nutrients

Protocol
Effect of an Education Program on Nutrition
Knowledge, Attitudes toward Nutrition, Diet Quality,
Lifestyle, and Body Composition in Polish Teenagers.
The ABC of Healthy Eating Project: Design, Protocol,
and Methodology
Jadwiga Hamulka 1, * , Lidia Wadolowska 2 , Monika Hoffmann 3 , Joanna Kowalkowska 2
and Krystyna Gutkowska 4
1 Department of Human Nutrition, Faculty of Human Nutrition and Consumer Sciences,
Warsaw University of Life Science (SGGW-WULS), 159C Nowoursynowska Street, 02-787 Warsaw, Poland
2 Department of Human Nutrition, Faculty of Food Sciences, University of Warmia and Mazury in Olsztyn,
45F Sloneczna Street, 10-718 Olsztyn, Poland; [email protected] (L.W.);
[email protected] (J.K.)
3 Department of Functional Food, Ecological Food and Commodities, Faculty of Human Nutrition and
Consumer Sciences, Warsaw University of Life Science (SGGW-WULS), 159C Nowoursynowska Street,
02-787 Warsaw, Poland; [email protected]
4 Department of Organization and Consumption Economics, Faculty of Human Nutrition and Consumer
Sciences, Warsaw University of Life Science (SGGW-WULS), 159C Nowoursynowska Street, 02-787 Warsaw,
Poland; [email protected]
* Correspondence: [email protected]; Tel.: +48-22-593-71-12




Received: 28 August 2018; Accepted: 28 September 2018; Published: 5 October 2018


Abstract: To increase teenagers’ nutrition knowledge is an important target and has the
potential to improve their dietary habits and lifestyle while reducing incidences of obesity-related
non-communicable diseases throughout the whole lifespan. This study protocol presents the
general approach and details of an assessment of nutritional knowledge, attitudes toward nutrition,
diet quality, lifestyle and body composition that have been used to comprehensively evaluate the
cross-behavioral patterns covering dietary and lifestyle behaviors in Polish teenagers. The study was
designed in two paths as: a cross-sectional study (covering 1569 students) and an education-based
intervention study (464 students) with a 9-month follow-up. We describe a short form of the food
frequency questionnaire (SF-FFQ4PolishChildren) used to collect data and details of diet-related
and lifestyle-related education program, which was developed and implemented by academic
researchers involved in the study. We also describe details of the data development and statistical
analysis, including multidimensional methods of clustering variables to identify cross-behavioral
patterns covering diet and lifestyle. The results of the study will provide evidence-based support for
preventive health care to promote normal growth and development of young population and reduce
the risk of diet-related diseases in adulthood, by early shaping of adequate dietary and lifestyle
behaviors. In the future, well-tailored education programs addressed to teenagers can be created as
an important public health action, based on our results.

Keywords: adolescents; dietary habits; education program; hand grip strength; nutrition
knowledge; obesity; physical activity; public health intervention; sedentary lifestyle; three-factor
eating questionnaire

Nutrients 2018, 10, 1439; doi:10.3390/nu10101439 www.mdpi.com/journal/nutrients

Nutrients 2018, 10, 1439 2 of 23

1. Introduction
Overweight and obesity are among the most serious public health challenges of the 21st century [1–3].
The prevalence of childhood obesity is increasing in most regions of the world, in both developed
and developing countries [4,5]. Recent studies suggest that these trends are gradually stabilizing or
have already reached a plateau in some countries [6]; however, the problem remains significant [7,8].
According to the World Health Organization (WHO), in Europe in 2009–2010, on average, one in every
three children aged six to nine years was overweight or obese, while among children aged 11 and
15 years the prevalence of overweight and obesity was 11–33% and 10–23%, respectively [9]. It is
expected that if the current trend continues, there will be 20% or more obese children and adolescents in
over 30 countries around the world in 2025 [3]. Among the European countries, Poland has one of the
highest rates of prevalence of overweight in children and adolescents; 12–25% of Polish children and
young people are overweight with a tendency toward an increase in the prevalence of both overweight
and abdominal obesity [5,10–13].
Being overweight in early childhood has been shown to increase the chance of being obese in later
childhood and also leads to adulthood obesity [2,7,11,14]. It is estimated that around 55% of obese
children and 70% of obese adolescents will experience adult obesity [15], which increases their risk of
obesity-related non-communicable diseases, such as diabetes, cardiovascular disease, various types
of cancer, as well as premature mortality [16–19]. Therefore, prevention and treatment of childhood
obesity is of the utmost importance, given the significant health and social consequences both in the
short and long terms [20,21].
Obesity has a multi-factorial and multi-level etiology with many factors such genetic,
physiological, environmental, and socioeconomic, including gender, family affluence, and education
level [12,22–24]. Although genetic and biological factors are important, they cannot fully explain the
current global childhood obesity trends. Environmental and societal factors associated with food
intake and physical activity should be the primary focus for understanding the macro-level impact
on obesity [5,9,11,24]. Rapid changes in ways of spending leisure time, a decline in physical activity,
the pressure of peers, and the fashion of using electronic devices and almost unlimited access to
food create an ”obesogenic” environment with two main factors affecting the positive energy balance
and obesity—sedentary lifestyle and unhealthy diet [10,25–28]. Food marketing also plays a role
by increasing children’s exposure to obesogenic foods because children and adolescents are more
susceptible to food marketing than adults [29]. The availability of snack vending machines, bars, and
kiosks with snack food and soft drinks in schools was also associated with lower fruits and vegetable
consumption, higher intake of calories from total and saturated fat, and a higher body mass index
(BMI) z-score [30,31].
As childhood obesity has a multi-factorial background, it should be tackled at multiple levels,
including individual, household, school, and social [32,33]. Knowledge and beliefs related to health
can improve health behaviours, especially when they are part of a targeted intervention, according
to the Integrated Theory of Health Behaviour Change [34,35]. Improving nutrition knowledge can
be assumed to contribute to the enhancement of dietary habits and food choices in those exposed to
education-based intervention [32,33,36].
Since poor dietary habits tend to be carried over from childhood to adulthood [37], childhood and
adolescence are periods when good nutritional diet quality is important to establish healthy dietary
behaviours. The family environment is a well-known factor influencing the food consumption of
children, especially in the early years of life. The influence of peer environment, mass-media, celebrities
and schools (due to school nutrition programs) increases rapidly during adolescence [38]. In this
period, an awareness of one’s own sexuality and gender differences with regard to diet and health as
well as physical abilities arise. The precise indication of the age at which changes start in the perception
of oneself as an independent individual and food consumer is difficult, but the age of 10 (12) years
old (depending on gender) is considered a breakthrough [39]. Thus, early adolescence may be the
Nutrients 2018, 10, 1439 3 of 23

last moment for implementing sustained healthy school culture and nutrition education addressed to
young people before determining their relatively stable eating habits.
Schools are considered a primary setting for implementing education programs—they have
the unique potential of involving a wide population of children and teenagers. A health education
curriculum that highlights the importance of nutrition and physical activity can help students adopt
and maintain healthy lifestyles regarding eating and physical activity [40,41]. Across the world, there
is a wide range of nutrition education initiatives addressed to school children, with the participation of
schools, government and health promotion agencies delivering knowledge about diet components [42]
and specific education to prevent or manage dietary-related and lifestyle-related diseases.
In Poland, multi-disciplinary programs covering the implementation of multi-component
activities are mainly directed at preschool children (e.g., ToyBox-study, Program Eating Healthy,
Growing Healthy) [43–45]. While programs conducted in schools are focused on single topics,
e.g., culinary workshops, nutrition education classes or physical activity education (POL-HEALTH;
“Zachowaj Równowag˛e”), there is a lack of a comprehensive approach with well-documented scientific
research [46–48]. There have been relatively few comprehensive educational programs in Poland,
which have been combined with a scientific assessment of nutrition knowledge, dietary habits and
lifestyle. Therefore, our project is aimed at filling the gap, with the potential that our research
conducted on Polish teenagers can be interpreted in a wider scope for other European teenagers living
in similar conditions.
The aim of the current paper is to describe the design, methodology and approach of the study
which was based on “ABC of Healthy Eating” project, and also to present the overall characteristics
of the study sample as assessed at baseline, prior to the implementation of a diet-related and
lifestyle-related education program.

1.1. Research Objective and Hypothesis

Our study is focused on teenagers aged 11–13 years, taking into account that early adolescence
may be the last moment for implementing nutrition-related and lifestyle-related education before
determining their relatively stable habits. We hypothesized that education focused on improving
teenagers’ dietary and lifestyle behaviours can increase the level of nutrition knowledge and positively
influence attitudes toward nutrition, dietary behaviours, lifestyle, and body composition. It was
assumed that education should be multi-component and use various forms of education tailored to
teenagers’ needs and their perceptions. However, the education effect can decrease over time and the
rate of this change depends on the initial level of nutrition knowledge and socio-demographic factors.
There were two aims of the study:

• To assess the short- and medium-term effect of nutrition-related and lifestyle-related education
program on nutrition knowledge, attitudes toward nutrition, diet quality, lifestyle, and body
composition of Polish teenagers in a socio-demographic context.
• To determine the association between teenagers’ nutrition knowledge, attitudes toward nutrition,
dietary and lifestyle behaviours, body composition, and socio-demographic factors.

1.2. Study Design

Data related to the study were collected as a part of:

• A national multicenter “ABC of Healthy Eating” project—the 1st edition in 2015–2016 (“ABC of
Healthy Nutrition”) and the 2nd edition in 2016 (“ABC of Kids’ Nutrition”) (Figure S1).
• The own research of academic centers involved in the study, carried out in parallel with those two
editions of the project.

The study had a wider scope and went beyond the activities carried out under the “ABC of
Healthy Eating” project.
Nutrients 2018, 10, 1439 4 of 23

The study was carried out on students of selected elementary schools from urban, sub-urban, and
rural areas covering
Nutrients 2018, 10, xthe
FORentire territory of Poland. The schools (not randomly selected) were
PEER REVIEW located a
4 of 24
convenient distance from nine Polish universities in ten locations. In 2015–2016, the study was carried
located a convenient distance from nine Polish universities in ten locations. In 2015–2016, the study
out in schools located near following cities: Gdynia, Cracow, Lublin, Olsztyn, Poznan, Przasnysz,
was carried out in schools located near following cities: Gdynia, Cracow, Lublin, Olsztyn, Poznan,
Warsaw,Przasnysz,
Wroclaw, and inWroclaw,
Warsaw, 2016, inand
schools
in 2016,located near:
in schools Biala
located Podlaska,
near: Bialystok,
Biala Podlaska, Gdynia,
Bialystok, Gdynia,Cracow,
Lublin, Cracow,
Olsztyn, Warsaw,
Lublin, andWarsaw,
Olsztyn, Wroclaw and(Figure
Wroclaw 1A).
(Figure 1A).

(A) The study (B)The validation study

Figure Figure
1. The1.location
The location of of academic centers
academic centers involved
involved in the study
in the (A) or(A)
study the validation study (B). study
or the validation A: (B).
Universities (cities) involved in the study: Warsaw University of Life
A: Universities (cities) involved in the study: Warsaw University of Life Sciences, WULS-SGGW Sciences, WULS-SGGW
(Warsaw, Przasnysz); Gdynia Maritime University (Gdynia); Medical University in Bialystok
(Warsaw, Przasnysz); Gdynia Maritime University (Gdynia); Medical University in Bialystok
(Bialystok); Josef Pilsudski University of Physical Education in Warsaw—The branch in Biala
(Bialystok); Josef Pilsudski University of Physical Education in Warsaw—The branch in Biala Podlaskiej
Podlaskiej (Biala Podlaska); University of Agriculture in Krakow (Cracow); University of Life Sciences
(Biala Podlaska); University
in Lublin (Lublin); of Agriculture
University in Krakow
of Life Sciences in Poznan(Cracow); University
(Poznan); University of Life and
of Warmia Sciences
Mazury in Lublin
(Lublin); University of Life Sciences in Poznan (Poznan); University of
in Olsztyn (Olsztyn); Wroclaw University of Environmental and Life Sciences (Wroclaw). B: Warmia and Mazury in
Universities (cities) involved in the validation study: Warsaw University of Life
Olsztyn (Olsztyn); Wroclaw University of Environmental and Life Sciences (Wroclaw). B: Universities Sciences, WULS-
SGGW (Warsaw,
(cities) involved in theTarnobrzeg);
validation Gdynia
study: Maritime
WarsawUniversity
University(Gdynia);
of LifeMedical
Sciences,University in Bialystok
WULS-SGGW (Warsaw,
(Bialystok); Josef Pilsudski University of Physical Education in Warsaw—The branch in Biala
Tarnobrzeg); Gdynia Maritime University (Gdynia); Medical University in Bialystok (Bialystok); Josef
Podlaskiej (Biala Podlaska); University of Agriculture in Krakow (Cracow); University of Life Sciences
Pilsudski University of Physical Education in Warsaw—The branch in Biala Podlaskiej (Biala Podlaska);
in Lublin (Lublin); University of Life Sciences in Poznan (Poznan); University of Warmia and Mazury
University of Agriculture
in Olsztyn (Olsztyn);in Krakow
Wroclaw (Cracow);
University University of and
of Environmental LifeLife
Sciences
Sciencesin(Wroclaw);
Lublin (Lublin); University
University
of Life Sciences
of Economyin (Bydgoszcz);
Poznan (Poznan); University
West Pomeranian of Warmia
University and Mazury
of Technology in Olsztyn (Olsztyn); Wroclaw
(Szczecin).
University of Environmental and Life Sciences (Wroclaw); University of Economy (Bydgoszcz); West
In April–May
Pomeranian University 2015of the research concept
Technology (Szczecin).and a short form of the food frequency questionnaire
(SF-FFQ4PolishChildren) dedicated to school children were developed. The main study was
In preceded
April–May by pilot
2015studies (in May 2015)
the research in which
concept and the research
a short procedure
form and tools
of the food were verified
frequency in
questionnaire
all centers, covering the 200 subjects above. Data collection for the main study started in June 2015.
(SF-FFQ4PolishChildren) dedicated to school children were developed. The main study was preceded
The data were collected and measurements were taken by well-trained researchers. The study started
by pilotatstudies
the same(in May
time 2015)
in all in involved
centers which the research
in the study. procedure and tools were verified in all centers,
covering theThe200study
subjects
was carried out in two paths: asfor
above. Data collection an the main studyintervention
education-based started in June 2015.
design and aThe data were
cross-
collected and measurements
sectional design. The study were taken according
is reported by well-trained researchers.
to the guidelines The study
of reporting started at
of observational the same
and
intervention studies of public health
time in all centers involved in the study. in nutritional epidemiology [49–51].
The study was carried out in two paths: as an education-based intervention design and
1.3. The Education-Based Intervention Study
a cross-sectional design. The study is reported according to the guidelines of reporting of observational
Two groups
and intervention studieswere established:
of public under
health intervention epidemiology
in nutritional (educated group)[49–51].
and without intervention
(control group) by accidental allocation of school classes to the educated or control group. In the
1.3. Theeducated group, a diet-related and lifestyle-related education program lasting three weeks was
Education-Based Intervention Study
implemented. The program consisted of 5 topics, each topic included various forms of education from
fungroups
Two to “scientific”
were cognition (Tableunder
established: 1). Each topic lasted approx.
intervention 180 min
(educated (4 h of
group) school
and lessons)
without and
intervention
was run by a minimum of 3–4 researchers. The program was developed and
(control group) by accidental allocation of school classes to the educated or control group. implemented by In the
academic researchers involved in the study. School teachers were not involved in the education
educated group, a diet-related and lifestyle-related education program lasting three weeks was
program, they were only present during educational activities. Apart from the study, students from
implemented. The program consisted of 5 topics, each topic included various forms of education from
fun to “scientific” cognition (Table 1). Each topic lasted approx. 180 min (4 h of school lessons) and was
run by a minimum of 3–4 researchers. The program was developed and implemented by academic
researchers involved in the study. School teachers were not involved in the education program, they
Nutrients 2018, 10, 1439 5 of 23

were only present during educational activities. Apart from the study, students from both educated
and control groups took part in regular school activities containing some content related to nutrition
and a healthy lifestyle.

Table 1. Topics and details of the education program.

Nutrition Topic
Goal Shaping pro-healthy dietary habits.
Nutrients important in the diet of young people. Health consequences of uncontrolled consumption of
Scope energy drinks and dietary supplements. A “Good snack” as an alternative to chips, sticks and sweets.
A Pyramid of Healthy Nutrition and Physical Activity [52] and dietary guidelines for teenagers.
Methods Talk; discussion; workshops.
Participants propose three various “healthy” breakfasts to take at school by selecting foods from a levels
Activities
of the Pyramid of Healthy Nutrition and Physical Activity; Participants prepare an “ideal sandwich”.
Tools Brochure; puzzles; crosswords; website.
Dietary topic
Goal Supporting well-being, physical, and intellectual development through a healthy lifestyle.
General recommendations for a healthy lifestyle (healthy eating and physical activity) at school age.
The influence of nutrition and physical activity on physical and intellectual development and healthy
Scope
well-being. The Student Menu—The brain cannot live only on chocolate—A discussion on the most
important nutrients in the diet of young people.
Methods Talk; discussion; workshops.
Using a pedometer to measure the number of calories consumed during the various activities.
Determination of time needed to “burn” the calories contained in the selected product. “Nutrition
Activities
detective”—measuring the amounts of oil, sugar and salt corresponding to fat, sucrose and salt content in
various foods (three sets).
Tools Brochure; puzzles; crosswords; website.
Sensory-consumer topic
Goal The world of senses. How to read the food labels? What is important when choosing food?
Recognition of the basic flavors in aqueous solutions. Discovering the different taste sensations in
Scope selected foods. “What kind of a consumer am I”—what do I like and why? Qualification of the
consumer’s personality.
Methods Talk; discussion; workshops.
Use of sensory memory to identify eight coded odor samples of natural spices, vegetables and fruits.
Activities Recognition of selected foods with masked/closed eyes, based on sensory perceptions in the mouth.
Preparation of colorful, tasty and healthy snacks from provided foods according to one’s own ideas.
Tools Brochure; recipes of ‘healthy’ snacks; website.
Hygiene topic
Goal Food safety. Hygiene during the preparation and consuming of meals.
The world of microorganisms, pathogens and probiotics. The rules of proper food storage and hygiene
Scope
during meal preparation and consumption.
Methods Talk; discussion; workshops.
Microscopic observation of selected microorganisms—Lactic acid bacteria (Lactobacillus) as an example of
a microorganism with healthy properties, E. coli (Escherichia coli) as an example of a pathogen. Mapping
Activities
the observed microorganisms’ cells in the prepared templates. Practicing proper hand washing according
to instructions. Checking hand cleanness with a test indicator, before and after hand washing.
Tools Brochure; microscope; test indicator of hand washing; puzzles; crosswords; website.
Culinary topic
Goal How to prepare healthy, cheap and tasty meals?
The impact of culinary processes on sensory quality, nutritional value and food safety. The phenomenon
Scope
of enzymatic browning of fruit and vegetables and ways to prevent this process.
Methods Talk; discussion; workshops.
Culinary experiments—preventing the darkening of fruits and vegetables. Checking the impact of storage
Activities temperature on the quality of frozen foods, e.g., vegetables. Preparing low-budget healthy meals and
low-sweetened beverages.
Tools A “healthy meals” recipe book; brochure; website.
Nutrients 2018, 10, 1439 6 of 23

The overall content of the intervention-based educational study is presented in Table 2. The data
was collected 4 times: (i) before education (at baseline), (ii) 3 weeks from baseline at the end of the
education program (in the educated group only), (iii) after 3 months (±2 weeks; 3-month follow-up)
to measure the short-term effect of education, (iv) after 9 months (±2 weeks; 9-month follow-up) to
measure the medium-term effect of education.

Table 2. The overall content of the education-based intervention study.

Group
Timing Activities
Educated Control

• Data collection with the

SF-FQ4PolishChildren questionnaire
(nutrition knowledge, attitudes A A
June 2015 towards nutrition, diet quality,
Before education (Baseline) lifestyle, socio-demographic factors).
• Measurements of body composition
(body weight, height, waist A A
circumference, hand grip strength).

• The implementation of diet-related

June 2015 and lifestyle-related education
Education program program with 5 topics: Nutrition, A NA
(lasting 3 weeks) Dietary, Sensory-consumer,
Hygiene, Culinary.

June 2015 • Data collection with the

After education SF-FQ4PolishChildren questionnaire A NA
(3 weeks from baseline) (nutrition knowledge).

• Data collection with the

September 2015 SF-FQ4PolishChildren questionnaire
After 3 months from baseline (nutrition knowledge, attitudes A A
(3-month follow-up) towards nutrition, diet quality,
lifestyle, socio-demographic factors).

• Data collection with the

SF-FQ4PolishChildren questionnaire
(nutrition knowledge, attitudes A A
March 2016
After 9 months from baseline towards nutrition, diet quality,
(9-month follow-up) lifestyle, socio-demographic factors).
• Measurements of body composition
(body weight, height, waist A A
circumference, hand grip strength).

SF-FQ4PolishChildren—A short form of food frequency questionnaire; A—applied; NA—not applied.

1.4. The Cross-Sectional Study

In this study design, pooled data were used (see Figure 2 in Section 2.2) covering the data collected
in June 2015 within the intervention-based educational study and June 2016 within the 2nd edition of
the project ‘ABC of Kids’ Nutrition’ (Table 3). In this database, all data were collected before education
program in those participants who have been subjected to the education, i.e., all participants from
educated group in 2015 (at baseline) and all participants in 2016. In the 2nd edition of the project,
no effect of education was measured and no control group was established.
Nutrients 2018, 10, 1439 7 of 23
Nutrients 2018, 10, x FOR PEER REVIEW 9 of 24

Figure 2. Flow chart of sample collection and study design.

Figure 2. Flow chart of sample collection and study design.
Table 3. The overall content of the cross-sectional study.
2.3. Questionnaire for Data Collection
Group
Timing Activities
Data related to nutrition knowledge, attitudes toward nutrition, diet,Educated lifestyle and socio-
Control
demographic factors were collected with 50-item, a short form of food frequency questionnaire (SF-
FFQ4PolishChildren)•developed Data collection with the SF-FQ4PolishChildren
by Kowalkowska, Wadolowska, and Hamulka for the “ABC of
questionnaire (nutrition knowledge, attitudes A A
Healthy Eating” project (Table S1; Table S2).
towards nutrition, diet quality, lifestyle,
AJune 2015
validation procedure of the short form of food
socio-demographic factors). frequency questionnaire (FFQ) was carried out,
Before education
regardless of this study, in cooperation with academic researchers from 12 universities covering the
• Measurements of body composition (body weight,
entire territory of Poland (data not
height, published,
waist paper in
circumference, preparation)
hand (Figure 1B). InAbrief, the internal
grip strength).
A
compatibility of the questionnaire was tested by 630 teenagers, aged 11–15. The questionnaire was
completed by teenagers •
twice (test and retest after 2 weeks). Fleiss’ Kappa for breakfast consumption
Data collection with the SF-FQ4PolishChildren
was 0.54, school meal consumption was
questionnaire 0.53, pro-Healthy
(nutrition knowledge,Diet Index categories A
attitudes (see in thisNA
section
below)June
was2016
0.44 and the non-Healthy Diet diet
towards nutrition, Index categories
quality, was 0.35, Family Affluence Scale items
lifestyle,
were to 0.83, nutritionsocio-demographic
0.69education
Before knowledge levelfactors).
was 0.36, screen time categories was 0.46 and physical
activity level was 0.52.• Compatible classification
Measurements of subjects (into
of body composition (bodythe same category in test and retest)
weight,
A NA
height,meal,
was for breakfast, 77.3%; school waist79.8%;
circumference, hand Diet
pro-Healthy grip Index,
strength).
75.8%; non-Healthy Diet Index,
92.0%; Family Affluence Scale items, 83.7% to 95.8%; nutrition knowledge level, 72.7%; screen time,
A—applied; NA—not applied.
64.8%; total physical activity level, 74.8%. The internal compatibility of the short form of FFQ
dedicated for teenagers was considered acceptable to good.
2. Materials and Methods
2.4. Nutrition Knowledge
2.1. Ethical Aspects
The nutrition knowledge level was determined based on eighteen questions (Q22 to Q39; Table
The project followed
S1). Participants the ethical
were asked standards
about nutrition recognized
based on questionsbydeveloped
the Declaration
by Whatiof Helsinki
[53] and was
and adapted
approved by conditions
to Polish the Bioethics Committee
and education of the
based Faculty
on the Polish of Medical
Food Sciences,[52].
Guide Pyramid University of Warmia
Correct answers
were scored
and Mazury with 1 point
in Olsztyn andJune
on 17 wrong or “don’t
2010, know”No.
Resolution answers and missing
20/2010. data were
The study was scored with 0to the
explained
participants before the start (verbally and by leaflets), to the pupils at the school lessons and to the
parents or legal guardians during the meetings. Parental or legal guardians’ written informed consent
was obtained for the participation of their children. The consent form was distributed for parents or
legal guardians during the meetings. All data regarding the study were collected at the time designated
Nutrients 2018, 10, 1439 8 of 23

for school classes instead of regular school lessons. Participants who refused to participate in the study
attended other school activities.

2.2. Participant Selection

A convenient sample selection was applied in both the education-based intervention study and
the cross-sectional study. A school class was the smallest unit in the sample selection. All students
from a given class were invited to take part in the study. Recruitment was conducted in selected
elementary schools (see in Section 1.2). Table 4 presents the criteria of schools and participants’
inclusion or exclusion.

Table 4. School and participant eligibility criteria.

School inclusion criteria:

• School located at a convenient distance from the academic centers (up to 50 km).
• The school master permitted their school to participate in the study.
• The school did not previously participate in other nutrition-health education programs.

Participant inclusion criteria—in general:

• 4th and 5th grade classes of elementary school.
• No disability, self-declared by parent or legal guardian or teacher.
• Written consent of parents or legal guardians to participate.

Participant exclusion criteria—for the education-based intervention study:

• Data cleaning: age <11 or >12 years,
• Not attending in all stages of the study with a 9-month follow-up, including all activities of the education
program lasting 3 weeks.

Participant exclusion criteria—for the cross-sectional study:

• Data cleaning: age <11 or >12 years in 2015, and age <11 or >13 years in 2016.

Students from 4th- and 5th-grade classes were invited to attend. The expected ages of the students
were 11–12 years in 2015 and 11–13 years in 2016 due to changes in the Polish law concerning the age
of starting education by children (obligatory or optional starting from six years). We decided to start
recruitment based on school classes because students were subject to the same school education and
would be at a similar stage of development. In 2015, 48 classes were invited (educated group/control
group: 32/16 classes) and in 2016, 68 classes were invited (Figure 2).
In the education-based intervention study, 668 students (educated/control: 405/263) were initially
recruited. Due to age (12/5) or not attending all stages of the study (74/113), 208 participants were
excluded from the analyses. Finally, the study included 464 teenagers (319/145) with a baseline age of
11–12 years, including 216 boys (46.6%) and 248 girls (53.4%).
In the cross-sectional study, 1,678 students were initially recruited. Due to age (in 2015/in 2016:
17/92), 109 participants were excluded from analyses. Finally, the study included 1569 teenagers aged
11–13 years, 760 boys (48.4%) and 809 girls (51.6%).

2.3. Questionnaire for Data Collection

Data related to nutrition knowledge, attitudes toward nutrition, diet, lifestyle and
socio-demographic factors were collected with 50-item, a short form of food frequency questionnaire
(SF-FFQ4PolishChildren) developed by Kowalkowska, Wadolowska, and Hamulka for the “ABC of
Healthy Eating” project (Tables S1 and S2).
A validation procedure of the short form of food frequency questionnaire (FFQ) was carried out,
regardless of this study, in cooperation with academic researchers from 12 universities covering the
entire territory of Poland (data not published, paper in preparation) (Figure 1B). In brief, the internal
compatibility of the questionnaire was tested by 630 teenagers, aged 11–15. The questionnaire was
Nutrients 2018, 10, 1439 9 of 23

completed by teenagers twice (test and retest after 2 weeks). Fleiss’ Kappa for breakfast consumption
was 0.54, school meal consumption was 0.53, pro-Healthy Diet Index categories (see in this section
below) was 0.44 and the non-Healthy Diet Index categories was 0.35, Family Affluence Scale items
were 0.69 to 0.83, nutrition knowledge level was 0.36, screen time categories was 0.46 and physical
activity level was 0.52. Compatible classification of subjects (into the same category in test and retest)
was for breakfast, 77.3%; school meal, 79.8%; pro-Healthy Diet Index, 75.8%; non-Healthy Diet Index,
92.0%; Family Affluence Scale items, 83.7% to 95.8%; nutrition knowledge level, 72.7%; screen time,
64.8%; total physical activity level, 74.8%. The internal compatibility of the short form of FFQ dedicated
for teenagers was considered acceptable to good.

2.4. Nutrition Knowledge

The nutrition knowledge level was determined based on eighteen questions (Q22 to Q39; Table S1).
Participants were asked about nutrition based on questions developed by Whati [53] and adapted
to Polish conditions and education based on the Polish Food Guide Pyramid [52]. Correct answers
were scored with 1 point and wrong or “don’t know” answers and missing data were scored with
0 points. Points were summed up for each respondent. Based on tertile distribution (data from the
cross-sectional study), the respondents were divided into three nutritional knowledge levels labeled as
follows: the lowest (0–4 points), moderately low (5–7 points), and higher (8–18 points).

2.5. Attitudes toward Nutrition

Based on a three-factor eating questionnaire (TFEQ13), dedicated to children, attitudes toward
nutrition were determined. The questionnaire was adapted and validated to the Polish conditions by
Dzielska et al. [54]. Originally, TFEQ13 consisted of 13 statements, but a weak understanding of three
statements was found during pilot studies. Therefore, we used a shorter version of the questionnaire
(TFEQ10) comprising 10 statements (Q12 to Q21; Table S1. According to Dzielska et al. [54], three
subscales, in shortened version, were developed: Emotional Eating (Q12, Q14), Uncontrolled Eating
(Q13, Q15, Q16, Q17, Q20), and Cognitive Restraint of Eating (Q18, Q19, Q21). The scores of each
subscale were calculated as a sum of points assigned to answers given by respondent. Answers were
scored as follows (points):

• Q12 to Q20: respondents could choose one of four answers: “definitely yes” (3 points), “rather
yes” (2 points), “rather not” (1 point), “definitely not” (0 points),
• Q21: respondents could choose one of eight points on the graphical scale: 1 or 2 (0 points), 3 or 4
(1 point), 5 or 6 with (2 points), 7 or 8 (3 points).

2.6. Diet Quality

The questionnaire was self-administered by teenagers in the classroom and supervised by
well-trained researchers. Explanations were given if necessary. The completion of the questionnaire
took the teenagers approx. 40 min.
Respondents were asked to specify their usual frequency consumption within the last 12 months
for breakfast and school meal and nine food items such as dairy products (e.g., milk, yogurt, cottage
cheese, cheese), fish (e.g., baked, smoked, fried, in vinegar, canned), vegetables (e.g., fresh, boiled,
baked, stewed), fruit (fresh or frozen), fruit or mixed fruit-veggie juices, fast foods (e.g., chips, pizza,
hamburgers), sweetened carbonated drinks (e.g., cola-type, water with syrup, tea-type with sugar),
energy drinks, sweets or confectionery (e.g., cookies, sweets, cake, chocolate bars, chocolate).
Breakfast was considered as consuming solid foods with or without beverages at the first eating
episode of a day [55,56]. Drinking only beverages was not considered as breakfast. Respondents could
choose one from four categories of breakfast consumption (number of days/week): 0/week, 1–3/week,
4–6/week, and 7/week. Some categories have been combined after distribution analysis. Finally,
breakfast consumption was analyzed in three categories as follows: every day (7/week), irregular
(4–6/week), and skipping (0–3/week).
Nutrients 2018, 10, 1439 10 of 23

The school meal was considered as consuming solid foods with or without beverages at the second
eating episode of a day while at school as lunch or a so-called second breakfast (more typical in Poland).
Drinking beverages only was not considered as a school meal. Respondents could choose one from four
categories of school meal consumption (number of days/week): 0/week, 1–2/week, 3–4/week, and
5/week. Some categories were combined after distribution analysis. Finally, school meal consumption
was considered in three categories as follows: every day (5/week), irregular (3–4/week), and skipping
(0–2/week).
Breakfast and school meal consumption considered together were analyzed in three categories as
follows: every day (breakfast: 7/week; school meal: 5/week), skipping (breakfast: 0–3/week; school
meal: 0–2/week), and irregular (other frequencies).
For food frequency consumption, respondents could choose one from the following seven
categories (converted into daily frequency, times/day): never or almost never (0 times/day), less
than once a week (0.06 times/day), once a week (0.14 times/day), 2–4 times/week (0.43 times/day),
5–6 times/week (0.79 times/day), every day (1 time/day), and several times a day (2 times/day).
Two diet quality scores were used as follows: a pro-Healthy Diet Index (pHDI) and a non-Healthy
Diet Index (nHDI). Both diet quality scores were established (a priori approach) on the basis of usual
food frequency consumption within the 12 last months. The pHDI included four food items as follows:
dairy products, fish, vegetables and fruit; the nHDI included four food items as follows: fast foods,
sweetened carbonated drinks, energy drinks and sweets or confectionery. Diet quality scores were
calculated by summing up the daily frequencies of four food items (as mentioned above) and expressed
in % points (range: 0 to 100). Two various ideas in categorizing diet quality scores (based on data from
the cross-sectional study) were applied:

• A posteriori approach—three levels of each diet quality score based on tertile distribution:
• pHDI: bottom tertile (<20.625% points), middle tertile (20.625–32.125% points), upper tertile
(≥32.125% points).
• nHDI: bottom tertile (<7.875% points), middle tertile (7.875–16.000% points), upper tertile
(≥16.000% points).
• A priori approach—three levels of each diet quality score: low (<33.33% points), moderate
(33.33–66.66% points), and high (≥66.66% points).

2.7. Lifestyle
Four measures were applied to assess lifestyle: screen time, total physical activity, physical activity
at school, and leisure time.
Using the question “How much time do you spend watching TV or in front of a computer on
an average day of the week?” (Q48; Table S1), screen time (ST) was assessed. The respondents could
choose one of six answers (with assigned scores): <2 h/day (0 points), 2 to <4 h/day (1 point), 4 to
<6 h/day (2 points), 6 to <8 h/day (3 points), 8 to <10 h/day (4 points), and ≥10 h/day (5 points).
Screen time expressed in points was calculated for each participant. The recommendation of the
American Academy of Pediatrics of a maximum of two hours of ST per day for children and youth
was used as a reference [57]. Three categories of screen time were established after combining some
answers: <2 h/day, 2 to 4 h/day, and ≥4 h/day.
Total physical activity (PA) was assessed using two questions regarding physical activity at
school and leisure time (Q49–Q50; Table S1). The respondents could choose one of three answers
describing their physical activity at school (low, moderate, vigorous) and leisure time (low, moderate,
vigorous). Many examples for each answer were given. Finally, after combining some categories of
both questions, the respondents were divided into three total physical activity levels: low, moderate
and high, with assigned scores from 0 point to 5 points (Table 5). Total PA expressed in points for each
participant was calculated. Vigorous PA at school combined with vigorous PA at leisure time (scored
with 5 points) was considered as adherence to the World Health Organization recommendation on
Nutrients 2018, 10, 1439 11 of 23

physical activity [58]. The WHO recommends that children and adolescents aged 5–17 years have a
minimum of 2.5 h/day of moderate-intensity PA. Finally, two categories of total physical activity were
considered: with the adherence to WHO recommendation on physical activity or no adherence.

Table 5. Categorizing and scoring (points) of total physical activity based on physical activity at school
and leisure time.

Physical Activity at Leisure Time

Physical Activity at School
Low Moderate Vigorous
Low Low (0) Low (1) Moderate (2)
Moderate Low (1) Moderate (3) Moderate (4)
High (5)
Vigorous Moderate (2) Moderate (4)
Adherence to WHO recommendation
Notes: Physical activity at school: low (most of the time in a sitting position, in class or on breaks), moderate (half the
time in a sitting position and half the time in motion), vigorous (most of the time on the move or on classes related
to high physical exertion). Physical activity at leisure time: low (more time spent sitting, watching TV, in front of a
computer, reading, light housework, a short walk to 2 h a week), moderate (walking, cycling, gymnastics, working
at home or other light physical activity performed 2–3 h/week), vigorous (cycling, running, working at home or
other sports activities requiring physical effort over 3 h/week).

2.8. Body Composition

The measurements of body weight (kg), height (cm), waist circumference (WC, cm), and hand
grip strength (HGS, kg) were taken with a precision of 0.1 kg, 0.1 cm, 0.1 cm, or 0.5 kg, respectively,
(between 8 and 12 a.m.) according to the ISAK International Standards for Anthropometric Assessment
guidelines [59]. Professional equipment and measuring tape were used, the same type across all the
research centers. Table 6 presents the details relating to procedures for body composition measurements.
Body mass index (BMI, kg/m2 ) and waist-to-height ratio (WHtR) were calculated. Body mass index
was categorized according to age-sex-specific BMI cut-offs for teenagers [60]. A BMI ≥25 kg/m2 was
used as the overweight measure (general adiposity) and WHtR ≥0.5 as the central obesity measure
(central adiposity) [61]. Z-scores of WC, BMI, and WHtR as well as HGS were calculated to achieve
mean equal 0 and standard deviation (SD) equal 1. Z-scores were categorized as follows: <−1, −1 to 1
and >1 SD.

Table 6. Procedures for body composition measurements.

Parameter (Units) Procedure, Accuracy, and Equipment

• Measurement with the head in horizontal Frankfort plane,

Height (H) (cm) • Recorded with a precision of 0.1 cm,
• A porTable Stadiometer (SECA 220, Hamburg, Germany).

• Measurement in light indoor clothes without shoes,

• Recorded with a precision of 0.1 kg,
Body weight (BW) (kg)
• Electronic digital scale—the same type across all research centers (SECA 799,
Hamburg, Germany).

• Measurement at the point midway between the iliac crest and the costal margin
(lower rib) on the anterior axillary line in a resting expiratory position,
Waist circumference (WC) (cm) • Recorded with a precision of 0.1 cm,
• A stretch-resistant tape that provides a constant 100 g tension (SECA 201,
Hamburg, Germany).

• Measurement in the standing position, the arm was allowed to move from 180◦
of flexion to near 0◦ with maximal effort,
Hand grip strength (HGS) (kg) • Recorded with a precision of 0.5 kg,
• hydraulic hand dynamometer—the same type across all research centres
(SAEHAN Corporation, Masan-Korea-type SH 5001).
Nutrients 2018, 10, 1439 12 of 23

2.9. Socio-Demographic Data

Respondents were divided into two categories of residence (rural or urban), based on question
Q43 (Table S1. The socioeconomic status was determined by using the Family Affluence Scale (FAS).
The FAS was a measure of material affluence derived from the characteristics of the household.
The scale was developed for the Health Behavior of School-aged Children (HBSC) cross-national study
including 44 countries. We used a scale composed of four questions described by the Polish HBSC
study team [62]. For each answer, points were assigned (Table S1). Some of the answers were scored
with the same number of points.

• Q44: “Does your family own a car, van or truck?” Answers: no (0 points); yes, one (1 point); yes,
two or more (2 points).
• Q45: “During the past year, how many times did you travel away on holiday with your family?”
(necessary examples were given) Answers: not at all (0 points); once (1 point); twice (2 points);
more than twice (2 points).
• Q46: “Do you have your own bedroom for yourself?” Answers: no (0 points); yes (1 point).
• Q47: “How many computers, laptops or tablets do your family own?” Answers: none (0 points);
one (1 point); two (2 points); more than two (2 points).

The points were summed up for each respondent (range: 0 to 7). Based on quartile distribution,
the respondents were divided into three FAS categories labeled as: low (0–4 points; <25th quartiles),
moderate (5–6 points) and high (7 points; ≥75th quartiles).

2.10. Sample Size Calculation

Two calculations of minimum sample size were made separately for two study designs.
For the education-based intervention study, the minimum sample size was calculated in regards
to nutrition knowledge as a key measure of the study. The calculation was based on the expected
increase in nutrition knowledge score after the 9-month follow-up and the expected difference
between the educated and control groups. With 5% significance level and 80% power, the sample
size required approximately 304 respondents (assuming 2/1 ratio for educated/control group, i.e.,
200/104 respondents and an equal number of respondents in each research center) at the end of the
study, to detect a 50% difference between the educated group (increase by 30%) and the control group
(increase by 15%) in the increase of nutrition knowledge score, including a 50% drop out rate (at the
end of the study) and recoding error or missing data of 10%.
For the cross-sectional study, the minimum sample size was calculated in regards to consuming
breakfast, which we chose as a key dietary habit and overweight occurrence. The calculation was based
on the expected proportion of respondents not consuming breakfast every day (25%) and the difference
in overweight occurrence between respondents consuming and not consuming breakfast every day
(20% vs. 30%, respectively). With a 5% significance level and 80% power, the minimum sample size
required is approximately 904 respondents (assuming a 75/25 ratio for respondents consuming and
not consuming breakfast every day, i.e., 678/226 and an equal number of respondents in each research
center) to detect a difference between groups in overweight occurrence, including recoding error or
missing data of 10%.

2.11. Statistical Analysis

Categorical variables are presented as a sample percentage (%), and continuous variables as
means with 95% confidence interval (95% confidence interval (CI)) or medians with inter-quartile
range for variables with normal or non-normal distribution, respectively. The normality of variables
distribution was verified with Kolmogorov-Smirnov test before the statistical analysis. Table 7 presents
the general characteristics of the sample across all key variables under study.
Nutrients 2018, 10, 1439 13 of 23

Table 7. Sample characteristics by study design (% or means (95% confidence interval)).

Education-Based Intervention Study (At Baseline) Cross-Sectional Study

Parameters
Total Sample Boys Girls Total Sample Boys Girls
Sample size 464 216 248 1569 760 809
Sample percentage 100.0 46.6 53.4 100.0 48.4 51.6
Age (years) 11.9 (11.9, 12.0) 11.9 (11.9, 12.0) 11.9 (11.9, 11.9) 11.9 (11.9, 12.0) 11.9 (11.9, 12.0) 11.9 (11.9, 12.0)
Age (categories)
11 years 7.1 5.1 8.9 16.6 15.3 17.9
12 years 92.9 94.9 91.1 73.6 75.9 71.3
13 years 0.0 0.0 0.0 9.8 8.8 10.8
Residence
rural 34.9 36.1 33.9 40.3 40.3 40.3
urban 65.1 63.9 66.1 59.7 59.7 59.7
Family Affluence Scale (points) 5.3 (5.1, 5.5) 5.2 (4.9, 5.4) 5.4 (5.2, 5.6) 5.4 (5.3, 5.4) 5.3 (5.2, 5.4) 5.4 (5.3, 5.5)
Nutrition knowledge
Nutrition knowledge score (points) 6.0 (5.7, 6.2) 5.5 (5.2, 5.9) 6.3 (6.0, 6.7) 6.1 (5.9, 6.2) 5.7 (5.5, 5.9) 6.5 (6.3, 6.6)
Nutrition knowledge (categories)
the lowest 30.6 35.6 26.2 30.9 35.7 26.3
moderately low 41.4 42.1 40.7 39.3 40.1 38.6
higher 28.0 22.2 33.1 29.8 24.2 35.1
Attitudes toward nutrition
Emotional Eating (points) 1.3 (1.2, 1.5) 1.3 (1.1, 1.5) 1.4 (1.2, 1.5) 1.3 (1.3, 1.4) 1.3 (1.2, 1.3) 1.4 (1.3, 1.5)
Uncontrolled Eating (points) 5.3 (5.0, 5.5) 5.5 (5.1, 5.9) 5.1 (4.8, 5.4) 5.3 (5.2, 5.4) 5.3 (5.1, 5.5) 5.3 (5.1, 5.5)
Cognitive Restraint of Eating (points) 4.2 (4.0, 4.4) 4.0 (3.7, 4.3) 4.3 (4.1, 4.6) 4.3 (4.2, 4.4) 4.3 (4.1, 4.4) 4.4 (4.2, 4.5)
Diet quality
Every day consumption of
breakfast 71.0 70.7 71.3 70.0 72.7 67.4
school meal 67.5 59.8 74.2 69.0 64.7 72.9
pHDI (% points) 27.7 (26.4, 29.0) 24.4 (22.8, 26.1) 30.5 (28.6, 32.4) 28.4 (27.7, 29.1) 27.4 (26.3, 28.4) 29.4 (28.4, 30.4)
pHDI (categories)
low 71.0 81.8 61.7 69.2 72.9 65.8
moderate 28.6 17.8 37.9 29.8 25.9 33.5
high 0.4 0.5 0.4 1.0 1.2 0.7
nHDI (% points) 14.3 (13.3, 15.3) 15.1 (13.8, 16.4) 13.6 (12.2, 15.1) 14.1 (13.5, 14.7) 15.0 (14.1, 15.9) 13.3 (12.5, 14.0)
nHDI (categories)
low 93.9 94.0 92.7 93.8 93.8 93.8
moderate 6.7 6.0 7.3 5.9 5.5 6.2
high 0.0 0.0 0.0 0.3 0.7 0.0
Nutrients 2018, 10, 1439 14 of 23

Table 7. Cont.

Education-Based Intervention Study (At Baseline) Cross-Sectional Study

Parameters
Total Sample Boys Girls Total Sample Boys Girls
Lifestyle
Screen time (points) 0.84 (0.74, 0.94) 0.91 (0.76, 1.06) 0.78 (0.65, 0.91) 0.86 (0.81, 0.92) 0.98 (0.89, 1.06) 0.76 (0.69, 0.82)
Screen time (categories)
<2 h/day 46.5 41.1 51.2 46.3 42.1 50.3
2 to 4 h/day 34.7 40.7 29.4 34.4 35.2 33.6
≥4 h/day 18.8 18.2 19.4 19.3 22.7 16.1
Total physical activity (points) 3.65 (3.53, 3.78) 3.66 (3.48, 3.85) 3.65 (3.49, 3.81) 3.68 (3.61, 3.74) 3.73 (3.63, 3.82) 3.63 (3.54, 3.71)
Total physical activity (categories)
low 12.3 13.4 11.3 9.8 10.0 9.5
moderate 54.6 49.5 59.1 58.9 53.3 58.9
high 33.0 37.0 29.6 31.3 36.7 26.3
Body composition
Waist circumference (cm) 64.4 (63.5, 65.3) 65.7 (64.3, 67.0) 63.3 (62.2, 64.5) 66.1 (65.6, 66.5) 67.8 (67.1, 68.6) 64.4 (63.8, 65.0)
z-Waist circumference >1 SD 13.5 18.5 9.2 15.1 20.1 10.4
WHtR 0.42 (0.42, 0.43) 0.43 (0.42, 0.44) 0.42 (0.41, 0.42) 0.43 (0.43, 0.43) 0.44 (0.44, 0.45) 0.42 (0.42, 0.42)
Central obesity a 10.0 11.4 8.8 12.2 15.9 8.7
BMI (kg/m2 ) 19.6 (19.2, 19.9) 19.7 (19.1, 20.2) 19.5 (19.0, 20.0) 19.4 (19.2, 19.6) 19.5 (19.2, 19.7) 19.3 (19.1, 19.6)
BMI category b
underweight 10.2 9.0 11.3 9.8 8.4 11.1
normal weight 63.4 62.3 64.4 65.6 64.8 66.2
overweight 26.4 28.8 24.3 24.7 26.8 22.7
Hand grip strength c (kg) 20.1 (19.5, 20.8) 20.7 (19.8, 21.6) 19.6 (18.7, 20.5) 20.8 (20.4, 21.1) 21.4 (20.9, 21.8) 20.2 (19.7, 20.6)
z-Hand grip strength c >1 SD 10.0 11.9 8.3 12.5 14.6 10.4
Sample size may vary in variables due to missing data. pHDI: pro-Healthy Diet Index. nHDI: non-Healthy Diet Index. a Central obesity identified as waist-to-height ratio ≥0.5 according
to Ashwell et al. [61]. b Body mass index (BMI) categorized according to age-sex-specific cut-offs for teenagers [60]: underweight BMI <18.5 kg/m2 ; normal weight BMI = 18.5 to 24.9
kg/m2 ; overweight BMI ≥25 kg/m2 . c Hand grip strength for dominant hand.
Nutrients 2018, 10, 1439 15 of 23

A multi-dimensional statistical analysis will be applied to derive dietary-lifestyle patterns (DLPs).

We will aim to cluster dietary and lifestyle variables in teenagers in order to identify specific sets of
dietary habits existing in combination with lifestyle behaviors (cross-behavioral patterns covering diet
and lifestyle). We intend to apply two methods as alternative procedures: Cluster Analysis (CA) and
Principal Component Analysis (PCA) [63]. In both analyses, the input variables will be 14 variables
(all standardized before) including:

• Eleven dietary, i.e., frequency consumption of breakfast, school meal and nine food items such
as dairy products, fish, vegetables, fruit, fruit or vegetable juices, fast foods, sweetened drinks,
energy drinks, sweets (in times/day).
• Three lifestyles, i.e., screen time, physical activity at school, physical activity at leisure time
(in scores).

A k-means clustering algorithm will be used in CA, and subjects will be grouped into clusters
based on the Euclidean distances. The analysis will be conducted several times in order to identify the
optimal number of clusters. After selecting clusters, the correctness of cluster identification will be
verified by comparing the dietary and lifestyle components of DLPs between clusters with one-way
analysis of variance (ANOVA).
A varimax rotation will be used in PCA. The following three criteria will be considered in order
to identify the number of PCA-derived patterns: (i) the eigenvalues from the correlation matrix of
the standardized variables >1.0, (ii) the break-point identified in the scree plot of the eigenvalues and
(iii) the total variance explained. Rotated factor loadings >|0.30| will be considered as significantly
contributing to each DLP, interpreted as follows: the higher factor loadings, the stronger association
between dietary and lifestyle components of DLPs, and the DLP. For each subject, a DLP score reflecting
subject’s adherence to the DLP will be calculated as a sum of the product of the input variables and
its factor loadings. In each PCA-derived pattern, subjects will be divided into tertiles based on the
DLP scores.
For continuous variables, changes in nutrition knowledge score, Emotional Eating score,
Uncontrolled Eating score, Cognitive Restraint of Eating score, pHDI, nHDI, screen time, total physical
activity, z-BMI, z-WHtR, z-Waist circumference and z-HGS (all expressed in scores or points or SDs)
after 3- or 9-month follow-up in respect to baseline will be verified with ANOVA or Mann-Whitney
test, for variables with normal or non-normal distribution, respectively.
For categorical variables, logistic regression modeling will be applied. The odds ratios (ORs) and
95% CIs will be calculated. The significance of ORs will be verified with Wald’s statistics. Crude model
and models with an adjustment for confounders will be created.
The following categories of modeled variables in the logistic regression analysis will be included:
nutrition knowledge score (low, moderate low, higher), breakfast consumption (skipping, irregular,
every day), school meal consumption (skipping, irregular, every school day), food frequency
consumption (≤1/week, several times a week, every day), pHDI (tertiles: bottom, middle, upper),
nHDI (tertiles: bottom, middle, upper), screen time (<2, 2 to 4, ≥4 h/day), physical activity at school
(low, moderate, vigorous), physical activity at leisure time (low, moderate, vigorous), total physical
activity (low, moderate, high), CA-derived the DLPs (clusters), PCA-derived the DLPs (tertiles: bottom,
middle, upper), BMI (underweight, normal weight, overweight), WHtR (lack of central obesity, central
obesity) or z-BMI, z-WHtR, z-Waist circumference, z-HGS (categories for z-scores: <−1, −1 to 1,
>1 SD).
The following confounders in the logistic regression analysis will be included: gender, age (years),
residence (rural, urban), FAS (points), nutrition knowledge score (points), Emotional Eating score
(points), Uncontrolled Eating score (points), Cognitive Restraint of Eating score (points), pHDI (%
points), nHDI (% points), screen time (points), and total physical activity (points). For each analysis,
the set of confounders will be selected according to the modeled research question.
We will consider assessing four main applications of the logistic regression modeling:
Nutrients 2018, 10, 1439 16 of 23

• For the education-based intervention study: (i) a chance to fall in the modeled category after 3-
or 9-month follow-up in respect to baseline as reference, (ii) the chance to fall in the modeled
category in the educated group in respect to the control group as reference,
• For the cross-sectional study: (iii) adherence to chosen DLP by nutrition knowledge, attitudes
toward nutrition and sociodemographic factors in respect to referent DLP, (iv) the chance to fall
in the modeled category of body composition in respect to normal body composition category
as reference.

All tests will be two-tailed, p-values <0.05 will be considered as significant. Analyses will
be performed using Statistica software (version 12.0 PL; StatSoft Inc., Tulsa, OK, USA; StatSoft,
Krakow, Poland).

3. Discussion
This study protocol presents a general approach and details of assessment of nutrition
knowledge, attitudes toward nutrition, diet quality, lifestyle, and body composition that were used
to comprehensively evaluate the cross-behavioral patterns covering dietary and lifestyle behaviors
in teenagers. The protocol also presents a scope of nationwide nutrition education activities taken in
Polish teenagers as a part of public health interventions.
Reducing childhood obesity is not a straightforward task. There are many factors that influence
this phenomenon, making it hard to create intervention programs that have a significant impact
on obesity reduction. Schools are considered a primary setting for implementing education
programs—they have the unique potential to involve a wide population of children and adolescents.
However, there are many variables influencing the effectiveness of school-based programs, including
school staff who need to be well-trained and familiarized with the program’s aims, procedures, and
tools. The creation of education programs by highly-specialized researchers and scientific support for
educators at the first stage of implementation of such programs can increase their effectiveness [43,64].
As mentioned before, in Poland a multi-disciplinary approach covering the implementation
of multi-component programs are directed mainly to preschool children while relatively few
comprehensive educational programs have been combined with scientific assessment of nutrition
knowledge, dietary habits, and lifestyle which have targeted teenagers [44,45]. Such programs, focused
on reduction of body mass through changing lifestyle and diet have been widely introduced in United
States, Australia, China, Brazil, Turkey, and many European countries, e.g., Spain, France, Belgium,
Great Britain, Ireland, Germany, Netherlands, Greece, and Norway. Their recapitulation has been
presented in several meta-analysis [21,41,65]. There is convincing evidence from this analysis that
school-based prevention interventions can lead to an improvement in dietary behaviors by increasing
the consumption of healthy foods and decreasing the consumption of unhealthy foods, as well as
by changing lifestyle patterns to be more physically active and less sedentary. However, recently
published systematic reviews have shown that school-based intervention programs have been at
least mildly effective in reducing BMI in children [64–66]. It was explained that these new studies
tended to have longer observation periods and be more comprehensive, including more factors under
study and confounders. Furthermore, more pronounced results were reported in older than younger
school children.
Scientific research shows that linking nutrition knowledge to dietary patterns or diet quality
scores would then seem most effective for assessing the relationship between dietary intake and
health outcomes. It seems that investing in high-quality research to measure nutrition knowledge
is a far-sighted approach, as the burden of diet-related diseases continues to rise worldwide [38,42].
To our knowledge, current intervention and cross-sectional studies should lead to evidence-based
initiatives in the field of nutrition education and public health policy. The effectiveness of a public
health campaign should be optimized to reduce the prevalence of obesity-related non-communicable
diseases, especially in children and adolescents as a vulnerable group.
Nutrients 2018, 10, 1439 17 of 23

We anticipate that the results of our comprehensive and multi-factorial statistical analysis will
be of great importance. Currently, considering breakfast consumption as a key dietary characteristic
along with total physical activity as the key lifestyle characteristic and overweight as the key health
outcome, we can discuss these variables based on initial cross-sectional results (in students 11–13 years
old). We estimated that 29% of students did not consume breakfast every day. A higher or wider range
of the percentage of Polish adolescents (9 to <19 years) not consuming breakfast every day (28–52%)
was previously reported [67–70]. Across Europe, generally more children and adolescents from central
(e.g., Slovenia 51–52%) or southern (e.g., Greece 46–48%) than northern countries (e.g., Finland 20%)
did not consume breakfast every day [71–73]. We found a higher percentage of students (33%) who
reached the WHO recommendation on physical activity [58] in comparison with the latest reports
(24.2% of Polish children 11–15 years old) [10]. In a further analysis, we will analyze whether this
is a positive trend, for example, resulting from the fashion for a healthy lifestyle, or if there are
mixed lifestyle behaviors combining greater physical activity with a longer time spent sitting [74].
Overweight was found in 26% of students under study, and this percentage was above the range
previously reported in Polish children and young people (12–25%) [5,10]. Thus, our initial findings
support previous studies, which have reported that more and more Polish children are overweight
or obese.

3.1. Strengths and Limitations in General

The main limitation of the study is the use of a questionnaire to collect dietary and lifestyle data.
The questionnaire contained a short list of food items and simple questions related to sedentary and
active behaviors. However, there is evidence of many advantages of using brief tools, for example,
high reproducibility of simple questions, the possibility to rank respondents into categories of habitual
food consumption and identify dietary patterns, the possibility to assess compliance with dietary or
lifestyle recommendations as well as low-cost, quick and easy administration [75].
The strengths of the study, regarding this questionnaire, are the validation procedure carried out
on a large sample of Polish teenagers (630 subjects 11–15 years old). Second, the questionnaire was
filled out by teenagers to reduce possible bias, which could be introduced by parents [76]. Future
studies should consider the use of a long form of FFQ and/or applying other methods of dietary and
lifestyle assessment to fully describe the teenagers’ dietary behaviors and measure physical activity
(e.g., use of accelerometry) instead of self-reported data on lifestyle [77–79]. However, to date, there is
no validated long form of FFQ, which has been developed for Polish children or adolescents.
To measure adiposity, we applied simple obesity measures (BMI, WHtR) with some interpretative
limitations but proper for comprehensive assessment of high adiposity levels [60,61,80,81]. In many
large epidemiological studies, both measures are widely used. Furthermore, it is easy to interpret
results and compare with others due to well-established procedures and strong documented cut-offs
developed for children and adolescents [60,61,80]. A fat-free mass was not directly measured,
we applied hand grip strength as an indirect measure of muscle mass. There are no cut-offs for
the hand grip strength of teenagers, including Polish individuals, so we used z-score of hand grip
strength to assess the differences between groups. Future studies should consider the use of more
advanced methods of body composition measurement.
To measure socioeconomic status and attitudes toward nutrition, we applied scales dedicated to
children (the Family Affluence Scale, FAS, and three-factor eating questionnaire, TFEQ, respectively),
both previously validated and used in the Polish population [54,62]. Although we have slightly
changed FAS and shortened TFEQ, both new versions have been tested in pilot and validation studies.
Data collection and all measurements were taken by well-trained researchers with the same type
of equipment in all scientific centers to minimize inter-center differences. The main study was preceded
by a pilot study that was carried out in all centers (covering above 200 subjects).
Nutrients 2018, 10, 1439 18 of 23

3.2. Strengths and Limitations of the Education-Based Intervention Study

The sample was relatively large (over 460), taking into account that complete dietary and weight
status data were obtained at all stages of the study. We applied a very rigorous selection of the sample,
each participant had to take part in lectures and workshops within all five topics of education program
lasting three weeks—students absent at least one school day for the lectures or workshops were
excluded. Next, we used appropriate statistical methods to interpret the differences between the
educated and control groups in regards to changes from the baseline to a 3- or 9-month follow-up.
The main limitation is a lack of random subjects to be allocated into the educated and control
groups. Unfortunately, we were unable to apply the random approach for several reasons. First,
for organizational reasons, we wanted to choose schools located at a convenient distance from the
academic centers (up to 50 km). Second (surprisingly), many schoolmasters did not permit their
school to participate in the study. Third, many of the schools had previously participated in other
nutrition-health education programs, so they could not be included in our study. There was an element
of randomness in our study because assigning classes to the educated or control group was accidental
and the schools represented a wide social spectrum.

3.3. Strengths and Limitations of the Cross-Sectional Study

The main strength of the study was the clustering of dietary and lifestyle variables and identifying
sets of dietary habits existing in teenagers in combination with lifestyle behaviors (DLPs). To date,
few studies have been aimed at identifying cross-behavioral patterns covering both dietary and
lifestyle variables. Although there is limited data available for teenagers from some developed
countries [74,76,82–86], there is the gap for Eastern or Central Europe, including Poland.
The main limitation is the sample itself, which was not randomly selected, although it is
relatively large (above 1500), has a narrow age range (11–13 years) and covers the entire territory
of Poland—widely reflecting the social diversity of Poles (Figure 1A) and providing a good basis
for generalizations.

4. Conclusions
The study will be able to assess the effectiveness of a nutrition-related and lifestyle-related
education program in the medium-term perspective. The weaknesses and strengths of such education
programs in respect to teenagers, as a target group, could be identified. Based on the results, in future,
well-tailored education programs addressed to teenagers can be created. We think that our results can
be interpreted in a wider scope, at least for European teenagers living in similar conditions.
The findings may contribute to identifying teenage subpopulations at risk for excessive body
weight and adiposity as well as poor muscling and determine predictors related to diet, lifestyle,
nutrition knowledge, attitudes toward nutrition and socio-demographic factors which influence
improper body composition and increase obesity risk.
The study provides evidence-based support for preventive health care to promote normal growth
and development of the young population and reduce the risk of diet-related diseases in adulthood
by the early shaping of adequate dietary and lifestyle behaviors. The results of the study can be
implemented as an important public health action.

Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6643/10/

10/1439/s1, Figure S1: Planned results of the implementation of the national multicenter project ‘ABC
of Healthy Eating’, Table S1: The short form of the Food Frequency Questionnaire for Polish Children
(SF-FFQ4PolishChildren)—translation into English, Table S2: The short form of the Food Frequency Questionnaire
for Polish Children (SF-FFQ4PolishChildren)—Polish language version.
Author Contributions: J.H., L.W., M.H., J.K., and. K.G. were responsible for the conception and design of the main
study. J.H., M.H. and K.G. were involved in the funding acquisition in respect to the project and managing of the
project. J.K. and L.W. were responsible for development of design of statistical analysis and results interpretation
for the study. J.K. was responsible for the data cleaning and statistical analysis for the main study. J.H. and
L.W. were responsible for writing the manuscript. M.H. was responsible for visualization. K.G. was responsible
Nutrients 2018, 10, 1439 19 of 23

for revising the manuscript critically for important intellectual content. The manuscript has been revised by
all co-authors.
Funding: The study was financially supported by Carrefour Foundation (Agreement ABC No.1/2014; Agreement
ABC No. 2/2016) and each scientific centre from sources of the Polish Ministry of Sciences and Higher Education.
Acknowledgments: Thanks are expressed to the participants for their contributions to the study and the academic
researchers from: Warsaw University of Life Sciences (WULS-SGGW), Gdynia Maritime University, Faculty of
Health Sciences, Medical University in Bialystok, Faculty of Physical Education and Sport in Biala Podlaska of
the Josef Pilsudski University of Physical Education in Warsaw, University of Agriculture in Krakow, University
of Life Sciences in Lublin, University of Life Sciences in Poznan, University of Warmia and Mazury in Olsztyn,
Wroclaw University of Environmental and Life Sciences.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Ng, M.; Fleming, T.; Robinson, M.; Thomson, B.; Graetz, N.; Margono, C.; Mullany, E.C.; Biryukov, S.;
Abbafati, C.; Abera, S.F.; et al. Global, regional, and national prevalence of overweight and obesity in
children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013.
Lancet 2014, 384, 766–781. [CrossRef]
2. Lobstein, T.; Jackson-Leach, R. Planning for the worst: Estimates of obesity and comorbidities in school-age
children in 2025. Pediatr. Obes. 2016, 11, 321–325. [CrossRef] [PubMed]
3. World Health Organization. Global Nutrition Monitoring Framework: Operational guidance
for tracking progress in meeting targets for 2025. Geneva, Switzerland. 2017. Available
online: http://apps.who.int/iris/bitstream/handle/10665/259904/9789241513609-eng.pdf;jsessionid=
004DBB115D5F666C3EE9F022BFD99D86?sequence=1 (accessed on 20 July 2018).
4. Ogden, C.L.; Carroll, M.D.; Kit, B.K.; Flegal, K.M. Prevalence of childhood and adult obesity in the United
States, 2011–2012. JAMA 2014, 311, 806–814. [CrossRef] [PubMed]
5. Suder, A.; Gomula, A.; Koziel, S. Central overweight and obesity in Polish schoolchildren aged 7–18 years:
Secular changes of waist circumference between 1966 and 2012. Eur. J. Pediatr. 2017, 176, 909–916. [CrossRef]
[PubMed]
6. Olds, T.; Maher, C.; Zumin, S.; Péneau, S.; Lioret, S.; Castetbon, K.; Bellisle, F.; de Wilde, J.; Hohepa, M.;
Maddison, R.; et al. Evidence that the prevalence of childhood overweight is plateauing: Data from nine
countries. Int. J. Pediatr. Obes. 2011, 6, 342–360. [CrossRef] [PubMed]
7. Sahoo, K.; Sahoo, B.; Choudhury, A.K.; Sofi, N.Y.; Kumar, R.; Bhadoria, A.S. Childhood obesity: Causes and
consequences. J. Family Med. Prim. Care 2015, 4, 187–192. [CrossRef] [PubMed]
8. De Onis, M.; Blössner, M.; Borghi, E. Prevalence and trends of stunting among pre-school children, 1990–2020.
Public Health Nutr. 2012, 15, 142–148. [CrossRef] [PubMed]
9. World Health Organization. European Food and Nutrition Action Plan 2015–2020, REGIONAL COMMITTEE
FOR EUROPE 64th SESSION. Copenhagen, Denmark. 2014. Available online: http://www.euro.who.int/__
data/assets/pdf_file/0008/253727/64wd14e_FoodNutAP_140426.pdf (accessed on 28 July 2018).
10. Mazur, J. (Ed.) Zdrowie i Zachowania Zdrowotne Młodzieży Szkolnej w Polsce na tle Wybranych Uwarunkowań
Socjodemograficznych. Wyniki. Badań HBSC 2014 [Health and Health Behaviour of School Children in Poland against
the Background of Selected Sociodemographic Conditions. HBSC 2014 Results]; IMiD: Warsaw, Poland, 2015.
11. Chobot, A.; Górowska-Kowolik, K.; Sokołowska, M.; Jarosz-Chobot, P. Obesity and diabetes-Not only a
simple link between two epidemics. Diabetes Metab. Res. Rev. 2018, 21, e3042. [CrossRef] [PubMed]
12. Kowalkowska, J.; Wadolowska, L.; Wuenstel, J.W.; Słowińska, M.A.; Niedźwiedzka, E. Socioeconomic status
and overweight prevalence in polish adolescents: The impact of single factors and a complex index of
socioeconomic status in respect to age and sex. Iran J. Public Health 2014, 43, 913–925. [PubMed]
13. Wadolowska, L.; Kowalkowska, J.; Czarnocinska, J.; Jezewska-Zychowicz, M.; Babicz-Zielinska, E.
Comparing dietary patterns derived by two methods and their associations with obesity in Polish girls aged
13–21 years: The cross-sectional GEBaHealth study. Perspect. Public Health 2017, 137, 182–189. [CrossRef]
[PubMed]
14. Faienza, M.F.; Wang, D.Q.; Frühbeck, G.; Garruti, G.; Portincasa, P. The dangerous link between childhood
and adulthood predictors of obesity and metabolic syndrome. Intern. Emerg Med. 2016, 11, 175–182.
[CrossRef] [PubMed]
Nutrients 2018, 10, 1439 20 of 23

15. Simmonds, M.; Llewellyn, A.; Owen, C.G.; Woolacott, N. Predicting adult obesity from childhood obesity:
A systematic review and meta-analysis. Obes. Rev. 2016, 17, 95–107. [CrossRef] [PubMed]
16. World Health Organization. Prevention of Cardiovascular Disease. Geneva, Switzerland. 2007. Available
online: http://www.who.int/cardiovascular_diseases/guidelines/en/ (accessed on 2 August 2018).
17. Vucenik, I.; Stains, J.P. Obesity and cancer risk: Evidence, mechanisms, and recommendations. Ann. N.Y.
Acad. Sci. 2012, 1271, 37–43. [CrossRef] [PubMed]
18. Llewellyn, A.; Simmonds, M.; Owen, C.G.; Woolacott, N. Childhood obesity as a predictor of morbidity in
adulthood: A systematic review and meta-analysis. Obes. Rev. 2016, 17, 56–67. [CrossRef] [PubMed]
19. Li, L.; Pérez, A.; Wu, L.T.; Ranjit, N.; Brown, H.S.; Kelder, S.H. Cardiometabolic Risk Factors among Severely
Obese Children and Adolescents in the United States, 1999–2012. Child Obes. 2016, 12, 12–19. [CrossRef]
[PubMed]
20. Estrada, E.; Eneli, I.; Hampl, S.; Mietus-Snyder, M.; Mirza, N.; Rhodes, E.; Sweeney, B.; Tinajero-Deck, L.;
Woolford, S.J.; Pont, S.J.; Children’s Hospital Association. Children’s Hospital Association consensus
statements for comorbidities of childhood obesity. Child Obes. 2014, 10, 304–317. [CrossRef] [PubMed]
21. Sbruzzi, G.; Eibel, B.; Barbiero, S.M.; Petkowicz, R.O.; Ribeiro, R.A.; Cesa, C.C.; Martins, C.C.; Marobin, R.;
Schaan, C.W.; Souza, W.B.; et al. Educational interventions in childhood obesity: A systematic review with
meta-analysis of randomized clinical trials. Prev. Med. 2013, 56, 254–264. [CrossRef] [PubMed]
22. Chaput, J.P.; Pérusse, L.; Després, J.P.; Tremblay, A.; Bouchard, C. Findings from the quebec family study on
the etiology of obesity: Genetics and environmental highlights. Curr. Obes. Rep. 2014, 3, 54–66. [CrossRef]
[PubMed]
23. Serra-Majem, L.; Bautista-Castaño, I. Etiology of obesity: Two ‘key issues’ and other emerging factors.
Nutr. Hosp. 2013, 28 (Suppl. 5), 32–43. [CrossRef] [PubMed]
24. Wadolowska, L.; Kowalkowska, J.; Lonnie, M.; Czarnocinska, J.; Jezewska-Zychowicz, M.; Babicz-Zielinska, E.
Associations between physical activity patterns and dietary patterns in a representative sample of Polish girls
aged 13–21 years: A cross-sectional study (GEBaHealth Project). BMC Public Health 2016, 16, 698. [CrossRef]
[PubMed]
25. Popkin, M.P.; Adair, L.S.; Ng, S.W. The Global Nutrition Transition: The pandemic of obesity in developing
countries. Nutr. Rev. 2012, 70, 3–21. [CrossRef] [PubMed]
26. Xu, H.; Wen, L.M.; Rissel, C. Associations of parental influences with physical activity and screen time
among young children: A systematic review. J. Obes. 2015, 2015, 546925. [CrossRef] [PubMed]
27. Van Hecke, L.; Loyen, A.; Verloigne, M.; van der Ploeg, H.P.; Lakerveld, J.; Brug, J.; De Bourdeaudhuij, I.;
Ekelund, U.; Donnelly, A.; Hendriksen, I.; et al. Variation in population levels of physical activity in
European children and adolescents according to cross-European studies: A systematic literature review
within DEDIPAC. Int. J. Behav. Nutr. Phys. Act. 2016, 28, 13–70. [CrossRef] [PubMed]
28. Dumith, S.C.; Gigante, D.P.; Domingues, M.R.; Kohl, H.W. Physical activity change during adolescence:
A systematic review and a pooled analysis. Int. J. Epidemiol. 2011, 40, 685–698. [CrossRef] [PubMed]
29. NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in body-mass index, underweight,
overweight, and obesity from 1975 to 2016: A pooled analysis of 2416 population-based measurement
studies in 128·9 million children, adolescents, and adults. Lancet 2017, 390, 2627–2642. [CrossRef]
30. Kubik, M.Y.; Lytle, L.A.; Hannan, P.J.; Perry, C.L.; Story, M. The association of the school food environment
with dietary behaviors of young adolescents. Am. J. Public Health 2003, 93, 1168–1173. [CrossRef] [PubMed]
31. Fox, M.K.; Dodd, A.H.; Wilson, A.; Gleason, P.M. Association between school food environment and practices
and body mass index of US public school children. J. Am. Diet Assoc. 2009, 109 (Suppl. 2), 108–117. [CrossRef]
[PubMed]
32. Singh, R. Childhood obesity: An epidemic in waiting? Int. J. Med. Public Health 2013, 3, 2–7. [CrossRef]
33. Trude, A.C.B.; Kharmats, A.Y.; Jones-Smith, J.C.; Gittelsohn, J. Exposure to a multi-level multi-component
childhood obesity prevention community-randomized controlled trial: Patterns, determinants, and
implications. Trials 2018, 19, 287. [CrossRef] [PubMed]
34. Ryan, P. Integrated theory of health behavior change: Background and intervention development.
Clin. Nurse Spec. 2009, 23, 161–172. [CrossRef] [PubMed]
35. Jezewska-Zychowicz, M.; Wadolowska, L.; Kowalkowska, J.; Lonnie, M.; Czarnocinska, J.; Babicz-Zielinska, E.
Perceived Health and Nutrition Concerns as Predictors of Dietary Patterns among Polish Females Aged
13–21 Years (GEBaHealth Project). Nutrients 2017, 9, 613. [CrossRef] [PubMed]
Nutrients 2018, 10, 1439 21 of 23

36. Alkaed, D.; Ibrahim, N.; Ismail, F.; Barake, R. Validity and Reliability of a Nutrition Knowledge Questionnaire
in an Adult Student Population. J. Nutr. Educ. Behav. 2018, 50, 718–723. [CrossRef] [PubMed]
37. Birch, L.; Jennifer, S.; Savage, J.S.; Ventura, A. Influences on the development of children’s eating behaviours:
From infancy to adolescence. Can. J. Diet Pract. Res. 2007, 68, S51–S56.
38. Spronk, I.; Kullen, C.; Burdon, C.; O’Connor, H. Relationship between nutrition knowledge and dietary
intake. Br. J. Nutr. 2014, 111, 1713–1726. [CrossRef] [PubMed]
39. Baxter, S.D. Cognitive processes in children’s dietary recalls: Insight from methodological studies. Eur. J.
Clin. Nutr. 2009, 63 (Suppl. 1), 19–32. [CrossRef] [PubMed]
40. Story, M.; Kaphingst, K.M.; French, S. The role of schools in obesity prevention. Future Child 2006, 16, 109–142.
[CrossRef] [PubMed]
41. Hung, L.S.; Tidwell, D.K.; Hall, M.E.; Lee, M.L.; Briley, C.A.; Hunt, B.P. A meta-analysis of school-based
obesity prevention programs demonstrates limited efficacy of decreasing childhood obesity. Nutr. Res. 2015,
35, 229–240. [CrossRef] [PubMed]
42. Worsley, A. Nutrition knowledge and food consumption: Can nutrition knowledge change food behaviour?
Asia Pac. J Clin. Nutr. 2002, 11, S579–S585. [CrossRef] [PubMed]
43. Manios, Y.; Grammatikaki, E.; Androutsos, O.; Chinapaw, M.J.; Gibson, E.L.; Buijs, G.; Iotova, V.; Socha, P.;
Annemans, L.; Wildgruber, A.; et al. A systematic approach for the development of a kindergarten-based
intervention for the prevention of obesity in preschool age children: The ToyBox-study. Obes. Rev. 2012, 13,
3–12. [CrossRef] [PubMed]
44. Manios, Y.; Androutsos, O.; Katsarou, C.; Iotova, V.; Socha, P.; Geyer, C.; Moreno, L.; Koletzko, B.; De
Bourdeaudhuij, I.; on behalf of the ToyBox-study group. Designing and implementing a kindergarten-based,
family-involved intervention to prevent obesity in early childhood: The ToyBox-study. Obes. Rev. 2014, 15,
5–13. [CrossRef] [PubMed]
45. Myszkowska-Ryciak, J.; Harton, A. Eating Healthy, Growing Healthy: Impact of a Multi-Strategy Nutrition
Education on the Assortments of Beverages Served in Preschools, Poland. Int. J. Environ. Res. Public Health
2018, 15, 1355. [CrossRef] [PubMed]
46. World Health Organization. Cancer Control: Knowledge into Action. Geneva, Switzerland. 2007. Available
online: http://www.who.int/cancer/modules/en/ (accessed on 2 August 2018).
47. World Health Organization. Framework for Care and Control of Tuberculosis and Diabetes. Geneva.
2011. Available online: https://www.worlddiabetesfoundation.org/files/collaborative-framework-care-
and-control-tuberculosis-and-diabetes (accessed on 27 July 2018).
48. Available online: http://www.zachowajrownowage.pl/pl/o-projekcie/ (accessed on 11 September 2018).
49. Hoffmann, T.C.; Glasziou, P.P.; Boutron, I.; Milne, R.; Perera, R.; Moher, D.; Altman, D.G.; Barbour, V.;
Macdonald, H.; Johnston, M.; et al. Better reporting of interventions: Template for intervention description
and replication (TIDieR) checklist and guide. BMJ 2014, 7, 348. [CrossRef] [PubMed]
50. Hörnell, A.; Berg, C.; Forsum, E.; Larsson, C.; Sonestedt, E.; Åkesson, A.; Lachat, C.; Hawwash, D.;
Kolsteren, P.; Byrnes, G.; et al. Perspective: An Extension of the STROBE Statement for Observational
Studies in Nutritional Epidemiology (STROBE-nut): Explanation and Elaboration. Adv. Nutr. 2017, 8,
652–678. [CrossRef] [PubMed]
51. Reporting Guidelines for Main Study Types. Available online: http://www.equator-network.org (accessed
on 23 July 2018).
52. Piramida Zdrowego Zywienia i Aktywnosci Fizycznej. Available online: http://www.izz.waw.pl/pl/
zasady-prawidowego-ywienia (accessed on 23 July 2018).
53. Whati, L.H.; Senekal, M.; Steyn, N.P.; Nel, J.H.; Lombard, C.; Norris, S. Development of a reliable and
valid nutritional knowledge questionnaire for urban South African adolescents. Nutrition 2005, 21, 76–85.
[CrossRef] [PubMed]
54. Dzielska, A.; Mazur, J.; Małkowska-Szkutnik, A.; Kołoło, H. Adaptation of the Three-Factor Eating
Questionnaire (TFEQ-13) for school-aged adolescents in a population study. Probl. Hig. Epidemiol. 2009, 90,
362–369.
55. Rampersaud, G.C.; Pereira, M.A.; Girard, B.L.; Adams, J.; Metzl, J.D. Breakfast habits, nutritional status,
body weight, and academic performance in children and adolescents. J. Am. Diet. Assoc. 2005, 105, 743–760.
[CrossRef] [PubMed]
Nutrients 2018, 10, 1439 22 of 23

56. Pereira, M.A.; Erickson, E.; McKee, P.; Schrankler, K.; Raatz, S.K.; Lytle, L.A.; Pellegrini, A.D. Breakfast
frequency and quality may affect glycemia and appetite in adults and children. J. Nutr. 2011, 141, 163–168.
[CrossRef] [PubMed]
57. AAP Council on communications and media. Media Use in School-Aged Children and Adolescents. Pediatrics
2016, 138, e20162592. [CrossRef] [PubMed]
58. World Health Organization. Global Recommendations on Physical Activity for Health. Geneva, Switzerland.
2010. Available online: http://www.who.int/dietphysicalactivity/factsheet_recommendations/en/
(accessed on 29 July 2018).
59. ISAK. International Standards for Anthropometric Assessment. Potchefstroom, South Africa: International
Society for the Advancement of Kinanthropometry. 2001. Available online: http://www.ceap.br/material/
MAT17032011184632.pdf (accessed on 17 July 2018).
60. Cole, T.J.; Lobstein, T. Extended international (IOTF) body mass index cut-offs for thinness, overweight and
obesity. Pediatr. Obes. 2012, 7, 284–294. [CrossRef] [PubMed]
61. Ashwell, M.; Gunn, P.; Gibson, S. Waist-to-height ratio is a better screening tool than waist circumference
and BMI for adult cardiometabolic risk factors: Systematic review and meta-analysis. Obes. Rev. 2012, 13,
275–286. [CrossRef] [PubMed]
62. Mazur, J. Family Affluence Scale—validation study and suggested modification. Hygeia Pub. Health 2013, 48,
211–217.
63. Armitage, P.; Berry, G.; Matthews, J.N.S. Statistical Methods in Medical Research, 4th ed.; Blackwell Science:
Oxford, UK, 2008.
64. Centeio, E.E.; McCaughtry, N.; Moore, E.W.G.; Kulik, N.; Garn, A.; Martin, J.; Shen, B.; Somers, C.L.;
Fahlman, M. Building healthy communities: A comprehensive school health program to prevent obesity in
elementary schools. Prev. Med. 2018, 111, 210–215. [CrossRef] [PubMed]
65. Sobol-Goldberg, S.; Rabinowitz, J.; Gross, R. School-based obesity prevention programs: A meta-analysis of
randomized controlled trials. Obesity 2013, 21, 2422–2428. [CrossRef] [PubMed]
66. Lycett, K.; Miller, A.; Knox, A.; Dunn, S.; Kerr, J.A.; Sung, V.; Wake, M. ‘Nudge’ interventions for improving
children’s dietary behaviors in the home: A. systematic review. Obes. Med. 2017, 7, 21–33. [CrossRef]
67. Zielińska, M.; Hamułka, J.; Gajda, K. Family influences on breakfast frequency and quality among primary
school pupils in Warsaw and its surrounding areas. Rocz. Panstw. Zakl. Hig. 2015, 66, 143–149. [PubMed]
68. Inchley, J.; Currie, D.; Young, T.; Samdal, O.; Torsheim, T.; Augustson, L.; Mathison, F.; Aleman-Diaz, A.;
Molcho, M.; Weber, M.; et al. Growing up unequal: Gender and socioeconomic differences in young people’s
health and wellbeing. Health Behaviour in School-aged Children (HBSC) study: International report from
the 2013/2014 survey. WHO Regional Office for Europe: Copenhagen, Denmark, 2016. Available online:
http://www.euro.who.int/en/hbsc-report-2016 (accessed on 9 September 2018).
69. Wunstel, J.W.; Kowalkowska, J.; Wadolowska, L.; Slowinska, M.A.; Niedzwiedzka, E.; Kurp, L. Habitual
eating breakfast consumption. Dev. Period Med. 2015, 19, 193–201.
70. Kollataj, W.; Sygit, K.; Sygit, M.; Karwat, I.D.; Kollataj, B. Eating habits of children and adolescents from
rural regions depending on gender, education, and economic status of parents. Ann. Agric. Environ. Med.
2011, 18, 393–397.
71. Zakrzewski, J.K.; Gillison, F.B.; Cumming, S.; Church, T.S.; Katzmarzyk, P.T.; Broyles, S.T.; Champagne, C.M.;
Chaput, J.P.; Denstel, K.D.; Fogelholm, M.; et al. Associations between breakfast frequency and adiposity
indicators in children from 12 countries. Int. J. Obes. 2015, 5 (Suppl. 2), 80–88. [CrossRef] [PubMed]
72. Brug, J.; van Stralen, M.M.; te Velde, S.J.; Chinapaw, M.J.M.; De Bourdeaudhuij, I.; Lien, N.; Bere, E.;
Maskini, V.; Singh, A.S.; Maes, L. Differences in weight status and energy-balance related behaviors among
schoolchildren across europe: The energy-project. PLoS ONE 2012, 7. [CrossRef] [PubMed]
73. Hallstrom, L.; Vereecken, C.A.; Labayen, I.; Ruiz, J.R.; Le Donne, C.; Garcia, M.C.; Gilbert, C.C.; Martínez, S.G.;
Grammatikaki, E.; Huybrechts, I. Breakfast habits among European adolescents and their association with
sociodemographic factors: The HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence) study.
Public Health Nutr. 2012, 15, 1879–1889. [CrossRef] [PubMed]
74. Gubbels, J.S.; Kremers, S.P.; Stafleu, A.; Goldbohm, A.; Vries, N.K.; Thijs, C. Clustering of energy
balance-related behaviors in 5-year-old children: Lifestyle patterns and their longitudinal association with
weight status development in early childhood. Int. J. Behav. Nutr. 2012, 9, 77. [CrossRef] [PubMed]
Nutrients 2018, 10, 1439 23 of 23

75. Food and Agriculture Organization. Dietary Assessment: A Resource Guide to Method Selection and
Application in Low Resource Settings. Rome. 2018. Available online: http://www.fao.org/3/i9940en/
I9940EN. (accessed on 17 July 2018).
76. Gubbels, J.S.; van Assema, P.; Kremers, S.P.J. Physical activity, sedentary behavior, and dietary patterns
among children. Curr. Nutr. Rep. 2013, 2, 105–112. [CrossRef] [PubMed]
77. Ferrari, P.; Friedenreich, C.; Matthews, C.E. The role of measurement error in estimating levels of physical
activity. Am. J. Epidemiol. 2007, 166, 832–840. [CrossRef] [PubMed]
78. Saunders, T.J.; Prince, S.A.; Tremblay, M.S. Clustering of children’s activity behaviour: The use of self-report
versus direct measures. Int. J. Behav. Nutr. Phys. Act. 2011, 8. [CrossRef] [PubMed]
79. Verbestel, V.; De Henauw, S.; Bammann, K.; Barba, G.; Hadjigeorgiou, C.; Eiben, G.; Konstabel, K.;
Kovács, E.; Pitsiladis, Y.; Reisch, L.; et al. Are context-specific measures of parental-reported physical
activity and sedentary behaviour associated with accelerometer data in 2–9-year-old European children?
Public Health Nutr. 2015, 18, 860–868. [CrossRef] [PubMed]
80. Martin-Calvo, N.; Moreno-Galarraga, L.; Martinez-Gonzalez, M.A. Association between body mass index,
waist-to-height ratio and adiposity in children: A systematic review and meta-analysis. Nutrients 2016, 8,
512. [CrossRef] [PubMed]
81. Ashwell, M.; Gibson, S. Waist-to-height ratio as an indicator of ‘early health risk’: Simpler and more predictive
than using a ‘matrix’ based on BMI and waist circumference. BMJ Open 2016, 6, e010159. [CrossRef] [PubMed]
82. Leech, R.M.; McNaughton, S.A.; Timperio, A. Clustering of diet, physical activity and sedentary behaviour
among Australian children: Cross-sectional and longitudinal associations with overweight and obesity.
Int. Obes. 2015, 39, 1079–1085. [CrossRef] [PubMed]
83. Pérez-Rodrigo, C.; Gil, A.; González-Gross, M.; Ortega, R.M.; Serra-Majem, L.; Varela-Moreiras, G.;
Aranceta-Bartrina, J. Clustering of Dietary Patterns, Lifestyles, and Overweight among Spanish Children
and Adolescents in the ANIBES Study. Nutrients 2016, 8, 11. [CrossRef] [PubMed]
84. Mandic, S.; Bengoechea, E.G.; Coppell, K.J.; Spence, J.C. Clustering of (Un) Healthy Behaviors in Adolescents
from Dunedin, New Zealand. Am. J. Health Behav. 2017, 41, 266–275. [CrossRef] [PubMed]
85. Nuutinen, T.; Lehto, E.; Ray, C.; Roos, E.; Villberg, J.; Tynjälä, J. Clustering of energy balance-related
behaviours, sleep, and overweight among Finnish adolescents. Int. J. Public Health 2017, 62, 929–938.
[CrossRef] [PubMed]
86. Laxer, R.E.; Brownson, R.C.; Dubin, J.A.; Cooke, M.; Chaurasia, A.; Leatherdale, S.T. Clustering of risk-related
modifiable behaviours and their association with overweight and obesity among a large sample of youth in
the COMPASS study. BMC Public Health 2017, 17, 102. [CrossRef] [PubMed]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).