A Conceptual Framework for Healthy Eating Behavior in

Ecuadorian Adolescents: A Qualitative Study

Roosmarijn Verstraeten1,2*, Kathleen Van Royen2, Angélica Ochoa-Avilés2,3, Daniela Penafiel2,4,
Michelle Holdsworth5, Silvana Donoso3, Lea Maes6, Patrick Kolsteren1,2
1 Department of Public Health, Institute of Tropical Medicine, Antwerp, Belgium, 2 Department of Food Safety and Food Quality, Ghent University, Ghent, Belgium,
3 Food, Nutrition and Health program, Universidad de Cuenca, Cuenca, Ecuador, 4 Rural Research Centre, Escuela Superior Politécnica del Litoral, Guayaquil, Ecuador,
5 Public Health Section, School of Health and Related Research (ScHARR) - The University of Sheffield, Sheffield, United Kingdom, 6 Department of Public Health, Ghent
University, Ghent, Belgium

Abstract
Objective: The objective of this study was to identify factors influencing eating behavior of Ecuadorian adolescents - from
the perspective of parents, school staff and adolescents - to develop a conceptual framework for adolescents’ eating
behavior.

Study design: Twenty focus groups (N = 144 participants) were conducted separately with adolescents aged 11–15 y (n
(focus groups) = 12, N (participants) = 80), parents (n = 4, N = 32) and school staff (n = 4, N = 32) in rural and urban Ecuador.
A semi-structured questioning route was developed based on the ‘Attitude, Social influences and Self-efficacy’ model and
the socio-ecological model to assess the relevance of behavioral and environmental factors in low- and middle-income
countries. Two researchers independently analyzed verbatim transcripts for emerging themes, using deductive thematic
content analysis. Data were analyzed using NVivo 8.

Results: All groups recognized the importance of eating healthily and key individual factors in Ecuadorian adolescents’ food
choices were: financial autonomy, food safety perceptions, lack of self-control, habit strength, taste preferences and
perceived peer norms. Environmental factors included the poor nutritional quality of food and its easy access at school. In
their home and family environment, time and convenience completed the picture as barriers to eating healthily. Participants
acknowledged the impact of the changing socio-cultural environment on adolescents’ eating patterns. Availability of
healthy food at home and financial constraints differed between settings and socio-economic groups.

Conclusion: Our findings endorse the importance of investigating behavioral and environmental factors that influence and
mediate healthy dietary behavior prior to intervention development. Several culture-specific factors emerged that were
incorporated into a conceptual framework for developing health promotion interventions in Ecuador.

Citation: Verstraeten R, Van Royen K, Ochoa-Avilés A, Penafiel D, Holdsworth M, et al. (2014) A Conceptual Framework for Healthy Eating Behavior in Ecuadorian
Adolescents: A Qualitative Study. PLoS ONE 9(1): e87183. doi:10.1371/journal.pone.0087183
Editor: Michel Botbol, University of Western Brittany, France
Received April 17, 2013; Accepted December 23, 2013; Published January 29, 2014
Copyright: ß 2014 Verstraeten et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This qualitative research study was part of a programme funded by a Flemish Inter-University Collaboration (VLIR-IUC, http://www.vliruos.be/).
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have read the journal’s policy and have the following conflicts. RV and AO received a PhD grant from VLIR-IUC and SD
received an honorarium from VLIR-IUC during the duration of this programme.
* E-mail: roosmarijn.verstraeten@ugent.be

Introduction School-based interventions targeting physical inactivity and

unhealthy eating are an important strategy in obesity prevention
Obesity and chronic diseases are no longer exclusive to affluent [11]. However, evidence is needed from LMICs of the pathways
societies, but are now the leading cause of morbidity and mortality through which school-based interventions mediate physical
in low- and middle-income countries (LMICs) [1]. A staggering activity and dietary behavior [12]. To increase our understanding,
rise in unhealthy body weight has been observed in children in intervention studies incorporating theoretical models to address
LMICs across all levels of socio-economic status [2,3]. This rise is population-specific behavioral and environmental influences on
associated with rapid economic and societal changes [4,5] and has dietary and physical activity behavior are required [13]. Current
led to obesity prevalence estimates in some LMICs as high as those models may not be transferable to LMICs because culture-specific
in high-income countries (HICs) [6]. Prevention is crucial, as influences on these behaviors, such as social values/norms and
childhood obesity is associated with several chronic conditions in physical environment may be different from HICs.
adulthood [7–9] and premature mortality [10] thereby aggravat- To develop a conceptual framework for health promotion
ing the burden on health systems and hindering economic interventions in Ecuadorian adolescents that accounts for its
development. cultural context, we solicited opinions of adolescents, parents and

PLOS ONE | www.plosone.org 1 January 2014 | Volume 9 | Issue 1 | e87183

 Healthy Eating Behaviour in Ecuadorian Adolescents

school staff, using focus groups to explore factors of adolescents’ refined. The issues addressed were designed to solicit information
eating behavior. about the individual, physical and social eating environment of
adolescents, consistent with the models selected. Open-ended
Methods questions were followed by more specific probes to clarify and
extend responses. Adolescent focus groups opened with a visual
Ethics statement listing of healthy and unhealthy foods which was then referred to
Focus groups were conducted between April - September 2008. during the group discussion. After each focus group, a debriefing
They were framed within a larger research study and the study was held with the moderator and observer.
protocol was approved by both the Ethics Committees of Quito
and the Ghent University Hospital (CBM/cobi-001; Participants
B67020084010; 2008/462). The different audiences included in Focus groups were conducted in 5 schools, 3 from Cuenca
these focus groups were asked for their consent. Adolescents who (urban) and 2 from Nabón (rural), which were selected by
returned signed parental consent forms and gave written assent to convenience sampling. Each of the 3 schools in Cuenca
participate were included in the study; parents and school staff represented a distinct socio-economic level, i.e. low, middle, and
needed to provide written consent. The ‘Consolidated criteria for high. There were only schools of low socio-economic level in
reporting qualitative research checklist’ was used to report the Nabón. Schools were categorized into these different levels based
results [14]. on the type of school (public/private) and school fees. From each
of these 5 schools, 20 adolescents (grade 8–11) were randomly
Theoretical framework selected. Convenience sampling was used to recruit participants to
Dietary behavior in young people is determined by the complex the parent and school staff focus groups. To be eligible, parents
interplay of factors at both individual and environmental level. To needed to have a child (aged 11–15 y) at one of the participating
better understand these factors in Ecuadorian adolescents, we used schools and school staff had to be employed at one of the schools.
a theoretical framework to conceptualize and analyze the findings
of focus group discussions. To ensure the cultural appropriateness School setting
of this framework, the cognitive variables from the ‘Attitude, Schools had either contact hours in the morning (7 am–1 pm;
Social influences and Self-efficacy’ ASE-model [15] were nestled n = 4) or in the afternoon (12 am–6 pm; n = 1) and both had one
within the socio-cultural and physical context of adolescents’ break of approximately 30 minutes. Food service was provided
environment, as elaborated by the socio-ecological model [16]. through a privately owned tuck shop, i.e. a small food-selling
The ASE-model poses that dietary behavior is a function of the retailer, based either in school (urban) or outside school (rural).
intention to perform the behavior that, in turn, can be explained Adolescents have easy access to street foods nearby school and
through 3 cognitive factors: attitudes, social influence (including sometimes street food vendors enter the school premises.
subjective norms, modeling and support) and self-efficacy.
Additionally, barriers and lack of skills might limit the possibility Data coding and analyses
to put the intention into practice [17]. As adolescents’ dietary Records were transcribed verbatim, translated into English and
behavior is strongly influenced by their environments [18], we cross-checked by 3 researchers. We used a deductive thematic
complemented our framework with a socio-ecological perspective. content analysis [24] which was based on both the literature and
In this model, dietary behavior is viewed as the interaction the theoretical framework of this study. This enabled us to identify
between, and interdependence of, factors within and across themes and factors influencing dietary behavior of adolescents.
multiple levels of influence. In other words, it highlights people’s The purpose of identifying these themes and factors was to build
interactions with their physical and socio-cultural environments up a model, i.e. a conceptual framework explaining the dietary
[16]. Both models have been used extensively to study dietary behavior of our participants. Using this analysis, 2 investigators
behaviors in young people [19–21]. independently read the transcripts and identified emergent
themes. For each participant group, a codebook based on these
Focus groups factors was developed independently by 2 researchers. If no
The protocol incorporated theoretical and practical guidelines agreement was reached on coding, a third researcher was
[22,23]. A double layer design using setting (urban/rural) as the consulted. The codebook was further validated on different
first layer and different audiences (adolescents, parents and school transcripts. NVivo software (QSR international – version 8.0)
staff) as the second layer, allowed for comparison and/or was used to code, manage and analyze the data. Summary reports
verification of results between these different layers [23]. The were written for each participant group according to identified
number of focus groups was defined prior to the start of the survey factors and themes. Moreover, focus group attributes, such as
[23] and considered sufficient as data saturation was reached. We socio-economic status and school setting were cross-linked with
conducted 20 focus groups, of which 12 were with adolescents constructs and themes for each participant group. For triangula-
separated by age group (6 for grade 8–9; 6 for grade 10–11) to tion of the data we took into account the non-verbal behavior,
produce homogenous groups, since ability and level of compre- group interactions and data from the parent and school staff
hension differs substantially with age [22]. In addition, 4 focus groups. Findings from the focus groups were grouped into
groups with parents and 4 with school staff were conducted. individual and environmental factors influencing eating behavior,
Participants received healthy refreshments as an incentive to which were subdivided into specific factors according to the
participate and completed a socio-demographic questionnaire; a literature and the theoretical framework used. Inclusion of factors
verbal record was taken in case of illiteracy. Audio-recorded focus was based on the frequency, specificity, emotion and extensiveness
groups, lasting 32 minutes on average, were conducted in Spanish of the quotes related to the factor [23]. Data from all participant
and led by a trained interviewer (AO). A silent observer (RV) was groups are presented for each selected factor and related quotes
present to take notes on non-verbal individual behavior and group are shown in Tables S1 and S2 (online material). The differences
interactions [22,23]. Using the theoretical framework, a semi- in these factors among the socio-economic levels and settings are
structured questioning route was developed, pre-tested and only presented where relevant.

PLOS ONE | www.plosone.org 2 January 2014 | Volume 9 | Issue 1 | e87183

 Healthy Eating Behaviour in Ecuadorian Adolescents

Results portion sizes, having breakfast, and eating regularly at set times are
important. Like adolescents, they associated eating healthily with
Twelve adolescent focus groups (N = 80) were conducted and traditional, home-grown and hygienically prepared food and not
group size ranged from 6–8 individuals. In addition, 4 parent necessarily with nutritional quality. Parents expressed their
(N = 32) and 4 school staff groups (N = 32) with an average group concerns about food safety in school tuck shops.
size of 8 were performed (Table 1). Attitudes. Overall, adolescents reported positive attitudes
The results are presented according to the two broad levels of towards healthy eating, with some of them associating healthy
individual and environmental influences, identified in the analysis. eating with a positive body image and health benefits, such as
Furthermore, environmental influences are presented according to looking good and being healthy. Nevertheless, they reported liking
the influences at school, family, and physical and societal level. ‘‘unhealthy food’’ so much that they could not resist it, even
though they were aware of its poor nutritional value. Parents and
Individual factors influencing eating behavior school staff in the study generally had positive attitudes towards
Awareness. Adolescents mainly discussed healthy eating by healthy eating but anticipated that adolescents would hold
identifying stereotype foods or food groups they perceived as negative attitudes.
(un)healthy, naming many more ‘‘unhealthy’’ than ‘‘healthy’’ Taste. Overall, adolescents were enthusiastic when talking
foods. Fruit and vegetables were perceived as healthy, while about the taste of sweet and fatty foods, while vegetables or salads
French fries, potato chips, candies and ‘junk food’ (referred to as were associated with unpleasant and negative taste experiences,
such by participants) were most frequently mentioned as particularly in the school environment. As such taste had an
‘‘unhealthy foods’’. On the other hand they mentioned, but less important impact on their preferences and consumption. This was
frequently, that eating healthily includes a balanced diet with a low re-iterated by parents and school staff.
amount of fat and lots of vitamins. Adolescents reported that they Self-efficacy. Many adolescents felt they would not succeed
were aware of the general health benefits of eating healthily. They in eating healthily and associated this inability with lack of self-
believed that traditional and home-grown foods are ‘always’ control and the abundance of tasty, yet ‘‘unhealthy food’’ at school
healthy as these were hygienically prepared at home. In contrast, and/or at home. Only a few adolescents indicated that they are or
street or restaurant foods and food out-of-home in general were would be capable of eating healthily.
perceived as unhealthy because preparation methods were School staff groups acknowledged their responsibility in
unknown. educating adolescents about healthy eating, but also stressed the
Parent and school staff groups reported that a healthy diet importance of parental responsibility. Surprisingly, parents did not
includes balanced and varied dietary practices in which moderate recognize their responsibility for their children’s dietary behavior,
but placed it with school, the environment or their children
themselves.
Table 1. Participant characteristics. Financial autonomy. Adolescents reported having financial
autonomy to choose food, generally originating from pocket
money received from parents/grandparents or money earned by
Total Urban Rural P-value
them. This pocket money was mainly used to purchase foods of
Adolescents (n = 80) poor nutritional quality at school. Even though no differences were
Gender (% male) 47.9 46.1 50.0 0.63 noted among adolescents from different socio-economic groups,
Age (mean (SD) yrs) 13.7 (1.2)* 13.7 (1.3)* 13.8 (1.1)* 0.67 parents from low socio-economic groups reported that their
School (% public) 62.5 31.2 68.8 ,0.01
children did not receive any/much money and mostly took food
from home to eat at school.
Socio-economic level based on schools
Habit strength. Most adolescent groups noted that their food
Low (%) 67.5 35 100 ,0.01 consumption was influenced by habit, which they reported has
Medium (%) 17.5 35 0 ,0.01 become less healthy since moving to secondary school. They
High (%) 15 30 0 ,0.01 identified the increased availability of ‘‘unhealthy food’’ and
Parents (n = 32) (financial) autonomy as main influences on their habits. A strong
habitual pattern was reported with regard to eating out at
Gender (% male) 25 11 43 0.04
weekends.
Age (mean (SD) yrs) 41.2 (10.7) 38.5 (6.5) 44.6 (14.1) 0.16
Parents and school staff groups also saw habit strength as a key
No. of children (mean (SD)) 2.9 (1.4) 2.7 (0.9) 3.1 (1.8) 0.39 influence. They expressed concern about the changes adolescents
Education face, such as increased (financial) autonomy and less parental
Illiterate (%) 6.5 5.5 5.6 0.001 control, and the transition from primary to secondary school
Primary (%) 45.2 84.6 16.6 0.001 accentuated the changes that have occurred in the socio-cultural
Secondary (%) 32.3 0 55.6 0.001
environment over recent years.
Subjective norm. Views on the pervasiveness of subjective
University (%) 16.1 7.7 22.2 0.001
norms on healthy eating varied among adolescent groups. Most
School staff (n = 32) adolescents reported being afraid of what others might think if
Gender (% male) 58.1 41.2 78.5 0.04 they ate healthily, such as embarrassment, being called ‘‘freaky’’,
Age (mean (SD) yrs) 36.7 (11.0) 39.6 (12.6) 33.1 (7.4) 0.09 ‘‘weird’’ or ‘‘not willing to spend money’’ or the possibility of being
Experience (mean (SD) yrs) 7.0 (8.7) 9.6 (11.6) 4.6 (4.0) 0.15 mocked by their peers. Positive perceptions were reported less
often and generally these adolescents felt confident and did not
*Date of birth was missing for 5 adolescents. care what their peers or other people thought.
P-values for urban-rural differences (two sample t-test, Chi square or Fisher Parents and school staff groups also emphasized the fear of
Exact test).
doi:10.1371/journal.pone.0087183.t001 embarrassment held by adolescents regarding eating healthily,

PLOS ONE | www.plosone.org 3 January 2014 | Volume 9 | Issue 1 | e87183

 Healthy Eating Behaviour in Ecuadorian Adolescents

indicating some strong social norms were operating in the peer ‘salchipapas’ (French fries with sausage), fried ‘empanada’ (deep-
environment. fried pastry snack) or rice with chicken/meat.
Perceived barriers. Adolescents from low socio-economic Environment outside home and school: socio-cultural
schools described the cost of healthy food as a barrier to eating changes and availability. Parents frequently stated that ‘junk
healthily, which was also stressed by parents. Furthermore, rural food’ is available everywhere, not only at school, but also outside
adolescents reported that availability was a barrier to eating school. In addition, parents from higher socio-economic groups
healthily. These 2 key factors were distinct for urban adolescents emphasized that media has a large impact on their children’s
who reported (as did parents) that food is readily available and cost eating habits, as food advertisements are specifically targeted
was not an issue. Some adolescents reported lack of time as a towards children. Parents and school staff believed that the
barrier for eating breakfast at home; this view was shared by availability of sweets and processed foods had increased since they
parents. All school staff and parent groups described the impact of were young. Both evoked the impact of the changing socio-cultural
the changing society and environment on lifestyles. Significant environment on traditional diets, food availability and family meal
barriers to eating healthily at home were: having less time to patterns. All these factors have led to increased portion sizes and a
prepare (healthy) meals, challenges of organizing their schedules variety of palatable foods with poor nutritional quality.
around family meals, and choosing convenient ready-to-eat dishes
which are preferred over ‘‘healthy foods’’. Conceptual framework
Based on our findings a composite conceptual framework was
Environmental influences on eating behavior proposed, in which adolescent eating behavior is conceptualized as
Family environment: parental rules, role modeling and a function of the identified individual and environmental
availability. Three key factors - parental rules, availability and influences (Figure 1). The framework emphasizes the interaction
role modeling – were identified. Some parents reported they try to of factors within and across these levels of influence. All of these
be a good role model for their children and include rules about factors may directly or indirectly influence adolescents’ dietary
healthy eating. Nevertheless, they confirmed that they inconsis- behavior. In addition to the more traditional influencing factors,
tently enforced rules about healthy eating and do not always set a the following culture-specific key factors were identified for our
good example for their children. They acknowledged that it is population: perceived food safety, lack of self-control, financial
difficult to expect their children to eat healthily if they do not do so autonomy, habit strength and changes in socio-cultural environ-
themselves. Parent groups reported that these inconsistencies arose ment. Furthermore, as acknowledged previously [13], our findings
from the fact that preparation and consumption of healthy food at indicated that the influence of these factors on behavior may differ
home is very often a negotiation process with adolescents. Due to according to socio-economic status and setting. This multilevel,
this constant struggle to encourage their children to eat healthily, interactive framework is useful for understanding and explaining
parents reported often giving in and adapting meals to children’s the factors influencing dietary behavior in Ecuadorian adolescents.
wishes. These inconsistencies were reflected in adolescents’
responses who stated that they tend to disobey rules on healthy Discussion
eating, particularly away from home. Nevertheless, adolescents
indicated that the availability of healthy food at home had an Eating behavior is influenced by inter-related factors reflecting
influence on their eating pattern, because they eat what is served ones’ personal, social and cultural experiences and environment
and available at home. Rural parents were most likely to evoke [25,26]. In addition, the reasons for choosing particular foods are
their dependency on their own food production to ensure that closely associated with concerns over identity, image and social
healthy food is available at home, whereas for urban participants belonging [26], which is ubiquitous in adolescence. Several
this was more related to availability in shops. culture-specific key factors - perceived food safety, lack of self-
School environment: school rules and availability. At control (attribution error), financial autonomy, habit strength and
school level, rules and availability were the 2 most important changes in socio-cultural environment - emerged from focus
factors. Urban adolescents reported food restrictions at school, e.g. groups endorsing the importance of the development of a
soft drinks and French fries. However, some adolescents did not conceptual framework in this population for future interventions.
feel constrained by these school rules and purchased their First, participants often associated eating healthily with food
preferred food outside school. This was different for rural safety issues and home-grown or -prepared food rather than with
adolescents, where no restrictions on food were in place, as the nutritional quality of their diet as a whole, which had an impact on
tuck shop was external to the school. School staff confirmed which foods were prohibited at school and on adolescents’ food
adolescents’ views on food restrictions at school and stated that choices. The importance of food safety in defining eating healthily
these were guided by food hygiene and safety practices, rather has been noted in previous research as an important factor for
than by nutritional quality. Food availability at school was viewed LMICs [27].
by adolescents as a key factor influencing their consumption, i.e. Whilst school staff recognized their role in shaping adolescents’
they eat what is available. Parent and school staff groups dietary behavior, they minimized their responsibility. They saw
confirmed the abundance of ‘junk food’ and poor availability of parents as gatekeepers of adolescents’ poor eating habits,
fresh fruit at school. However, they explained that food available suggesting that parents need to act as positive role models.
in the tuck shop is a reflection of adolescents’ preference for However, parents evoked their work commitments, the changing
processed food. Even when fresh fruit was available at the school socio-cultural environment, schools and their children’s food
tuck shop, it was not sold to adolescents as it was often seen as preferences as key influences on food choice. This concept of
unpalatable to them. However, all participant groups believed that attribution error, i.e. blaming situational factors when justifying
if fresh fruit looked appealing, was ready-to-eat and sold at an one’s behavior, has been demonstrated previously [28].
acceptable price then adolescents would be more willing to buy it. Increasing financial autonomy, which coincides with the
These tuck shops typically sell confectionery food, such as sweets, transition from primary to secondary school, played a large part
crisps, ice cream and soft drinks. In addition to these foods, some in adolescents’ food choices. This confirms previous findings in
of them offered warm snacks or meals during the break such as Vietnamese adolescents, where pocket money increased eating out

PLOS ONE | www.plosone.org 4 January 2014 | Volume 9 | Issue 1 | e87183

 Healthy Eating Behaviour in Ecuadorian Adolescents

Figure 1. Conceptual framework for eating behavior in Ecuadorian adolescents.

doi:10.1371/journal.pone.0087183.g001

frequency [29]. Starting secondary school is a critical period of previous focus groups in LMICs [27]. These differences might
increasing independence as the extent of parental support for explain why participants from rural and low socio-economic
eating healthily decreases and the desire to fit in with peer norms schools reported lower availability of healthy food at home and
increases [18]. This process, in conjunction with easy access and could not afford to buy ‘‘healthy foods’’. Previously, the
constant exposure to tasty and unhealthy food in schools, explains importance of socio-demographic factors as moderating factors
the adolescents’ indication of their deteriorating eating habits. This or effect modifiers of behavior has been established [37,38]. This
deterioration is accentuated further by the increasing difficulty means that influencing factors may have differential effects on
adolescents have to eat healthily within the rapidly changing socio- behavior with respect to socio-economic status [13,39,40] and
cultural environment, which has impacted on family life and food setting [41], which supports the inclusion of these as moderating
availability, mirroring the ongoing nutrition transition in Ecuador factors in the conceptual framework.
[30] and other LMICs. Ready-to-eat meals in large portion sizes Adolescent participants might have experienced difficulties in
are now the norm, due to busy family work schedules. A similar sharing their views within the focus groups due to social
pattern has occurred in other countries where women’s roles have desirability and peer pressure. Yet, we do not believe this
changed, leading to a loss of cooking skills and an increasing influenced our results to a great extent, as the moderator tried
reliance on convenience foods [31]. A daunting prospect, as eating to establish a friendly and comfortable environment encouraging
out and relying on convenience foods has been associated with active participation and secondly, and more importantly, findings
poor dietary intake in LMICs [32]. did not differ across adolescent groups. We aimed at minimizing
In line with previous findings from HICs, taste [33], availability bias by using triangulation and standardized data collection
and accessibility [27,34], self-efficacy, financial constraints, time procedures. Since parents and school staff re-iterated the findings
and convenience [25,34] emerged as important features in of the adolescent focus groups we can assume these findings are
adolescents’ food choices. In addition, strong subjective peer valid. Furthermore, despite the accumulating evidence of
norms were present - choosing to eat healthily was often associated unhealthy dietary practices, dietary behavior remains poorly
with an untrendy image leading to teasing from others and understood in young people in LMICs [37]. Few attempts have
marginalization - supporting the preferences for unhealthy foods of been made to use theory to guide the development and evaluation
adolescents. Similarly, Stead et al. (2011) found that ‘‘it’s emotionally of interventions [12]. Additionally, testing the validity of these
and socially risky to be seen to be interested in healthy eating’’ for adolescents theories, i.e. their appropriateness to specific cultures and local
in school and peer contexts [26]. To conclude, rules at home and contexts, is rarely undertaken [42]. This study adds to the current
at school were inconsistent, so adolescents were likely to receive evidence-base, by identifying key factors influencing Ecuadorian
contradictory messages that they regarded as marginal and they adolescents’ eating behavior and developing a composite concep-
developed strategies for buying their preferred food elsewhere. tual framework. The factors identified within this framework
This might be an indirect indication that parental influence is less should be investigated using culturally appropriate scales with
important in this group than peer influence. Similar associations good psychometric properties. Doing so would allow this
between mixed messages and adolescent eating preferences have framework to be tested by evaluating the inter-relationships and
been found in previous research [35]. Nevertheless, the impact of association of these factors with dietary behaviors. Additionally, it
parents might differ across behaviors (e.g. fruit and vegetable facilitates tailoring of intervention strategies towards these factors,
consumption versus sugary drink intake) [36]. and could be used to identify pathways of behavior change when
Few socio-demographic differences emerged. Availability and evaluating interventions [42].
financial constraints clearly differed between the rural and urban Our conceptual framework indicates that future interventions
area and the socio-economic groups, supporting findings from should not only consider individual, peer and family influences

PLOS ONE | www.plosone.org 5 January 2014 | Volume 9 | Issue 1 | e87183

 Healthy Eating Behaviour in Ecuadorian Adolescents

when aiming to change adolescent eating habits, but should target Supporting Information
the physical school and social environment as well, which is
consistent with findings from other studies [43]. A particular focus Table S1 Quotes on individual factors influencing
on school policies including regulation on food sold at the tuck eating behavior in adolescents, parents and school staff.
shop based on its nutritional value and control of food practices is A: adolescents; P: parents; S: school staff; …: short silence; […]:
needed. Such strategies need to be tailored to the specific settings overlapping speech; __: emphasis; £ £: smiley voice; 11: laughing;
and socio-economic conditions, even though this might be / /: irony; (()): transcribers’ comments.
challenging [12]. Specifically, the intervention should take into (DOCX)
account the issue of attribution error amongst parents and school Table S2 Quotes on environmental factors influencing
staff. Despite the possible relative importance of parents, they still eating behavior in adolescents, parents and school staff.
play an important role in the daily life and dietary behavior of A: adolescents; P: parents; S: school staff; …: short silence; […]:
adolescents and should be included when designing interventions overlapping speech; __: emphasis; £ £: smiley voice; 11: laughing;
[38], particularly in LMICs [12]. On a positive note, all / /: irony; (()): transcribers’ comments.
participant groups requested practical advice on how to eat
(DOCX)
healthily and develop skills.
Acknowledgments
Conclusion
DA and SA made substantial contributions to acquisition of data in the
Focus groups provided a clear insight into the factors that field. In addition, we would like to thank the schools, adolescents and
influence adolescents’ dietary behavior. Adolescents, parents and parents who participated in this study.
school staff identified financial autonomy, food safety, self-efficacy,
habit strength and socio-cultural changes as key cultural factors in Author Contributions
adolescent’s food choices. As a consequence, a conceptual
framework for adolescents’ eating behaviors emerged, which Conceived and designed the experiments: RV PK LM SD MH. Performed
highlights points of leverage for developing future interventions. the experiments: AO DP RV. Analyzed the data: RV KVR. Contributed
reagents/materials/analysis tools: N/A. Wrote the paper: RV KVR AO
Interactions between the identified factors in the conceptual
DP MH SD LM PK.
framework and eating behaviors should be studied using structural
equation or mediation analysis.

References
1. Abegunde DO, Mathers CD, Adam T, Ortegon M, Strong K (2007) Chronic Senior Farmers’ Market Nutrition Pilot Program. Prev Chron Dis [serial online]
diseases 1 - The burden and costs of chronic diseases in low-income and middle- 1: 1–11.
income countries. Lancet 370: 1929–1938. 18. Story M, Neumark-Sztainer D, French S (2002) Individual and environmental
2. Goldstein J, Jacoby E, del Aguila R, Lopez A (2005) Poverty is a predictor of influences on adolescent eating behaviors. J Am Diet Assoc 102: S40–S51.
non-communicable disease among adults in Peruvian cities. Prev Med 41: 800– 19. Brug J, de Vet E, de Nooijer J, Verplanken B (2006) Predicting fruit
806. DOI 10.1016/j.ypmed.2005.06.001. consumption: Cognitions, intention, and habits. Journal of Nutrition Education
3. Monteiro CA, Conde WL, Popkin BM (2007) Income-specific trends in obesity and Behavior 38: 73–81.
in Brazil: 1975–2003. Am J Public Health 97: 1808–1812. DOI 10.2105/ 20. Sandvik C, Gjestad R, Brug J, Rasmussen M, Wind M, et al. (2007) The
AJPH.2006.099630. application of a social cognition model in explaining fruit intake in Austrian,
4. Baillie K (2008) Health implications of transition from a planned to a free- Norwegian and Spanish schoolchildren using structural equation modelling.
market economy - an overview. Obes Rev 9: 146–150. International Journal of Behavioral nutrition and physical activity 4.
5. Ebrahim S, Garcia J, Sujudi A, Atrash H (2007) Globalization of behavioral risks 21. Townsend N, Foster C (2013) Developing and applying a socio-ecological model
needs faster diffusion of interventions. Prev Chron Dis [serial online]. to the promotion of healthy eating in the school. Public Health Nutrition 16:
6. Wang Y, Lobstein T (2006) Worldwide trends in childhood overweight and 1101–1108.
obesity. Int J Pediatr Obes 1: 11–25. DOI 10.1080/17477160600586747. 22. Kennedy C, Kools S, Krueger R (2001) Methodological considerations in
7. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS (1999) The relation of children’s focus groups. Nurs Res 50: 184–187.
overweight to cardiovascular risk factors among children and adolescents: The 23. Krueger R, Casey MA (2000) Focus groups: A practical guide for applied
Bogalusa heart study. Pediatrics 103: 1175–1182. research. US: Sage Publications, Inc.
8. Must A (1996) Morbidity and mortality associated with elevated body weight in 24. Elo S, Kyngas H (2008) The qualitative content analysis process. J Adv Nurs 62:
children and adolescents. Am J Clin Nutr 63: S445–S447. 107–115.
9. Virdis A, Ghiadoni L, Masi S, Versari D, Daghini E, et al. (2009) Obesity in the 25. Bisogni CA, Jastran M, Seligson M, Thompson A (2012) How People Interpret
childhood: A link to adult hypertension. Curr Pharm Design 15: 1063–1071. Healthy Eating: Contributions of Qualitative Research. J Nutr Educ Behav 44:
10. Lee IM, Shiroma EJ, Lobelo F, Puska P, Blair SN, et al. (2012) Effect of physical 282–301.
inactivity on major non-communicable diseases worldwide: an analysis of 26. Stead M, McDermott L, MacKintosh AM, Adamson A (2011) Why healthy
burden of disease and life expectancy. Lancet 380: 219–229. eating is bad for young people’s health: Identity, belonging and food. Social
11. World Health Organization (2002) Diet, Nutrition, and the Prevention of Science & Medicine 72: 1131–1139.
Chronic Diseases: Report of a joint WHO/FAO Expert Consultation. WHO 27. Nago E, Verstraeten R, Lachat C, Dossa R, Kolsteren P (2012) Food safety is a
technical report series: No 916. key determinant of fruit and vegetable consumption in urban Beninese
12. Verstraeten R, Roberfroid D, Lachat C, Leroy JL, Holdsworth M, et al. (2012) adolescents. J Nutr Educ Behav 44: 548–555.
Effectiveness of preventive school-based obesity interventions in low- and 28. Power TG, Bindler RC, Goetz S, Daratha KB (2010) Obesity Prevention in
middle-income countries: a systematic review. Am J Clin Nutr 96: 415–438. Early Adolescence: Student, Parent, and Teacher Views. J Sch Health 80: 13–
13. Kremers SPJ, De Bruijn G-J, Visscher TLS, Van Mechelen W, De Vries NK, et 19.
al. (2006) Environmental influences on energy balance-related behaviors: a dual- 29. Lachat C, Le NBK, Nguyen CK, Nguyen QD, Nguyen DVA, et al. (2009)
process view. Int J Behav Nutr Phys Act 3: 9. Eating out of home in Vietnamese adolescents: socioeconomic factors and
14. Tong A, Sainsbury P, Craig J (2007) Consolidated criteria for reporting dietary associations. Am J Clin Nutr 90: 1648–1655.
qualitative research (COREQ): a 32-item checklist for interviews and focus 30. Ochoa-Avilés A, Andrade S, Huynh TTT, Verstraeten R, Lachat C, et al.
groups. Qual Assur Health Care 19: 349–357. (2012) Prevalence and socioeconomic differences of risk factors of cardiovascular
15. De Vries H, Dijkstra M, Kuhlman P (1988) Self-efficacy: the third factor besides disease in Ecuadorian adolescents. Pediatr Obes 7: 274–283.
attitude and subjective norm as a predictor of behavioural intentions. Health 31. Pettinger C, Holdsworth M, Gerber M (2006) Meal patterns and cooking
Educ Res 3: 273–282. practices in Southern France and Central England. Public Health Nutr 9: 1020–
16. Green LW, Richard L, Potvin L (1996) Ecological foundations of health 1026.
promotion. Am J Health Promot 10: 270–281. 32. Lachat C, Nago E, Verstraeten R, Roberfroid D, Van Camp J, et al. (2012)
17. Smith LT, Johnson DB, Beaudoin S, Monsen ER, LoGerfo JP (2004) Eating out of home and its association with dietary intake: a systematic review of
Qualitative assessment of participant utilization and satisfaction with the Seattle the evidence. Obes Rev 13: 329–346.

PLOS ONE | www.plosone.org 6 January 2014 | Volume 9 | Issue 1 | e87183

 Healthy Eating Behaviour in Ecuadorian Adolescents

33. Haerens L, De Bourdeaudhuij I, Barba G, Eiben G, Fernandez J, et al. (2009) 12-year-old children: a systematic review within the ENERGY (EuropeaN
Developing the IDEFICS community-based intervention program to enhance Energy balance Research to prevent excessive weight Gain among Youth)
eating behaviors in 2-to 8-year-old children: findings from focus groups with project. Public Health Nutr 15: 1380–1395.
children and parents. Health Educ Res 24: 381–393. 39. Brug J, Kremers SP, van Lenthe F, Ball K, Crawford D (2008) Environmental
34. Kubik MY, Lytle L, Fulkerson JA (2005) Fruits, vegetables, and football: determinants of healthy eating: in need of theory and evidence. Proc Nutr Soc
Findings from focus groups with alternative high school students regarding 67: 307–316.
eating and physical activity. J Adolesc Health 36: 494–500. 40. Abraham C, Sheeran P, Henderson M (2011) Extending social cognition models
35. Krolner R, Rasmussen M, Brug J, Klepp KI, Wind M, et al. (2011) of health behaviour. Health Education Research 26: 624–637.
Determinants of fruit and vegetable consumption among children and 41. Popkin BM, Paeratakul S, Zhai FY, Ge KY (1995) A Review of Dietary and
adolescents: a review of the literature. Part II: qualitative studies. International Environmental Correlates of Obesity with Emphasis on Developing-Countries.
Journal of Behavioral nutrition and physical activity 8. Obesity Research 3: S145–S153.
36. Lau RR, Quadrel MJ, Hartman KA (1990) Development and Change of 42. Cerin E, Barnett A, Baranowski T (2009) Testing Theories of Dietary Behavior
Young-Adults Preventive Health Beliefs and Behavior - Influence from Parents Change in Youth Using the Mediating Variable Model with Intervention
and Peers. Journal of Health and Social Behavior 31: 240–259.
Programs. Journal of Nutrition Education and Behavior 41: 309–318.
37. Rasmussen M, Krolner R, Klepp KI, Lytle L, Brug J, et al. (2006) Determinants
43. Neumark-Sztainer D, Story M, Perry C, Casey MA (1999) Factors influencing
of fruit and vegetable consumption among children and adolescents: a review of
food choices of adolescents: Findings from focus-group discussions with
the literature. Part I: quantitative studies.
adolescents. J Am Diet Assoc 99: 929–937.
38. Verloigne M, Van Lippevelde W, Maes L, Brug J, De Bourdeaudhuij I (2012)
Family- and school-based correlates of energy balance-related behaviours in 10-

PLOS ONE | www.plosone.org 7 January 2014 | Volume 9 | Issue 1 | e87183