The Mediterranean dietary pattern has been used to describe the traditional dietary practices of countries bordering the Mediterranean Sea, such as Greece, southern Italy, Spain, Cyprus and Turkey, amongst others. This dietary pattern is rich in plant-derived products such as fruits, vegetables, breads and cereals (mainly wholegrains), beans, nuts and seeds, with olive oil as the main culinary fat used. It is also characterised by a low consumption of red meat and low-to-moderate intake of fish, poultry and wine, as well as high intakes of MUFA, fibres and antioxidants, such as vitamins E and C, resveratrol, polyphenols and Se(Reference Davis, Bryan and Hodgson1). There is a growing evidence of the numerous health benefits of this diet, including a reduction in the risk of CVD and a lengthened life expectancy(Reference Bonaccio, Iacoviello and De Gaetano2). Adherence to the Mediterranean diet (MD) was inversely associated with the metabolic syndrome(Reference Godos, Zappalà and Bernardini3), and with lower risks of cancer(Reference Schwingshackl and Hoffmann4), and mental diseases such as depression(Reference Aridi, Walker and Wright5). These health benefits are mainly due to the synergistic effect of the diet’s nutritional components, as well as other healthier habits like physical activity and social interaction such as sharing meals with others.
In recent decades, the majority of Mediterranean countries have more and more frequently abandoned this dietary pattern and substituted it with a high-energy, Westernised diet that is abundant in saturated fats while delivering a low amount of micronutrients. As a consequence of this nutrition transition, the obesity epidemic, along with other metabolic disorders, has risen dramatically in these countries(Reference Musaiger, Al-Hazzaa and Takruri6). The age group that has been most affected by this phenomenon is children and adolescents, where the traditional MD has been abandoned amongst youth in Turkey(Reference Ċomert, Cerkez and Tekin7), Italy(Reference Roccaldo, Censi and D’Addezio8) and Greece(Reference Mazaraki, Tsioufis and Dimitriadis9), and was mainly related to poor consumption of fruits, vegetables and legumes and an overconsumption of commercially baked goods and sweets. Thus, in Lebanon, similar observations have been reported amongst a nationally representative sample, where 34·8 % of children and adolescents aged 6–19 years were overweight, and 13·2 % were obese(Reference Nasreddine, Naja and Akl10).
The roots of most chronic diseases encountered during adulthood are largely linked to the nutritional choices adopted during adolescence(Reference Cömert, Çerkez and Tekin11). Obese adolescents were at higher risk of becoming obese adults(Reference The, Suchindran and North12), and developing many chronic diseases in adulthood, including CVD, type 2 diabetes, certain types of cancers, in addition to psychological disorders(Reference Black, Victora and Walker13,Reference Park, Falconer and Viner14) . Therefore, it is crucial to target this age group in order to decrease disease risks. The KIDMED index was developed by a group of Spanish researchers in order to evaluate youth (2–24 years old) compliance with the MD(Reference Serra-Majem, Ribas and Ngo15). The results of a systematic review indicated that the KIDMED index was the most commonly used scoring index for assessing adherence to the MD amongst children and adolescents when compared with other tools such as the Mediterranean Diet Score and the food frequency-based Mediterranean Diet Score(Reference Idelson, Scalfi and Valerio16). The KIDMED questionnaire has shown good validity and reliability for evaluating adherence to MD amongst young population from Mediterranean(Reference Rei, Severo and Rodrigues17,Reference Štefan, Prosoli and Juranko18) and non-Mediterranean countries(Reference Atencio-Osorio, Carrillo-Arango and Correa-Rodríguez19,Reference Carrillo and Ramírez-Vélez20) . It has demonstrated adequate psychometric properties for assessing adherence to the MD in schoolchildren(Reference Carrillo and Ramírez-Vélez20). A review of the current available literature indicates that data are lacking in respect of factors associated with adolescents’ adherence to the MD in North Lebanon. To our knowledge, the dietary pattern adopted by Lebanese teenagers has been assessed in the past using the FFQ and 24-h recall; however, none used the KIDMED index(Reference Naja, Hwalla and Itani21). Hence, the objective of this study was to assess the level of MD adherence using the KIDMED test in a sample of Lebanese adolescents. The secondary objective was to explore the association between the adherence to the MD and the sociodemographic, eating habits and lifestyle factors of the participants, and to compare their dietary practices with similar studies conducted amongst adolescents in the Mediterranean region.
Methods
Study design and participants
This cross-sectional study was conducted between January and March 2019. The study included students aged 11–18 years enrolled in public and private schools in Tripoli for the academic years 2018–2019. Published studies regarding the adherence to the MD demonstrated that similar methodologies were used in the younger age group(Reference Archero, Ricotti and Solito22,Reference Muros, Cofre-Bolados and Arriscado23) .
Sampling procedure
Based on a previously published paper conducted on Turkish adolescents(Reference Sahingoz and Sanlier24), a total sample size of 780 students, distributed over the three categories of the score, was required to allow the detection of an 11 % decrease in KIDMED with a 5 % significance level and a 90 % study power. Turkey is a country with borders on the Mediterranean Sea, whose adolescents were also affected by this nutritional transition similar to the Lebanese youth(Reference Ċomert, Cerkez and Tekin7).
Records provided from the Ministry of Education were used to compile a sampling frame of all schools in the City of Tripoli. Schools were randomly selected based on the stratified cluster sampling method. The strata were public and private schools, and complementary and secondary classes. Clusters were defined as the schools within each stratum, from which two schools were randomly selected. In total, our sample was composed of eight schools. The end units were all the students within the selected schools aged between 11 and 18 years and agreeing to participate. Students suffering from physical disabilities or who were absent on the days of the data collection were not invited to participate in the study.
Ethical approval
The study was approved by the Institutional Review Board of Beirut Arab University. Consent from students’ parents was obtained regardless of whether the student was attending a private or public school.
Data collection
Each student completed a 25-min questionnaire, which was administered during school hours under the supervision of dietitians who received previous training to standardise the data collection procedures. A training session consisting of 2 h was performed prior to the data collection, where the objective of the study and sampling procedure were addressed, and special attention was provided to the standardisation of the anthropometric measurements procedure. The multicomponent questionnaire, which was completed through a face-to-face interview, evaluated the eating habits, sociodemographic characteristics, physical activity and quality of life of the participants. Each student was interviewed separately to ensure his privacy. The questionnaire was piloted in advance on a group of thirty students from different ages, in order to ensure the clarity of the questions. The data collection was carried out in the classroom in the presence of a teacher who helped in the questionnaire interpretation if needed. The results of piloting the questionnaire revealed that most of the questions were clear, except for two sentences that were paraphrased. Field monitoring was executed on the days of the school visits.
Measures
Eating habits
The KIDMED index was used to assess adherence to the MD and general meal habits (such as skipping breakfast). The index comprises sixteen yes-or-no questions with a total score ranging from –4 to +12(Reference Serra-Majem, Ribas and Ngo15). Questions with a negative connotation (skipping breakfast, going to a fast-food restaurant more than once a week, eating pastries or baked goods for breakfast, eating sweets or candies several times a day) were given a score of –1, while the rest were given a score of +1. The total score was organised into three categories: ≤3, poor adherence; 4–7, average level of adherence and ≥8–12, good adherence(Reference Kontogianni, Vidra and Farmaki25). The KIDMED index was followed by questions about the number of meals consumed daily and consumed with the family, as well as about the frequency of meals that were skipped.
Sociodemographic characteristics
Each adolescent provided information regarding his or her age, sex, grade level and parent’s educational status (primary, complementary, secondary or university).
Quality of life
The KIDSCREEN-27, an internationally validated questionnaire, was used to assess the health-related quality of life of the participants. This tool consists of twenty-seven questions combining five dimensions of well-being: physical wellbeing (five items), psychological well-being (seven items), parent relations and autonomy (seven items), social support and peers (four items) and school environment (four items). The final score is the sum of all item responses. Better well-being was reflected through a high KIDSCREEN score(Reference Ravens-Sieberer, Auquier and Erhart26).
The level of physical activity was evaluated using the physical activity questionnaires for children (below the age of 14 years) and adolescents (above the age of 14 years)(Reference Kowalski, Crocker and Faulkner27).
Anthropometric measurements
Weight, height and waist circumference were measured for each student in duplicates using standardised techniques, and the average of two values was adopted(Reference Lee and Nieman28). Height was assessed in a standing position without shoes to the nearest 0·1 cm using a stadiometer. Feet were put together with buttocks, heels, scapulae and back of the head against the vertical board of the stadiometer during maximal expiration. Participants were weighed with light indoor clothing using an electronic digital calibrated scale to the nearest 0·1 kg. Waist circumference was measured in a standing position with non-stretchable flexible tape to the nearest 0·1 cm. BMI was calculated using the following formula: weight (kg)/height (m2). The level of obesity was then estimated based on the Center for Disease Control and Prevention’s age- and sex-specific growth charts. A value greater than the 85th percentile reflects overweight or obesity(Reference Kuczma and rski29). Central adiposity was defined according to the waist:height ratio (WHtR). A value of 0·5 or greater was used to detect children with elevated WHtR(Reference McCarthy and Ashwell30).
Statistical analysis
Means and standard deviations were calculated for continuous variables, while frequencies and percentages were generated for categorical variables. ANOVA test for the comparison of three means and χ 2 test were used to significantly identify the factors associated with adherence to MD amongst Lebanese adolescents. Multiple linear regression analysis using a stepwise technique was applied to identify the factors associated with adherence to the MD amongst Lebanese adolescents in a multivariate approach, where variables included in the equation were presented with their corresponding coefficients and P-values. To control for school’s type, as children in the same school are likely to be correlated on a number of key variables, we performed a ‘mixed model’ analysis for the PAQ score and the KIDSCREEN-27 total score and the different subscales (physical well-being, psychological well-being, parent relations and autonomy, social support and peers and school environment), considering KIDMED categories as fixed effects and school as random effects. Data analysis was carried out using the Statistical Package for the Social Sciences (SPSS/PC version 22.0) software. All tests were two-tailed, and a P-value <0·05 was considered statistically significant.
Results
The baseline characteristics of the study sample (n 798) are presented in Table 1. The mean age of the participants was 15 years. Participants were distributed between complementary (56·4 %) and secondary classes (43·6 %). Those from private schools represented 52·5 % (n 419) of the whole sample. In addition, more than half of the parents attained a high educational level, especially the mothers (65·4 %). Prevalence rates of generalised overweight/obesity (according to BMI-for-age) and those of central obesity (evaluated by WHtR) were, respectively, 39·5 and 38·7 %. Moreover, skipping meals was highly prevalent, affecting almost 80 % of the sample.
Table 1. Characteristics of the adolescents by adherence to Mediterranean diet, health-related quality of life, physical activity, sociodemographic and anthropometric measurements*
(Mean values and standard deviations; numbers and percentages, n = 798)
PAQ, Physical Activity Questionnaire.
* Means and standard deviations were calculated for continuous variables, while frequencies and percentages were generated for categorical variables.
Table 2 represents the factors associated with adherence to the MD amongst Lebanese adolescents in bivariate analysis. Adherence to the MD was high in 13·3 %, average in 57·5 % and poor in 29·2 % of the sample. Higher adherence to MD was reported amongst adolescents attending private schools when compared with those attending public schools (P = 0·021). Furthermore, students whose parents had a higher educational level benefited from a better adherence to the MD (P = 0·048). Skipping meals more often was significantly associated with a poor diet quality, while adolescents who consumed more meals per day with their families had a significantly higher KIDMED score (P = 0·001 and P < 0·001, respectively). A higher level of physical activity was also associated with a better adherence to the MD (P < 0·001). Finally, adolescents with a higher KIDMED score enjoyed a better general well-being (based on the KIDSCREEN total score) (P < 0·001), with significantly higher levels of physical and psychological well-being and better parental relationships, autonomy and school environment.
Table 2. Factors associated with adherence to Mediterranean diet amongst Lebanese adolescents
(Numbers and percentages; mean values and standard deviations, bivariate analysis, n = 798)
PAQ, Physical Activity Questionnaire.
The difference between groups was examined using the χ 2 test for categorical variables and ANOVA test for continuous variables.
* P < 0·05 is considered as significant.
When all significant determinants following the bivariate analysis are taken into account in the multivariate analysis, the variables that remain positively associated with a higher KIDMED score are consuming more meals with the family (P < 0·001), engaging in a higher level of physical activity (P < 0·001) and a better physical well-being according to the KIDSCREEN score (P < 0·01), while skipping meals tends to be negatively correlated with a low KIDMED score (P = 0·001) (Table 3). The results of the mixed models revealed that three variables similar to the multivariate analysis came to be significant (skipping meals, meals with the family and physical activity), in addition to the school environment which emerged as a new variable instead of the physical well-being, which was significant in the multivariate analysis (P < 0·001) (Table 4).
Table 3. Factors associated with KIDMED score – multiple regression analysis (n = 798)
VIF: variance inflation factor.
Table 4. Factors associated with KIDMED score – mixed model approach (n = 798)
Discussion
The current study reported data from 798 adolescents attending public and private schools in Tripoli, the capital of the North Lebanon Governorate, and evaluated the factors associated with adherence to the MD. The level of adherence to the MD was considered as poor amongst the participants; in fact, only 13·3 % of the adolescents had a good adherence to this traditional dietary pattern. A low adherence to the MD was also reported amongst Lebanese adolescents from private and public schools across Mount Lebanon and Beirut (15·8 %) in a recent study of 607 adolescents aged between 15 and 18 years(Reference Mounayar, Jreij and Hachem31).
Significant decreases in adherence to the MD were observed between 1997 and 2015 (using two different indexes: the composite MD and the Lebanese MD). Projections in 2030 reveal additional decreases, with less than a quarter of Lebanese adolescents continuing to follow this dietary pattern(Reference Naja, Hwalla and Hachem32). Compared with other Mediterranean countries where the KIDMED index was used, our results were above those reported amongst Turkish (7·7 %)(Reference Cömert, Çerkez and Tekin11), Italian (5 %)(Reference Roccaldo, Censi and D’Addezio8) and Greek adolescents (6·8 %)(Reference Mazaraki, Tsioufis and Dimitriadis9), but lower when compared with Spanish (30·9 %)(Reference Grao-Cruces, Nuviala and Fernández-Martínez33) and Cretan adolescents (28·3 %)(Reference Chatzi, Apostolaki and Bibakis34).
In the present study, skipping meals was associated with poor adherence to the MD. Similar findings were reported amongst Lebanese adolescents in Mount Lebanon and Beirut, where skipping breakfast in particular was correlated with poor adherence to the MD(Reference Mounayar, Jreij and Hachem31). Skipping meals is associated with unnecessary snacking, particularly unhealthy food choices and energy-dense foods rather than the nutrient-dense items that are part of the MD(Reference Grosso, Marventano and Buscemi35,Reference Mithra, Unnikrishnan and Thapar36) . Snacking is correlated with a reduction in the consumption of fruits and vegetables, which are essential food items of the MD(Reference Falbe, Willett and Rosner37). The PANACEA study carried out in Greece revealed that the consumption of unhealthy salty snacks was associated with poor eating habits such as low adherence to the MD(Reference Arvaniti, Priftis and Papadimitriou38).
Another finding of this study relates to physical activity, which was strongly associated with adherence to MD. Having a healthy lifestyle involves making the right food choices and being physically active. Amongst Spanish adolescents, engagement in physical activities or, in contrast, spending more than 4 h daily watching television were main predictors of adherence to the MD(Reference Bibiloni, Fernandez-Blanco and Pujol-Plana39). Likewise, Greek and Italian adolescents who had low adherence to the MD were found to lead a more sedentary lifestyle(Reference Papadaki and Mavrikaki40,Reference Santomauro, Lorini and Tanini41) . The level of physical activity mediated the relationship between children’s BMI and their adherence to the MD in the PANACEA study(Reference Priftis, Panagiotakos and Anthracopoulos42).
The current study demonstrated the beneficial influence of sharing meals with the family on improving adherence to the MD. Indeed, home-prepared meals are mostly based on traditional Lebanese cuisine, which focuses on various healthy and balanced dishes that are part of the MD. In addition, students living with their parents had a higher consumption of fruits and vegetables when compared with their peers who lived in a dorm(Reference El Ansari, Stock and Mikolajczyk43). Inversely, a transition to a more Westernised dietary pattern characterised the diet of young Greeks who had left their family home. Their new diet consisted of a higher intake of snack foods, fast foods, sugar and alcohol, with a reduced intake of vegetables, fruits, pulses and olive oil(Reference Kremmyda, Papadaki and Hondros44,Reference Papadaki, Hondros and Scott45) .
This study showed a strong association between adherence to an MD and self-perceived physical health, based on the KIDSCREEN questionnaire. The MD is abundant in vitamins, minerals and antioxidants, n-3 long-chain PUFA and MUFA (from olive oil), and the diet’s positive effects on physical and mental health have been proven in many studies, lowering the incidence of diabetes, CVD and certain types of cancers(Reference Schwingshackl and Hoffmann4,Reference Mattioli, Palmiero and Manfrini46) . In a cohort study amongst 11 015 Spanish adults, better adherence to the MD was found to be significantly correlated with improved mental and physical health(Reference Priftis, Panagiotakos and Anthracopoulos42). Several other authors have associated the MD constituents with a better health-related quality of life(Reference Sánchez, Ruano and De Irala47–Reference Panagiotakos, Chrysohoou and Pitsavos53).
Better adherence to the MD has been found to be associated with lower adiposity markers (such as BMI and WHtR) in some studies(Reference Panagiotakos, Chrysohoou and Pitsavos53,Reference Schröder, Marrugat and Vila54) . In some cases, this correlation was mediated by socio-economic status(Reference Lazarou, Panagiotakos and Matalas55), as well as the educational level of the parents(Reference Antonogeorgos, Panagiotakos and Grigoropoulou56). However, other investigators found no association at all(Reference Mazaraki, Tsioufis and Dimitriadis9,Reference Trichopoulou, Naska and Orfanos57) . The results of the current study could not find any association with BMI, nor with the WHtR, which could be explained by the fact that more than half of the adolescents (55·4 %) had a normal BMI-for-age and normal WHtR (61·3 %).
The findings of the present study should be seen in the light of their strengths and limitations. First, the authors used questionnaires that have been internationally validated amongst adolescents, such as the KIDMED, KIDSCREEN and the PAQ. Another main strength of this study is that data collection was carried out through interviews, which could reduce the bias of self-reported data. In addition, the study was representative of adolescents attending schools in Tripoli. On the other hand, one of the limitations is the cross-sectional design, which prevents the investigation of causal relationships. Consequently, longitudinal studies should be undertaken to further investigate the association between those components in regard to adherence to the MD.
Conclusion and future implications
The results obtained from our study confirm previous findings reporting low adherence to the traditional MD amongst Lebanese adolescents. Consuming meals with the family, being physically active and physical well-being was positively associated with better adherence to the MD, whereas skipping meals tends to be negatively correlated.
Since adolescence is a critical phase in shaping health and nutritional choices, future interventions targeting Lebanese adolescents should be planned. These interventions should orient adolescents towards the benefits of the MD and should offer practical tips to improve their adherence. These types of interventions have shown a positive influence on shaping children’s eating habits by increasing their fruits, vegetables, yogurt and/or cheese intake, while decreasing skipping breakfast, consumption of bakery products for breakfast and consumption of sweets(Reference Bibiloni, Fernandez-Blanco and Pujol-Plana39). Schools can act as a mediator in changing the adolescents’ eating behaviours and, ultimately, in preventing obesity(Reference Li, Yang and Lowry58). Intervention programmes that target nutrition and physical activity showed health benefits in preventing childhood obesity(Reference Amini, Djazayery and Majdzadeh59). A school-based intervention programme delivered by a trained schoolteacher was successful in improving Lebanese children eating behaviours, such as increasing their breakfast consumption, while reducing crisp intake(Reference Habib-Mourad, Ghandour and Maliha60). Another intervention delivered by nutritionist reduced the purchase of high-energy-dense snacks and beverages and raised the students’ nutritional knowledge(Reference Habib-Mourad, Ghandour and Moore61). Furthermore, parents should participate in these programmes through educating them on how to make healthier choices when buying or preparing food at home.
Acknowledgements
The authors would like to thank the directors of the schools in Tripoli and the students for their participation in the study.
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
R. M. conceived and designed the study; she contributed to the questionnaire design, data collection supervision and coordination, data analysis and interpretation and writing and finalising the manuscript. F. Z. provided expertise in sampling methods, statistical analysis and revised and approved the final draft of the manuscript. C. B. assisted in the interpretation of data and revised and approved the final draft of the manuscript.
There are no conflicts of interest.
