiden/Groningen,1960. p 1-29 (In Dutch)
Wilcox AJ. Birth weight and perinatal mortality: The effect of maternal smoking. Am
J Epidemiol 1993;137:1098-104
Wilcox MA, Chang AM, Johnson IR. The effects of parity on birth weight using
successive pregnancies. Acta Obstet Gynecol Scand 1996;75(5):459-63
Wilkinson PW, Parkin JM, Pearlson J, Philips PR, Sykes P. Obesity in childhood: a
community study in Newcastle upon Tyne. Lancet 1977;1:350-2
Williams AF. Growth charts for breastfed babies. Acta Paediatr 2002;91:1282-95
Williams SR, Jones E, Bell W, Davies B, Bourne MW. Body habitus and coronary
heart disease in men. A review with reference to methods of body habitus
assessment. Eur Heart J 1997;18:376-93
Wolanski N. Parent-offspring similarity in body size and proportions. Studies in
Human Ecology 1979;3:7-26
Wolanski N. Ecological aspects of the growth and development of man.
Coll.Antropol. 1988;12(1):7-21
Wolanski N, Dickinson F, Siniarska A. Biological traits and living conditions of Maya
Indian and non-Maya girls from Merida, Mexico. Int J Anthropol 1993;8:23346
Woods KA, Camacho-Hubner C, Savage MO, et al. Intrauterine growth retardation
and postnatal growth failure associated with deletion of the insulin-like growth
factor I gene. N Engl J Med 1996;335:1363-7
World Health Organization. Methodology of nutritional surveillance. Twenty-seventh
report of a joint FAO/UNICEF/WHO expert committee. Technical Report
Series No. 593. Geneva: WHO,1976
World Health Organisation. New trends and approaches in the delivery of maternal
and child care in health services. Sixth report of the WHO expert committee on
maternal and child care. WHO Technical Report Series No. 600. Geneva:
WHO,1976
World Health Organization. Measuring change in nutritional status. Geneva:
WHO,1983
WHO Working Group. Use and interpretation of anthropometric indicators of
nutritional status. Bull WHO 1986;64:929-41
130
World Health Organization. Physical status: the use and interpretation of
anthropometry. Report of a WHO expert committee. Technical Report Series
No. 854. Geneva: WHO,1995
World Health Organisation. Obesity, preventing and managing the global epidemic:
Report of the WHO consultation of obesity. Geneva, WHO,1997
WHO Working Group on the Growth Reference Protocol and WHO Task Force
Methods for the Natural Regulation of Fertility. Growth patterns of breastfed
infants in seven countries. Acta Paediatr 2000;89:215-22
WHO Child Growth Standards. Acta Paediatr 2006;95 Suppl 450
Worthman CM. Biocultural interactions in human development. In Juvenile Primates:
Life History, Development, and Behavior, ed. ME Perieira & LA Fairbanks.
New York: Oxford University Press,1993. p 339-57
Wright CM, Parker L, Lamont D, Craft AW. Implications of childhood obesity for
adult health: findings from thousand families cohort study. BMJ
2001;323:1280-4
Wretlind EW. Iakttagelser rörande helsotillståndet i några af Göteborgs flickskolor.
(Observations of the health situation in some of the schools for girls in
Gothenburg). Eira 1878;2(24):745-64 (In Swedish)
Wright CM, Waterston A, Aynsley-Green A. Effect of deprivation on weight gain in
infancy. Acta Paediatr 1994;83:357-9
Wright CM, Booth IW, Buckler JMH, Cameron N, Cole TJ, Healy MJR, Hulse JA,
Preece MA, Reilly JJ, Williams AF. Growth reference charts for use in the
United Kingdom. Arch Dis Child 2002;86:11-4
Zhang X, Huang Z. The Second National Growth and Development Survey of
Children in China, 1985:children 0 to 7 years. Ann Hum Biol 1988;15(4):289305
Åkerman S, Högberg U, Danielsson M. Height, health and nutrition in early modern
Sweden. In Brändstöm A, Tedebrand LG. Society, health and population.
During the Demographic Transition,1988
131
Paper I-V
Appendix
133
134
I
135
136
137
138
139
140
141
142
II
143
144
145
146
147
148
149
150
151
152
153
154
155
156
III
157
158
159
160
161
162
163
164
IV
165
166
167
168
169
170
171
172
OBESITY IN SWEDISH SCHOOLCHILDREN
IS INCREASING IN BOTH PREVALENCE
AND SEVERITY
Bo Werner1 and Lennart Bodin2
1
Division of Social Medicine, Department of Public Health Sciences,
Karolinska Institutet, Stockholm, Sweden
2
Statistical and Epidemiological Unit, University Hospital, Örebro,
Sweden and Department of Statistics, Örebro University, Örebro,
Sweden
Correspondence: Bo Werner, Child Health Care Unit, Örebro County
Council, Box 1613, SE-701 16 Örebro, Sweden. Tel. +46 70328 1847,
e-mail [email protected]
V
173
Abstract
Purpose: To monitor and describe, on a national level, the development of body mass index
(BMI), overweight, and obesity of schoolchildren in Sweden aged 7–18 y over a period of 8
years.
Methods: Longitudinal and cross-sectional studies of two nationally representative cohorts.
A representative sample of 3749 individuals from a birth cohort of 109,663 individuals born
in 1973, and another representative sample of 3158 individuals from a birth cohort of 94,064
individuals born in 1981; 4.5% and 1.6% of those born in 1973 and 1981, respectively, were
missing from the sample. Data regarding height and weight from school health records.
Results: From age 7–18 y, a strong positive secular change in BMI is found at all ages, and
the rate of overweight and obesity is increasing for both boys and girls. Furthermore, obesity
is growing more severe.
Conclusion: Nationally representative longitudinal BMI data for two cohorts, in which nonresponse bias is minimized, permitted monitoring and revealed a nation-wide strongly positive
secular change in BMI in Sweden, over a period of 8 years for individuals aged 7–18 y.
Keywords: longitudinal study, national, obesity, schoolchildren, secular change
174
:KLOHPHDVXUHPHQWLQFRQVLVWHQFLHVPDNHLWGLI¿FXOWWRSURYLGHDQRYHUYLHZRIWKHJOREDO
prevalence of overweight and obesity in children, studies generally report a high and
increasing prevalence of these conditions worldwide [1,2]. Current estimates (1998) of the
childhood obesity prevalence in the USA suggest that 21.5% of African-Americans, 21.8% of
Hispanics, and 12.3% of non-Hispanic whites are overweight, and that overweight increased
rapidly between 1986 and 1998 [3]. In Australia, most recent estimates suggest that at ages
7–12 y, 16.1–16.9% of boys are overweight and 5.1–6.9% are obese, and 17.4–20.4% of girls
are overweight and 5.7–7.0% are obese [4]. In the UK, reports suggest that the prevalence of
obesity among children of all ages is increasing [5-7]. Data from the National Study of Health
and Growth indicate a dramatic rise in the proportion of overweight primary school children
(aged 4–11 y) in the UK between 1984 and 1994: overweight increased from 5.4% to 9.0% in
English boys, from 6.4% to 10.0% in Scottish boys, from 9.3% to 13.5% in English girls, and
from 10.4% to 15.0% in Scottish girls [8]. Data from a large survey of young children (aged 1
mo to 4 y) in England indicated a similar rise in the prevalence of overweight [9].
In a summary of cross-sectional, nationally representative school-based surveys in 19971998 in 15 countries, based on self-reported data, the highest prevalence of overweight for
13 and 15 years old boys and girls was found in the United States and the lowest prevalence
in Lithuania. The highest rate in Europe was found in Greece [10]. Prevalence data from 21
surveys in Europe revealed a tendency for higher prevalence of overweight among children in
western and especially southern Europe [11].
The prevalence of childhood overweight has increased in almost all countries for which
data are available. Exceptions are found among school-age children in Russia and to some
extent in Poland during the 1990’s. Exceptions are also found among infant and pre-school
children in some lower-income countries. Both obesity and overweight have increased more
dramatically in economically developed countries and in urbanized populations [12]. The
prevalence of overweight and obesity among children is rising in the European region, and the
annualised rates of increase are themselves increasing [13].
'HYHORSLQJFRQVLVWHQWDSSURDFKHVWRPHDVXULQJFKLOGKRRGREHVLW\LVDSULRULW\LQWKLV¿HOG
[14], and recently international standardized cut-off points have been proposed [15]. For
GH¿QLQJRYHUZHLJKWLQFKLOGUHQUHIHUHQFHYDOXHVIRUERG\PDVVLQGH[%0,DUHDYDLODEOH
from the US Centers for Disease Control and Prevention (CDC) [16] and the IOTF
(International Obesity Task Force) [15]. Current expert opinion supports the use of BMI cutoff points for children and adolescents, but experts are divided as to which centiles should be
used for purposes of comparison [17,18].
In Sweden, studies of overweight time trends are either local [19], do not compare nationally
representative, cross-sectional materials that could reveal national trends [20,21], or examine
only 7-y-old children in Stockholm from 1940 to 1990 [22]. The most recent study examining
time trends compares population-based material from Gothenburg for subjects born in 1973–
1975 with material from parts of Stockholm for subjects born in 1985–1987 [23]. It must be
asked, however, whether such a comparison could actually capture a time trend, or rather a
local difference between Gothenburg and Stockholm?
The ongoing worldwide epidemic of overweight and obesity among children and adolescents
is a public health crisis [24]. We need a reliable and valid tool for epidemiological
surveillance to genuinely capture the development. It is important both to describe the depth
and velocity of the time change and to discover when the trend changes. What kinds of survey
let us monitor this epidemic? How can national situations be monitored?
It is rare for surveys to be statistically representative of larger populations, and they are
very seldom nationally representative. Data regarding body weight are collected either by
weighing individuals in dedicated studies, or more often by asking individuals about their
body weight in questionnaires. The response rate in questionnaire surveys is often less than
60% [4,29]. National epidemiological surveillance and the proper study of time changes
175
require representative studies with ignorable selection bias and repeatable later under
similar circumstances. Growth data should also be based on measurements. Sometimes
one encounters a time period covered by register data of national coverage, but just for a
single age, as is the case with conscription data regarding 18-y-old males in Sweden [25].
Longitudinal studies of nationally representative samples make it possible to conduct analyses
both within and between cohorts. So far, to the best of our knowledge, there has been only
one such study, but that was a mixed longitudinal study [26]. Examples of growth monitoring
treating large samples do exist in Australia, the UK, the Netherlands, China, Finland, Canada,
Norway, and Sweden [4,26-33].
The present study monitors and describes, on a national level, the development of body mass
index (BMI), overweight, and obesity of schoolchildren in Sweden aged 7–18 y over a period
of 8 y.
What kind of knowledge is missing?
Repeated comparable surveys of a whole country with longitudinal data that makes it possible
to reveal changing patterns of overweight and obesity from age 7y to 18 y with minimal
selection bias.
Material and methods
From two target populations, all born in 1973 and 1981 and living in Sweden as of 31
December 1989, two study populations were sampled, including all individuals born on the
15th of any month. Data for these two samples were collected from school health records from
all of Sweden, where trained nurses had taken all measurements. Table 1 presents the study
populations, data sources, and sample sizes.
Table 1 about here
The study population for the 1973 cohort consists of 3749 children and the number of missing
records was 170 (4.5%), resulting in the records of 3579 individuals being available (1855
ER\VDQGJLUOV7KHFRUUHVSRQGLQJ¿JXUHVIRUWKHFRKRUWZHUHLQGLYLGXDOV
in the study population, 51 missing records (1.6%), and records being available for 3107
LQGLYLGXDOVER\VDQGJLUOV7KHGDWDZHUHFROOHFWHGLQWKUHHZDYHVWKH¿UVWLQ
1989 and then up to 1994 for the 1973 cohort and up to 2002 for the 1981 cohort. For more
information concerning methods and data collection, see Werner et al. [34,35]. Combining all
height and all weight measurements resulted in 25,977 BMI values for all individuals in the
1973 cohort and 31,551 BMI values for all individuals in the 1981 cohort.
Statistics
Data were analyzed both cross-sectionally and longitudinally. For the cross-sectional
approach, more thoroughly described in Werner et al>@ZHGH¿QHGLQWHUYDOVZLWK
midpoints at whole years from 7 y and up to and including 18 y. The width of the intervals
was (180 d. In each interval, one randomly chosen observation for each individual (when
available) was used to form a linear regression of the outcome (height and BMI) on age
measured in days; this was done separately for each cohort. The linear regression was used to
estimate the outcome corresponding to the midpoint of the interval, since observations were
randomly distributed in calendar time. In addition to means, we described the variation and
distribution in each age group, i.e., SD, skewness, kurtosis, and percentiles were calculated.
Kurtosis indicates whether the distribution around the mean is thick-tailed (a higher
proportion of subjects are found at the extremes of the distribution, kurtosis >0) or thin-tailed
(the opposite case, kurtosis <0).
In the longitudinal approach, the two cohorts were analyzed together. We formed a regression
model with outcome (height or BMI) as a function of two sets of categorical variables, one
for cohort (1973 or 1981) and the other for age in whole years (7, 8, 9 … 18 y). Interaction
between these two variables was also introduced and tested. The parameters in this model
correspond to means for each age category and for each cohort estimated simultaneously
176
using all selected data from the age groups in both cohorts.
The longitudinal model was estimated using a mixed-model approach and the SAS statistical
package (Version 8.2, SAS Institute, Cary, NC, USA), with a covariance structure according
WRDQDXWRUHJUHVVLYHVWUXFWXUHRIWKH¿UVWRUGHUGHJUHHVRIIUHHGRPDFFRUGLQJWRWKH
Satterthwaite approximation, and using the restricted maximum likelihood estimation method
(REML). We calculated the marginal means at each age and contrasted these means for the
two cohorts; p-values were adjusted for multiple comparisons. The cross-sectional analysis
was done using the SPSS statistical package (Version 13, SPSS Inc., Chicago, IL, USA),
VXSSOHPHQWHGZLWKXVHUGH¿QHGFRPPDQGV
The Ethics Committee of the Örebro County Council, Sweden, approved this study.
Results
Table 2 presents the cross-sectionally based comparison of the 1973 and 1981 cohorts.
Table 2 about here
Height
Mean heights of both boys and girls are higher in the 1981 than the 1973 cohort, with a
difference of roughly 0.4 cm for boys and 0.8 cm for girls. An exception is the mean height
at age 18 y for boys, in which case the 1973 cohort has a marginally higher value. For boys
born in 1981, SDs are higher at all ages except at age 15 y; skewness and kurtosis are close to
zero for both cohorts and the differences between the cohorts are small and in both directions.
For girls, the SD is marginally higher in the 1973 cohort; as well, there is a tendency towards
negative skewness in the 1973 cohort, which is not present in the 1981 cohort except at
higher ages. Kurtosis is lower in the 1981 than the 1973 cohort. For both boys and girls the
distribution in each age category is close to normal, though the means are somewhat higher in
the 1981 cohort.
BMI
Mean BMIs of both boys and girls are higher for those born in 1981 than in 1973, except for
girls aged 17 and 18 y. For both boys and girls, SDs are higher in the 1981 cohort, indicating
a more heterogeneous distribution. For boys born in 1973, skewness and kurtosis at age 7–14
y are higher than in the 1981 cohort, and the difference in kurtosis is especially high at ages
7, 8, and 10 y. At the higher ages, i.e., 16–18 y, the distributional shape has shifted and both
skewness and kurtosis are higher in the 1981 than the 1973 cohort. Girls display somewhat
higher skewness in the 1981 cohort and an irregular pattern for the difference in kurtosis
between the 1973 and 1981 cohorts. BMI is not normally distributed for either boys or girls,
since both skewness and kurtosis differ substantially from zero.
,Q7DEOHPHDQKHLJKWVDQG%0,VIRUERWKFRKRUWVVWUDWL¿HGIRUER\VDQGJLUOVDUHFRPSDUHG
LQDORQJLWXGLQDODQDO\VLVRIWKHUHSHDWHGPHDVXUHPHQWVIRUDOOFKLOGUHQ6WDWLVWLFDOVLJQL¿FDQFH
IRUDJHVSHFL¿FGLIIHUHQFHVEHWZHHQWKHFRKRUWVEDVHGRQDPL[HGPRGHODQDO\VLVDQG
FRUUHFWHGIRUPXOWLSOHVLJQL¿FDQFHWHVWLQJLVSUHVHQWHG6LQFHQRVWDWLVWLFDOVLJQL¿FDQFHVZHUH
found for the interactions between age and cohort, they were removed from the model; age
and cohort were the only remaining variables.
Table 3 about here
The means obtained from the longitudinal analysis correspond very well, although not exactly,
to the values obtained from the cross-sectional analysis presented in Table 2. Using the
longitudinal model, it was possible to test simultaneously for statistical differences between
WKHFRKRUWV7KHUHZDVDVLJQL¿FDQWGLIIHUHQFHLQKHLJKWIRUER\VRQO\DWDJH\EXWIRUJLUOV
at almost every age, except at ages 10, 12, and 18 y. For BMI, on the other hand, there were
VWDWLVWLFDOO\VLJQL¿FDQWGLIIHUHQFHVDWDOODJHVIRUER\VDQGDWDOODJHVH[FHSW±\IRUJLUOV
Another feature that could be explicitly tested with the longitudinal analysis was the secular
177
trend in BMI from the 1973 to the 1981 cohort. Performing the analysis exactly as above,
but contrasting the “age + 1” age group in the 1973 cohort with “age” in the 1981 cohort,
DOOVLJQL¿FDQWGLIIHUHQFHVGLVDSSHDUHGIRUERWKER\VDQGJLUOV,IWKHFRQWUDVWZDV³DJH´
versus “age”±a difference of two years±WKHVWDWLVWLFDOVLJQL¿FDQFHZDVDJDLQKLJKDVLQWKH
³DJH´YHUVXV³DJH´DQDO\VLV7KXVWKHUHZHUHQRVWDWLVWLFDOO\VLJQL¿FDQWGLIIHUHQFHVLQ%0,
IRUDQDJHGLIIHUHQFHRIRQH\HDUORZHULQWKHFRKRUWEXWVLJQL¿FDQWGLIIHUHQFHVZHUH
found for all other age comparisons.
The 5th, 10th, 50th, 90th, and 95th percentiles for the two cohorts and for boys and girls,
respectively, are presented in Figure 1.
Figure 1 about here
The lower percentiles, i.e., 5th and 10th, do not indicate any substantial difference between
the 1973 and 1981 cohorts. The median, i.e., the 50th percentile, has slightly higher values in
all age groups for boys, whereas for girls not all ages have higher values in the 1981 cohort,
since the 14, 17 and 18-y age groups have higher values in the 1973 cohort, in particular
the oldest group. For the higher percentiles, i.e., 90th and 95th, both boys and girls display
substantially higher values in the 1981 cohort. For the 90th percentile, the differences are
from 0.3 to 1.3 units, the greatest relative difference being for the 12 and 13-y age groups. The
95th percentiles are similar in appearance to the 90th percentiles. A notable peak with high
values for the highest percentiles is present for boys at age 12 y, in particular for the 1973
FRKRUWWKLVLVDOVRUHÀHFWHGLQWKHWKSHUFHQWLOHGDWDQRWVKRZQZKHUHWKHFRKRUW
has higher values than the 1981 cohort does for boys in this age group. Another common
feature of the 99th percentile is that the 1981 cohort has higher values than the 1973 cohort
does, the maximum difference being 4.7 and 3.0 units at age 17 y for boys and 15 y for girls,
respectively. However, these highest percentiles are based on comparably few individuals
DOORZLQJIRUVXEVWDQWLDOVDPSOLQJÀXFWXDWLRQV
The reference values for overweight and obesity suggested by Cole et al. [15] have been
applied to the 1973 and 1981 cohorts (see Table 4).
Table 4 about here
Boys
The percentage of boys with overweight ranges from 7.3 to 12.6 in the 1973 cohort and from
11.9 to 18.3 in the 1981 cohort. The percentage of boys with obesity ranges from 0.7 to 2.4
in the 1973 cohort and from 1.7 to 4.4 in the 1981 cohort. The difference in percentage of
overweight between the cohorts varies between 4.1 and 6.3, while the difference in percentage
of obesity varies between 0.4 and 2.3.
Girls
The percentage of girls with overweight ranges from 9.2 to 13.0 in the 1973 cohort and from
11.8 to 17.9 in the 1981 cohort, the difference between cohorts tending to decrease with age.
The percentage of girls with obesity ranges from 1.2 to 2.3 in the 1973 cohort and from 1.6 to
3.5 in the 1981 cohort; there are very small percentage differences (i.e., 0.1<0.6) at age 13, 14,
16, and 17 y.
Discussion
&RPSDULQJ%0,¿JXUHVIRUWZRQDWLRQDOO\UHSUHVHQWDWLYHVDPSOHVRIFKLOGUHQERUQLQ
and 1981 revealed a positive secular change at all ages. Over eight years, the prevalence of
both overweight and obesity increased, and obesity has grown more severe among both boys
and girls. A notable feature is that the mean BMI achieved at one age in the 1973 cohort is
generally achieved one year earlier in the 1981 cohort. To the best of our knowledge this is
WKH¿UVWWLPHH[FHSWLQWKHFDVHRIH[DPSOHVIURPWKH8.>@WKDWKHLJKWDQGZHLJKW
have been monitored epidemiologically over time, on a national level, using two longitudinal
samples collected in the same way with the same data source.
178
The strength of the present study lies in its use of longitudinal datasets with high national
coverage and minimal selection bias. We were also able to obtain some information regarding
selection bias when this was suspected. This was the case for girls in the 1981 cohort at age
17 and 18 y, where we were able to control for the selection bias recognized in the course of
the study. Mean BMI at these ages was probably underestimated by approximately 0.5 units
[35].
Thus the materials are comparable in almost every respect. Even though both datasets have
a small number of missing individuals, the fact that the percentage of missing cases differs
(4.5% in the 1973 cohort versus 1.6% in the 1981 cohort) could mean that differences in
BMI are over-estimated because of selection bias. However, the external missing cases
were not caused by individuals unwilling to take part in the investigation, but rather by the
inability of the investigators to collect all the existing records in the school health system. We
demonstrated in an earlier paper [34] that excluding the 10 heaviest individuals substantially
reduced the high kurtosis found in the 1973 cohort, especially at ages 7 and 8 y. Therefore,
the high kurtosis found in the 1973 cohort speaks against any selection bias, in terms of the
heaviest boys not being measured.
Another possible objection to considering the cohorts comparable is that because of increasing
obesity, obese individuals may be measured more often. In our analysis, every individual
contributed only one measurement in each age interval. Even if an increasing prevalence of
obesity results in more frequent measurement and thus possibly distorts the comparison, we
do not think this is a crucial problem. Besides, our longitudinal statistical analysis would be
less affected by this change in the measurement pattern.
Two different statistical approaches to comparing the two cohorts were used, one crosssectional and one longitudinal. The cross-sectional method was applied separately to each age
group and cohort, whereas the longitudinal method used the same data but in a simultaneous
estimation treating all the data together. The cross-sectional results were more informative as
to the spread and variation in each group, but took no account of the growth and correlated
data for each child over the entire age range of 7<18 y. In contrast, an ability to consider the
growth and correlated data for each child is the strength of the longitudinal model, providing
the basis for a simultaneous estimation using all ages and both cohorts. The varying number
RIPHDVXUHPHQWVLQHDFKDJHLQWHUYDODOVRH[HUWVOHVVLQÀXHQFHLQWKHORQJLWXGLQDODQDO\VLV,W
should be noted, however, that the longitudinal model uses a longitudinal analysis approach
QRWGHSHQGHQWRQDQ\VSHFL¿FIXQFWLRQDOIRUPIRUJURZWKVXFKDVH[SRQHQWLDORUORJDULWKPLF
forms. This could be either a strength or potential weakness depending on whether an inferior
RUDJRRGIXQFWLRQDOVSHFL¿FDWLRQLVXVHGLQDPRUHHODERUDWHPDWKHPDWLFDOPRGHO
So far, the development of relative body weight in Sweden has been best described by data
collected at conscription. Conscript data cover approximately 90% of 18-y-old males, yearly,
from 1962 to 2000,2 that is, for cohorts born in 1944 up to 1982. The development of mean
BMI among 18-y-old conscripts in 1962, 1972, 1982, 1992, and 2000 is represented by values
of 20.9, 21.3, 21.7, 22.1, and 22.7, respectively [25], a shift being evident in approximately
1972. Perhaps this indicates that boys born as late as approximately 1954 still had a body
weight that could be considered “normal.”
The two previously mentioned reviews [10, 11] were examined by Lissau and she concluded
that the year of data collection, methods and use of appropriate statistics are of critical
importance for the conclusion drawn from comparative epidemiological surveys on the
SUHYDOHQFHRIRYHUZHLJKW>@,QWKHRYHUYLHZE\-DFNVRQ/HDFK>@WKH¿JXUHVIRU6ZHGHQ
DUHEDVHGRQWKHVXUYH\VE\(NEORP>@DQG3HWHUVHQ>@7KH¿UVWVXUYH\KDVGHFUHDVLQJ
heights over time for the populations compared although we know that height has increased
over time in Sweden [25], and a positive secular trend for height is also found in the present
study. The second survey covers only the city of Umeå in the northernmost part of Sweden,
thus a regional rather than a national survey. This emphasizes the importance of national
representative samples when making comparisons over time for revealing national trends in
overweight and obesity.
179
Conclusion
To monitor changes in growth patterns among children and adolescents in a large population,
it is important to have data that are representative; in this study we have achieved
representativity in many basic respects. We can conclude that there is a marked positive
change in BMI among both boys and girls from age 7 to 18 y; furthermore, a visible positive
secular change in the height of both boys and girls is still evident in Sweden. In the same two
cohorts, we earlier found a time change of an increasing rate of weight loss, especially for
girls [38]; even so, there is a strong general increase of overweight and obesity.
It is thus possible to monitor a national situation using data collected from school health
records, which is more cost-effective than performing a dedicated study to examine the
national growth situation among children and adolescents.
Acknowledgements
First, we wish to thank the initiators of the work with the Swedish National Growth Studies,
Gösta Samuelsson and Bengt-Erik Ginsburg. The late Anders Ericson at the National Board of
Health and Welfare is remembered for supporting our group over many years. Thanks are also
GXHWR$QQLND6WUDQGHOODQG.DULQ*UXQGEHUJIRURUJDQL]LQJDQGVXSSRUWLQJWKH¿UVWZDYHRI
data collection.
We would like to acknowledge all the nurses in the Child Health Care and School Health
systems all over the country, and all the personnel at archives in local communities and
county councils who helped us retrieve data from records. We are grateful to Lars Ekholm
who computerized the database. Sven Bremberg is thanked for his valuable comments on the
manuscript.
gUHEUR&RXQW\&RXQFLODQGWKH1DWLRQDO%RDUGRI+HDOWKDQG:HOIDUHKDYH¿QDQFLDOO\
supported the data collection.
180
References
[1] World Health Organization. Obesity, preventing and managing the global epidemic:
Report of a WHO consultation. Geneva, World Health Organisation, 1997.
[2] Livingstone MBE. Childhood obesity in Europe: a growing concern. Public Health Nutr
2001;4(1A):109<16.
[3] Strauss RS, Pollack HA. Epidemic increase in childhood overweight, 1986<1998. JAMA
2001;286:2845<8.
[4] Lazarus R, Wake M, Hesketh K, Waters E. Change in body mass index in Australian
primary school children, 1985<1997. Int J Obes 2000;24(6):679<84.
[5] Reilly JJ, Dorosty AR, Emmett PM. Prevalence of overweight and obesity in British
children: cohort study. BMJ 1999;319:1039.
[6] Reilly JJ, Dorosty AR. Epidemic of obesity in UK children. Lancet 1999;354:1874<5.
[7] Rudolf MCJ, Sahota P, Barth JH, Walker J. Increasing prevalence of obesity in primary
school children: cohort study. BMJ 2001;322:1094<5.
[8] Chinn S, Rona RJ. Prevalence and trends in overweight and obesity in three cross
sectional studies of British children, 1974<94. BMJ 2001;322:24<6.
[9] Bundred P, Kitchiner D, Buchan I. Prevalence of overweight and obese children between
1989 and 1998: population based series of cross-sectional studies. BMJ 2001;322:1<4.
[10] Lissau I. Overweight and obesity epidemic among children. Answer from European
countries. Int J Obes 2004;28:S10-S15
[11] Lobstein T, Frelut M-L. Prevalence of overweight among children in Europe. Obes Rev
2003;4:195-200
[12] Wang Y, Lobstein T. World trends in childhood overweight and obesity. Int J Pediatr
Obes 2006;1:11-25
[13] Jackson-Leach R, Lobstein T. Estimated burden of paediatric obesity in Europe. Part 1.
The increase in the prevalence of child obesity in Europe is itself increasing. Int J Peadiatr
Obes 2006;1:26-32
[14] Barlow SE, Dietz WB. Obesity evaluation and treatment: Expert Committee
recommendations. The Maternal and Child Health Bureau, Health Resources and Services
Administration and the Department of Health and Human Services. Pediatrics 1998;102(3):
Available at: www.pediatrics.org/cgi/content/full/102/3/e29.98393765.
>@&ROH7-%HOOL]]L0&)OHJDO.0'LHW]:+(VWDEOLVKLQJDVWDQGDUGGH¿QLWLRQIRUFKLOG
overweight and obesity worldwide: international survey. BMJ 2000;320:1240<53.
[16] Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, et al.
CDC growth charts: United States. Adv Data 2000;8:1<27.
[17] Bellizzi MC, Dietz WH. Workshop on childhood obesity: summary of the discussion. Am
J Clin Nutr 1999;70(1):173S<175S.
[18] Reilly JJ, Wilson M, Summerbell CD, Wilson D. Obesity in childhood and adolescence:
evidence based answers to common questions. Arch Dis Child 2002;86:302<5.
[19] Petersen S, Brulin C, Bergström E. Increasing prevalence of overweight in young
schoolchildren in Umeå, Sweden, from 1986 to 2001. Acta Paediatr 2003;92:848<53.
>@(NEORPg2GGVVRQ.(NEORP%+HDOWKUHODWHG¿WQHVVLQ6ZHGLVKDGROHVFHQWVEHWZHHQ
1987 and 2001. Acta Paediatr 2004;93:681<6.
[21] Westerstahl M, Barnekow-Bergkvist M, Hedberg G, Jansson E. Secular trends in body
GLPHQVLRQVDQGSK\VLFDO¿WQHVVDPRQJDGROHVFHQWVLQ6ZHGHQIURPWR6FDQG-
Med Sci Sports 2003;13:128<37.
[22] Cernerud L. Height and body mass index of seven-year-old Stockholm school children
from 1940 to 1990. Acta Paediatr 1993;82:304<5.
[23] Eriksson M, Rasmussen F, Nordqvist T. Changes in shape and location of BMI
distribution of Swedish children. Acta Paediatr 2005;94:1558<65.
[24] Lobstein T, Baur L, Uany R for the IOTF Childhood Obesity Working Group. Obesity in
children and young people: A crisis in public health. Obesity Reviews 2004;5(Suppl):4<85.
[25] MSCR, The Swedish Military Conscript Registry, Pliktverket Karlstad, Sweden
(Mönstringsregistret, Pliktverket Karlstad) URL: Available at: http://www.pliktverket.se/sv/
Statistik/Rapporter/ (In Swedish).
[26] Hughes JM, Chinn LL, Rona RJ. Trends in growth in England and Scotland, 1972 to
1994. Arch Dis Child 1997;76:182<9.
181
[27] Fredriks AM, van Buuren S, Burgmeijer RJ, Meulmeester JF, Beuker RJ, Brugman E,
Roede MJ, Verloove-Vanhorick SP, Wit JM. Continuing positive secular growth change in the
Netherlands 1955<1997. Pediatr Res 2000;47(3):316<23.
[28] Luo J, Hu FB. Time trends of obesity in pre-school children in China from 1989 to 1997.
Int J Obes 2002;26:553<8.
[29] Kautiainen S, Rimpelä A, Vikat A, Virtanen SM. Secular trends in overweight and
obesity among Finnish adolescents in 1977<1999. Int J Obes 2002;26:544<52.
[30] Booth M, Wake M, Armstrong T, Chey R, Hesketh K, Mathur S. The epidemiology of
overweight and obesity among Australian children and adolescents, 1995<1997. Aus N Z J
Public Health 2001;25(2):162<9.
[31] Tremblay MS, Katzmarzyk PT, Willms JD. Temporal trends in overweight and obesity in
Canada, 1981<1996. Int J Obes 2002;26:538<43.
[32] Andersen L, Lillegaard ITL, Overby N, Lytle L, Klepp K-I, Johansson L. Overweight
and obesity among Norwegian schoolchildren: changes from 1993 to 2000. Scand J Public
Health 2005;33:99<106.
[33] Rasmussen F, Johansson M, Hansen HO. Trends in overweight and obesity among 18year-old males in Sweden between 1971 and 1995. Acta Paediatr 1999;88:431<7.
[34] Werner B, Bodin L, Bremberg S. Data of height and weight from school health records
as a national public health surveillance tool: the case of Sweden. Scand J Public Health
2006;34:406<13.
[35] Werner B, Bodin L. Growth from birth to age 19 for children in Sweden born in 1981:
descriptive values. Acta Paediatr 2006;95:600<13.
[36] Thomas PW, Peters TJ, Golding J, Haslum MN. Weight-for-height in two national
cohorts with particular reference to 10-year-old children. Ann Hum Biol 1989;16:109<19
[37] Lissau I, Overpeck D, Ruan J, Due P, Holstein BE, Hediger ML, and the Health
Behaviour in School-aged Children Obesity Working Group. Body mass index and
overweight in adolescents in 13 European countries, Israel and the United States. Arch Pediatr
Adolesc Med 2004;158:27-33
[38] Werner B, Magnusson A, Bodin L. An increasing rate of weight loss among
schoolchildren, especially girls, in Sweden. J Adolesc Health 2007;40:238-44
182
Body mass index (BMI)
1973 5%
1973 10%
1973 50% (Median)
1973 90%
1973 95%
1981 5%
1981 10%
1981 50% (Median)
1981 90%
1981 95%
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
Boys
Median
7
8
9
10 11 12 13 14 15 16 17 18
Body mass index (BMI)
Age
1973 5%
1973 10%
1973 50% (Median)
1973 90%
1973 95%
1981 5%
1981 10%
1981 50% (Median)
1981 90%
1981 95%
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
Girls
Median
7
8
9
10 11 12 13 14 15 16 17 18
Age
Figure 1. Percentiles at the low and high end, as well as median, for the two
FRKRUWVVWUDWL¿HGIRUER\VDQGJLUOV
183
184
109,663
(56,411 / 53,252)
3749
3579
(1855 / 1724)
94,064
(48,131 / 45,933)
3158
3107
(1548 / 1559)
All children born in Sweden in 1973
All children born on the 15th of any month in 1973, living in Sweden on
31 December 1989
All children born on the 15th of any month in 1973, living in Sweden on
31 December 1989
All children born in Sweden in 1981
All children born on the 15th of any month in 1981, living in Sweden on
31 December 1989
All children born on the 15th of any month in 1981, living in Sweden on
31 December 1989
Statistics Sweden
Statistics Sweden
School health
records
Statistics Sweden
Statistics Sweden
School health
records
n
(boys / girls)
Study populations
Available data
sources
Table 1. Study populations, data sources, and sample sizes.
185
1973
1328
909
815
1203
918
760
1270
1058
1144
1235
523
1539
n
1981
1118
1306
1065
1162
918
888
1101
899
977
1067
377
1083
n
1973
Mean
1981
Mean
1973
1981
1973
1981
7
1255
1078
8
871
1325
9
770
1118
10
1106
1217
11
868
939
12
794
920
13
1159
1188
14
1033
953
15
1084
1063
16
1155
1117
17
487
387
18
831
350
1RWH¨ GLIIHUHQFHEHWZHHQDQG¿JXUHV
Age
Height, Girls
7
8
9
10
11
12
13
14
15
16
17
18
Age
Height, Boys
¨
¨
0
1973
1973
1981
SD
1981
SD
¨
0
¨
0
1973
1973
¨
0
0
0
0
0
0
1981
¨
0
Skewness
1981
Skewness
1973
1973
Table 2.+HLJKWDQG%0,IRUWKHWZRFRKRUWVDQGVWUDWL¿HGIRUER\VDQGJLUOV0HDQVWDQGDUGGHYLDWLRQ6'VNHZQHVVDQGNXUWRVLV
DWHDFKDJHEDVHGRQFURVVVHFWLRQDOGDWD
Kurtosis
1981
Kurtosis
1981
¨
¨
0
186
1973
1325
889
762
1171
892
713
1188
1015
1097
1180
492
1527
n
n
1981
1111
1301
1056
1156
912
880
1080
878
957
1042
360
1083
1973
15.8
16.1
16.5
17.0
17.4
18.2
18.5
19.5
20.2
20.9
21.3
22.1
Mean
1981
16.1
16.4
16.9
17.5
17.9
18.8
19.1
20.0
20.8
21.4
21.8
22.6
Mean
1973
1981
1973
1981
7
1251
1071
15.7
16.0
8
856
1319
16.2
16.5
9
740
1110
16.6
17.0
10
1086
1213
17.1
17.5
11
844
924
17.5
17.9
12
750
907
18.3
18.9
13
1101
1163
19.1
19.6
14
983
919
20.2
20.5
15
1046
1040
20.7
21.2
16
1106
1088
21.3
21.6
17
456
373
21.6
21.5
18
809
347
22.0
21.9
1RWH¨ GLIIHUHQFHEHWZHHQDQG¿JXUHV
Age
BMI, Girls
7
8
9
10
11
12
13
14
15
16
17
18
Age
BMI, Boys
Table 2. (continued).
¨
0.3
0.3
0.4
0.4
0.4
0.6
0.5
0.3
0.5
0.3
-0.1
-0.1
¨
0.3
0.3
0.4
0.5
0.5
0.6
0.6
0.5
0.6
0.5
0.5
0.5
1973
1.7
2.0
2.2
2.4
2.5
2.8
2.8
3.0
2.9
2.9
2.9
2.8
1973
1.6
1.8
1.8
2.1
2.4
2.7
2.5
2.7
2.8
2.7
2.5
3.0
1981
2.0
2.2
2.5
2.8
3.0
3.3
3.2
3.2
3.3
3.2
2.9
3.4
SD
1981
1.7
1.9
2.2
2.5
2.6
3.0
2.8
3.0
3.3
3.2
3.3
3.7
SD
¨
0.3
0.2
0.3
0.4
0.5
0.5
0.4
0.2
0.4
0.3
0
0.6
¨
0.1
0.1
0.4
0.4
0.2
0.3
0.3
0.3
0.5
0.5
0.8
0.7
1973
1.1
1.4
1.4
1.2
1.3
1.1
1.1
1.2
1.2
1.3
1.6
1.1
1973
2.2
2.3
1.7
2.0
1.9
1.8
1.7
1.6
1.6
1.3
1.6
1.4
1981
1.2
1.4
1.2
1.3
1.4
1.3
1.2
1.3
1.3
1.4
1.7
1.7
Skewness
1981
1.3
1.4
1.5
1.4
1.4
1.2
1.3
1.3
1.6
1.6
2.0
1.6
Skewness
¨
0.1
0
-0.2
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.1
0.6
¨
-0.9
-0.9
-0.2
-0.6
-0.5
-0.6
-0.4
-0.3
0
0.3
0.4
0.2
1973
2.6
3.5
3.7
2.4
2.3
1.2
2.0
2.7
2.1
3.0
5.2
2.4
1973
12.9
12.4
5.6
9.8
6.5
4.8
5.6
5.0
5.3
3.0
5.9
3.5
1981
2.4
2.5
1.9
2.2
3.0
1.8
2.3
2.4
2.7
3.0
5.6
5.7
Kurtosis
1981
3.1
3.7
3.4
3.1
3.0
2.0
2.5
2.7
4.4
4.4
7.7
4.4
Kurtosis
¨
-0.2
-1.0
-1.8
-0.2
0.7
0.6
0.3
-0.3
0.6
0
0.4
3.3
¨
-9.8
-8.7
-2.2
-6.7
-3.5
-2.8
-3.1
-2.3
-0.9
1.4
1.8
0.9
187
1973
Age
7
8
9
11
12
13
15
16
17
18
!
!
!
!
!
Height
1981
SYDOXH
%R\V
1981
1973
BMI
SYDOXH
1973
Height
1981
SYDOXH
*LUOV
1973
BMI
1981
!
!
SYDOXH
Table 3.+HLJKWDQG%0,IRUWKHWZRFRKRUWVDQGVWUDWL¿HGIRUER\VDQGJLUOV0HDQYDOXHVIRUHDFKDJHEDVHGRQDORQJLWXGLQDO
DQDO\VLVRIUHSHDWHGPHDVXUHVIRUDOOFKLOGUHQ3YDOXHVDGMXVWHGIRUPXOWLSOHFRPSDULVRQVIRUDJHVSHFL¿FGLIIHUHQFHVEHWZHHQFRKRUWVEDVHG
RQDPL[HGPRGHODQDO\VLV
188
Boys
Obesity
(%)
Age
¨
¨
(y)
±
±
4.6
1.2
1.1
12.1
2.6
2.1
10
14.5
2.5
11
4.6
0.5
12
5.5
5.1
0.4
14
11.0
15.6
4.6
1.6
1.2
15
10.6
5.2
16
11.6
4.1
1.5
5.6
1.0
12.6
5.6
2.4
4.4
2.0
1RWH¨ 'LIIHUHQFHEHWZHHQWKHSHUFHQWDJHVIRUDQG
Overweight
(%)
12.0
11.6
11.1
11.5
10.1
12.1
10.5
16.5
16.2
15.0
14.5
15.0
Overweight
%)
¨
±
5.6
6.0
6.4
2.4
Girls
2.0
1.6
1.2
1.6
1.5
2.6
1.6
Obesity
(%)
¨
±
1.4
1.6
1.4
0.6
1.4
0.4
0.1
1.5
Table 4. Overweight and obesity according to limits by IOTF, International Obesity Task Force, Cole et al. (11), applied to the two cohorts,
VWUDWL¿HGIRUER\VDQGJLUOV3HUFHQWDJHRIFKLOGUHQDERYHOLPLWVLQHDFKDJHJURXSDQGWKHGLIIHUHQFHV¨LQSHUFHQWDJHVEHWZHHQWKH
DQGFRKRUWV
Appendix
189
190
191
192
193
194
Tryckverksta’n i Örebro 2007