Mediterranean diet, overweight and body
composition in children from eight European
countries: Cross-sectional and prospective
results from the IDEFICS study
G. Tognona,*,1, A. Hebestreitb,1, A. Lanferb,1, L.A. Morenoc,1,
V. Palad,1, A. Sianie,1, M. Tornaritisf,1, S. De Henauwg,1,
T. Veidebaumh,1, D. Molna ´ri,1, W. Ahrensb,j,1, L. Lissnera,1
aPublic Health Epidemiology Unit, Department of Public Health and Community Medicine,
University of Gothenburg, Gothenburg, Sweden
bLeibniz Institute for Prevention Research and Epidemiology - BIPS GmbH, Bremen, Germany
cGENUD (Growth, Exercise, Nutrition and Development) Research Group, University of Zaragoza,
dNutritional Epidemiology Unit, Department of Preventive & Predictive Medicine, Fondazione IRCCS
Istituto Nazionale dei Tumori, Milan, Italy
eInstitute of Food Sciences, Unit of Epidemiology and Population Genetics, National Research Council,
fResearch and Education Institute of Child Health, Strovolos, Cyprus
gDepartment of Public Health, Ghent University, Faculty of Medicine and Health Sciences, Ghent,
hNational Institute for Health Development, Tallin, Estonia
iDepartment of Pediatrics, Medical Faculty, University of Pe ´cs, Hungary
jUniversity of Bremen, Bremen, Germany
Received 26 October 2012; received in revised form 17 April 2013; accepted 27 April 2013
Available online 17 July 2013
inversely associated with many diseases, but its role in early obesity prevention is not clear.
We aimed to determine if this pattern is common among European children and whether it
is associated with overweight and obesity.
Background & aims: A Mediterranean-like dietary pattern has been shown to be
Abbreviations: fMDS, frequency-based Mediterranean Diet Score; OR, Odds ratio; BMI, Body Mass Index.
* Corresponding author. Department of Public Health and Community Medicine, University of Gothenburg, Sahlgrenska Academy, Box 454,
SE 405 30 Go ¨teborg, Sweden. Tel.: þ46 31 7866541; fax: þ46 31 7781704.
E-mail address: email@example.com (G. Tognon).
1On behalf of the IDEFICS Consortium.
0939-4753/$ - see front matter ª 2013 Elsevier B.V. All rights reserved.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nmcd
Nutrition, Metabolism & Cardiovascular Diseases (2014) 24, 205e213
Change in adiposity
Methods and results: The IDEFICS study recruited 16,220 children aged 2e9 years from study
centers in eight European countries. Weight, height, waist circumference, and skinfolds were
measured at baseline and in 9114 children of the original cohort after two years. Diet was eval-
uated by a parental questionnaire reporting children’s usual consumption of 43 food items.
Adherence to a Mediterranean-like diet was calculated by a food frequency-based Mediterra-
nean Diet Score (fMDS).
The highest fMDS levels were observed in Sweden, the lowest in Cyprus. High scores were
inversely associated with overweight including obesity (OR Z 0.85, 95% CI: 0.77; 0.94) and
percent fat mass (b Z ?0.22, 95% CI: ?0.43; ?0.01) independently of age, sex, socioeconomic
status, study center and physical activity. High fMDS at baseline protected against increases in
BMI (OR Z 0.87, 95% CI: 0.78; 0.98), waist circumference (OR Z 0.87, 95% CI: 0.77; 0.98) and
waist-to-height ratio (OR Z 0.88, 95% CI: 0.78; 0.99) with a similar trend observed for percent
fat mass (p Z 0.06).
Conclusions: Although a Mediterranean dietary pattern is inversely associated with childhood
obesity, it is not common in children living in the Mediterranean region and should therefore be
advocated as part of EU obesity prevention strategies.
ª 2013 Elsevier B.V. All rights reserved.
Greater adherence to a Mediterranean-like dietary pattern
is associated with a significant improvement in health status
. It may reduce the metabolic syndrome risk , major
chronic disease morbidity [3,4] and mortality  as well as
total mortality [6,7]. An inverse association between this
pattern and overweight, particularly in childhood, could
have short- and long-term health implications, but only few
papers describe its relationship with childrens’ weight sta-
tus, BMI, waist circumference or waist-to-height ratio
Dietary patterns in south Europe are changing, especially
among the young . Animal products and fats are increas-
ingly consumed, while the intake of vegetable-based food-
stuffs declines . The IDEFICS study has shown that taste
preference for fat and sugar are associated to children’s
high-risk TV behaviour, also associated to overweight .
Several indices to assess adherence to a Mediterranean-
like dietary pattern exist for adults, while for children the
KIDMED index has been used in association with obesity
indicators [8,10,11]. We propose a modified version of the
Mediterranean Diet Score (fMDS), adapted to the IDEFICS
food frequency questionnaire. We aimed to evaluate
whether a Mediterranean-like dietary pattern (i.e. rich in
cereals, vegetables, fruit and nuts, fish and low in meat and
dairy products) is adopted by European young children and
in which geographical areas. We examined both cross-
sectionally and prospectively whether a high adherence
level to this pattern was associated with overweight or
Study population and design
IDEFICS, an 8-centre intervention study including a non-
representative sample of 16,220 children (2e9 years) from
eight different European countries (Sweden, Germany,
Hungary, Italy, Cyprus, Spain, Belgium, Estonia), aimed to
investigate the risks and long-term consequences of over-
weight and obesity in children. Started at the end of 2006
and run for five years, it also offered health promotion
activities in kindergartens and schools.
The baseline survey (2007e2008), was followed by an
intervention phase and repeated after two years on 9114
children. The intervention was conducted in selected re-
gions of each country with a similar region serving as the
control population where no obesity prevention strategies
were implemented. Focused on diet, physical activity and
stress-coping capacity, it was community-based and inte-
grated throughout various levels of society, for a total of 10
modules: three at community level, six at school level and
one for parents. All materials for the interventions were
centrally developed and culturally adapted .
assessed from a standardized parental questionnaire
including both the highest education attained in the
household and the highest income level. To facilitate cross-
country comparisons of education, questions were trans-
formed to the International Standard Classification of Ed-
ucation (ISCED) . Country-specific income levels were
assigned with reference to the average net equivalence
income, considering the median income and poverty line.
Sampling and study basic characteristics have been
described in detail elsewhere .
The parents also reported both the time their children
spent playing outdoors and in sport clubs . Daily means
from the total value were calculated for each child and
considered a proxy variable for children’s physical activity.
Dietary assessment and fMDS definition
A baseline parental questionnaire on diet together with
physical examination data were collected for 14,972 chil-
dren (49.1% girls). The dietary questionnaire (self-admin-
istered Children’s Eating Habits Questionnaire, CEHQ-FFQ),
intended to cover only food under parental control, was
used to assess frequency of consumption of 43 foods and
other diet-related habits. The same general description of
206G. Tognon et al.
each food group was used in every study center, although
country-specific food examples for the given food items
were also used. The questionnaire was tested before the
baseline examination in all the study centers and repro-
ducibility was assessed in a sub-sample of parents, who
filled in a second questionnaire. Consumption frequencies
of the two questionnaires were compared, showing repro-
ducible estimates . The relative validity of the food
questionnaire was also performed, comparing estimates
from CEHQ-FFQ with the mean intakes from two separate
24-h dietary recalls. Moderately strong correlations were
observed for the most frequently consumed foods .
Since the questionnaire did not include quantitative in-
takes, adherence to a Mediterranean-like diet was assessed
by a food frequency-based Mediterranean Diet Score
(fMDS). Daily frequencies from different food groups (see
below) were divided by the total daily frequency of all food
items included in the CEHQ-FFQ to obtain relative fre-
quencies of each food group. The latter values were cate-
gorized according to sex-specific and age-specific (6e9
years/2e5 years) medians (calculated on the whole
cohort), to identify high and low intakes. One point was
given for intakes higher than the median relative frequency
for (1) vegetables and legumes (cooked/raw vegetables,
potatoes, legumes and vegetable-based meat replacements
such as tofu, tempeh, etc.), (2) fruit and nuts (fruit added
or not with sugar, nuts, seeds, dried fruit), (3) cereals
(breakfast cereals, white and wholemeal bread, rolls and
crispbread, pasta, noodles and rice) or (4) fish (fresh or
frozen fish, and fried fish and fish fingers) and one point if
intakes were below the median for: (5) dairy (milk, yoghurt,
cheese including spreadable cheese) and (6) meat products
(fried and non-fried meat, hamburgers, falafel, kebab,
etc.). The final fMDS added up to a maximum of 6 points.
High adherence levels to a Mediterranean-like dietary
pattern were considered for subjects characterized by an
fMDS > 3.
A sub-sample of parents (N Z 9082) also completed a
single 24-h self-administered and computer-assisted di-
etary recall of their children’s diet (Self-Administered
Children and Infant Nutrition Assessment, SACINA). SACINA
has been recently validated with doubly labelled water,
showing to be a valid instrument to assess energy intake on
group level . It includes pictures of portion sizes of a list
of most common country-specific foods and probing ques-
tions for usual combinations, e.g. cereals and milk or bread
and butter. Fatty acid intakes, estimated from SACINA
based on country-specific food databases, were used to
calculate a ratio between unsaturated (MUFA þ PUFA) and
saturated (SFA) fatty acids, used for supplementary
Anthropometric measures were taken at both examina-
tions, with children wearing underwear and T-shirt. Weight
was measured at the nearest 0.1 kg by means of a TANITA
BC 420 SMA electronic scale. Waist circumference was
countries differences in fMDS.
Percent of subjects with high/low intakes of each food item included in the fMDS, high fMDS levels (>3) and between
(N Z 1744)
(N Z 2200)
(N Z 1979)
(N Z 1460)
(N Z 2486)
(N Z 1820)
(N Z 1644)
(N Z 1637)
Percent of subjects with high intakesaof:
Percent of subjects with low intakesbof:
61.147.0 49.672.0 50.744.855.8 52.8
11.611.8 6.712.86.016.4 34.524.7
Proportion of children with a high adherence to the Mediterranean dietd:
% fMDS > 356.737.5 35.1 31.8 32.2 32.726.3 24.2
Difference (95% CI) with fMDS in Swedenc
aAbove sex-specific and age-specific median.
bBelow sex-specific and age-specific median.
cEstimated by means of an analysis of variance adjusted for age, parental education and income as well as sex, coupled to a post-hoc
analysis based on the Dunnet test, in order to estimate differences to the study center characterized by the highest fMDS (Sweden,
score Z 4.2) versus all the others.
dNumbers of children can slightly vary by each analysis.
Mediterranean diet and overweight in European children.207
measured at the midpoint between the iliac crest and the
lower coastal border or tenth rib and recorded at the
nearest 0.1 cm. Triceps and subscapular skinfold thickness
were measured twice on the right side using a Holtain
caliper , and mean values were calculated. Percent fat
mass was estimated by means of Slaughter equations ,
which have been recently shown to be a reliable way to
estimate percent fat mass :
Boys : 0:783 ? ðtriceps þ subscapularÞ ? 1:7
Girls : 0:546 ? ðtriceps þ subscapularÞ þ 9:7
In cross-sectional analyses, dichotomous outcomes
were defined as: WtHR > 0.5 (a value considered higher
than normal ) as well as overweight plus obesity
based on Body Mass Index (BMI) cut-offs calculated by
Cole et al. in the International Obesity Task Force (IOTF)
database which includes 0e25-year-old subjects from six
different countries . Continuous outcomes were
defined as waist circumference and percent fat mass. In
prospective analyses, dichotomous outcomes were the
highest quintile of change during follow up (versus all the
other quintiles) in: sex- and age-specific BMI; BMI
z-scores (according to Cole [28,29]); waist circumfer-
ence; WtHR; and percent fat mass. BMI change was
adjusted for sex and age by subtracting the change in
sex-specific and age-specific (6e9 years/2e5 years) me-
dians of BMI from each child’s observed BMI change for
the same time period.
by the highest adherence, followed by Italy and Germany, Spain, Hungary and Belgium (medium grey tone) and by Estonia and Cyprus
(light grey tone).
Ecological distribution of adherence to a Mediterranean-like dietary patterna.aThe eight IDEFICS countries were catego-
208 G. Tognon et al.
Differences in fMDS levels between study centers were
assessed by an analysis of variance, adjusted for sex, age,
parental education and income. A post-hoc analysis, using
the Dunnett test, was done in order to estimate differences
between the center characterized by the highest fMDS
(Sweden) and each of the other centers.
The cross-sectional association between high fMDS levels
and the dichotomous anthropometric outcomes (see above)
was assessed in a logistic regression analysis, adjusted for
the above-mentioned covariates plus study center. For
continuous outcomes, linear regression analyses were used.
Both center-specific and combined estimates of the asso-
ciation between overweight including obesity and fMDS > 3
were depicted in a forest plot. Effect modification of each
covariate included in the model with each exposure vari-
able was evaluated by including an interaction term be-
tween the exposureand
estimates were produced for each level of those variables
which significantly interacted with fMDS.
The prospective associations of baseline fMDS with the
highest sex-specific and age-specific quintiles of change
during follow up of different anthropometric outcomes
(BMI, BMI z-scores, waist circumference, WtHR and percent
fat mass) were tested by logistic regression. The latter was
adjusted for sex, baseline age, study center, parental in-
come and education, baseline age-specific and sex-specific
BMI z-scores (or baseline waist circumference in the ana-
lyses on waist circumference change, etc.). A further
adjustment for a variable describing inclusion or not in the
intervention study was also performed.
Sex, age, parental education and income were included
simultaneously in a logistic regression analysis adjusted
for centre and tested as possible determinants of
fMDS > 3. To assess possible differences in children eating
lunch at school (not covered by the food questionnaire),
an extra analysis was adjusted for the number of meals
consumed at home. Other sensitivity analyses included the
exclusion of subjects in the highest 5% of BMI, waist
circumference, WtHR or percent fat mass (according to
the analyses) and adjusting for having at least one immi-
grant parent, a possible determinant of different food
culture and socioeconomic status (Tables 1S and 2S).
No information about fatty acid subtypes was available
in the CEHQ. However, a ratio between unsaturated (MUFA
þ PUFA) and saturated fat (SFA) was calculated in the
subsample with 24 h recall data. A possible increasing trend
of this ratio was tested across different fMDS categories
(0e2; 3; 4e6) containing an equal number of subjects. P for
trend was calculated in a linear regression analysis,
adjusted for sex, age, study center as well as parental
education and income.
A final confirmatory analysis was done, adjusting for a
variable summarizing the average number of hours per day
the children spent either playing outdoors or in sport clubs,
as a proxy for physical activity. Values greater than an
average of 7 h/day (N Z 209) were considered implausible
Both children and parents gave their consent to the ex-
amination and could accept or refuse participation in any
single protocol component. The study protocol
approved by local ethics committees.
Sweden had the highest proportion of children with high
intake frequencies of vegetables, fruit plus nuts and cereals
(Table 1). 56.7% of Swedish children had a fMDS level > 3
followed by the Italians (37.5%) and the Germans (35.1%).
The lowest fMDS levels were observed in Cyprus (24.2% of
children with fMDS > 3). Between-center differences in the
fMDS > 3 versus fMDS ? 3) and different overweight indicators.
Models Odds ratios (95% CIs) of categorical outcomes
Cross-sectional associationabetween high adherence levels to a Mediterranean-like dietary pattern (categorized as
b (95% CIs) of continuous outcomes
Overweight including obesityb
WtHR > 0.5c
Baseline waist circumference (cm)Percent fat mass
(N Z 13,256)
0.83 (0.76; 0.91)***
0.83 (0.76; 0.91)***
0.83 (0.76; 0.91)***
0.83 (0.75; 0.91)***
0.85 (0.77; 0.94)**
(N Z 8964)
0.92 (0.83; 1.02)
0.92 (0.83; 1.02)
0.91 (0.82; 1.01)
0.97 (0.87; 1.08)
1.00 (0.89; 1.12)
(N Z 8964)
?0.48 (?0.76; ?0.20)***
?0.48 (?0.76; ?0.20)***
?0.47 (?0.72; ?0.22)***
?0.26 (?0.51; ?0.01)*
?0.20 (?0.46; 0.06)
(N Z 8685)
?0.42 (?0.70; ?0.14)**
?0.34 (?0.54; ?0.13)**
?0.34 (?0.54; ?0.13)**
?0.27 (?0.48; ?0.07)**
?0.22 (?0.43; ?0.01)*
*p ? 0.05, **p ? 0.01, ***p ? 0.001.
aFrom logistic (overweight including obesity and WtHR) and linear regression (waist circumference and percent fat mass) analyses.
bDefined according to Cole  and compared with the rest of the population.
cCompared with subjects characterized by WtHR ? 0.5.
eAdjusted for sex.
fAlso adjusted for age.
gAlso adjusted for study center.
hAlso adjusted for parental education and high parental income.
Mediterranean diet and overweight in European children. 209
score levels, measured by an analysis of variance, were
statistically significant across all centers (p < 0.0001).
Cypriot children showed the highest difference compared
to the Swedes (?1.16, 95% CIs: ?1.03; ?0.90). The
ecological distribution of fMDS levels (Fig. 1) indicated no
clear geographical pattern. Parental education was a
determinant of high fMDS levels (OR Z 1.21, 95% CI: 1.11;
1.32) and high parental income tended to the same direc-
tion (OR Z 1.07, 95% CI: 0.98; 1.17). Children’s age and
gender were not associated with fMDS.
The association of high fMDS levels with overweight
including obesity, WtHR, waist circumference and percent
fat mass was tested, adjusting for an increasing number of
covariates, i.e. sex, age, study center and socioeconomic
status (evaluated as parental education and family in-
come). After adjustment for covariates, the association
remained stable for both overweight including obesity
(OR Z 0.85, 95% CIs: 0.77; 0.94) and percent fat mass
(b Z ?0.22%, 95% CIs: ?0.43; ?0.01) (Table 2). Center-
specific estimates (Fig. 2, panel A), showed that the in-
verse association between high levels of fMDS (>3) and
overweight was statistically significant only in Hungarian
children (OR Z 0.73, 95% CIs: 0.57; 0.93). No cross-
sectional association was found with a WtHR > 0.5
(OR Z 1.00, 95% CIs: 0.89; 1.12). An inverse association
with waist circumference
significance when the analyses were adjusted for parental
income and education.
A statistically significant interaction was found between
fMDS and age (p Z 0.03), with a significant association with
percent fat mass only in school aged (?6 years) (b Z ?0.38,
95% CI: ?0.72; ?0.05) and not in pre-school aged children.
The fMDS was inversely associated with the highest change
in all outcomes with the association of percent fat mass
(p Z 0.06). Results from all models were unchanged when
adjusting for a variable describing inclusion in the inter-
vention study. Center-specific estimates of the association
between diet and the highest quintile of BMI z-score change
(Fig. 2, panel B) showed that only in Belgian children the
association was significant (OR Z 0.59, 95% CI: 0.41; 0.85)
fully adjusted model
Sensitivity analyses were also run (Tables 1S and 2S).
Adjustment for the number of meals consumed at home
(thus under parental control) or for having one immigrant
parent did not affect either cross-sectional or longitudinal
associations nor did the exclusion of subjects with the
highest 5% BMI, waist circumference, WtHR or percent fat
mass (according to the outcome analyzed).
A positive statistically
found (p for trend 0.001), showing an increasing proportion
of unsaturated vs saturated fats at higher score levels.
The adjustment for the average number of hours/day
the children spent either playing outdoors or in sport clubs,
confirmed all the previous associations. The prospective
association with percent fat mass change became statisti-
p Z 0.045).
95% CI: 0.76;0.98,
Adherence to a Mediterranean-like dietary pattern was
assessed through a modified version of the Mediterranean
diet score used in adults, based on food intake frequencies.
To date, two papers in children found no association
timates, of the association between high fMDS levels (>3) and
either baseline overweight including obesity (cross-sectional
analyses, panel A) or the highest sex-specific and age-specific
quintile of change in BMI z-scores (longitudinal analyses,
panel B)a.aThe results were obtained from logistic regression
analyses adjusted for sex, age, center and parental education
level and income (plus baseline BMI and a variable describing
inclusion in the intervention study in longitudinal analyses).
Center-specific and combined center-adjusted es-
210G. Tognon et al.
between this pattern and weight or BMI [8,9], while two
others suggested an association with BMI  or other
adiposity indicators . Three of these studies were based
on the KIDMED index, recently developed by a Spanish
group  and based on food frequencies as fMDS. KIDMED
includes additional questions mainly related to breakfast
and not necessarily related to a Mediterranean pattern
which were not included in the fMDS.
Studying internationally what was previously investi-
gated at a national level, we found that a Mediterranean-
like dietary pattern at baseline was inversely associated
with baseline obesity and percent fat mass and with a 2-
year change in BMI, BMI z-scores, waist circumference and
WtHR. These results were robust and not affected by
adjustment for different potential confounders. Cross-
sectionally, WtHR was unrelated to fMDS, while the asso-
ciation with waist circumference did not resist to adjust-
ment for socioeconomic status. Diet at baseline predicted
changes in both. This dietary pattern seems to be inversely
associated with overweight and obesity, but not clearly
associated with fat distribution per se.
Children from Mediterranean countries have the highest
overweight and obesity prevalence in Europe , possibly
linked to a “westernized” diet  and lifestyle . The
traditional Mediterranean diet has probably been aban-
doned by Cypriot children who have gradually adopted
western-like dietary habits . Although this research was
not based on representative population samples, the high-
est overweight and obesity levels were indeed observed in
the three Mediterranean cohorts .
The study’s strengths include that IDEFICS is one of the
largest European children’s cohort established to date,
comprising standardized measurements in eight centers
from different European countries with a 2-year follow up
for prospective analyses. However, the fact that low-
income groups were not equally distributed in all centers
could have limited the possibility to adjust for confounding
by socioeconomic status. Other limitations include that the
CEHQ-FFQ was limited to 43 items, with the risk to under-
estimate intakes, and that it was not designed to assess
portion sizes, which might have led to further over- or
underestimation. We measured a relative adherence to a
pattern characterized by a frequent intake of those food
items that are representative of a Mediterranean dietary
pattern and a reduced frequency of consumption of animal
products. Additional supplementary analyses showed a
positive trend in the ratio between unsaturated vs satu-
rated fats (calculated from 24-h recall data) across
increasing fMDS levels.
In conclusion, using a measure of relative adherence to a
Mediterranean-like diet, it appears that this pattern is not
common in Mediterranean children, while the Swedish
children paradoxically showed the highest adherence
levels. This pattern was inversely associated with over-
weight including obesity, both cross-sectionally and longi-
tudinally. Promotionof dietary
Mediterranean dietary pattern should thus be considered
for inclusion in EU obesity prevention strategies.
All authors have no financial disclosures and no conflict of
The authors’ responsibilities were as follows: GT performed
the data analysis and wrote the manuscript, AH, AL, LAM,
TV, DM, AS, SDH, VP, MT, WA supported the statistical an-
alyses and their interpretation and provided comments on
the manuscript, LL supervised this research, contributed to
the interpretation of results and to the writing of the
manuscript, WA coordinated the IDEFICS study.
This work was carried out as part of the IDEFICS study
(www.idefics.eu). We acknowledge the financial support of
the European Union within the Sixth RTD Framework Pro-
gram Contract No. 016181 (FOOD), by the Swedish Council
on Working Life and Social Research (FAS) EpiLife Center
and the Swedish Research Council. AL received funding
from the Hans-Bo ¨ckler-Stiftung.
Mediterranean-like dietary pattern and the highest quintile of change from baseline to follow up in age, BMI z-score, WtHR,
waist circumference and percent fat mass.
Odds ratios and 95% confidence limits of the prospective associationabetween baseline high level of adherence to a
Models Highest quintile
of BMI z-score change
of WtHR change
Highest quintile of waist
of percent fat mass changeb
(N Z 9196)(N Z 8796)(N Z 8796)(N Z 8387)
0.81 (0.73; 0.90)***
0.86 (0.77; 0.96)**
0.87 (0.78; 0.98)*
0.82 (0.73; 0.91)***
0.86 (0.77; 0.96)**
0.88 (0.78; 0.99)*
0.82 (0.73; 0.91)***
0.85 (0.76; 0.95)**
0.87 (0.77; 0.98)*
0.86 (0.77; 0.96)**
0.87 (0.78; 0.98)**
0.89 (0.78; 1.00)yy
*p ? 0.05, **p ? 0.01, ***p ? 0.001. yyp Z 0.06.
aFrom logistic regression analyses.
bMean difference at follow up: 1.8% ? 3.7.
cAdjusted for baseline BMI z-scores, WtHR, waist circumference, fat mass (according to the analyses).
dAlso adjusted for sex, age and study center.
eAlso adjusted for parental income and education.
Mediterranean diet and overweight in European children.211
Appendix A. Supplementary data
Supplementary data related to this article can be found at
 Sofi F, Cesari F, Abbate R, Gensini GF, Casini A. Adherence to
Mediterranean diet and health status: meta-analysis. BMJ
 Kesse-Guyot E, Ahluwalia N, Lassale C, Hercberg S, Fezeu L,
Lairon D. Adherence to Mediterranean diet reduces the risk of
metabolic syndrome: a 6-year prospective study. Nutr Metab
Cardiovasc Dis 2012;May 25.
 Sofi F, Abbate R, Gensini GF, Casini A. Accruing evidence on
benefits of adherence to the Mediterranean diet on health: an
updated systematic review and meta-analysis. Am J Clin Nutr
 Dilis V, Katsoulis M, Lagiou P, Trichopoulos D, Naska A,
Trichopoulou A. Mediterranean diet and CHD: the Greek Eu-
ropean prospective investigation into cancer and nutrition
cohort. Br J Nutr Aug 2012;108(4):699e709.
 Hodge AM, English DR, Itsiopoulos C, O’Dea K, Giles GG. Does a
Mediterranean diet reduce the mortality risk associated with
diabetes: evidence from the Melbourne Collaborative Cohort
Study. Nutr Metab Cardiovasc Dis Sep 2011;21(9):733e9.
 Trichopoulou A, Kouris-Blazos A, Wahlqvist ML, et al. Diet and
overall survival in elderly people. BMJ311(7018):1457e60.
Does the Mediterranean diet predict longevity in the elderly? A
Swedish perspective. Age (Dordr) Sep 2011;33(3):439e50.
 Farajian P, Risvas G, Karasouli K, Pounis GD, Kastorini CM,
Panagiotakos DB, et al. Very high childhood obesity preva-
lence and low adherence rates to the Mediterranean diet in
Greek children: the GRECO study. Atherosclerosis Aug 2011;
 Jennings A, Welch A, van Sluijs EM, Griffin SJ, Cassidy A. Diet
quality is independently associated with weight status in
children aged 9e10 years. J Nutr Mar 2011;141(3):453e9.
 LydakisC, StefanakiE,
stiffness in children. Eur J Pediatr Sep 2012;171(9):1373e82.
 Schroder H, Mendez MA, Ribas-Barba L, Covas MI, Serra-
Majem L. Mediterranean diet and waist circumference in a
representative national sample of young Spaniards. Int J
Pediatr Obes Dec 2010;5(6):516e9.
 Van Diepen S, Scholten AM, Korobili C, Kyrli D, Tsigga M, Van
Dieijen T, et al. Greater Mediterranean diet adherence is
observed in Dutch compared with Greek university students.
Nutr Metab Cardiovasc Dis Jul 2011;21(7):534e40.
 Martinez E, Llull R, Del Mar Bibiloni M, Pons A, Tur JA.
Adherence to the Mediterranean dietary pattern among
Balearic Islands adolescents. Br J Nutr Jun 2010;103(11):
 Lanfer A, Knof K, Barba G, Veidebaum T, Papoutsou S, de
Henauw S, et al. Taste preferences in association with dietary
habits and weight status in European children: results from
the IDEFICS study. Int J Obes (Lond) Jan 2012;36(1):27e34.
 Lissner L, Lanfer A, Gwozdz W, Olafsdottir S, Eiben G,
Moreno LA, et al. Television habits in relation to overweight,
diet and taste preferences in European children: the IDEFICS
study. Eur J Epidemiol Sep 2012;27(9):705e15.
 De Henauw S, Verbestel V, Marlid S, Barba G, Bammann K,
Eiben G, et al. The IDEFICS community-oriented intervention
programme: a new model for childhood obesity prevention in
Europe? Int J Obes (Lond) Apr 2011;35(Suppl. 1):S16e23.
 UNESCO. International Standard Classification of Education
(ISCED). Available from: http://www.unesco.org/education/
 Ahrens W, Bammann K, Siani A, Buchecker K, De Henauw S,
Iacoviello L, et al. The IDEFICS cohort: design, characteristics
and participation in the baseline survey. Int J Obes (Lond) Apr
 Burdette HL, Whitaker RC, Daniels SR. Parental report of
outdoor playtime as a measure of physical activity in
preschool-aged children. Arch Pediatr Adolesc Med Apr 2004;
 Lanfer A, Hebestreit A, Ahrens W, Krogh V, Sieri S, Lissner L,
et al. Reproducibility of food consumption frequencies
derived from the Children’s eating habits questionnaire used
in the IDEFICS study. Int J Obes (Lond) Apr 2011;35(Suppl. 1):
 Bel-Serrat S, Mouratidou T, Pala V, Huybrechts I, Bo ¨rnhorst C,
Ferna ´ndez-Alvira JM, et al. Relative validity of the children’s
eating habits questionnaire-food frequency section among
young European children: the IDEFICS Study. Public Health
Nutr 2013;Jan 4:1e11.
 Bo ¨rnhorst C, Bel-Serrat S, Pigeot I, Huybrechts I, Ottavaere C,
Comparison of proxy-reported energy intakes with measured
energy expenditure. Clin Nutr 2013. [Epub ahead of print].
 Stomfai S, Ahrens W, Bammann K, Kova ´cs E, Ma ˚rild S,
Michels N, et al. Intra- and inter-observer reliability in
anthropometric measurements in children. Int J Obes (Lond)
Apr 2011;35(Suppl. 1):S45e51.
 Janz KF, Nielsen DH, Cassady SL, Cook JS, Wu YT, Hansen JR.
Cross-validation of the Slaughter skinfold equations for chil-
dren and adolescents. Med Sci Sports Exerc Sep 1993;25(9):
 Michels N, Huybrechts I, Bammann K, Lissner L, Moreno L,
Peeters M, et al. Caucasian children’s fat mass: routine
anthropometry v. air-displacement plethysmography. Br J
Nutr 2012;Oct 11:1e10.
 Weili Y, He B, Yao H, Dai J, Cui J, Ge D, et al. Waist-to-height
ratio is an accurate and easier index for evaluating obesity in
children and adolescents. Obesity (Silver Spring) Mar 2007;
 Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a
standard definition for child overweight and obesity world-
wide: international survey. BMJ May 6, 2000;320(7244):
 Cole TJ, Freeman JV, Preece MA. British 1990 growth refer-
ence centiles for weight, height, body mass index and head
circumference fitted by maximum penalized likelihood. Stat
 Cole TJ, Green PJ. Smoothing reference centile curves: the
LMS method and penalized likelihood. Stat Med Jul 1992;
 Serra-Majem L, Ribas L, Ngo J, Ortega RM, Garcı ´a A, Pe ´rez-
Rodrigo C, et al. Food, youth and the Mediterranean diet in
Spain. Development of KIDMED, mediterranean diet quality
index in children and adolescents. Public Health Nutr Oct
 Hakanen M, Lagstrom H, Kaitosaari T, Niinikoski H, Na ¨nto ¨-
Salonen K, Jokinen E, et al. Development of overweight in
an atherosclerosis prevention trial starting in early child-
hood. The STRIP study. Int J Obes (Lond) Apr 2006;30(4):
 Kontogianni MD, Vidra N, Farmaki AE, Koinaki S, Belogianni K,
Sofrona S, et al. Adherence rates to the Mediterranean diet
are low in a representative sample of Greek children and
adolescents. J Nutr Oct 2008;138(10):1951e6.
212G. Tognon et al.
 Rey-Lopez JP, Vicente-Rodriguez G, Biosca M, Moreno LA. Download full-text
Sedentary behaviour and obesity development in children and
adolescents. Nutr Metab Cardiovasc Dis Mar 2008;18(3):
 TornaritisM,Savva S,
Hadjigeorgiou C. The Cyprus experience. In: Matalas AL,
Zampelas A, Stavrinos V, Wolinsky I, editors. The Mediterra-
nean diet: constituents and health promotion. Washington,
DC: CRC Press; 2001.
 Hense S, Barba G, Pohlabeln H, De Henauw S, Marild S,
Molnar D, et al. Factors that influence weekday sleep duration
in European children. Sleep May 2011;34(5):633e9.
Mediterranean diet and overweight in European children.213