Sports Med 2006; 36 (12): 1067-1086

REVIEW ARTICLE 0112-1642/06/0012-1067/$39.95/0

 2006 Adis Data Information BV. All rights reserved.

The Physical Activity Patterns of

European Youth with Reference to
Methods of Assessment
Neil Armstrong and Joanne R. Welsman
Children’s Health and Exercise Research Centre, University of Exeter, Exeter, UK

This material is
Contents

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1067
1. Physical Activity Guidelines for Youth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1068

the copyright of the

2. Measurement Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1069
3. Subjective Measures of Physical Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1070
4. Objective Measures of Physical Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1073
4.1 Doubly Labelled Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1073
4.2 Indirect Calorimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1074
4.3 Direct Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1074

original publisher.
4.4 Heart-Rate Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1075
4.5 Pedometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1079
4.6 Accelerometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1081
5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1082

Unauthorised copying
Abstract This article reviews the habitual physical activity of children and adolescents
from member countries of the European Union in relation to methods of assessing
and interpreting physical activity. Data are available from all European Union

and distribution countries except Luxembourg and the trends are very similar. European boys of
all ages participate in more physical activity than European girls and the gender
difference is more marked when vigorous activity is considered. The physical
activity levels of both genders are higher during childhood and decline as young

is prohibited.
people move through their teen years. Physical activity patterns are sporadic and
sustained periods of moderate or vigorous physical activity are seldom achieved
by many European children and adolescents. Expert committees have produced
guidelines for health-related physical activity for youth but they are
evidence-informed rather than evidence-based and where there is evidence of a
relationship between physical activity during youth and health status there is little
evidence of a particular shape of that relationship. The number of children who
experience physical activity of the duration, frequency and intensity recommend-
ed by expert committees decreases with age but accurate estimates of how many
girls and boys are inactive are clouded by methodological problems. If additional
insights into the promotion of health through habitual physical activity during
1068 Armstrong & Welsman

youth are to be made, methods of assessment need to be further refined and

recommended guidelines re-visited in relation to the existing evidence base.

This article reviews the physical activity patterns empirically based guidelines that can be used by
of youth from member countries of the European clinicians in their counselling, as well as by policy
Union (EU). Relevant studies for review were locat- makers with responsibility for youth health promo-
ed through computer searches of MEDLINE, tion.”[6] A systematic review of the scientific paedia-
SportDiscus and personal databases, supplemented tric literature relating physical activity to blood
with an extensive search of bibliographies of ac- lipids,[7] blood pressure,[8] body fatness,[9] skeletal
cessed studies and through personal contacts with health[10] and psychological health[11] was presented
colleagues in other European countries. Where data to the predominantly North American members of
from specific countries were sparse, embassies were the conference and the following guidelines

This material is
requested to provide appropriate contacts within
their home countries.
Physical activity is defined as a complex set of
behaviours that encompass any bodily movement
emerged:
• All adolescents should be physically active daily
or nearly every day, as part of play games, sports,
work, transportation, recreation, physical educa-
produced by skeletal muscles that result in energy

the copyright of the

tion or planned exercise, in the context of family,
expenditure.[1] The habitual physical activity of chil- school and community activities.
dren and adolescents is extremely difficult to assess
and interpret in the context of promoting health.
• Adolescents should engage in three or more ses-
sions per week of activities that last 20 minutes or
More than 30 different methods of assessing physi- more at a time and that require moderate to

original publisher.
cal activity have been identified but the reliability,
objectivity and validity of many of these methods
have not been established with children and adoles-
cents.[2,3] Several expert committees have published
vigorous levels of exertion. Moderate to vigorous
activities were defined as those that require at
least as much effort as brisk or fast walking.[12]
In the first guideline, in contrast with previous
guidelines for young people’s physical activity but

Unauthorised copying
the scientific rationales underlying the recommen-
dations have been challenged.[4] To evaluate the
physical activity of European youth, we therefore
guidelines, the intensity and duration of activity was
not emphasised. The rationale was that daily weight-
bearing activities, of even brief duration, during
adolescence are critical for enhancing bone develop-
need to consider both measurement techniques and
ment that affects skeletal health throughout life, and

and distribution
current physical activity guidelines.
that substantial daily energy expenditure is expected
to reduce risks of obesity, and may also have other
1. Physical Activity Guidelines for Youth
positive health effects. The rationale for the second
The earliest formal physical activity guidelines guideline was that regular participation in continu-

is prohibited.
for youth were provided by the American College of ous moderate to vigorous physical activity during
Sports Medicine (ACSM) who developed an opin- adolescence enhances psychological health, in-
ion statement on the amount of physical activity creases high-density lipoprotein cholesterol and in-
needed for optimal functional capacity and health. creases cardiorespiratory fitness. Although other
The ACSM based their proposals on guidelines for physical activity guidelines were published, mainly
adults and recommended that children and adoles- in North America,[13,14] internationally the ICC
cents should achieve 20–30 minutes of vigorous guidelines informed youth physical activity studies
exercise each day.[5] In 1993, an International Con- throughout the 1990s.
sensus Conference on Physical Activity Guidelines In 1998 in England, the Health Education Au-
for Adolescents (ICC) was convened “to develop thority (UKHEA) commissioned a similar series of

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
Physical Activity of European Youth 1069

reviews of the scientific paediatric literature to those • a relationship where significant health benefits
that informed the ICC.[15] Following a consensus can be gained only in the lower part of the
conference of principally British scientists, different physical activity scale;
recommendations for the physical activity of young • a relationship where significant health benefits
people were proposed, although they were founded can be gained only in the upper part of the
on the same evidence base as the ICC. The primary physical activity scale;
recommendation was that all young people should • an S-shaped curve, which suggests that there is a
participate in physical activity of at least moderate threshold value somewhere on the scale.
intensity for 1 hour per day and that young people He explored the scientific paediatric literature
who currently do little activity should participate in linking physical activity with cardiovascular, skele-
tal and psychological health seeking dose-response
physical activity of at least moderate intensity for at
relationships or threshold values and demonstrated
least half an hour per day. A secondary recommen-
that where there is evidence of a relationship be-
dation was that at least twice a week, some of these

This material is
activities should help to enhance and maintain mus-
cular strength and flexibility and bone health.[15] The
UKHEA Conference’s primary recommendation
tween physical activity during youth and health sta-
tus there is hardly any evidence of a particular
pattern of that relationship. He stated that there are
different patterns of relationships for different
shifted the emphasis from vigorous to moderate health outcomes and that there is only marginal

the copyright of the

intensity physical activity, and from sustained peri-
ods of activity to activity accumulated over a day.
In 2005, a panel of experts from North America
evidence to support the guidelines proposed by ex-
pert committees, which tend to be ‘evidence-in-
formed’ rather than ‘evidence-based’. Twisk[4] con-
systematically evaluated evidence of the effects of cluded that based on scientific evidence, the current

original publisher.
physical activity on several health and behavioural guidelines are as valid as stating that every increase
outcomes in US school-age youth and concluded in physical activity can have some beneficial health
that they should participate daily in ≥1 hour of effects for young people. He suggested that the
advantage of this ‘guideline’ is that the goal is much
moderate to vigorous physical activity that is devel-
easier to achieve and probably leads to the same

Unauthorised copying
opmentally appropriate, enjoyable and involves a
health benefits as satisfying the guidelines proposed
variety of activities.[16] However, at the time of
by expert committees.
writing, no published study of European youth has
adopted this recommendation and the UKHEA rec-
ommendation has been the most influential in the 2. Measurement Issues

and distribution
conclusions drawn from recent studies of young
people’s physical activity.
Twisk[4]critically reviewed the confusion in the
Physical activity assessment techniques must be
socially acceptable, should not burden the child with
field of physical activity guidelines for youth. He cumbersome equipment and should only minimally

is prohibited.
elegantly argued that before guidelines can be devel-
oped, the pattern of the possible relationship(s) be-
tween physical activity and health status must be
influence the person’s normal physical activity pat-
tern. Ideally the intensity, duration, frequency and
mode of activity should be monitored and, if a true
picture of habitual physical activity is required,
considered. He discussed hypothetical dose-re-
some account should be taken of day to day varia-
sponse relationships between physical activity and
tion. Although empirical evidence is not available,
health such as: Bar-Or[17] has recommended a minimum monitoring
• a linear dose-response relationship where every period of 3 days, whereas others[3,18,19] have suggest-
increase in physical activity level will have simi- ed 4–9 days as appropriate depending on the mea-
lar health consequences; surement instrument used.

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1070 Armstrong & Welsman

Habitual physical activity is a behaviour, but it occupational activity, making up a greater propor-
can only occur as a result of skeletal muscle activity tion of total physical activity in children.
that is supported by energy expenditure. Physical
Self-administered questionnaires are less accu-
activity is therefore interrelated with energy expen-
rate than those administered by an interviewer[23]
diture; however, it may cause an elevation in meta-
and large discrepancies have been demonstrated be-
bolic rate that persists long after cessation of observ-
tween the two methods.[24,25] It is difficult to ascer-
able movement. Different techniques may therefore
tain at what age children become able to produce
be measuring different dimensions of physical activ-
meaningful questionnaire data but it has been sug-
ity and this could, at least in part, account for the
gested that children under the age of 12 years cannot
often weak relationships between different measure-
recall activities accurately and are unable to quantify
ment techniques used on the same children and
the time-frame of activity.[26]
adolescents. To interpret adequately the habitual
physical activity of European youth, measurement Physical activity diaries have been reported to be

This material is
issues must be considered and the data evaluated in superior to retrospective questionnaires,[27,28] but
the context of the methodology used. some studies[24,29] have found that the quality of
completed diaries is inconsistent with children. Dia-
ry measures place a heavy burden on the subjects[20]
3. Subjective Measures of and keeping a diary may in itself influence physical

the copyright of the

Physical Activity

Self-report of physical activity is the most widely

used method in epidemiological research due to the
activity habits.[30] Saris[31] concluded that physical
activity diaries were unsuitable for use with children
<15–16 years of age.
Proxy reports are sometimes used to assess the
ease and low costs of implementation. Self-report

original publisher.
methods include retrospective questionnaires, inter-
view-administered recall, activity diaries and mail
surveys. Proxy reports by parents and/or teachers
physical activity of children <10 years of age, who
would have difficulty completing a questionnaire,
and they are normally provided by parents or teach-
ers. The validity of proxy reports is limited especial-
have been employed in studies with children and
ly when they concern activities outside the home or

Unauthorised copying
some studies have estimated the level of physical
activity through self-report of surrogate measures
such as sports participation. Considerable demands
are placed on the child’s cognitive abilities to recall
school.[32,33] For example, Harro[34] observed paren-
tal report of activity to provide values 1.8 times
higher than physical activity estimated from heart-
rate recordings of the same children.
specific events from the past and Sallis[20] has point-

and distribution
ed out actual behaviour is not directly assessed by
self-reports and the data obtained are “memories of
the behaviour of interest that have decayed, been
Several research teams have attempted, on the
basis of self-report data, to classify activities accord-
ing to their estimated energy expenditure using
filtered through perceptions and biases, and have physiological data from other studies and often ex-

is prohibited.
been tainted by competing memories, social desira-
bility and misunderstanding of instructions.”[20]
Children are less time-conscious than adults and
tend to engage in physical activities at sporadic
trapolated from data on adults.[35,36] This approach is
fraught with inaccuracies due to the methodological
limitations of directly determining the energy chil-
dren expend in specific activities. The use of adult
times and intensities rather than consistent energy costs of activities to estimate energy expen-
bouts.[21,22] The self-recall of the intensity, frequen- diture in children introduces substantial errors. In
cy and duration of bouts of activity by children is young children, energy costs may be underestimated
therefore even more problematic than with adults. by up to 40% using adult values. At 10 years of age,
The problem is further confounded by leisure time the underestimation is about 20% decreasing to
activity, which is more difficult to quantify than about 5% at 16 years.[37,38]

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
Physical Activity of European Youth 1071

Self-report and objective measures of physical mark,[68,69] Lithuania,[50,70] Cyprus,[71] Latvia,[50,72]

activity should be compared cautiously. Children Czechoslovakia,[73] the Czech Republic,[74,75] Mal-
have been reported to overestimate their recall of ta,[55] Slovenia,[76,77] Slovakia,[70] Estonia[78,79] and
vigorous physical activity[39] and ‘dramatic discrep- The Netherlands.[29,80] There are no data on young
ancies’ have been reported between self-reported people from Luxembourg. Levels of physical activi-
and objectively measured estimates of vigorous ty cannot be confidently compared across studies,
physical activity.[40] However, when moderate-in- but age- and gender-related trends are consistent and
tensity physical activity is considered, children tend we will use a large WHO multinational survey to
to underestimate their level of physical activity com- illustrate general trends.
pared with objective measures.[41,42] The probable Several multinational WHO surveys of young
explanation for this is that moderate activity tends to people’s physical activity have been carried
be sporadic, non-planned, less memorable and quan- out.[50,51,55,70] The first in 1983/84 involved four EU
tifiable, and therefore less likely to be collected by countries and the latest in 2001/02 included children

This material is
self-report methods.[43] Ekelund et al.,[44] however,
compared the use of heart-rate monitoring and activ-
ity diaries in 30 15-year-old adolescents and report-
ed no significant difference between methods in
and adolescents from 22 EU countries. The surveys
have selected samples representative of national
populations and provide across-country insights into
the physical activity of European youth. A potential
time spent in moderate to vigorous physical activity weakness in national comparisons, however, is that

the copyright of the

(MVPA). They concluded that heart-rate monitoring
and activity diaries are comparable for group assess-
ment of total daily energy expenditure and its com-
the time of year the questionnaires were adminis-
tered varied between countries, although it was quite
tightly defined within countries.
ponents, and for estimating time spent in MVPA. The 2001/02 survey[55] involved 22 EU countries

original publisher.
However, the activity diary underestimated time
spent in MVPA for inactive subjects and conse-
quently overestimated highly active subjects. Kem-
per[45] attempted to validate physical activity mea-
and 115 981 young people aged 11, 13 or 15 years.
The national sample sizes varied from 1980 in Malta
to 14 372 in the UK. The German sample of 5650
participants was regional and selected from Berlin,
sures during the first 4 years of the Amsterdam

Unauthorised copying
Hessen, North Rhine-Westphalia and Saxony. The
Growth Study by relating mean daily heart-rate data were collected from October 2001 to June
measures and mean pedometer counts to the mean 2002. This survey cannot be compared directly with
results of a habitual physical activity interview. the earlier reports as it introduced revised questions
None of the calculated correlation coefficients ex- that focused on physical activity of at least moderate

and distribution
ceeded r = 0.20 and he concluded that the various
methods for assessing physical activity should be
considered to measure different aspects of physical
activity.
intensity, carried out at school and/or in free time
during both the previous week and a typical week.
The participants were provided with a definition of
physical activity as “any activity that increases your
heart rate and makes you get out of breath some of

is prohibited.
The use of self-report techniques is common in
studies of children’s physical activity; however, in
anything but large studies with high statistical pow-
er, the data need to be interpreted cautiously. Studies
the time. Physical activity can be done in sports,
school activities, playing with friends, or walking to
school. Some examples of physical activity are run-
of generally non-representative samples of young ning, brisk walking, rollerblading, biking, dancing,
people are widely available and data have emerged skateboarding, swimming, soccer, basketball, foot-
from Belgium,[46,47] Ireland,[48,49] Austria,[50,51] the ball and surfing.”[55]
UK,[52,53] Italy,[54,55] Spain,[51,56] Finland,[28,57] Hun- The methodology was informed by the UKHEA
gary,[50,51] Greece,[58,59] France,[60,61] Sweden,[62,63] physical activity guidelines[18] and based on the
Poland,[50,51] Germany,[64,65] Portugal,[66,67] Den- screening measure developed by Prochaska et al.[81]

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1072 Armstrong & Welsman

a 11 years
70 13 years
15 years

60

50

40

30

20

10
No. of children (%)

This material is 0

b
70

the copyright of the 60

50

original publisher. 40

30

Unauthorised copying
20

10

and distribution
o ia
th ia

ot a

Po ep
Fi nd

C ngl d
ch d

M d
N W ta
rla s
Sw nds
G den
en e
un rk
La ry
Sp a
Es ain
a
m l
y
Be ly

Fr m
ce
Au nd

er a
Sc ani

he le

an
E an
ze an
Sl str
Li ven

D ec

Po oni

Ita
tv

G rtug
al

H ma
ga

iu
R

an
la

la
la

et a

re

lg
nl
u

t
Ire

Fig. 1. Percentage of (a) boys and (b) girls meeting the primary UK Health Education Authority guideline on physical activity.[55]

is prohibited.
so that the participants were asked to “add up all the
time you spend in physical activity each day.” The
first question asked about physical activity under-
response categories for both questions were 0 days,
1 day, 2 days… 7 days and scores were calculated by
averaging the results of the two items with a score of
taken in the previous week, and the second about a ≥5 classifying the respondent as meeting the recom-
typical week as follows: (i) over the past 7 days, on mendation of >60 minutes of moderate activity a
how many days were you physically active for at day on most days.
least 60 minutes per day?; and (ii) over a typical or The data suggest that about one-third of young
usual week, on how many days are you physically people meet the UKHEA primary recommendation
active for a total of at least 60 minutes per day? The (figure 1). There are, however, wide variations

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
Physical Activity of European Youth 1073

across countries and although across-country com- derived from indirect calorimetry for subjects living
parisons must be made cautiously, in Austria, Fin- in metabolic chambers.[83] It provides a powerful
land, Malta, Wales, Greece, Hungary, Latvia, Esto- tool for the accurate measurement of daily energy
nia, Portugal, Germany, Italy, Belgium and France expenditure but it has several limitations. The iso-
<20% of 15-year-old girls self-reported meeting the topes are expensive and difficult to obtain, accurate
recommendation. Comparisons within countries are dietary records must be kept and the data only
more secure and in all countries more boys than girls provide a measure of total daily energy expenditure
reported being physically active for 60 minutes a (TDEE) over the study period. Studies generally
day at least five times a week, although the gender have very small sample sizes and no information is
differences were small in some countries (e.g. provided on the participants’ physical activity pat-
among 15-year-olds, 1% in The Netherlands and 5% terns, therefore making it impossible to identify
in Italy) but marked in others (e.g. among 15-year- activity of moderate or vigorous intensity.
olds, 25% in Malta and 22% in Wales). In all coun- In DLW studies, physical activity is usually ex-

This material is
tries, except boys in France and the Czech Republic, pressed as the physical activity level (PAL) or the
fewer 15-year-olds than 11-year-olds met the activity-related energy expenditure (AEE). James et
UKHEA criterion and in France and the Czech al.[84] defined PAL as the index TDEE over basal
Republic more 13-year-olds than 15-year-olds met metabolic rate (BMR). TDEE and BMR are used to
the guideline. calculate AEE as (0.9 × TDEE) – BMR, assuming a

the copyright of the

The findings that physical activity declines with
age and that girls are less likely to be physically
active than boys are remarkably consistent across
diet-induced thermogenesis of 10%. Normal levels
of PAL and AEE for children are the subject of
ongoing debate.
European countries. There is no evidence to indicate
Davies[85] reviewed TDEE data on children aged

original publisher.
major differences in the level of physical activity of
1–6 years and concluded that there was a consistent
youth living in Europe compared with those living
pattern across the three laboratories, which had gen-
in North America.[50,55]
erated the world database on 225 children. His re-
view revealed that TDEE was, on average, 20%
4. Objective Measures of
below current international recommendations. He

Unauthorised copying
Physical Activity

4.1 Doubly Labelled Water

postulated that this was probably due to changes in
levels of habitual physical activity.
Hoos et al.[86] systematically reviewed the litera-
Doubly labelled water (DLW) is recognised as ture and combined data from 17 studies, including

and distribution
the reference method or ‘gold standard’ for the
assessment of energy expenditure in free-living sub-
jects. This technique assesses energy expenditure by
studies of European children,[87,88] with the aim of
constructing a reference line for PAL and AEE as a
function of age and gender. The children were aged
from 3 to 16 years and whereas there was no gender
estimating carbon dioxide production using isotope
difference in PAL values, boys had higher AEE

is prohibited.
dilution during a minimum of 3 days.[82] However, it
should be noted that energy expenditure is a physio-
logical consequence of physical activity and the two
are distinct constructs, which limits attempts to use
values than girls. Low values were found in young
children increasing to adult values with age. Hoos et
al.[89] explained this as a function of adults’ higher
DLW to validate measures of physical activity. For body mass and less time spent sleeping. They sug-
example, physical activity may cause an elevation in gested that as children grow older, they sleep less
metabolic rate that persists long after cessation of and spend more time on physical activities, resulting
observable movement. in higher PAL and AEE values.
The DLW technique has low reactivity and it DLW has been used quite extensively to validate
appears to be accurate to within 5% relative to data other methods of estimating energy expenditure.

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1074 Armstrong & Welsman

Accelerometers have been reported to provide sig- (9.16MJ) at 8 years of age to 3901 calories
nificant relationships with TDEE, PAL and AEE in (16.31MJ) at 17 years of age. Girls’ ‘heat produc-
6-[89] and 9-year-old children,[88] although O’Connor tion’ increased from 2223 calories (9.29MJ) at 9
et al.[90] found no significant correlation between years of age to 3214 calories (13.43MJ) at 17 years
DLW measures and either accelerometer or activity of age. Although difficult to interpret, her results
diary estimates of energy expenditure. Emons et indicated that boys had a higher ‘heat production’ at
al.[91] showed heart-rate monitoring to overestimate all ages studied.
DLW determined TDEE by about 12% in 8- to Recent technology has allowed the introduction
9-year-old children. Livingstone et al.[92] described a of portable, lightweight metabolic systems; howev-
FLEX HR technique, which uses an individually er, despite this advance, the equipment is still too
determined heart rate measured in conjunction with cumbersome to use indirect calorimetry with chil-
oxygen uptake (V̇O2), that can be used to distin- dren under free-living conditions.
guish between resting and AEE. Resting metabolic

This material is
rate is substituted for periods when the heart rate 4.3 Direct Observation
falls below the FLEX HR. Using this technique,
they accurately predicted group DLW estimates of The assessment of physical activity through the
TDEE in 7- to 15-year-olds. use of direct observation has inherent appeal and
recent technological advances permitting complex

the copyright of the

4.2 Indirect Calorimetry

Indirect calorimetry has been used to validate

heart-rate monitoring, pedometry and accelerome-
observational codes to be entered, stored and
analysed by microcomputers have stimulated re-
search into children’s activity patterns using direct
observation techniques. Sirard and Pate[2] critically
try. Emons et al.[91] compared 24-hour energy ex- reviewed seven direct observation systems and con-

original publisher.
penditure determined during a 1-day stay in a calo-
rimeter with energy expenditure estimated from
continuous heart-rate monitoring and demonstrated
cluded that all seven techniques attained satisfactory
inter-observer agreement (84–99%) among simulta-
neous observations of the same child.
that heart-rate monitoring overestimated energy ex- Direct observation is, however, labour-intensive,

Unauthorised copying
penditure by about 10%. Eston and his col-
leagues[93,94] investigated energy expenditure during
a series of activities with 8- to 10-year-old children
using indirect calorimetry, heart-rate monitoring,
time consuming and therefore costly. Events studied
must be observable and codeable and observers or
video cameras need to be in the same environment
as the subject. The extent to which even well trained
accelerometry and pedometry. Across all activities, observers affect subject behaviour (subject reactivi-

and distribution
both heart rate and accelerometry counts were
strongly correlated with V̇O2 and the results also
suggested that pedometry offers potential for large
ty) is problematic. Puhl et al.[96] reported that 16.6%
of the children in their study reacted to the observ-
ers. Direct observation can capture valuable short-
studies of young people’s physical activity. term patterns and sudden changes in children’s

is prohibited.
Historically, Bedale’s[95] analysis of ‘heat pro-
duction’ during selected activities was probably the
first attempt to classify the daily energy expenditure
of European children. Her work was limited by the
physical activity,[22] but it is normally impossible to
follow a child for a full day[19,97] and few long-term
direct observation studies have been undertaken
with children.
available technology as she used Douglas bags to The series of studies by Sleap and
collect expired air, which she subsequently analysed Warburton[98-100] of English children appear to be
for oxygen and carbon dioxide content using a the most comprehensive direct observation studies
Haldane apparatus. Bedale[95] reported her results in of the physical activity of European youth. They
terms of ‘total heat production in 24 hours’ and observed 93 girls and 86 boys aged 5–11 years, on
concluded that boys’ values rose from 2191 calories separate occasions, during school break times, lunch

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
Physical Activity of European Youth 1075

times and physical education lessons. Further obser- of storing minute by minute data for several days
vations were undertaken on one weekday evening and therefore allow long duration monitoring of
and one 4-hour period on either a weekend or during physical activity. However, the activity patterns of
a school vacation. In total, each child was observed young children are characterised by very short
for an average of 418 minutes. During this time, bursts of MVPA and therefore shorter recording
children were engaged in MVPA for 29.3% of total intervals may be required in some circumstances;
time observed. Twenty one percent of children re- this will reduce the overall duration of monitoring.
corded at least one sustained 20-minute period of Telemetry systems are used widely, but they are not
MVPA. Ninety five percent of children engaged in without problems and heart rate traces often need
at least one 5-minute period of MVPA and no signif- individual editing because of electrical interference
icant differences in MVPA were observed between causing spuriously high or low datapoints.[18,101]
boys and girls. Sleap and Warburton[100] concluded The interpretation of heart-rate data is complex
that their results were disturbing since preadolescent and Harro and Riddoch[18] have identified 24 differ-

This material is
children appeared to be engaging in very little sus- ent methods of data reduction. This makes direct
tained, playful physical activity during their free comparisons between studies very difficult. Gener-
time outside of school. ally, heart rate has been used to either estimate
energy expenditure or to provide ‘thresholds’ equat-
4.4 Heart-Rate Monitoring ing to moderate and/or vigorous physical activity.

the copyright of the

Heart-rate monitoring provides an objective esti-
mate of physical activity that is based on the as-
sumption that children who spend longer periods of
The direct prediction of children’s daily energy
expenditure from heart rate[29,103] is fraught with
problems[101] and more recent studies have tended to
time with elevated heart rates are generally more rely on the FLEX HR technique.[92] Numerous heart-

original publisher.
active than those whose heart rate remains in the
lower ranges. Heart-rate monitoring is accepted as a
reliable and valid measure of physical activity over
rate indices or thresholds can be found in the litera-
ture. Some investigators[31,104] have suggested the
use of a heart-rate threshold corresponding to 50%
extended time periods but it is an indirect measure of each individual’s peak V̇O2; however, as this

Unauthorised copying
that indicates the relative stress placed on the cardio- requires individual calibration of each subject, it is a
respiratory system.[101,102] Several other factors be- burdensome technique in large studies. Other re-
side physical activity can influence heart rate, par- searchers have defined moderate and vigorous activ-
ticularly during low-intensity physical activity. Ex- ity as brisk walking and jogging, respectively, and
amples include anxiety, emotional stress, fatigue, directly determined the heart rate corresponding to

and distribution
body position, active muscle group, type of muscle
contraction, training status, level of fitness, food
intake, state of hydration, ambient temperature and
these activities to use as a threshold.[101,102] Several
studies have used a threshold of 140 beats/min to
define moderate physical activity.[18,101]
humidity.[101] Despite the problems with its assessment and

is prohibited.
A number of self-contained, computerised telem-
etry systems have been developed for the unobtru-
sive monitoring of heart rate. Typically, these sys-
tems consist of a lightweight transmitter that is fixed
interpretation, heart-rate monitoring has provided
valuable insights into young people’s physical activ-
ity patterns. Bradfield et al.[105] appear to have been
the first to continuously monitor the heart rates of
to the chest with electrodes or a belt, and a receiver/ European boys and, following calibration of heart
microcomputer that is worn as a watch on the wrist. rates against V̇O2, they reported the mean energy
They are socially acceptable, they permit freedom of expenditure of 54 7- to 10-year-old boys as 9.2 kJ/
movement, they are not immediately noticeable, and min during lunchtime and play periods. Seliger et
therefore should not unduly influence the child’s al.[73] monitored the heart rates of 11 12-year-old
normal activity pattern. Some systems are capable Czech boys for 24 hours and reported that heart rates

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1076 Armstrong & Welsman

>150 beats/min were rare and only fleetingly en- ical activity (equivalent to brisk walking) as generat-
countered. They noted that “the daily activity heart ing a heart rate ≥140 beats/min and vigorous physi-
rate response implied that very little circulation re- cal activity (equivalent to jogging) as generating a
sponse was required to support the daily activity”, heart rate ≥160 beats/min.
but they did not describe their method of monitoring For the purpose of this review, these data have
in any detail. been merged and re-analysed in three categories: (i)
Saris[29] developed an eight-level heart-rate inte- first school (mean age 7.2 years; 93 girls, 95 boys);
grator that stored heart rates within appropriate (ii) middle school (mean age 11.0 years; 263 girls,
ranges over periods of 24 hours. He subsequently 271 boys); and (iii) high school (mean age 13.3
used the instrument in a bi-annual analysis of 217 years; 265 girls, 220 boys). Specific details of sam-
boys and 189 girls from The Netherlands over a pling procedures are available in the original papers
6-year period beginning when the children were 6 but the participants were generally representative of
years old. Saris[29] reported that, at all ages, boys had the South of England. The boys in each category

This material is
a higher total energy expenditure and spent more
time above an exercise intensity equivalent to 50%
of peak V̇O2, than girls.
spent a significantly greater percentage of time with
their heart rate ≥140 and 160 beats/min than the
corresponding girls.
Verschuur and Kemper[103,106,107] used Saris’ In total, 305 of the high-school children and 114

the copyright of the

heart-rate integrator to annually monitor Dutch chil- of the middle-school children also had their heart
dren for 48 hours over a 4-year period, beginning at rates monitored on a Saturday. The high-school girls
age 12–13 years. The daily energy expenditure of spent significantly less time above the 140 beats/min
girls in relation to body mass was found to decrease threshold than high school boys. No significant dif-
from 12–13 to 17–18 years of age. Boys’ daily ferences were detected between middle-school boys

original publisher.
energy expenditure in relation to body mass was
almost constant from age 12–13 to 14–15 years and
then decreased to 17–18 years of age. Boys demon-
strated significantly higher energy expenditures than
and girls.
To provide a more comprehensive picture of
young people’s physical activity patterns, the num-
ber of 5-, 10- and 20-minute periods with heart rate
girls at ages 12–13 and 14–15 years.

Unauthorised copying
By the end of the 1980s, heart-rate monitoring
technology had advanced to the point where unob-
trusive telemetry systems allowed monitoring over
sustained ≥140 and 160 beats/min was determined.
Five-minute periods of moderate physical activity
were common, particularly among the younger chil-
dren, where 93% of the boys and 78% of the girls
several days. Studies of European children and ado- achieved at least one 5-minute period of moderate

and distribution
lescents that have employed heart-rate monitoring
over at least a 3-day period and from which relevant
data can be extracted have been summarised and
collated in table I.
physical activity per day. Eighty two percent of
older (high- plus middle-school) boys and 63% of
older girls achieved a daily 5-minute period of mod-
erate physical activity, but 34% of older boys and

is prohibited.
Over a 10-year period and in a series of studies
Armstrong and his associates[21,102,108-110,113,115-117]
monitored the heart rates of 1227 young people aged
5–16 years from South West England over three
47% of older girls did not experience a single
10-minute period of moderate physical activity over
3 days of monitoring. Ten-minute periods of moder-
ate physical activity were more common among the
schooldays. To interpret their data, they exercised first school children, but 11% of boys and 31% of
98 young people, aged 5–16 years, at various speeds girls failed to raise their heart rate ≥140 beats/min
on a horizontal treadmill and noted that, regardless for a sustained 10-minute period. Twenty-minute
of age, brisk walking and jogging elicited steady- periods of either moderate or vigorous physical ac-
state heart rates of about 140 and 160 beats/min, tivity were sparse in all age groups, illustrating that
respectively. They therefore defined moderate phys- sustained periods of physical activity do not

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
 Unauthorised copying
 2006 Adis Data Information BV. All rights reserved.

Physical Activity of European Youth

Table I. Habitual physical activity of European youth measured over at least 3 days using heart rate monitoring

Study Participants Physical activity Outcomes

the copyright of the

measure

original publisher.
Armstrong et al.[102] 163 girls, 103 boys; aged 4d HR monitoring Boys had HR >139 bpm for significantly greater percentage of time than girls
11–16y; England (6.2% vs 4.3%) during 3 weekdays and a Saturday, boys (n = 92) 5.6% vs girls
(n = 120) 2.6%. A significant negative correlation with age was found with girls

and distribution
(weekday r = –0.27, Saturday r = –0.21), but not with boys. 51.5% of girls and

This material is
35.9% of boys did not exhibit a single 10 min sustained period with HR >139
bpm over 3 weekdays. On Saturday the corresponding figures were 70.7% of

Armstrong and Bray[108]

is prohibited.
11 girls, 13 boys; aged 10–11y; 6d HR monitoring
boys and 93.3% of girls
Children spent 8.8% of time with HR >139 bpm and 3.6% of time with HR >159
England bpm. 33.4% of children did not exhibit a single 10 min sustained period with HR
>139 bpm and 62.5% did not exhibit a single 10 min sustained period with HR
>159 bpm. No significant differences were detected between boys and girls
Armstrong and Bray[21] 65 girls, 67 boys; aged 10.7y; 4d HR monitoring No significant gender differences in time spent with HR >139 bpm during 3
England weekdays (boys 9.4% vs girls 8.2%) or a Saturday (16 boys [5.2%] vs 23 girls
[6.0%]). Boys spent significantly more time with HR >159 bpm during weekdays
(4.5% vs 3.5%) but not during Saturdays (1.8% vs 1.8%). 25% of girls and 19%
of boys did not exhibit a single 10 min sustained period with HR >139 bpm
during weekdays. On Saturday the corresponding figures were 75% of boys
and 65% of girls
Biddle et al.[109] 20 girls and 26 boys; aged 3d HR monitoring No significant gender differences in time spent with HR >139 bpm during 3
12.2y; England weekdays, but boys spent significantly more time with their HR >159 bpm
(2.7% vs 1.9%). 45.9% of boys and 34.3% girls did not sustain a single 10 min
period with HR >139 bpm. The corresponding figures for 10 min periods with
HR >159 bpm were 64.9% of boys and 91.4% of girls
Riddoch et al.[104] 22 girls and 23 boys; aged 2–4d HR monitoring No significant gender difference in daily time spent with HR >50% V̇O2peak
11–16y; Northern Ireland (boys, 24 min; girls, 17 min). Boys spent significantly more time than girls with
HR >70% V̇O2peak (8 vs 4 min). Younger boys engaged in significantly more
total activity than older boys and in more vigorous activity than younger girls. A
significant negative correlation (r = –0.48) was found between age and total
activity in boys but not in girls
McManus and 100 girls, 100 boys; aged 3d HR monitoring Boys spent significantly more time than girls with HR >139 bpm (9.1% vs 7.3%)
Armstrong[110] 11.1y; England and HR >159 bpm (4.3% vs 3.0%). 22% of boys and 30% of girls did not
sustain a single 10 min period with HR >139 bpm and 49% of boys and 57% of
girls did not sustain a single 10 min period with HR >159 bpm
Sports Med 2006; 36 (12)

Falgairette et al.[111] 34 girls, 30 boys; aged 6–11y; 7d HR monitoring Daily HR and percentage of HR reserve varied little with age, although there
France was a tendency for a decrease in girls. PA was higher during weekdays than
weekend days

Continued next page

1077
 Unauthorised copying
 2006 Adis Data Information BV. All rights reserved.

1078
Table I. Contd
Study Participants Physical activity Outcomes

the copyright of the

measure
Sallo and Silla[112] 29 girls, 25 boys; aged 7y; 4d HR monitoring The children spent 7.9% of monitored time with HR >139 bpm. No gender

original publisher.
Estonia differences were detected. 55% of girls and 64% of boys did not exhibit a 10
min sustained period with HR >139 bpm

and distribution

This material is
Welsman and 31 girls, 26 boys; aged 6–9y; 3d HR monitoring Boys spent significantly more time than girls with HR >139 bpm (11.1% vs
Armstrong[113] England 8.7%) and HR >159 bpm (4.7% vs 3.3%). 23% of boys and 31% of girls did not
is prohibited. exhibit a 10 min sustained period with HR >139 bpm, corresponding figures for
HR >159 bpm were 69% of boys and 83% of girls. Time with HR >139 and 159
bpm declined significantly with age in both boys and girls

Gavarry et al.[114] 38 girls, 28 boys; aged 11–16y; 7d HR monitoring 32% of the children were classified as active according to ACSM[5] guidelines.
France No gender differences were noted, but PA decreased with age

Maniôs et al.[59] 22 girls, 17 boys; aged 6y; 3d HR monitoring Children from urban areas had significantly longer activity intervals than those
Greece from rural areas. The time spent in activity intervals was significantly longer
during weekdays than weekend days

Welsman and 22 girls, 17 boys; aged 5–7y; 3d HR monitoring Boys spent significantly more time than girls with HR >139 bpm (13.6% vs
Armstrong[115] England 8.5%) and HR >159 bpm (6.1% vs 2.9%). 32% of girls did not exhibit a 10 min
sustained period with HR >139 bpm, although all boys did so. Corresponding
figures for HR >159 bpm were 77% of girls and 53% of boys

Armstrong et al.[116] 98 girls, 104 boys; aged 3d HR monitoring In the first year of the study, 76.2% of boys and 73.5% of girls accumulated 30
11–13y; England min with HR >139 bpm. In the final year the figures were 44.4% and 31.6% for
boys and girls, respectively. At age 11y, 22.1% of boys and 29.6% of girls did
not experience a single 10 min sustained period with HR >139 bpm at age 13y
the figures were 28.4% and 51.9% for boys and girls, respectively. PA declined
with age with a consistent gender difference reflecting the lower PA of girls

Welsman and 40 girls, 42 boys; aged 14–15y; 4d HR monitoring 67.3% of the boys and 59.5% of the girls accumulated a daily 30 min with HR
Armstrong[117] England >139 bpm. 28.8% of boys and 54.8% of girls did not experience a single 10
min sustained period with HR >139 bpm

Ekelund et al.[118] 40 girls, 42 boys; aged 14–15y; 3d HR monitoring No gender differences were observed in time spent engaged in MVPA, but
Sweden boys’ activity energy expenditure (total-sedentary energy expenditure) was
significantly higher than girls

Sleap and Tolfrey[42] 36 girls, 43 boys; aged 10.6y; 4d HR monitoring Girls accumulated 105 min with their HR >120 bpm and 38 min >140 bpm.
England Corresponding data for boys were 118 min and 50 min. Boys spent significantly

Armstrong & Welsman

more time with HR >140 bpm than girls
Sports Med 2006; 36 (12)

Vermorel et al.[61] 29 girls, 31 boys; aged 12–16y; 5d HR monitoring Mean DEE increased significantly with age in boys, but not in girls. PA level did
France not vary with gender and age. Mean DEE was 21% higher on non-schooldays
in the active children but 7% lower in the sedentary children

ACSM = American College of Sports Medicine; bpm = beats/min; DEE = daily energy expenditure; HR = heart rate; MVPA = moderate to vigorous physical activity; PA = physical
activity; V̇O2peak = peak oxygen uptake.
Physical Activity of European Youth 1079

characterise young people’s physical activity pat- Pedometers are relatively simple motion sensors
terns. Less than 3% of boys were reported to experi- that record the acceleration and deceleration of
ence the equivalent of a daily 20-minute period of movement in one direction[120] and are normally
vigorous physical activity. Of the 641 girls moni- used to estimate mileage walked or the number of
tored for 3 weekdays and the 195 girls monitored on steps taken over a period of time. Six days of pe-
a Saturday, not a single girl experienced the dometer data have been reported as adequately relia-
equivalent of a daily 20-minute period of vigorous ble for research into children’s physical activity with
physical activity. the inclusion of both weekday and weekend scores
In what appears to be the only longitudinal study recommended.[121] Disadvantages of pedometry in-
of 3-day heart-rate monitoring, Armstrong et al.[116] clude the inability to measure intensity or pattern of
assessed annually the physical activity patterns of 98 activities performed, record counts during cycling or
girls and 104 boys, aged 11 years at the onset of the increases in energy expenditure due to carrying ob-
study. Sustained 10- or 20-minute periods of moder- jects or walking/running uphill. Ankle-mounted pe-

This material is
ate or vigorous physical activity were not character-
istic of the physical activity patterns and both mea-
sures declined with age with a consistent gender
difference reflecting the lower physical activity
dometers overcome the problem of recording during
cycling, but ankle pedometers are less reliable than
waist-mounted pedometers[120] and most recent
pedometry studies have used waist-mounted instru-
levels of girls. Using multilevel modelling, they ments.

the copyright of the

demonstrated age and gender differences in time
spent above 139 and 159 beats/min. Seventy six
percent of 11-year-old children accumulated a daily
There are no well accepted guidelines for young
people for the number of steps that equate to 1 hour
of moderate physical activity per day and recom-
30 minutes with their heart rate >139 beats/min, but mendations for adults (e.g. 10 000 steps per day) are

original publisher.
by age 13 years 44% of the children met this target. likely to be too low for children. Recommendations
The decline in the percentage of girls being particu- for appropriate step counts during childhood and
larly marked (from 74% to 32%). adolescence should take into account factors such as
Most, but not all (see table I) heart-rate monitor- age, stature and stride length. Predictions and pro-
ing studies have reported boys to be more physically motion of daily moderate physical activity engage-

Unauthorised copying
active than girls. Boys appear to engage in moderate
and vigorous, sustained periods of physical activity
more often than girls, but this type of activity does
not appear to be characteristic of European youth’s
ment during youth based on pedometer counts per
day should therefore be made with caution.[122]
Nevertheless, Rowlands et al.[123] provide a de-
tailed examination of the use of pedometry and
physical activity patterns. Studies consistently show

and distribution
a decline in physical activity with age, at least
during the teen years. Children tend to be more
active during schooldays than at weekends.
suggest that as pedometers are inexpensive, re-usa-
ble, objective and non-reactive they are useful tools
for large scale studies and well suited to the assess-
ment of children’s physical activity.

is prohibited.
4.5 Pedometry

As almost all forms of physical activity require

movement of the trunk or limbs, the assessment of
Pedometry studies of European youth that have
assessed physical activity over a minimum of 3 days
have been collated in table II. The measurements are
less precise than with other objective measures of
‘movement’ is appealing. The most common physical activity, but the large sample sizes
mechanical device for measuring movement is the strengthen the results. Findings are very consistent
pedometer, which was first used in this context by across the countries surveyed and show that boys are
Lauter,[119] although Leonardo da Vinci designed a more active than girls at all ages from 7 to 18 years.
pedometer to measure distance by counting steps Two recent large Swedish studies[124,125] across the
somewhat earlier.[33] age range 7–14 years indicate that girls’ activity

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
 Unauthorised copying
 2006 Adis Data Information BV. All rights reserved.

1080
Table II. Habitual physical activity of European youth measured over at least 3 days using pedometry

Study Participants Physical activity measure Outcomes

the copyright of the

Verschuur et al.[126] 215 girls, 195 boys; 3d pedometer monitoring Boys spent significantly more time on ‘heavy activities’ than girls. Boys’ median

original publisher.
aged 13–14y; The 99 min, girls’ median 1 min

and distribution
Netherlands

This material is
Skalik et al.[125] 58 girls, 68 boys; aged 7d accelerometer and Mean DEE in PA was 9.7 kcal/kg/d in girls and 11.2 kcal/kg/d in boys. In all
is prohibited.
16–18y; Czech pedometer monitoring regions boys had a higher level of PA. 17% of girls failed to reach 6 kcal/kg/d
Republic, Poland and
Sweden

Vincent et al.[127] 324 girls, 356 boys; 4d pedometer monitoring The 12-year-old boys in the most active tertile accumulated more steps than
aged 7–12y; Sweden the 7-year-old boys. In the least active tertile the younger boys accumulated
more steps than the older boys. In both the most active and the least active
tertiles the younger girls accumulated more steps than the older girls

Loucaides et al.[71,128] 127 girls, 129 boys; 4d pedometer monitoring in Urban schoolchildren were significantly more active in winter than rural
aged 11–12y; Cyprus winter schoolchildren and rural schoolchildren were more active in the summer. Boys
4d pedometer monitoring in acquired significantly higher mean daily step counts than girls in winter
summer

Raustorp et al.[124] 435 girls, 457 boys; 4d pedometer monitoring Boys were significantly more active than girls across all age groups. For both
aged 7–14y; Sweden boys and girls the most active age groups were 10-year-olds. With the
exception of 10-year-olds, activity levels were quite stable among children aged
7–11y but there tended to be a drop-off in step counts among adolescents in
the 13–14y age groups

Cardon and De 41 girls, 51 boys; aged 6d pedometer monitoring Mean daily step counts were significantly higher in boys than in girls (16 628 ±

Armstrong & Welsman

3527 steps vs 13 002 ± 3644 steps). Daily step counts during the week did not
Sports Med 2006; 36 (12)

[122]
Bourdeaudhuij 6–13y; Belgium
differ significantly from the weekend step count. Step counts were significantly
higher in younger children regardless of gender

DEE = daily energy expenditure; PA = physical activity.

Physical Activity of European Youth 1081

declines with age, although the decline in boys’ approaches used to calibrate the devices against
physical activity with age may not commence until directly measured energy expenditure. They report-
the early teens. ed that the cut points (counts per minute) used to
define the boundaries of young people’s moderate (3
4.6 Accelerometry metabolic equivalents [METs]) and vigorous (6
METs) physical activity varied substantially for a
Accelerometers are more sophisticated motion given accelerometer, even though most studies had
sensors than pedometers and they record movement included similar activities in the protocol. They ac-
through piezo-electric transducers and microproces- knowledged that calibration should be fit for pur-
sors that convert recorded accelerations to a quanti-
pose with, for example, studies using accelerome-
fiable digital signal referred to as ‘counts’. Acceler-
ters to examine the impact of activity on obesity
ometers are normally attached to a belt at the hip and
relating counts to energy expenditure. However,
relative to heart-rate monitors they are less burden-
they suggested that for studies where the purpose is

This material is
some to children. Accelerometers provide an objec-
to characterise or track physical activity behaviour,
tive, non-reactive and re-usable technique with
which to assess habitual physical activity but they the most appropriate accelerometer measure may be
are insensitive to cycling, locomotion on a gradient the raw counts.
or other activities with limited torso movement.[2,3] Relatively few accelerometry studies of the phys-

the copyright of the

Early accelerometers were uniaxial, but more
recent models measure movement in three planes
(triaxial). Trost[3] critically reviewed uniaxial and
triaxial models and stated that they “provide compa-
ical activity of European youth have been published,
but those located that have monitored activity over
at least 3 days are collated in table III.
The most substantial study of European chil-
rable assessments of free-living physical activity in

original publisher.
children.” He noted that triaxial models tend to be
bigger than uniaxial models and that this is impor-
tant in field-based studies where children are more
dren’s physical activity using accelerometry was
carried out as part of the European Youth Heart
Study.[43] Subjects aged 9 and 15 years were recruit-
ed to broadly represent children either side of puber-
likely to wear a device that is unobtrusive. They can, ty. Data were collected from well defined popula-

Unauthorised copying
however, be worn underneath clothes when necessa-
ry and do not prevent participation in common activ-
ities and sports. Trost concluded that the develop-
ment of a compact and highly reliable three-dimen-
tions from four countries and 2185 children had
their physical activity assessed over either 4 days
(70%) or 3 days (30%).
sional accelerometer remains a research priority. The results confirmed significant gender differ-

and distribution
There are a number of commercial accelerome-
ters on the market and Sirard and Pate[2] have tabu-
lated available studies of young people’s physical
ences in physical activity with 9-year-old boys being
21% more active than girls and 15-year-old boys
being 26% more active than similarly aged girls.
Gender differences in time spent in activity of at
activity where accelerometry has been compared

is prohibited.
with energy expenditure (determined by DLW or least moderate intensity were even more marked
indirect calorimetry) and direct observation of activ- (20% and 36% difference, respectively). Similarly,
ity. Correlation coefficients across the various mod- 9-year-old children of both genders were considera-
els ranged from r = –0.09 to r = 0.95. bly more active than 15-year-olds (27% more active
The calibration of accelerometer output to esti- in boys, 32% in girls). The age difference was even
mate the physical activity of children and adoles- more marked with time spent in moderate activity
cents is, however, a challenge. Freedson et al.[129] (94% more active in boys, 129% more active in
rigorously reviewed calibration studies of the four girls). Ninety seven percent of 9-year-old children
accelerometers used most frequently to assess phys- accumulated 60 minutes of moderate activity as
ical activity in this population and critiqued the recommended by the UKHEA,[15] but fewer boys

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1082 Armstrong & Welsman

boys and 97.6% of girls accumulated 60 min of moderate intensity activity. At age 15y,
(82%) and, particularly, girls (62%) achieved this at

gender differences in favour of boys were apparent at age 9 and 15y. At 9y, 97.4% of
days of the week (weekdays or weekend days). Short-term stability of PA during late

Mean DEE in PA was 9.7 kcal/kg/d in girls and 11.2 kcal/kg/d in boys. In all regions,
Between-day reproducibility of accelerometer monitoring was higher between similar

Boys were more active than girls at age 9y and at age 15y. 9-year-olds were more
age 15 years.

accumulated at least 60 min of MVPA per day. The time spent in MVPA increased
Boys engaged in significantly more MVPA per day. 94% of boys and 79% of girls

active than 15-year-olds. With respect to time spent in moderate-intensity activity,

Riddoch et al.[43] commented on the remarkable
consistency of the results across Denmark, Portugal,
Estonia and Norway. Physical activity levels and
boys had a higher level of PA. 17% of girls failed to reach 6 kcal/kg/d

Children accumulated >60 min of at least moderate activity every day

age and gender differences were mirrored across the

four countries despite the wide differences in geog-
raphy, socio-economic circumstances, culture and
climate. The authors suggested that physical activity
habits in children may be determined by biological
In general, girls did less physical activity than boys

factors as much as by environmental factors, a con-

cept supported by others.[136-138]
across age groups in both boys and girls

81.9% of boys and 62.0% of girls did so

Other accelerometry studies of the physical activ-

This material is ity of European youth provide further support for

gender differences;[132,135] however, Santos et al.[133]
observed that the time spent in MVPA by Portu-
guese children increased with age in direct contrast
childhood was moderate

with the findings of the European Heart Study[43]

the copyright of the

Table III. Habitual physical activity of European youth measured over at least 3 days using accelerometry

whose study contained a large Portuguese sample

DEE = daily energy expenditure; MVPA = moderate to vigorous physical activity; PA = physical activity.

covering the same age range.

Outcomes

5. Conclusions

original publisher. The habitual physical activity of children and

7d accelerometer and
pedometer monitoring

adolescents is difficult to assess and interpret in the

4d accelerometer

4d accelerometer

3d accelerometer

5d accelerometer

4d accelerometer

context of promoting health. Several methods of

Physical activity

Unauthorised copying assessing physical activity are available but all have
monitoring

monitoring

monitoring

monitoring

monitoring

weaknesses. Nevertheless, despite the lack of preci-

measure

sion, self-report surveys of large representative sam-

ples have provided informative overviews of young
people’s level of activity and more objective meth-

and distribution
aged 7–10.5y, England
23 girls, 19 boys; aged

58 girls, 68 boys; aged

93 girls, 64 boys; aged

ods have allowed insights in the patterns of physical

2185 children; aged 9
6 girls, 10 boys; aged

Portugal, Estonia and

Republic, Poland and

and 15y, Denmark,

activity during the progression through childhood

95 girls, 120 boys;
8–16y; Portugal

and adolescence. Data are available from all Euro-

16–18y; Czech
8–9y; Estonia

7.5y; Sweden
Participants

pean Countries except Luxembourg, although ob-

is prohibited.
Sweden

jective studies have been mainly centred in specific

Norway

countries, principally Estonia, Sweden, France, Por-

tugal, The Netherlands and the UK. Trends in differ-
ent countries are very similar and European boys
Raudsepp and Päll[130]

appear to participate in more physical activity than

European girls, with the gender difference being
Guerra et al.,[132]
Santos et al.,[133]

al.[43]
Nilsson et al.[131]

Mallam et al.[135]
Skalik et al.[125]

Mota et al.[134]

more marked in comparisons of vigorous physical

Riddoch et

activity. The physical activity levels of both genders

decline as they move through adolescence. Physical
Study

activity patterns are sporadic and sustained periods

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
Physical Activity of European Youth 1083

of either moderate or vigorous physical activity are 6. Sallis JF, Patrick K, Long BJ. Overview of the international
consensus conference on physical activity guidelines for ado-
not part of the lifestyle of most European children. lescents. Pediatr Exerc Sci 1994; 6: 299-302
It is well documented that the number of young 7. Armstrong N, Simons-Morton B. Physical activity and blood
lipids in adolescents. Pediatr Exerc Sci 1994; 6: 381-405
people who experience physical activity of the type, 8. Alpert BS, Wilmore JH. Physical activity and blood pressure in
duration, frequency and intensity recommended by adolescents. Pediatr Exerc Sci 1994; 6: 361-80
expert committees decreases with age, but accurate 9. Bar-Or O, Baranowski T. Physical activity, adiposity, and obesi-
ty among adolescents. Pediatr Exerc Sci 1994; 6: 348-60
estimates of how many girls and boys can be classi- 10. Bailey DA, Martin AD. Physical activity and skeletal health in
fied as inactive according to current recommenda- adolescents. Pediatr Exerc Sci 1994; 6: 330-47
tions are clouded by methodological inconsisten- 11. Calfas KJ, Taylor WC. Effects of physical activity on psycho-
logical variables in adolescents. Pediatr Exerc Sci 1994; 6:
cies. For example, arguably the two most important 406-23
recent contributions to the European youth physical 12. Sallis JF, Patrick K. Physical activity guidelines for adolescents:
activity literature are the studies by Riddoch et al.[43] a consensus statement. Pediatr Exerc Sci 1994; 6: 302-14
13. Cale L, Harris J. Exercise recommendations for young people:
and Currie et al.[55] Both studies were rigorously an update. Health Educ J 2001; 101: 126-38

This material is
designed and executed with large well defined sam-
ples. Data were collected from the same four Euro-
pean countries, Estonia, Portugal, Norway and Den-
mark, at the same time. Riddoch et al.,[43] using
14. Pate RR, Trost SG, Williams CA. Critique of existing guidelines
for physical activity in young people. In: Biddle S, Sallis J,
Cavill N. editors. Young and active? London: Health Educa-
tion Authority, 1998: 162-76
15. Biddle S, Sallis J, Cavill N. Young and active? London: Health
accelerometry, reported 81.9% of 15-year-old boys Education Authority, 1998

the copyright of the

and 62.0% of 15-year-old girls satisfied the guide-
lines of the UKHEA expert committee,[15] whereas
Currie et al.,[55] using self-report methods, noted
16. Strong WB, Malina RM, Blimkie CJR, et al. Evidence based
physical activity for school-age youth. J Pediatr 2005; 146 (6):
732-7
17. Bar-Or O. Pediatric sports medicine for the practitioner. New
York: Springer-Verlag, 1983
27.9% and 18.5% of 15-year-old boys and girls, 18. Harro M, Riddoch C. Physical activity. In: Armstrong N, van

original publisher.
respectively, to satisfy the same criterion. Mechelen W. editors. Paediatric exercise science and
medicine. Oxford: Oxford University Press, 2000: 77-84
Trends describing the physical activity patterns
19. Janz KF, Witt J, Mahoney LT, et al. The stability of children’s
of European youth are consistent but to obtain addi- physical activity measured by accelerometry and self-report.
tional insights into the role of physical activity in Med Sci Sports Exerc 1995; 27: 1326-32
20. Sallis JF. Self-report measures of children’s physical activity. J
promoting health during childhood and adolescence,

Unauthorised copying
Sch Health 1991; 61: 215-9
methods of assessment need to be further refined 21. Armstrong N, Bray S. Physical activity patterns defined by
and recommended guidelines regularly re-visited in continuous heart rate monitoring. Arch Dis Child 1991; 66:
245-7
relation to the increasing evidence base. 22. Bailey RC, Olson J, Pepper SL, et al. The level and tempo of
children’s physical activities: an observational study. Med Sci
Acknowledgements Sports Exerc 1995; 27: 1033-41

and distribution
This review was supported by a grant from the European
Union. The authors have no conflicts of interest that are
directly relevant to the content of this review.
23. Montoye HJ, Taylor HL. Measurement of physical activity in
population studies: a review. Hum Biol 1984; 56: 195-216
24. Watson AWS, O’Donovan DJ. Influences of level of habitual
activity on physical work capacity and body composition of
post pubertal school boys. Q J Exp Physiol 1977; 62: 325-32
25. Wessel JA, Montoye HJ, Mitchell H. Physical activity assess-

is prohibited.
References
1. Casperson CJ, Powell K, Christenson G. Physical activity, exer-
cise and physical fitness: definitions and distinctions of health-
related research. Public Health Rep 1985; 100: 126-31
2. Sirard JR, Pate RR. Physical activity assessment in children and
ment by recall method. Am J Public Health 1965; 55: 1430-6
26. Pate RR. Physical activity assessment in children and adoles-
cents. Crit Rev Food Sci Nutr 1993; 33: 321-6
27. Baranowski T, Dworkin RJ, Cieslik C, et al. Reliability and
validity of self report of aerobic activity: family health report.
adolescents. Sports Med 2001; 31: 439-54 Res Q 1984; 55: 309-17
3. Trost SG. Objective measurement of physical activity in youth: 28. Telama R, Viikari J, Valimaki I, et al. Atherosclerosis precur-
current issues, future directions. Exerc Sport Sci Rev 2001; 29: sors in Finnish children and adolescents leisure time physical
32-6 activity. Acta Paediatr Scand 1985; 318: 169-80
4. Twisk JWR. Physical activity guidelines for children and ado- 29. Saris WHM. Aerobic power and daily physical activity in chil-
lescents. Sports Med 2001; 31: 617-27 dren. Meppel (The Netherlands): Kripps Repro, 1982
5. American College of Sports Medicine. Physical fitness in chil- 30. Salonen JT, Lakka T. Assessment of physical activity in popula-
dren and youth. Med Sci Sports Exerc 1988; 20: 422-3 tion studies: validity and consistency of the methods in the

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1084 Armstrong & Welsman

Kuopio ischemic heart disease risk factor study. Scand J Sports 52. Heartbeat Wales. Welsh youth health survey 1986. Cardiff:
Sci 1987; 9: 89-95 Heartbeat Wales, 1986
31. Saris WHM. Habitual physical activity in children: methodolo- 53. Riddoch C. Northern Ireland health and fitness survey. Belfast:
gy and findings in health and disease. Med Sci Sports Exerc Sports Council for Northern Ireland and Department of Health
1986; 18: 253-63 and Social Services, 1990
32. Murphy JK, Alpert BS, Christman JV, et al. Physical fitness in 54. Marella M, Colli R, Faina M. Evaluation de l’aptitude physique:
children: a survey method based on parental report. Am J Eurofit, batterie experimentagle. Rome: Scuola Dello Sport,
Public Health 1988; 78: 708-10 1986
33. Saris WHM. The assessment and evaluation of daily physical 55. Currie C, Roberts C, Morgan A, et al., editors. Young people’s
activity in children: a review. Acta Paediatr Scand 1985; 318: health in context. Copenhagen: World Health Organisation,
37-48 2004
34. Harro M. Validation of a questionnaire to assess physical activi-
56. Cantera-Garde MA, Devís-Devís J. Physical activity levels of
ty of children ages 4-8 years. Res Q Exerc Sport 1997; 68:
secondary school Spanish adolescents. Eur J Phys Educ 2000;
259-68
5: 28-44
35. Bouchard C, Tremblay A, Leblanc C, et al. Method to assess
energy expenditure in children and adults. Am J Clin Nutr 57. Silvennionen M. Relations between different kinds of physical
1983; 37: 461-7 activity and motive types among Finnish comprehensive and
upper secondary school pupils. Scand J Sports Sci 1984; 6:
36. Durnin JVGA, Lonergan ME, Good J, et al. A cross-sectional
72-82

This material is
nutritional and anthropometric study, with an interval of 7
years on 611 young adolescent school children. Br J Nutr 58. Bouziotas C, Koutedaiks Y. A three year study of coronary heart
1974; 32: 169-79 disease risk factors in Greek adolescents. Pediatr Exerc Sci
37. Sallis JF, Buono MJ, Freedson PS. Bias in estimating caloric 2003; 15: 9-18
expenditure from physical activity in children. Sports Med 59. Maniôs Y, Kafatos A, Markakis G. Physical activity of 6-year-
1991; 11: 203-9 old children: validation of two proxy reports. Pediatr Exerc Sci
38. Torun B. Inaccuracy of applying energy expenditure rates of 1998; 10: 176-88

the copyright of the

39.

40.
adults to children. Am J Clin Nutr 1983; 38: 813-4
Sallis JF, Saelens BE. Assessment of physical activity by self-
report; status, limitations and future directions. Res Q Exerc
Sports 2000; 71: S1-14
Pate R, Freedson P, Sallis J, et al. Compliance with physical
60. Deheeger M, Rolland-Cachera MF, Fontvielle AM. Physical
activity and body composition in 10 year old French children:
linkages with nutritional intake? Int J Obes 1997; 21: 372-9
61. Vermorel M, Vernet J, Bitar A, et al. Daily energy expenditure,
activity patterns, and energy costs of the various activities in
activity guidelines: prevalence in a population of children and French 12–16-y-old adolescents in free living conditions. Eur J

original publisher.
41.

42.
youth. Ann Epidemiol 2002; 12: 303-8
Epstein LH, Paluch RA, Kalakanis LE, et al. How much activity
do youth get? A quantitative review of heart-rate measured
activity. Pediatrics 2001; 108: E44
Sleap M, Tolfrey K. Do 9-to-12-yr-old children meet existing
Clin Nutr 2002; 56: 819-29
62. Engstrom LM. Physical activity of children and youth. Acta
Paediatr Scand 1980; 283: 101-5
63. Sunnegardh J, Bratteby LE, Sjolin S, et al. The relation between
physical activity and energy intake of 8 and 13 year old
physical activity recommendations for health? Med Sci Sports children in Sweden. In: Binkhorst RA, Kemper HC G, Saris

Unauthorised copying
Exerc 2001; 33: 591-6 WHM, editors. Children and exercise XI. Champaign (IL):
43. Riddoch CJ, Andersen LB, Wedderkopp N, et al. Physical Human Kinetics, 1985: 183-93
activity levels and patterns of 9- and 15-yr-old European
64. Fuchs R, Semmer NK, Lippert P, et al. Patterns of physical
children. Med Sci Sports Exerc 2004; 36: 86-92
activity among German adolescents: the Berlin-Bremen study.
44. Ekelund U, Yngve A, Sjöström M. Total daily energy expendi- Prev Med 1988; 17: 746-63
ture and patterns of physical activity in adolescents assessed by
65. Rutenfranz J, Berndt I, Knauth P. Daily physical activity inves-

and distribution
two different methods. Scand J Med Sci Sports 1999; 9:
257-64 tigated by time budget studies and physical performance ca-
pacity of schoolboys. Acta Paediatr Belg 1974; 28: 79-86
45. Kemper HCG. The Amsterdam growth study. Champaign (IL):
Human Kinetics, 1995 66. Guerra S, Duarte J, Mota J. Physical activity and cardiovascular
46. Deforce B, Lefevre J, De Boudeaudhuig I, et al. Physical fitness disease risk factors in schoolchildren. Eur Phys Educ Rev
and physical activity in obese and non-obese Flemish youth. 2001; 7: 269-81
Obes Res 2003; 11: 434-41 67. Ribeiro J, Guerra S, Pinto A, et al. Overweight and obesity in
47.

48.
is prohibited.
Guillaume M, Lapidus L, Bjorntorp P, et al. Physical activity,
obesity and cardiovascular risk factors in children: the Belgi-
um Luxembourg child study II. Obes Res 1997; 5: 549-56
Hussey J, Gormley J, Bell C. Physical activity in Dublin chil-
dren aged 7-9 years. Br J Sports Med 2001; 35: 268-73
children and adolescents: relationship with blood pressure and
physical activity. Ann Hum Biol 2003; 30: 203-13
68. Andersen LB, Schelin B. Physical activity and performance in a
random sample of adolescents attending school in Denmark.
Scand J Med Sci Sports 1994; 4: 13-8
49. O’Sullivan SO. The physical activity of children: a study of 69. Hasselstrom H, Hansen SE, Frobert K, et al. Physical fitness and
1603 Irish school children aged 11-12 years. Ir Med J 2002; physical activity during adolescence as predictors of cardio-
95: 78-81 vascular disease risk in young adulthood: Danish youth and
50. Currie C, Hurrelmann K, Settertobulte W, et al. Health and sports study: an eight year follow-up. Int J Sports Med 2002;
health behaviour among young people. Copenhagen: World 23: S27-31
Health Organisation, 2000 70. King A, Wold B, Tudor-Smith C, et al., editors. The health of
51. King AJC, Coles B. The health of Canada’s youth. Ottawa: youth: a cross national survey. Copenhagen: World Health
Ministry of Health and Welfare, 1992 Organisation, 1996

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
Physical Activity of European Youth 1085

71. Loucaides LA, Chedzoy SM, Bennett N, et al. Correlates of 91. Emons HJG, Groenenboom DC, Westerterp KR, et al. Compari-
physical activity in a Cypriot sample of sixth-grade children. son of heart rate monitoring combined with indirect calorime-
Pediatr Exerc Sci 2004; 16: 25-36 try and the doubly labelled water method for the measurement
72. Aberberga-Augskalne L. Individual growth patterns and physi- of energy expenditure in children. Eur J Appl Physiol 1992;
cal fitness in Riga schoolchildren. Acta Medico-Historica 65: 99-103
Rigensia 2002; 25: 65-76 92. Livingstone MBE, Coward AW, Prentice AM, et al. Daily
73. Seliger VS, Trefny S, Bartenkova S, et al. The habitual physical energy expenditure in free-living children: comparison of heart
activity and fitness of 12 year old boys. Acta Paediatr Belg rate monitoring with the doubly labelled water method. Am J
1974; 28: 54-9 Clin Nutr 1992; 56: 343-52
74. Naul R, Telema R, Rychtécky A. Physical fitness and active 93. Eston RG, Rowlands AV, Ingledew DK. Validity of heart rate,
lifestyle of Czech, Finnish and German youth. Acta Univ Carol pedometry, and accelerometry for predicting the energy cost of
Kinanthrop 1997; 33: 5-16 children’s activity. J Appl Physiol 1998; 84: 362-71
75. Telema R, Naul R, Nupponen H, et al. Physical fitness, sporting 94. Louie L, Eston RG, Rowland AV, et al. Validity of heart rate,
lifestyles and Olympic ideals: cross-cultural studies on youth pedometry and accelerometry for estimating the energy cost of
sport in Europe. Schorndorf: Verlag Karl Hofmann, 2002 activity in Hong Kong Chinese boys. Pediatr Exerc Sci 1999;
76. Jurak G, KovaĉM, Strel J. How Slovenian primary school pupils 11: 229-39
spend their summer holidays. In: Jurak G, editor. Sports activi- 95. Bedale EM. Energy expenditure and food requirements of chil-
ties of Slovenian children and young people during their dren at school. Proc R Soc (London) 1923; 94: 368-404

This material is
summer holidays. Ljubljana: University of Ljubljana, 2003: 96. Puhl J, Greaves K, Hoyt M, et al. Children’s activity rating scale
23-38 (cars): description and calibration. Res Q Exerc Sport 1990;
77. Jurak G, KovaĉM, Strel J, et al. How Slovenian secondary 61: 26-36
school children spend their summer holidays. In Jurak G, 97. Armstrong N, Welsman JR. Young people and physical activity.
editor. Sports activities of Slovenian children and young peo- Oxford: Oxford University Press, 1997
ple during their summer holidays. Ljubljana: University of 98. Sleap M, Warburton P. Physical activity levels of 5-11 year old
Ljubljana, 2003: 39-58 children in England determined by continuous observation.

the copyright of the

78. Raudsepp L, Liblik R, Hannus A. Children’s and adolescents’
physical self-perception as related to moderate to vigorous
physical activity and physical fitness. Pediatr Exerc Sci 2002;
14: 97-106
79. Sallo M, Viru A. Aerobic capacity and physical activity in 4 to
Res Q Exerc Sport 1992; 63: 238-45
99. Sleap M, Warburton P. Physical activity levels of preadolescent
children in England. Br J Phys Educ Res Suppl 1994; 14: 2-6
100. Sleap M, Warburton P. Physical activity levels of 5-11-year-old
children in England: cumulative evidence from three direct
10-year-old children. Biol Sport 1996; 13: 211-9 observation studies. Int J Sports Med 1996; 17: 248-53

original publisher.
80. Kemper HCG. Growth, health and fitness of teenagers. Med
Sport Sci 1985; 20: 1-202
81. Prochaska JJ, Sallis JF, Long BA. A physical activity screening
measure for use with adolescents in primary care. Arch Pediatr
Adolesc Med 2001; 155: 554-9
101. Armstrong N. Young people’s physical activity patterns as
assessed by heart rate monitoring. J Sports Sci 1998; 16: S9-16
102. Armstrong N, Balding J, Gentle P, et al. Patterns of physical
activity among 11 to 16 year old British children. BMJ 1990;
301: 203-5
82. Ainslee PN, Reilly T, Westerterp KR. Estimating human energy 103. Verschuur R, Kemper HCG. Habitual physical activity in Dutch

Unauthorised copying
expenditure. Sports Med 2003; 33: 683-98 teenagers measured by heart rate. In: Binkhorst RA, Kemper
83. Saris WHM. New developments in the assessment of physical HCG, Saris WHM, editors. Children and exercise XI. Cham-
activity in children. In: Coudert J, Van Praagh E. editors. paign (IL): Human Kinetics, 1985: 194-202
Pediatric work physiology. Paris: Masson, 1992: 107-14 104. Riddoch C, Mahoney C, Murphy N, et al. The physical activity
84. James WPT, Ferro-Luzzi A, Waterlow JC. Definition of chronic patterns of Northern Irish schoolchildren ages 11 to 16 years.
energy deficiency in adults: report of a working party of the Pediatr Exerc Sci 1991; 3: 300-9

and distribution
international dietary energy consultancy group. Eur J Clin Nutr 105. Bradfield RB, Chan H, Bradfield NE, et al. Energy expenditures
1988; 42: 969-81 and heart rates of Cambridge boys at school. Am J Clin Nutr
85. Davies PSW. Total energy expenditure in young children. Am J 1971; 24: 1461-6
Hum Biol 1996; 8: 183-8 106. Verschuur R, Kemper HCG. Habitual physical activity. Med
86. Hoos MB, Gerver WJM, Kester AD, et al. Physical activity Sport Sci 1985; 20: 56-65
levels in children and adolescents. Int J Obes 2003; 27: 605-9 107. Verschuur R, Kemper HCG. The pattern of daily physical
87. Bratteby LE, Sandhagen B, Fan H, et al. Total energy expendi- activity. Med Sport Sci 1985; 20: 169-86

is prohibited.
ture and physical activity as assessed by the doubly labeled
water method in Swedish adolescents in whom energy intake
was underestimated by 7-d diet records. Am J Clin Nutr 1998;
67: 905-11
88. Ekelund U, Sjostrom M, Yngve A, et al. Physical activity
108. Armstrong N, Bray S. Primary schoolchildren’s physical activi-
ty patterns during autumn and summer. Bull Phys Educ 1990;
26: 23-6
109. Biddle S, Mitchell J, Armstrong N. The assessment of physical
activity in children: a comparison of continuous heart rate
assessed by activity monitor and doubly labeled water in monitoring, self-report and interview techniques. Br J Phys
children. Med Sci Sports Exerc 2001; 33: 275-81 Educ Res Suppl 1991; 10: 4-8
89. Hoos MB, Plasqui G, Gerver WJM, et al. Physical activity level 110. McManus A, Armstrong N. Patterns of physical activity among
measured by doubly labeled water and accelerometry in chil- primary schoolchildren. In: Ring FJ, editor. Children in sport.
dren. Eur J Appl Physiol 2003; 89: 624-6 Bath: University Press, 1995: 17-23
90. O’Connor J, Ball EJ, Steinbeck KS, et al. Measuring physical 111. Falgairette G, Bedu M, Fellmann N, et al. Bioenergetic profile
activity in children: a comparison of four different methods. in 144 boys aged from 6 to 15 years with special reference to
Pediatr Exerc Sci 2003; 15: 202-15 sexual maturation. Eur J Appl Physiol 1991; 62: 151-6

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)
1086 Armstrong & Welsman

112. Sallo M, Silla R. Physical activity with moderate to vigorous Ilmarinen J, Valimaki I. editors. Children and sport. New
intensity in preschool and first grade schoolchildren. Pediatr York: Springer-Verlag, 1984: 255-61
Exerc Sci 1997; 9: 44-54
127. Vincent SD, Pangrazi RP, Raustorp A, et al. Activity levels and
113. Welsman JR, Armstrong N. Physical activity patterns of 5 to body mass index of children in the United States, Sweden and
11-year-old children. In: Armstrong N, Kirby BJ, Welsman JR, Australia. Med Sci Sports Exerc 2001; 35: 1367-73
editors. Children and exercise XIX: promoting health and
well-being. London: E and FN Spon, 1997: 139-44 128. Loucaides CA, Chedzoy SM, Bennett N. Pedometer-assessed
physical (ambulatory) activity in Cypriot children. Eur Phys
114. Gavarry O, Bernard T, Giacomoni M, et al. Continuous heart Educ Rev 2003; 9: 43-55
rate monitoring over 1 week in teenagers aged 11-16 years. Eur
J Appl Physiol 1998; 77: 125-32 129. Freedson P, Pober D, Janz KF. Calibration of accelerometer
output for children. Med Sci Sports Exerc 2005; 37 (11 Sup-
115. Welsman JR, Armstrong N. Physical activity patterns of 5-to
pl.): S523-30
7-year-old children and their mothers. Eur J Phys Educ 1998;
3: 145-55 130. Raudsepp L, Päll P. Reproducibility and stability of physical
116. Armstrong N, Welsman JR, Kirby BJ. Longitudinal changes in activity in children. Pediatr Exerc Sci 1998; 10: 320-6
11-13-year-olds’ physical activity. Acta Paediatr 2000; 89: 131. Nilsson A, Ekelund U, Yngve A, et al. Assessing physical
775-80 activity among children with accelerometers using different
117. Welsman JR, Armstrong N. Physical activity patterns in secon- time sampling intervals and placements. Pediatr Exerc Sci

This material is
dary school children. Eur J Phys Educ 2000; 5: 147-57
118. Ekelund U, Poortvleit E, Nilsson A, et al. Physical activity in
relation to aerobic fitness and body fat in 14-to-15 year-old
boys and girls. Eur J Appl Physiol 2001; 85: 195-201
2002; 14: 87-96
132. Guerra S, Santos P, Ribeiro JC, et al. Assessment of children’s
and adolescent’s physical activity levels. Eur Phys Educ Rev
2003; 9: 75-85
119. Lauter S. Zur genese der fettsucht. Dtsch Arch Klin 1926; 150: 133. Santos P, Guerra S, Ribeiro JC, et al. Age and gender-related

the copyright of the

315-65 physical activity: a descriptive study in children using acceler-
120. Saris WHM, Binkhorst RA. The use of pedometer and actome- ometry. J Sports Med Phys Fitness 2003; 43: 85-9
ter in studying daily physical activity in man: part II -validity 134. Mota J, Santos P, Guerra S, et al. Differences of daily physical
of pedometer and actometer measuring the daily physical activity levels of children according to body mass index.
activity. Eur J Appl Physiol 1977; 37: 229-35 Pediatr Exerc Sci 2002; 14: 442-52
121. Rowe DA, Mahar MT, Raedeke TD, et al. Measuring physical
135. Mallam KM, Metcalf BS, Kirby J, et al. Contribution of timet-

original publisher.
activity in childhood with pedometers: reliability, reactivity
and replacement of missing data. Pediatr Exerc Sci 2004; 16:
343-54
122. Cardon G, De Bourdeaudhuij I. A pilot study comparing pedom-
eter counts with reported physical activity in elementary
abled physical education to total physical activity in primary
schoolchildren: cross-sectional study. BMJ 2003; 327: 592-3
136. Berman N, Bailey R, Barstow TJ, et al. Spectral and bout
detection analysis of physical activity patterns in healthy,
schoolchildren. Pediatr Exerc Sci 2004; 16: 355-67 prepubertal boys and girls. Am J Hum Biol 1998; 10: 289-97

Unauthorised copying
123. Rowlands AV, Eston RG, Ingledew DK. Measurement of physi-
cal activity in children with particular reference to the use of
heart rate and pedometry. Sports Med 1997; 24: 258-72
124. Raustorp A, Pangrazi RP, Stahle A. Physical activity level and
body mass index among schoolchildren in south eastern Swe-
137. Perusse L, Trembley A, Le Blanc C, et al. Genetic and environ-
mental influences on level of habitual physical activity and
exercise participation. Am J Epidemiol 1989; 129: 1012-22
138. Rowland TW. The biological basis of physical activity. Med Sci
Sport Exerc 1998; 30: 392-9
den. Acta Paediatr 2004; 93: 400-4

and distribution
125. Skalik K, Frömel K, Sigmund E, et al. Weekly physical activity
in secondary school students (a comparative probe into Czech,
Polish and Swedish conditions). Gymnica 2001; 31: 21-6
126. Verschuur R, Kemper HCG, Besseling CWM. Habitual physical
Correspondence and offprints: Prof. Neil Armstrong, Chil-
dren’s Health and Exercise Research Centre, St Luke’s
Campus, University of Exeter, Exeter, EX1 2LU, UK.
activity and health in 13 and 14 year old teenagers. In: E-mail: [email protected]

is prohibited.

 2006 Adis Data Information BV. All rights reserved. Sports Med 2006; 36 (12)