Content uploaded by Eva Brita Roos
Author content
All content in this area was uploaded by Eva Brita Roos on Aug 31, 2016
Content may be subject to copyright.
ORIGINAL ARTICLE
Health behaviors, waist circumference and
waist-to-height ratio in children
R Lehto
1
, C Ray
1,2
, M Lahti-Koski
3
and E Roos
1,2
1
Folkha¨lsan Research Center, Paasikivenkatu 4, Helsinki, Finland;
2
Hjelt Institute, the University of Helsinki, Helsinki, Finland
and
3
The Finnish Heart Association, Helsinki, Finland
Background: Waist circumference (WC) and waist-to-height ratio (WHtR) begin to gain attention as measures of adiposity and
as important cardiometabolic disease risk factors also among children. Still, little research has been done on behavioral
determinants of WC and WHtR in children. The purpose of this study was to examine associations between health behaviors, WC
and WHtR in children.
Methods: The study was a cross-sectional study conducted in Swedish-speaking schools in Helsinki region in 2006. In all, 1146
children were recruited, from which 55 % took part in the study. A total of 604 9–11-year-old children (312 girls, 292 boys) were
measured by research staff and completed a study questionnaire on their health behaviors, including breakfast intake, TV
viewing, sleep duration and physical activity, and a 16-item food frequency questionnaire. Covariance analysis was used as the
statistical analysis method.
Results: When controlling for other health behaviors, for example, irregular breakfast (B-coefficient 2.49 CI, 0.64–4.34;
Po0.01), TV viewing (B-coefficient 0.89 CI, 0.17–1.61; Po0.05), a TV in child’s room (B-coefficient 2.30 CI, 0.73-3.86; Po0.01)
and physical inactivity during school breaks (B-coefficient 0.78 CI, 0.19–1.37; Po0.01) were associated with larger WC. Results
were similar with WHtR.
Conclusions: Many health behaviors were related to children’s WC and WHtR. Determinants were associated to both WC and
WHtR similarly.
European Journal of Clinical Nutrition (2011) 65, 841–848; doi:10.1038/ejcn.2011.49; published online 13 April 2011
Keywords: health behavior; waist circumference; waist-to-height ratio; children; physical activity; food habits
Background
Although children’s overweight and obesity, based on body
mass index (BMI), has increased significantly in recent
decades, children’s waist circumference (WC) has increased
even more (Moreno et al., 2001; McCarthy et al., 2003). In
some studies, WC has been found to be an even stronger
indicator of cardiometabolic disease risk factors than BMI in
children (Lee et al., 2006; Sung et al., 2007; Reinehr and
Wunsch, 2010). However, a recent systematic review con-
cluded that WC and BMI were equally useful in identifying
cardiometabolic disturbances in children (Reilly et al., 2010).
One shortcoming of WC as an indicator of obesity is that it
does not take height into account. To solve this problem,
waist-to-height ratio (WHtR) has been used (McCarthy and
Ashwell, 2006; Cossio et al., 2009; Nambiar et al., 2009). In
adults, WHtR has been shown to be a stronger factor in
identifying cardiometabolic disease risk factors compared
with WC, waist-hip ratio and BMI, although the differences
were small (Lee et al., 2008). A few studies have shown
similar results in children (Kahn et al., 2005; Cossio et al.,
2009).
Although many studies have examined the relation-
ship between health behaviors and BMI in children,
studies on health behaviors and WC are few. In addition
to our knowledge, no studies have examined the
relationship between health behaviors and WHtR in
children.
In most of the previous studies on physical activity and
WC, physical activity has been found to be associated with
smaller WC (Klein-Platat et al., 2005; Delmas et al., 2007;
Ortega et al., 2007; Lazarou and Soteriades, 2010). Sedentary
behavior, that is TV viewing or total sedentary behavior, has
been associated with larger WC (Klein-Platat et al., 2005;
Received 28 October 2010; revised 1 February 2011; accepted 9 March 2011;
published online 13 April 2011
Correspondence: Dr E Roos, Folkha¨lsan Research Center, Paasikivenkatu 4,
Helsinki 00250, Finland.
E-mail: eva.roos@folkhalsan.fi
European Journal of Clinical Nutrition (2011) 65, 841– 848
&
2011 Macmillan Publishers Limited All rights reserved 0954-3007/11
www.nature.com/ejcn
Ortega et al., 2007; Lazarou and Soteriades, 2010). The
presence of a TV in a child’s bedroom was found to be
associated with larger WC and other obesity measures, but
only in boys (Delmas et al., 2007). Shorter sleep duration was
related to larger WC, at least in girls in two studies (Yu et al.,
2007; Hitze et al., 2009). Only few studies have reported
results on food habits and WC in children. In a study on
Cypriot children, an index describing adherence to a
Mediterranean diet was not associated with WC (Lazarou
and Soteriades, 2010). In another study on American
children, intake of some foods was associated with WC
(Bradlee et al., 2010). Few previous studies done on break-
fast intake and WC have shown that skipping breakfast
is associated with larger WC (Isacco et al., 2010;
Smith et al., 2010).
As BMI, WC and WHtR correlate strongly with each other,
the factors related to each of them are probably similar.
Nevertheless, differences may occur, especially as WC is
growing more rapidly than BMI among children. In
addition, these measures are indicators of different kinds of
fat distribution. Therefore, it is important to get more
knowledge about factors that are related to WC and WHtR.
The aim of this study was to examine the associations
between health behaviors, WC and WHtR among school
children in Finland. Studied health behaviors were food
consumption, the regularity of breakfast, physical activity,
screen time and sleep duration. In addition, we investigated
the possible associations with and without adjustment for
general obesity, defined by BMI, to see if health behaviors are
associated with WC and WHtR independently of BMI.
Methods
Participants
This study was performed as a part of a project called
Ha
¨lsoverkstaden (Health workshop), which studies the
health behaviors of 9–11-year-old children (Ray et al., 2009;
Westerlund et al., 2009; Lehto et al., 2010). The study
material was cross-sectional, and was collected in Swedish-
speaking elementary schools in the capital region of Finland
during 2006. All 44 Swedish-speaking schools with more
than 50 pupils in the region were asked to take part in the
study. The headmasters in 27 schools decided that their
school would participate. The participating and not partici-
pating schools did not vary according to the socio-economic
status of the neighborhood. The study was approved by the
ethical committee of the Department of Public Health of the
University of Helsinki.
In the spring 2006, all children in the grades three and four
(n¼1146) and their parents in the participating schools were
contacted. In all, 677 children and their parents gave their
informed consent for participation in the study. Data were
collected during two school visits. In total, 630 children were
measured and weighed by the research staff in the spring.
The 47 children who gave their consent but weren’t
measured, were either absent during the measurements or
declined to be measured. In the fall, 604 of those children
who were measured completed a questionnaire on their
health behavior. The questionnaire was administered in a
classroom setting, and a member of the research staff was
always present. These children (n¼604) form the sample of
our study, representing a participation rate of 53% of the
children who were first contacted.
Anthropometrics
The children’s height was measured without shoes to the
nearest 0.5 cm with the same study measure. WC was
measured on top of a t-shirt to the nearest 1 cm midway
between iliac crest and the lowest rib. WHtR was calculated
as WC (cm) divided by height (cm). The children were
weighed with the same study scale to the nearest 0.1kg
wearing only underwear and a t-shirt. The measurements
were always carried out before lunch-time.
The children’s health behaviors
The children’s research questionnaire included a 16-item
food frequency questionnaire (FFQ), and questions on meal
patterns, sleep duration, sedentary behavior and physical
activity. All questions on health behavior, except for the
question on school break activities, were taken from the
WHO’s Health Behavior in School-aged Children study
questionnaire (Currie et al., 2001). We only used data on
breakfast frequency, sleep duration and screen time during
weekdays, because in our view weekday activities represent
routines of everyday life.
The 16-item FFQ measured habitual intake of certain foods
and food groups. The answer choices in the FFQ were as
follows: never, less than once a week, once a week, 2–4 times
per week, 5–6 times per week, once a day and several times a
day. These were then scored afterwards by researchers as
0, 0.5, 1, 3, 5.5, 7 and 14 to represent average intake
occasions per week. Principal component analysis was also
used to form food indices from the FFQ. This has been
described in detail previously (Westerlund et al., 2009). Two
factors were found from which sum variables were formed by
adding up the weekly intake occasions. The first sum variable
consisted of pizza; hamburgers, hot dogs and meat pastry;
potato chips and popcorn; cookies; ice cream; sweets and
chocolate; and cola and other soft drinks. It was named the
energy-dense food index. The other sum variable consisted
of fresh vegetables, cooked vegetables, fruits and berries, and
dark bread and it was named the nutrient-dense food index.
The frequency of eating breakfast at home during the
school week was asked with answer alternatives ranging from
zero to five. A regular breakfast was then defined as having
breakfast usually on all 5 days during school week. Less than
5 days was defined as an irregular breakfast.
Sleep duration was calculated as the difference between
bed time and waking time for school days. Alternative
Health behaviors and central adiposity in children
R Lehto et al
842
European Journal of Clinical Nutrition
answer choices for bed time when next day was school day
included every half an hour starting from ‘at 8 pm at the
latest’ and ending with ‘midnight or later.’ For waking time
alternative answers included every half an hour between ‘at
6 am at the latest’ and ‘8.30 am or later’.
Total television viewing and computer screen time per day
during the school week was calculated through two ques-
tions, one on TV, video and DVD viewing time and the other
on time spent using a computer or playing with game
consoles. The seven alternative answers ranged from ‘not at
all’ to ‘approximately 5 h per day or more’. The children were
also asked two questions about the presence of a TV and a
computer or a game console in their room, with the answer
alternatives of yes or no.
The amount of free time physical activity per week in a
sports club or on one’s own was asked with six answer
alternatives ranging from ‘not at all’ to ‘7 h or more’. These
answer alternatives were then converted to hours equaling
7, 5, 2.5, 1, 0.5 and 0 h.
Physical activity during school breaks was assessed with
five statements about school break activities. The statements
were: ‘I am physically active when I play,’ ‘I play ball games,’
‘I walk,’ ‘I talk with friends,’ and ‘I stand still.’ The
alternative answer choices were ‘almost every break,’ ‘during
most breaks,’ ‘seldom,’ and ‘never during breaks’ and they
were given points from 1 to 4. Principal component analysis
(with varimax rotation) was employed to extract factors from
the school break activity statements (Stevens, 1992). Two
components had eigenvalues over 1, the first of which
consisted of the statements ‘I walk,’ ‘I talk with friends,’ and
‘I stand still’ (load over 0.5), and the second of the two
statements, ‘I am physically active when I play’ and ‘I play
ball games’ (load over 0.5). Two sum variables were formed
based on these factors by adding up the points from each
statement. The first variable was named physical inactivity
during school breaks and the second physical activity during
school breaks.
Statistical methods
Gender differences in WC, other anthropometrics and
health behaviors were tested with the t-test and w
2
-test.
Covariance analysis was used to examine the association of
health behaviors with WC and WHtR. Free time physical
activity, physical inactivity during school breaks (quartiles),
physical activity during school breaks (quartiles), sleep
duration, TV viewing and the energy-dense food index
(quartiles) were all used as continuous variables in these
analyses, as they appeared to have a linear association with
WC. The nutrient-rich food index and computer/game
console time were used as categorical variables, as they did
not appear to have a linear association with WC. SPSS for
Windows 17.0 was used for the analyses (SPSS Inc., Chicago,
IL, USA).
Three different models were used. The first model was
adjusted only for age and gender, and the second for the
aforementioned plus all other health behaviors. In the third
model, BMI was added to the model 2.
Results
A description of the study sample is seen in Table 1. The
average WC was larger in the boys than in the girls. The same
applied to WHtR, whereas no gender differences in BMI were
found. A larger proportion of the boys than the girls had a
TV or a computer/game console in their room, and the boys
also reported spending more time on a computer or playing
with a game console than the girls. The boys exercised more
than the girls during their free time and they were physically
Table 1 Description of the study sample
Girls (N¼312) Boys (N¼292) Total (N¼604)
Mean s.d. Mean s.d. Mean s.d.
Age (years)
a
9.6 0.6 9.7 0.6 9.6 0.6
WC
b
64.5 7.2 66.2 7.4 65.3 7.3
WHtR
b
0.449 0.04 0.459 0.04 0.454 0.04
Height (cm) 144 7.9 144 6.5 144 7.3
Weight (kg) 36.4 8.0 36.8 7.3 36.6 7.7
BMI (m
2
/kg) 17.5 2.6 17.5 2.5 17.5 2.6
TV viewing (h/school day) 1.3 1.0 1.3 1.1 1.3 1.0
Computer/game console use (h/school day)
b
0.8 0.8 1.3 1.1 1 1.0
Sleep duration (h/school day) 9.7 0.7 9.7 0.8 9.7 0.7
Physical activity (hours/week)
b
4.3 2.0 4.7 2.1 4.5 2.1
Regular breakfast (yes) (%) 88 85 87
TV in child’s room (yes) (%)
b
33 46 39
Computer/game console in child’s room (%)
b
28 53 40
Abbreviations: BMI, body mass index; WHtR, waist-to-height ratio; WC, waist circumference.
a
During anthropometric measurements.
b
Significant difference between boys and girls (P¼o0.05), t-test and w
2
-test.
Health behaviors and central adiposity in children
R Lehto et al
843
European Journal of Clinical Nutrition
more active also during school breaks. There were no gender
differences in the intake of energy-dense and nutrient-
rich foods.
In the first model, adjusted only for age and gender,
irregular breakfast, a TV in the child’s room, a computer or
game console in the child’s room, more TV viewing, less
frequent intake of energy-dense foods, less physical activity
in free time and during school breaks, shorter sleep duration
and more physical inactivity during school breaks were
associated with larger WC (Table 2). All of these variables,
except for physical inactivity during school breaks, were also
associated with WHtR. Frequency of intake of nutrient-rich
foods and time spent on a computer or playing with a game
console were not associated with WC or WHtR.
In the second model, when health behavior variables were
adjusted for each other, irregular breakfast, TV viewing, less
frequent intake of energy-dense foods, a TV in the child’s
room, more physical activity and less inactivity during
school breaks remained associated with larger WC and
WTHR (Table 2).
In the third model, when BMI was added to the model,
physical activity during free time was inversely related to
both WC and WHtR (Table 2). In addition, physical
inactivity during school breaks was associated with larger
WC and physical activity during school breaks with smaller
WHtR. Other associations were no longer significant.
Discussion
In this study on 9–11-year-old Finnish children, many health
behaviors were related to children’s WC and WHtR. When
adjusted with other health behaviors, irregular breakfast, less
frequent intake of energy-dense foods, more TV viewing, the
presence of a TV in the child’s room, more physical inactivity
and less physical activity during school breaks were asso-
ciated with larger WC. All of these variables except for
physical inactivity during school breaks were also associated
with larger WHtR. After controlling for BMI, only variables
concerning physical activity were associated with WC and
WHtR.
In concordance with our results, the association between
free time physical activity and WC has been found in other
studies (Ortega et al., 2007). In a large sample of 12-year-old
French children, the association also persisted after control-
ling for BMI (Klein-Platat et al., 2005). The reason for
physical activity and WC/WHtR associations might be that
physical activity may be more strongly associated with
measures of central adiposity than with BMI, because
physical activity might increase fat-free mass, which would
increase BMI but not WC.
In concordance with other studies on having breakfast and
WC (Isacco et al., 2010; Smith et al., 2010), we found that an
irregular pattern of eating breakfast was related to larger WC
and WHtR, but this association was no longer significant
after adjusting for BMI. A number of studies have also shown
that skipping breakfast is associated with higher BMI and
overweight in children and adolescents both in cross-
sectional (Dubois et al., 2009; Szajewska and Ruszczynski,
2010) and prospective settings (Albertson et al., 2007; Timlin
et al., 2008). A reason for this can be that breakfast eaters
have consistently been reported to have healthier diets than
breakfast skippers (Rampersaud et al., 2005). Breakfast eating
can also be an indicator of an overall healthy lifestyle
(Vereecken et al., 2009).
Contrary to our findings, in an American study, (Bradlee
et al., 2010) reported that the consumption of dairy, grains,
and fruits and vegetables was associated with smaller WC in
12–16-year-old adolescents. These results are not fully
comparable, as in Bredlee’s study the measure of diet was
24 h recall while we used a FFQ, and instead of separate food
groups we used a sum variable in our study. In a Cypriot
study, no association was found between adherence to a
Mediterranean diet and children’s WC (Lazarou and Soter-
iades, 2010). Problems in finding a valid and reliable method
to measure diet and eliciting reliable information from the
study subjects might be reasons for inconsistent findings.
Similar to our results, other studies have also found that
overweight children report eating sweets less often than
normal weight children (Andersen et al., 2005; Janssen et al.,
2005). One explanation for these odd findings can be more
underreporting by bigger children (Lanctot et al., 2008;
Singh et al., 2009). However, it is possible that children with
larger WC and WHtR truly do consume energy-dense foods
less often due to their own or a parental restriction.
In our study, sleep duration was related to WC and WHtR
only in the first model, which was adjusted for age and
gender. This can be due to a small statistical power or the fact
that meal patterns, TV viewing and food habits, which all
have been associated with sleep duration (Ray et al., 2007;
Hitze et al., 2009; Westerlund et al., 2009) were mediators of
this association. In two other studies, sleep duration has
been related to larger WC, but mostly only among girls (Yu
et al., 2007; Hitze et al., 2009). In many recent studies,
shorter sleep duration has been found to be associated with
overweight in children (Patel and Hu, 2008).
Similar to our study, associations have been found
between TV viewing or sedentary behavior and WC in
previous studies, but in most studies only among one gender
(Klein-Platat et al., 2005; Delmas et al., 2007; Ortega et al.,
2007; Lazarou and Soteriades, 2010). Ortega et al. (2007) also
found that the negative impact of TV viewing on WC can be
attenuated with more vigorous physical activity. However,
TV viewing’s association with childhood obesity after
controlling for physical activity has been stated a number
of times (Vandewater and Huang, 2006; Jackson et al., 2009).
In our study as well, the results were controlled for physical
activity.
In accordance with our study, the presence of a TV in a
child’s room was associated with larger WC and other obesity
measures in a sample of 12-year-old French school children,
but only in boys (Delmas et al., 2007). The mechanism by
Health behaviors and central adiposity in children
R Lehto et al
844
European Journal of Clinical Nutrition
which a TV in a child’s room could be associated with larger
WC is the duration of TV viewing time. It has been found
that a TV in child’s room is associated with TV viewing time
(Gorely et al., 2004; Delmas et al., 2007), although that was
not the case in our study (data not shown). Thus, it is
surprising that a TV in a child’s room is associated with a
larger WC even after controlling for TV viewing time. This
might be because children misreport TV viewing time and
children with a TV in their room in particular underestimate
their TV viewing.
One strength of our study was that we measured the
anthropometrics of the study children. Self-reported or
Table 2 The associations of health behaviors with WC and WHtR. Covariance analysis, B-coefficients and 95% confidence intervals (CI)
WC WHtR
Model B 95% CI B 95% CI
Regular breakfast (no vs yes) 0 0
1 2.47** 0.60–4.35 0.017*** 0.007–0.027
2 2.49** 0.64–4.34 0.015** 0.004–0.027
3 0.59 0.43–1.62 0.004 0.003–0.011
Energy-dense food index (quartiles as a continuous variable) 1 1.50*** 2.09 to (0.90) 0.006*** 0.009 to (0.003)
21.45*** 2.04 to (0.86) 0.009*** 0.013 to (0.005)
30.16 0.49–0.17 0.001 0.003–0.001
Nutrient-rich food index (quartiles as a categorical variable,
least often vs most often)
10.76 2.32–0.81 0.001 0.009–0.010
21.71 3.43–0.01 0.005 0.016–0.006
30.32 1.26–0.62 0.003 0.003–0.010
Sleep duration (h) 1 0.95* 1.77 to (0.14) 0.005* 0.01–0.00
20.61 1.50 to 0.28 0.002 0.007–0.004
30.21 0.69–0.28 0.001 0.003–0.004
TV viewing (h) 1 0.64* 0.07–1.2 0.004* 0.001–0.007
2 0.89* 0.17–1.61 0.006** 0.002–0.011
30.08 0.48–0.31 0.000 0.002–0.003
TV in child’s room (no vs yes) 0 0
1 2.24** 0.67–3.80 0.013** 0.0030.023
2 2.30** 0.73–3.86 0.014** 0.0040.024
3 0.63 0.22–1.49 0.004 0.002–0.010
Computer/game console time (tertiles as a
categorical variable, least vs most)
00
10.89 3.14–1.36 0.002 0.015–0.012
20.00 2.66–2.65 0.004 0.021–0.013
30.11 1.55–1.33 0.005 0.015–0.005
Computer/game console in child’s room (no vs yes) 0 0
1 1.33* 0.09–2.57 0.009* 0.001–0.016
2 0.53 1.08–2.14 0.004 0.006–0.014
30.21 1.08–0.66 0.000 0.006–0.005
Physical activity (h/week) 1 0.31* 0.59 to (0.03) 0.002* 0.004–0.000
20.21 0.53–0.11 0.001 0.003–0.001
30.26** 0.43–(0.09) 0.002** 0.003–0.000
Physical activity during school breaks 0 0
(quartiles as a continuous variable) 1 0.71* 1.28 to (0.14) 0.005** 0.008 to (0.002)
20.68* 1.29 to (0.07) 0.005** 0.009 to (0.001)
30.27 0.60–0.06 0.003* 0.005–0.000
Physical inactivity during school breaks 1 0.81** 0.25–1.37 0.003 0.000–0.007
(quartiles as a continuous variable) 2 0.78** 0.19–1.37 0.003 0.000–0.007
3 0.33* 0.01–0.65 0.000 0.0020.002
Abbreviations: WHtR, waist-to-height ratio; WC, waist circumference.
Models: adjusted for: model 1: age, gender; model 2: model 1 þall other health behaviors; model 3: model 2þBMI. *P¼o0.05; **P¼o0.01; ***P¼o0.001.
Health behaviors and central adiposity in children
R Lehto et al
845
European Journal of Clinical Nutrition
parent-reported anthropometrics would have been prone to
errors and underreporting (Goodman et al., 2000; Sherry
et al., 2007). Many lifestyle-related health behaviors were
studied, and therefore, we were able to study the association
between many health behaviors and WC or WHtR simulta-
neously.
The weaknesses of this study relate mostly to the study
sample. The study sample was selective as it represented a
language minority in the capital region of Finland. The
children came mostly from high socio-economic status
families. In that sense, it is possible that more associations
would have been found with a larger heterogeneity in
relation to children’s socio-economic status. Because the
study sample was quite small, covariance analyses were not
run separately for boys and girls. In any case, larger and more
representative studies are needed to confirm our results. The
response rate was quite low, because 20% of the contacted
parents and children did not fill in the consent form and due
to time limitations no reminder was sent.
Even though the ability of children of this age to answer
questions about their health behavior could be questioned,
10–12-year-old children are considered to be able to fill in
FFQs (Livingstone and Robson, 2000), and the FFQ used
in our study has been validated with 11–12-year-old children
(Vereecken and Maes, 2003). Most of the questions were also
used in WHO’s Health Behavior in School-aged Children
study with 11–15-year-old children (Currie et al., 2001), but
to our knowledge the questions have not been validated in
this age group. However, the questions on physical activity
have been validated among 13–18-year old adolescents
(Rangul et al., 2008), showing that these questions had good
validity. In our view, it would have been problematic to use
parents’ reports on their child’s health behavior, because
parents do not automatically know about their child’s food
intake, TV viewing, computer use or physical activity that
well. In Finland both parents usually work full-time. All
children also get a warm lunch at school, and therefore
parents are not that well aware of what their child eats at
lunch. Thus, as children of this age spend a large part of the
day without their parents being present, parents’ ability to
report for example, food intake of their child has been
reported to be limited (Livingstone and Robson, 2000). In a
subsample, we have data on the children’s health behaviors
reported by the parents as well. When comparing children’s
and parents’ reports on the children’s health behavior, it was
found that the reports were quite similar, but children
reported more screen time and less sleep (Roos et al., 2009).
By using a model including both BMI and WC, we wanted
to find out if health behaviors are associated with measures
of central obesity when BMI is constant. We are aware that
this is problematic, as the correlation between BMI and WC
was high (0.87). However, similar models have been used in
several other studies (Rexrode et al., 1998; Halkjaer et al.,
2004; Klein-Platat et al., 2005; Wildman et al., 2005).
The cross-sectional design of our study does not permit
any conclusions to be made of causality. It can be, as it is
proposed in many cross-sectional studies on BMI and
physical activity, that children with larger WC have become
more sedentary as a consequence of their size. The same
thing applies to other results, such as the intake frequency of
energy-dense foods. In addition, the 6-month delay in the
administration of the study questionnaire after the anthro-
pometrics was measured, poses additional challenges. To
overcome these problems, prospective studies on the matter
are needed.
Conclusions
In this study, we have examined multiple health behaviors
with regard to WC and WHtR among 9–11-year-old children.
Controlling for other health behaviors and later for BMI gave
us new knowledge about the independent contribution of
different factors to the associations with WC and WHtR. We
found that many health behaviors were associated with
children’s WC and WHtR. After controlling for BMI,
measures of physical activity were still associated with WC
and WHtR. Our study suggests that physical activity can
have a special association with WC and WHtR beyond BMI,
but further and especially prospective studies on the matter
are needed.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
We would like to thank all the study schools, children and
their parents for the participation in the study. This study
was supported by Juho Vainio Foundation, Pa
¨ivikki and
Sakari Sohlberg Foundation, Signe and Ane Gyllenberg
Foundation, Medicinska understo
¨dsfo
¨rening Liv och Ha
¨lsa.
References
Albertson AM, Franko DL, Thompson D, Eldridge AL, Holschuh N,
Affenito SG et al. (2007). Longitudinal patterns of breakfast eating
in black and white adolescent girls. Obesity 15, 2282–2292.
Andersen LF, Lillegaard IT, Overby N, Lytle L, Klepp KI, Johansson L
(2005). Overweight and obesity among Norwegian schoolchil-
dren: changes from 1993 to 2000. Scand J Public Health 33, 99–106.
Bradlee ML, Singer MR, Qureshi MM, Moore LL (2010). Food group
intake and central obesity among children and adolescents in the
Third National Health and Nutrition Examination Survey
(NHANES III). Public Health Nutr 13, 797–805.
Cossio S, Messiah SE, Garibay-Nieto N, Lopez-Mitnik G, Flores P,
Arheart KL et al. (2009). How do different indices of obesity
correlate with cardiometabolic disease risk factors in multiethnic
youths? Endocr Pract 15, 403–409.
Currie C, Samdal O, Boyce W, Smith B (2001). Health behaviours in
school-aged children: a World Health Organization cross-national
Health behaviors and central adiposity in children
R Lehto et al
846
European Journal of Clinical Nutrition
study. (eds). Research protocol for the 2001/2002 survey. Edinburgh:
Child and Adolescent Health Research Unit, University of
Edinburgh.
Delmas C, Platat C, Schweitzer B, Wagner A, Oujaa M, Simon C
(2007). Association between television in bedroom and adiposity
throughout adolescence. Obesity 15, 2495–2503.
Dubois L, Girard M, Potvin Kent M, Farmer A, Tatone-Tokuda F
(2009). Breakfast skipping is associated with differences in meal
patterns, macronutrient intakes and overweight among pre-school
children. Public Health Nutr 12, 19–28.
Goodman E, Hinden BR, Khandelwal S (2000). Accuracy of teen and
parental reports of obesity and body mass index. Pediatrics 106,
52–58.
Gorely T, Marshall SJ, Biddle SJ (2004). Couch kids: correlates
of television viewing among youth. Int J Behav Med 11,
152–163.
Halkjaer J, Sorensen TI, Tjonneland A, Togo P, Holst C, Heitmann BL
(2004). Food and drinking patterns as predictors of 6-year
BMI-adjusted changes in waist circumference. Br J Nutr 92, 735–748.
Hitze B, Bosy-Westphal A, Bielfeldt F, Settler U, Plachta-Danielzik S,
Pfeuffer M et al. (2009). Determinants and impact of sleep
duration in children and adolescents: data of the Kiel Obesity
Prevention Study. Eur J Clin Nutr 63, 739–746.
Isacco L, Lazaar N, Ratel S, Thivel D, Aucouturier J, Dore E et al.
(2010). The impact of eating habits on anthropometric charac-
teristics in French primary school children. Child Care Health Dev
36, 835–842.
Jackson DM, Djafarian K, Stewart J, Speakman JR (2009). Increased
television viewing is associated with elevated body fatness but not
with lower total energy expenditure in children. Am J Clin Nutr 89,
1031–1036.
Janssen I, Katzmarzyk PT, Boyce WF, Vereecken C, Mulvihill C,
Roberts C et al. (2005). Comparison of overweight and obesity
prevalence in school-aged youth from 34 countries and their
relationships with physical activity and dietary patterns. Obes Rev
6, 123–132.
Kahn HS, Imperatore G, Cheng YJ (2005). A population-
based comparison of BMI percentiles and waist-to-height
ratio for identifying cardiovascular risk in youth. J Pediatr 146,
482–488.
Klein-Platat C, Oujaa M, Wagner A, Haan MC, Arveiler D, Schlienger
JL et al. (2005). Physical activity is inversely related to waist
circumference in 12-y-old French adolescents. Int J Obes 29, 9–14.
Lanctot JQ, Klesges RC, Stockton MB, Klesges LM (2008). Prevalence
and characteristics of energy underreporting in African-American
girls. Obesity 16, 1407–1412.
Lazarou C, Soteriades ES (2010). Children’s physical activity, TV
watching and obesity in Cyprus: the CYKIDS study. Eur J Public
Health 20, 70–77.
Lee CM, Huxley RR, Wildman RP, Woodward M (2008). Indices of
abdominal obesity are better discriminators of cardiovascular risk
factors than BMI: a meta-analysis. J Clin Epidemiol 61, 646–653.
Lee S, Bacha F, Gungor N, Arslanian SA (2006). Waist circumference is
an independent predictor of insulin resistance in black and white
youths. J Pediatr 148, 188–194.
Lehto R, Ray C, Lahti-Koski M, Roos E (2010). Meal pattern and
BMI in 9–11-year-old children in Finland. Public Health Nutr Dec 6,
1–6.
Livingstone MB, Robson PJ (2000). Measurement of dietary intake in
children. Proc Nutr Soc 59, 279–293.
McCarthy HD, Ashwell M (2006). A study of central fatness using
waist-to-height ratios in UK children and adolescents over two
decades supports the simple message—‘keep your waist circum-
ference to less than half your height’. Int J Obes 30, 988–992.
McCarthy HD, Ellis SM, Cole TJ (2003). Central overweight and
obesity in British youth aged 11–16 years: cross sectional surveys
of waist circumference. BMJ 326, 624.
Moreno LA, Fleta J, Sarria A, Rodriguez G, Gil C, Bueno M (2001).
Secular changes in body fat patterning in children and adolescents
of Zaragoza (Spain), 1980–1995. Int J Obes Relat Metab Disord 25,
1656–1660.
Nambiar S, Truby H, Abbott RA, Davies PS (2009). Validating the
waist-height ratio and developing centiles for use amongst
children and adolescents. Acta Paediatr 98, 148–152.
Ortega FB, Ruiz JR, Sjostrom M (2007). Physical activity, overweight
and central adiposity in Swedish children and adolescents: the
European Youth Heart Study. Int J Behav Nutr Phys Act 4, 61.
Patel SR, Hu FB (2008). Short sleep duration and weight gain: a
systematic review. Obesity 16, 643–653.
Rampersaud GC, Pereira MA, Girard BL, Adams J, Metzl JD (2005).
Breakfast habits, nutritional status, body weight, and academic
performance in children and adolescents. J Am Diet Assoc 105,
743–760; quiz 761–762.
Rangul V, Holmen TL, Kurtze N, Cuypers K, Midthjell K (2008).
Reliability and validity of two frequently used self-administered
physical activity questionnaires in adolescents. BMC Med Res
Methodol 8, 47.
Ray C, Suominen S, Roos E (2009). The role of parents’ sense of
coherence in irregular meal pattern and food intake pattern of
children aged 10–11 in Finland. J Epidemiol Community Health 63,
1005–1009.
Ray C, Westerlund L, Roos E (2007). Do sleeping habits associate with
meal pattern among 10–11 years old children? Eur J Public Health
17, 143 (abstract).
Reilly JJ, Kelly J, Wilson DC (2010). Accuracy of simple clinical and
epidemiological definitions of childhood obesity: systematic
review and evidence appraisal. Obes Rev 11, 645–655.
Reinehr T, Wunsch R (2010). Relationships between cardiovascular
risk profile, ultrasonographic measurement of intra-abdominal
adipose tissue, and waist circumference in obese children. Clin
Nutr 29, 24–30.
Rexrode KM, Carey VJ, Hennekens CH, Walters EE, Colditz GA,
Stampfer MJ. et al. (1998). Abdominal adiposity and coronary
heart disease in women. JAMA 280, 1843–1848.
Roos E, Ray C, Stroor C (2009). Measuring children’s health
behaviours: comparing answers from 10- to 11year-old children
and their parents. In: 7th International Conference on Diet and
Activity Methods. Program and Abstracts:http://www.icdam.org/
ICDAM_Program_Abstracts_Book.pdf, accessed 27 January 2011.
Sherry B, Jefferds ME, Grummer-Strawn LM (2007). Accuracy of
adolescent self-report of height and weight in assessing over-
weight status: a literature review. Arch Pediatr Adolesc Med 161,
1154–1161.
Singh R, Martin BR, Hickey Y, Teegarden D, Campbell WW, Craig BA
et al. (2009). Comparison of self-reported, measured, metaboliz-
able energy intake with total energy expenditure in overweight
teens. Am J Clin Nutr 89, 1744–1750.
Smith KJ, Gall SL, McNaughton SA, Blizzard L, Dwyer T, Venn AJ
(2010). Skipping breakfast: longitudinal associations with cardio-
metabolic risk factors in the Childhood Determinants of Adult
Health Study. Am J Clin Nutr 92, 1316–1325.
Stevens J (1992). Applied multivariate statistics for the social sciences.
New Jersey: Lawrence Erlbaum Associates; 1992.
Sung RY, Yu CC, Choi KC, McManus A, Li AM, Xu SL et al. (2007).
Waist circumference and body mass index in Chinese children:
cutoff values for predicting cardiovascular risk factors. Int J Obes
31, 550–558.
Szajewska H, Ruszczynski M (2010). Systematic review demonstrating
that breakfast consumption influences body weight outcomes
in children and adolescents in Europe. Crit Rev Food Sci Nutr 50,
113–119.
Timlin MT, Pereira MA, Story M, Neumark-Sztainer D (2008).
Breakfast eating and weight change in a 5-year prospective
analysis of adolescents: Project EAT (Eating Among Teens).
Pediatrics 121, e638–e645.
Vandewater EA, Huang X (2006). Parental weight status as a
moderator of the relationship between television viewing and
childhood overweight. Arch Pediatr Adolesc Med 160, 425–431.
Health behaviors and central adiposity in children
R Lehto et al
847
European Journal of Clinical Nutrition
Vereecken C, Dupuy M, Rasmussen M, Kelly C, Nansel TR,
Al Sabbah H et al. (2009). Breakfast consumption and its
socio-demographic and lifestyle correlates in schoolchildren in
41 countries participating in the HBSC study. Int J Public Health 54
(Suppl 2), 180–190.
Vereecken CA, Maes L (2003). A Belgian study on the reliability
and relative validity of the Health Behaviour in School-Aged
Children food-frequency questionnaire. Public Health Nutr 6,
581–588.
Westerlund L, Ray C, Roos E (2009). Associations between sleeping
habits and food consumption patterns among 10–11-year-old
children in Finland. Br J Nutr 102, 1531–1537.
Wildman RP, Gu D, Reynolds K, Duan X, Wu X, He J (2005). Are waist
circumference and body mass index independently associated
with cardiovascular disease risk in Chinese adults? Am J Clin Nutr
82, 1195–1202.
YuY,LuBS,WangB,WangH,YangJ,LiZet al. (2007). Short sleep
duration and adiposity in Chinese adolescents. Sleep 30, 1688–1697.
Health behaviors and central adiposity in children
R Lehto et al
848
European Journal of Clinical Nutrition
