Hostname: page-component-586b7cd67f-t8hqh Total loading time: 0 Render date: 2024-11-25T19:28:49.016Z Has data issue: false hasContentIssue false

Is adherence to the Mediterranean diet associated with healthy habits and physical fitness? A systematic review and meta-analysis including 565 421 youths

Published online by Cambridge University Press:  09 December 2020

Antonio García-Hermoso*
Affiliation:
Navarrabiomed, Complejo Hospitalario de Navarra (CHN), Universidad Pública de Navarra (UPNA), IdiSNA, 31008 Pamplona, Spain Escuela de Ciencias de la Actividad Física, el Deporte y la Salud, Laboratorio de Ciencias de la Actividad Física, el Deporte y la Salud, Facultad de Ciencias Médicas, Universidad de Santiago de Chile, USACH, 71783-5 Santiago, Chile
Yasmin Ezzatvar
Affiliation:
Department of Physiotherapy, Exercise Intervention for Health Research Group (EXINH-RG), Universitat de València, 46003 Valencia, Spain
José Francisco López-Gil
Affiliation:
Departamento de Actividad Física y Deporte, Facultad de Ciencias del Deporte, Universidad de Murcia, 30720 San Javier, Spain
Robinson Ramírez-Vélez
Affiliation:
Navarrabiomed, Complejo Hospitalario de Navarra (CHN), Universidad Pública de Navarra (UPNA), IdiSNA, 31008 Pamplona, Spain
Jordi Olloquequi
Affiliation:
Instituto de Ciencias Biomédicas, Facultad de Ciencias de la Salud, Universidad Autónoma de Chile, 3460000 Talca, Chile
Mikel Izquierdo
Affiliation:
Navarrabiomed, Complejo Hospitalario de Navarra (CHN), Universidad Pública de Navarra (UPNA), IdiSNA, 31008 Pamplona, Spain
*
*Corresponding author: Antonio García-Hermoso, email [email protected]
Rights & Permissions [Opens in a new window]

Abstract

The relationship between adherence to the Mediterranean diet (MD), physical activity (PA), sedentary behaviour and physical fitness levels has been analysed in several studies; however, there is mixed evidence among youth. Thus, this study aimed to meta-analyse the associations between adherence to the MD, PA, sedentary behaviour and physical fitness among children and adolescents. Three databases were systematically searched, including cross-sectional and prospective designs with a sample of healthy youth aged 3–18 years. Random effects inverse-variance model with the Hartung–Knapp–Sidik–Jonkman adjustment was used to estimate the pooled effect size (correlation coefficient (r)). Thirty-nine studies were included in the meta-analysis, yielding a total of 565 421 youth (mean age, 12·4 years). Overall, the MD had a weak-to-moderate positive relationship with PA (r 0·14; 95 % CI 0·11, 0·17), cardiorespiratory fitness (r 0·22; 95 % CI 0·13, 0·31) and muscular fitness (r 0·11; 95 % CI 0·03, 0·18), and a small-to-moderate negative relationship with sedentary behaviour (r –0·15; 95 % CI –0·20, –0·10) and speed–agility (r –0·06; 95 % CI –0·12, –0·01). There was a high level of heterogeneity in all of the models (I 2 ≥ 75 %). Overall, results did not remain significant after controlling for sex and age (children or adolescents) except for PA. Improving dietary habits towards those of the MD could be associated with higher physical fitness and PA in youth, lower sedentary behaviours and better health in general.

Type
Systematic Review and Meta-Analysis
Copyright
© The Author(s), 2020. Published by Cambridge University Press on behalf of The Nutrition Society

The Mediterranean diet (MD) is derived from the traditional dietary habits of people living around the Mediterranean basin(Reference Iaccarino Idelson, Scalfi and Valerio1). because it is linked to major health and nutrition benefits, a substantial body of scientific evidence has demonstrated the importance of the MD in preventing a number of chronic non-communicable diseases and maintaining good health over the entire lifespan(Reference Martinez-Gonzalez and Bes-Rastrollo2).

Despite the well-known benefits of the MD, current lifestyle changes characterised by a high consumption of processed food, sugar, soda and meat products (i.e. Western dietary pattern) have become a growing public health concern and are a major contributory factor to the global epidemic of obesity(Reference Monteiro, Moubarac and Cannon3,Reference Tur, Romaguera and Pons4) . Another factor related to obesity and associated health conditions is the increase in sedentary behaviours, such as increasing time spent on watching television, playing video games and using the Internet and reductions in physical activity (PA)(Reference Andersen, Crespo and Bartlett5) and physical fitness levels(Reference Tomkinson, Lang and Tremblay6). In this regard, evidence suggests that food habits of youth are strongly influenced by several factors such as healthy lifestyles(Reference Iaccarino Idelson, Scalfi and Valerio1) and physical fitness(Reference Thivel, Aucouturier and Isacco7,Reference Cuenca-García, Ruiz and Ortega8) in a complex interactive manner.

Childhood and adolescence are critical periods in life for the adoption of lifestyle habits that are likely to persist into adulthood. Because many chronic diseases that manifest in adulthood stem from childhood(Reference Pearson and Biddle9), it is important to establish healthy habits early in life. Also, healthy behaviours do not act in isolation, and the effects of multiple healthy lifestyle behaviours may be greater than the sum of their individual impacts(Reference Rosi, Giopp and Milioli10). For this reason, the relationship between adherence to the MD, PA, sedentary behaviours and different parameters of physical fitness has been analysed in several studies, but inconsistent results have been reported, and a meta-analytic understanding of how the MD adherence contributes to healthy habits and physical fitness in youth remains unknown. Thus, the aim of the present systematic review and meta-analysis was to synthesise the evidence regarding these relationships and to quantify their associations in children and adolescents.

Methods

The present systematic review and meta-analysis was developed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and was registered in the International Platform of Registered Systematic Review and Meta-analysis Protocols (registration number: INPLASY202040032). The entire process from literature selection to data extraction was performed independently by two researchers (A. G.-H. and Y. E.). Any disagreements were resolved through consultation with a third researcher (R. R.-V.).

Selection criteria

To be eligible for inclusion in the meta-analysis, studies needed to meet the following criteria (using PICOS criteria): (i) participants: generally healthy population aged 3 up to 18 years (mean age); (ii) exposure: MD measured using the KIDMED test (Mediterranean Diet Quality Index); (iii) comparisons: exposed v. non-exposed youths; (iv) outcomes: PA, sedentary behaviour and physical fitness and (v) study design: cross-sectional and prospective cohort studies. The first and second reviewer (A. G.-H. and Y. E.) assessed the full-text articles for eligibility. If a single study assessed different sedentary behaviours (e.g. television watching, computer use, sitting time), all effect sizes were extracted.

Search strategy

Two investigators (A. G.-H. and Y. E.) systematically searched MEDLINE, Embase and SPORTDiscus electronic databases for articles, from inception to March 2020. We used variations of the terms MD (e.g. Mediterranean, diet, adherence), children and adolescents (e.g. child, children, adolescent, adolescents, youth), PA (e.g. active, exercise, physical inactivity), sedentary behaviour (e.g. sitting time, screen time, screen media, electronic media, Internet use, computer use, mobile phone use, television watching, video game) and physical fitness (e.g. cardiorespiratory fitness (CRF), aerobic fitness, muscular fitness, muscular strength) (online Supplementary material 1). Searching was restricted to published articles in the English and Spanish languages.

Data collection process and data items

The extracted data from the articles meeting the selection criteria included the following information: (i) study characteristics (the first author’s name, publication year, enrolment year, study location, sample size, study design); (ii) participants’ information (sex and age); (iii) measurement details (method of MD, PA, sedentary behaviours and physical fitness assessments) and (iv) analysis and study results (adjusted variables, outcome of interest and main results). We requested from the study authors any effect sizes that were missing from the original published papers.

Risk of bias in individual studies

The Quality Assessment Tool for Observational Cohort and Cross-sectional Studies was used to evaluate the risk of bias(11). The checklist comprised fourteen items for longitudinal research, of which only eleven could be applied to cross-sectional studies. Each item of methodological quality was classified as ‘yes’, ‘no’ or ‘not reported’.

Summary measures

We chose the correlation coefficient (r) as the main effect size for the present meta-analysis. Effect sizes reported by studies were standardised regression coefficients (β), unstandardised regression coefficients (beta), standardised mean differences (Cohen’s d) and OR. We converted all of these estimates to correlations according to their corresponding formulas(Reference Nieminen, Lehtiniemi and Vähäkangas12Reference Bring14). If adjusted and unadjusted effect sizes were provided, we selected adjusted. Correlation coefficients were entered along with the corresponding standard errors or sample size, and the software was set to produce pooled r values with 95 % CI using a random effects inverse-variance model with the Hartung–Knapp–Sidik–Jonkman adjustment. The pooled effect size for r was classified as small (≤0·10), moderate (0·10–0·37) or large (≥0·37)(Reference McGrath and Meyer15). All analyses were performed using the admetan routine(Reference Fisher16) within version 16.1 of STATA (STATA Corp.). A P value of <0·05 was considered a threshold for statistical significance.

Synthesis of results

For each meta-analysis, heterogeneity across studies was calculated using the total variance (Q), the df and the inconsistency index (I 2)(Reference Higgins, Thompson and Deeks17), considering I 2 values of <25, 25–75 and ≥75 % as small, moderate and high heterogeneity, respectively(Reference Higgins and Thompson18). Sensitivity analysis was performed to determine whether any single study with extreme findings had an undue influence on the overall results.

Risk of bias across studies

In the different meta-analyses performed, small-study effect bias was assessed using the extended Egger’s regression test(Reference Egger, Smith and Schneider19).

Additional analysis

We identified potential moderator variables a priori. The variables were sex and age (children <12 years and adolescents ≥12 years of mean age), by stratifying meta-analyses by each of these factors.

Results

Study selection

In total, thirty-nine studies met the inclusion criteria and were included in the systematic review and meta-analysis(Reference Rosi, Giopp and Milioli10,Reference Arnaoutis, Georgoulis and Psarra20Reference Galan-Lopez, Sánchez-Oliver and Ries57) . The reasons for exclusion based on full text are reported as online Supplementary material 2. The PRISMA flow diagram illustrating the number of studies excluded at each stage of the systematic review and meta-analysis is shown in Fig. 1.

Fig. 1. Study selection.

Study characteristics

A total of thirty-nine studies reported endpoints, and details of these are listed in Table 1. Most of the studies were cross-sectional in nature except for two that included a prospective design (however, we used only data from baseline)(Reference Grao-Cruces, Fernández-Martínez and Nuviala23,Reference Roccaldo, Censi and D’Addezio43) . The studies involved a total sample of 565 421 youth with ages between 8 and 17 years (mean age, 12·4 years). All studies included males and females (51 and 49 %, respectively). Sample sizes across studies ranged from 298(Reference Mieziene, Emeljanovas and Novak35) to 335 810 participants(Reference Arnaoutis, Georgoulis and Psarra20).

Table 1. Characteristics of the studies

PA, physical activity; CRF, cardiorespiratory fitness.

Seventeen studies included participants from Spain(Reference Galan-Lopez, Domínguez and Pihu21Reference Grao-Cruces, Nuviala and Fernández-Martínez24,Reference Rosa Guillamón, Carrillo López and García Cantó26,Reference Kontogianni, Vidra and Farmaki28,Reference Arriscado, Muros and Zabala31,Reference Magriplis, Farajian and Pounis32,Reference Mazaraki, Tsioufis and Dimitriadis34,Reference Mieziene, Emeljanovas and Novak35,Reference Muros, Cofre-Bolados and Arriscado38,Reference Papadaki and Mavrikaki40,Reference Santomauro, Lorini and Tanini44,Reference Bibiloni, Pich and Córdova53,Reference Fauquet, Sofi and López-Guimerà55Reference Galan-Lopez, Sánchez-Oliver and Ries57) , eight from Greece(Reference Arnaoutis, Georgoulis and Psarra20,Reference Grosso, Marventano and Buscemi25,Reference Monjardino, Lucas and Ramos36,Reference Moral García, Agraso López and Pérez Soto37,Reference Peng, Goldsmith and Berry41,Reference Chacón-Cuberos, Zurita-Ortega and Martínez-Martínez47,Reference Manzano-Carrasco, Felipe and Sanchez-Sanchez51) , four from Italy(Reference Rosi, Giopp and Milioli10,Reference Martínez, Llull and Del Mar Bibiloni33,Reference Pino-Ortega, De La Cruz-Sánchez and Martínez-Santos49,Reference Cabanas-Sánchez, Martínez-Gómez and Izquierdo-Gómez50) and one from Chile(Reference Tambalis, Panagiotakos and Moraiti45), Croatia(Reference Farajian, Risvas and Karasouli54), Iceland(Reference Rosa Guillamon, Garcia Canto and Rodríguez García27), Estonia(Reference Lazarou, Panagiotakos and Matalas29) and Lithuania(Reference Arcila-Agudelo, Ferrer-Svoboda and Torres-Fernàndez42). Two studies were multi-national (Lithuania and Serbia(Reference Agostinis-Sobrinho, Santos and Rosário46), Colombia and Portugal(Reference Bawaked, Gomez and Homs52)).

Measurements

Nine of the thirty-nine studies in the systematic review reported PA through several instruments(Reference García-Hermoso, Vegas-Heredia and Fernández-Vergara22,Reference Rosa Guillamón, Carrillo López and García Cantó26,Reference Magriplis, Farajian and Pounis32,Reference Martínez, Llull and Del Mar Bibiloni33,Reference Monjardino, Lucas and Ramos36,Reference Moral García, Agraso López and Pérez Soto37,Reference Cabanas-Sánchez, Martínez-Gómez and Izquierdo-Gómez50,Reference Manzano-Carrasco, Felipe and Sanchez-Sanchez51) (e.g. the Moderate-to-Vigorous Physical Activity Screening Measure), fifteen used validated questionnaires like the Physical Activity Questionnaire for Children(Reference Galan-Lopez, Domínguez and Pihu21,Reference Grao-Cruces, Fernández-Martínez and Nuviala23,Reference Grosso, Marventano and Buscemi25,Reference Mazaraki, Tsioufis and Dimitriadis34,Reference Muros, Cofre-Bolados and Arriscado38,Reference Tambalis, Panagiotakos and Moraiti45) or adolescents(Reference Rosi, Giopp and Milioli10,Reference Bibiloni, Pich and Córdova53) , the International Physical Activity Questionnaire(Reference Grao-Cruces, Nuviala and Fernández-Martínez24,Reference Novak, Štefan and Prosoli39,Reference Papadaki and Mavrikaki40,Reference Arcila-Agudelo, Ferrer-Svoboda and Torres-Fernàndez42,Reference Santomauro, Lorini and Tanini44,Reference Agostinis-Sobrinho, Santos and Rosário46) , the Krece Plus test(Reference Mieziene, Emeljanovas and Novak35) and accelerometry(Reference Galan-Lopez, Ries and Gisladottir56).

Regarding screen time, most studies reported data on overall screen media time or frequency, mainly television viewing, computer use and video game playing. Four studies used sitting time(Reference Agostinis-Sobrinho, Santos and Rosário46,Reference Cabanas-Sánchez, Martínez-Gómez and Izquierdo-Gómez50,Reference Farajian, Risvas and Karasouli54,Reference Galan-Lopez, Ries and Gisladottir56) as sedentary behaviour.

Finally, components of physical fitness used were the CRF(Reference Galan-Lopez, Domínguez and Pihu21,Reference Rosa Guillamon, Garcia Canto and Rodríguez García27Reference Lazarou, Panagiotakos and Matalas29,Reference Arriscado, Muros and Zabala31,Reference Muros, Cofre-Bolados and Arriscado38,Reference Tambalis, Panagiotakos and Moraiti45,Reference Manzano-Carrasco, Felipe and Sanchez-Sanchez51,Reference Bawaked, Gomez and Homs52,Reference Galan-Lopez, Sánchez-Oliver and Ries57) , muscular fitness(Reference Rosa Guillamon, Garcia Canto and Rodríguez García27Reference Arriscado, Muros and Zabala31,Reference Muros, Cofre-Bolados and Arriscado38,Reference Tambalis, Panagiotakos and Moraiti45,Reference Manzano-Carrasco, Felipe and Sanchez-Sanchez51,Reference Bawaked, Gomez and Homs52,Reference Galan-Lopez, Sánchez-Oliver and Ries57) and speed–agility(Reference Rosa Guillamon, Garcia Canto and Rodríguez García27Reference Lazarou, Panagiotakos and Matalas29,Reference Arriscado, Muros and Zabala31,Reference Muros, Cofre-Bolados and Arriscado38,Reference Manzano-Carrasco, Felipe and Sanchez-Sanchez51) measured with the EUROFIT(Reference Manzano-Carrasco, Felipe and Sanchez-Sanchez51,Reference Fauquet, Sofi and López-Guimerà55) or ALPHA fitness battery(Reference Rosa Guillamon, Garcia Canto and Rodríguez García27Reference Lazarou, Panagiotakos and Matalas29,Reference Arriscado, Muros and Zabala31,Reference Muros, Cofre-Bolados and Arriscado38,Reference Tambalis, Panagiotakos and Moraiti45,Reference Galan-Lopez, Ries and Gisladottir56,Reference Galan-Lopez, Sánchez-Oliver and Ries57) tests (Table 1).

Risk of bias within studies

All thirty-nine studies met at least four criteria and were considered to have moderate methodological quality. The average score was 7·5/14·0 (online Supplementary material 3).

Synthesis of results

Figs. 26 show the synthesis of results. There was a weak-to-moderate direct relationship between adherence to the MD and PA (r 0·14; 95 % CI 0·11, 0·17; I 2 88·6), CRF (r 0·22; 95 % CI 0·13, 0·31; I 2 95·7) and muscular fitness (r = 0·11; 95 % CI 0·03, 0·18; I 2 95·4), and a weak-to-moderate negative relationship with sedentary behaviour (r –0·15; 95 % CI –0·20, –0·10; I 2 97·3) and speed–agility (r –0·06; 95 % CI –0·12, –0·01; I 2 84·2).

Fig. 2. Forest plot showing the correlation of Mediterranean diet adherence and physical activity for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Fig. 3. Forest plot showing the correlation of Mediterranean diet adherence and sedentary behaviours for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Fig. 4. Forest plot showing the correlation of Mediterranean diet adherence and cardiorespiratory fitness for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Fig. 5. Forest plot showing the correlation of Mediterranean diet adherence and muscular fitness for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Fig. 6. Forest plot showing the correlation of Mediterranean diet adherence and speed–agility (lower values represent higher performance) for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

The results remained significant after controlling for sex and age (children or adolescents) in PA (Table 2). Also, there was a significant slightly higher correlation between adherence to the MD and sedentary behaviour in children (r –0·21; 95 % CI –0·29, –0·12; I 2 96·5) and CRF in adolescents (r 0·30; 95 % CI 0·12, 0·47; I 2 96·5).

Table 2. Subgroup analysis of the association between adherence to the Mediterranean diet and physical activity, sedentary behaviour and physical fitness components according to sex and age

(Correlation coefficients (r) and 95 % confidence intervals)

CRF, cardiorespiratory fitness.

Publication bias and sensitivity analysis

Funnel plots and the Egger asymmetry test indicated statistically significant publication bias for PA (effect size = 1·75; 95 % CI 0·58, 2·91; P < 0·001), sedentary behaviour (effect size = 5·22; 95 % CI 3·24, 7·32; P < 0·001) and CRF (effect size = 3·38; 95 % CI 1·48, 5·27; P = 0·002) (online Supplementary material 4).

Finally, sensitivity analyses were carried out by sequentially removing one study from the data set to elucidate the influence of individual studies on the analysis. Results remained consistent across all deletions.

Discussion

To our knowledge, the present systematic review and meta-analysis is the first to synthesise the evidence on the cross-sectional associations between adherence to the MD, PA, sedentary behaviour and physical fitness levels in children and adolescents. Our analyses showed small-to-moderate positive relationships between adherence to the MD, PA, CRF and muscular fitness and a negative relationship with sedentary behaviours and speed–agility. Overall, the meta-analysis indicates that youths with higher adherence to the MD are more likely to be physically active, fit and have a less sedentary lifestyle. However, the findings should be interpreted while keeping in mind the heterogeneity between studies in the associations, exposure and outcomes assessment and publication bias.

A recent review by Iaccarino Idelson et al. (Reference Iaccarino Idelson, Scalfi and Valerio1) suggested a positive association between adherence to the MD and PA levels. In this line, our meta-analytic findings suggest a moderate direct association between them, independent of sex and age. One might speculate about four possible hypotheses underlying the relationships between the two behaviours: (i) it is possible that MD adherence influences PA by allowing those youths who eat properly to be more physically active by providing them with the necessary nutrients(Reference Petrie, Stover and Horswill58). For example, it has been reported that the higher energy expenditure of active children and adolescents requires higher intake of essential nutrients(Reference Kelishadi, Ardalan and Gheiratmand59), which requires a higher consumption of carbohydrates, vitamins and quality proteins, all provided by adequate MD adherence; (ii) better eating habits seem to be linked with some personality traits such as self-efficacy and self-regulation, which is a motive to get more involved in PA among children and adolescents(Reference López-Gil, Oriol-Granado and Izquierdo60); (iii) specific behavioural and environmental factors such as the school environment and the influence of peer groups often result in the clustering of behavioural habits, meaning that youths who are more active tend to develop other healthy habits such as consuming a healthy diet(Reference Leech, McNaughton and Timperio61) and (iv) both behaviours and their relationship could be linked to the family environment(Reference Farajian, Risvas and Karasouli54). With respect to the latter, there is some evidence to indicate the importance of parental education in the healthy eating choices of their children(Reference Arcila-Agudelo, Ferrer-Svoboda and Torres-Fernàndez42). Likewise, it has been determined that socio-economic status is one of the most important determinants of both MD adherence(Reference Iaccarino Idelson, Scalfi and Valerio1) and PA among young people(Reference O’Donoghue, Kennedy and Puggina62).

With respect to sedentary behaviour, our analysis showed that higher adherence to the MD was related to lower sedentary lifestyle, usually assessed as screen time and predominantly television viewing and computer use. These relationships have been reiterated by other authors, suggesting that children with healthy eating habits, assessed with the question ‘Do you think this child typically eats healthy meals?’, are engaged in fewer sedentary behaviours(Reference Shi, Tubb and Fingers63). There are several possible explanations for this result: (i) screen time viewing, mainly television use, is inversely related to healthy foods such as fruits and vegetables and is positively associated with a higher intake of energy-dense snacks and drinks, fast foods, total energy intake and energy from fat in youth(Reference Avery, Anderson and McCullough64); (ii) television viewing and exposure to food advertising are connected, in a positive manner, to unhealthy food habits in children and adolescents(Reference Lobstein and Dibb65). In this context, some authors disagree between television and computer sedentarism, suggesting that using personal computers involves lower nutritional risk, as hands are usually engaged and therefore not free for the consumption of unhealthy foods(Reference Rey-López, Vicente-Rodríguez and Biosca66). In contrast to these earlier findings, our meta-analysis showed similar relationship between MD and television (r –0·14; 95 % CI –0·22, –0·05) and computer use (r –0·19; 95 % CI –0·33, –0·05) (data not shown).

Our study also shows that adherence to the MD is significantly associated with higher CRF, muscular fitness and speed–agility levels. These results are similar to those reported by several studies which used other eating habits questionnaires(Reference Thivel, Aucouturier and Isacco7,Reference Cuenca-García, Ruiz and Ortega8) . An example of this is the study carried out by Thivel et al. (Reference Thivel, Aucouturier and Isacco7) who reported that CRF and muscular fitness in French schoolchildren were associated with healthy eating habits. In a similar case in ten European cities, the authors suggested that adolescents who skipped breakfast were less likely to achieve elevated CRF measurements in comparison with those who consumed breakfast(Reference Cuenca-García, Ruiz and Ortega8). According to this, it could be proposed that a well-balanced diet, providing adequate energy sources throughout the day, rich in all essential nutrients and natural antioxidants, low in saturated fat, and based on an abundant consumption of fruits, vegetables, legumes, fish, nuts and olive oil, would yield great benefits on physical fitness during childhood and adolescence(Reference Arnaoutis, Georgoulis and Psarra20).

Practice implications

Improving dietary habits towards those of the MD could be associated with higher physical fitness and PA in youth, lower sedentary behaviours(Reference Peters, Sutton and Jones67) and better health in general. Understanding the behaviours associated with adherence to MD in the young population could be essential for the appropriate, specific design of public health interventions that will contribute to early adoption of healthy habits to reduce the impact of Western dietary patterns.

Limitations and strengths

To the best of our knowledge, this is the first study to determine the association between the MD and healthy habits among children and adolescents, including a total of 562 764 youth in the analyses.

The present meta-analysis has some limitations. First, the cross-sectional design of the included studies prevents causal inferences and can be more susceptible to biases (e.g. selection bias, information bias, etc.). Reverse causality could also be true. However, a prospective study included in our meta-analysis confirms results on PA and sedentary behaviours(Reference Grao-Cruces, Fernández-Martínez and Nuviala23). Second, in most studies, PA, dietary habits and sedentary time were self-reported and could potentially be subject to socially desirable reporting bias. Also, the included studies measured PA, sedentary behaviours and physical fitness using a wide variety of different tools. Third, most of the studies did not consider potential confounding factors such as socio-economic status and/or parental education. Finally, the results indicate publication bias for PA, sedentary behaviour and CRF. However, due to between-study heterogeneity, we are unable to conclude whether the asymmetry in the plot is due to heterogeneity between studies or publication bias(Reference García-Hermoso, Saavedra and Ramírez-Vélez68).

In conclusion, our findings have significant implications for the understanding of how MD is related to behavioural outcomes among youth. In general, it seems that lifestyle factors could interact with each other synergistically. The lack of experimental and prospective studies found in our meta-analysis suggests that the overall body of evidence is weak and further research is needed.

Acknowledgements

A. G. H. is a Miguel Servet Fellow (Instituto de Salud Carlos III – FSE, CP18/0150). R. R.-V. is funded in part by a Postdoctoral Fellowship Resolution ID 420/2019 of the Universidad Pública de Navarra.

This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

A. G. H., Y. E., J. F. L. G., R. R. V., J. O. and M. I. developed the concept and drafted the manuscript. A. G. H., Y. E. and R. R. V. prepared the tables and figures and completed the scoping review. A. G. H., Y. E. and M. I. contributed to data preparation and modelling. All other authors provided data, developed models, reviewed results, provided guidance on methodology or reviewed the manuscript and approved the final version of the manuscript.

The authors declare that there are no conflicts of interest.

Supplementary material

For supplementary materials referred to in this article, please visit https://doi.org/10.1017/S0007114520004894

References

Iaccarino Idelson, P, Scalfi, L & Valerio, G (2017) Adherence to the Mediterranean diet in children and adolescents: a systematic review. Nutr Metab Cardiovasc Dis 27, 283299.CrossRefGoogle Scholar
Martinez-Gonzalez, MA & Bes-Rastrollo, M (2014) Dietary patterns, Mediterranean diet, and cardiovascular disease. Curr Opin Lipidol 25, 2026.CrossRefGoogle ScholarPubMed
Monteiro, CA, Moubarac, J-C, Cannon, G, et al. (2013) Ultra-processed products are becoming dominant in the global food system. Obes Rev 14, 2128.CrossRefGoogle ScholarPubMed
Tur, JA, Romaguera, D & Pons, A (2004) Food consumption patterns in a Mediterranean region: does the Mediterranean diet still exist? Ann Nutr Metab 48, 193201.CrossRefGoogle Scholar
Andersen, RE, Crespo, CJ, Bartlett, SJ, et al. (1998) Relationship of physical activity and television watching with body weight and level of fatness among children results from the third National Health and Nutrition Examination survey. J Am Med Assoc 279, 938942.CrossRefGoogle ScholarPubMed
Tomkinson, GR, Lang, JJ & Tremblay, MS (2019) Temporal trends in the cardiorespiratory fitness of children and adolescents representing 19 high-income and upper middle-income countries between 1981 and 2014. Br J Sports Med 53, 478486.CrossRefGoogle Scholar
Thivel, D, Aucouturier, J, Isacco, L, et al. (2013) Are eating habits associated with physical fitness in primary school children? Eat Behav 14, 8386.CrossRefGoogle ScholarPubMed
Cuenca-García, M, Ruiz, JR, Ortega, FB, et al. (2014) Association of breakfast consumption with objectively measured and self-reported physical activity, sedentary time and physical fitness in European adolescents: the HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence) Study. Public Health Nutr 17, 22262236.CrossRefGoogle ScholarPubMed
Pearson, N & Biddle, SJH (2011) Sedentary behavior and dietary intake in children, adolescents, and adults: a systematic review. Am J Prev Med 41, 178188.CrossRefGoogle ScholarPubMed
Rosi, A, Giopp, F, Milioli, G, et al. (2020) Weight status, adherence to the Mediterranean diet, physical activity level, and sleep behavior of Italian junior high school adolescents. Nutrients 12, 478.CrossRefGoogle Scholar
National Heart, Lung, and Blood Institute. Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools (accessed March 2020).Google Scholar
Nieminen, P, Lehtiniemi, H, Vähäkangas, K, et al. (2013) Standardised regression coefficient as an effect size index in summarising findings in epidemiological studies. Epidemiol Biostat Public Health 10.Google Scholar
Peterson, RA & Brown, SP (2005) On the use of beta coefficients in meta-analysis. J Appl Psychol 90, 175181.CrossRefGoogle ScholarPubMed
Bring, J (1994) How to standardize regression coefficients. Am Stat 48, 209213.Google Scholar
McGrath, RE & Meyer, GJ (2006) When effect sizes disagree: the case of r and d. Psychol Methods 11, 386401.CrossRefGoogle ScholarPubMed
Fisher, DJ (2015) Two-stage individual participant data meta-analysis and generalized forest plots. Stata J 15, 369396.CrossRefGoogle Scholar
Higgins, J, Thompson, S, Deeks, J, et al. (2003) Measuring inconsistency in meta-analyses. BMJ 327, 557560.CrossRefGoogle ScholarPubMed
Higgins, JPT & Thompson, SG (2002) Quantifying heterogeneity in a meta-analysis. Stat Med 21, 15391558.CrossRefGoogle ScholarPubMed
Egger, M, Smith, G, Schneider, M, et al. (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315, 629634.CrossRefGoogle ScholarPubMed
Arnaoutis, G, Georgoulis, M, Psarra, G, et al. (2018) Association of anthropometric and lifestyle parameters with fitness levels in Greek schoolchildren: results from the EYZHN Program. Front Nutr 5, 10.CrossRefGoogle ScholarPubMed
Galan-Lopez, P, Domínguez, R, Pihu, M, et al. (2019) Evaluation of physical fitness, body composition, and adherence to Mediterranean diet in adolescents from Estonia: the AdolesHealth study. Int J Environ Res Public Health 16, 4479.CrossRefGoogle ScholarPubMed
García-Hermoso, A, Vegas-Heredia, ED, Fernández-Vergara, O, et al. (2019) Independent and combined effects of handgrip strength and adherence to a Mediterranean diet on blood pressure in Chilean children. Nutrition 60, 170174.CrossRefGoogle ScholarPubMed
Grao-Cruces, A, Fernández-Martínez, A & Nuviala, A (2014) Association of fitness with life satisfaction, health risk behaviors, and adherence to the Mediterranean diet in Spanish adolescents. J Strength Cond Res 28, 21642172.CrossRefGoogle Scholar
Grao-Cruces, A, Nuviala, A, Fernández-Martínez, A, et al. (2015) Relationship of physical activity and sedentarism with tobacco and alcohol consumption, and Mediterranean diet in Spanish teenagers. Nutr Hosp 31, 16931700.Google ScholarPubMed
Grosso, G, Marventano, S, Buscemi, S, et al. (2013) Factors associated with adherence to the Mediterranean diet among adolescents living in Sicily, southern Italy. Nutrients 5, 49084923.CrossRefGoogle Scholar
Rosa Guillamón, A, Carrillo López, PJ, García Cantó, E, et al. (2019) Mediterranean diet, weight status and physical activity in schoolchildren of the Region of Murcia. Clin Investig Arterioscler 31, 17.Google ScholarPubMed
Rosa Guillamon, A, Garcia Canto, E, Rodríguez García, PL, et al. (2017) Physical activity, physical fitness and quality of diet in schoolchildren from 8 to 12 years. Nutr Hosp 34, 12921298.Google ScholarPubMed
Kontogianni, MD, Vidra, N, Farmaki, A-E, et al. (2008) Adherence rates to the Mediterranean diet are low in a representative sample of Greek children and adolescents. J Nutr 138, 19511956.CrossRefGoogle Scholar
Lazarou, C, Panagiotakos, DB & Matalas, AL (2009) Level of adherence to the Mediterranean diet among children from Cyprus: the CYKIDS study. Public Health Nutr 12, 9911000.CrossRefGoogle Scholar
López-Gil, JF, Brazo-Sayavera, J, García-Hermoso, A, et al. (2020) Adherence to Mediterranean diet related with physical fitness and physical activity in schoolchildren aged 6–13. Nutrients 12, 567.CrossRefGoogle ScholarPubMed
Arriscado, D, Muros, JJ, Zabala, M, et al. (2014) Factors associated with low adherence to a Mediterranean diet in healthy children in northern Spain. Appetite 80, 2834.CrossRefGoogle ScholarPubMed
Magriplis, E, Farajian, P, Pounis, GD, et al. (2011) High sodium intake of children through ‘hidden’ food sources and its association with the Mediterranean diet: the GRECO study. J Hypertens 29, 10691076.CrossRefGoogle ScholarPubMed
Martínez, E, Llull, R, Del Mar Bibiloni, M, et al. (2010) Adherence to the Mediterranean dietary pattern among Balearic Islands adolescents. Br J Nutr 103, 16571664.CrossRefGoogle Scholar
Mazaraki, A, Tsioufis, C, Dimitriadis, K, et al. (2011) Adherence to the Mediterranean diet and albuminuria levels in Greek adolescents: data from the Leontio Lyceum ALbuminuria (3L study). Eur J Clin Nutr 65, 219225.CrossRefGoogle Scholar
Mieziene, B, Emeljanovas, A, Novak, D, et al. (2019) The relationship between social capital within its different contexts and adherence to a Mediterranean diet among Lithuanian adolescents. Nutrients 11, 1332.CrossRefGoogle ScholarPubMed
Monjardino, T, Lucas, R, Ramos, E, et al. (2014) Associations between a priori-defined dietary patterns and longitudinal changes in bone mineral density in adolescents. Public Health Nutr 17, 195205.CrossRefGoogle ScholarPubMed
Moral García, JE, Agraso López, AD, Pérez Soto, JJ, et al. (2019) Physical activity practice according to adherence to the Mediterranean diet, alcohol consumption and motivation in adolescents. Nutr Hosp 36, 420427.Google Scholar
Muros, JJ, Cofre-Bolados, C, Arriscado, D, et al. (2017) Mediterranean diet adherence is associated with lifestyle, physical fitness, and mental wellness among 10-y-olds in Chile. Nutrition 35, 8792.CrossRefGoogle ScholarPubMed
Novak, D, Štefan, L, Prosoli, R, et al. (2017) Mediterranean diet and its correlates among adolescents in non-Mediterranean European countries: a population-based study. Nutrients 9, 177.CrossRefGoogle ScholarPubMed
Papadaki, S & Mavrikaki, E (2015) Greek adolescents and the Mediterranean diet: factors affecting quality and adherence. Nutrition 31, 345349.CrossRefGoogle ScholarPubMed
Peng, W, Goldsmith, R & Berry, EM (2017) Demographic and lifestyle factors associated with adherence to the Mediterranean diet in relation to overweight/obesity among Israeli adolescents: findings from the Mabat Israeli national youth health and nutrition survey. Public Health Nutr 20, 883892.CrossRefGoogle Scholar
Arcila-Agudelo, AM, Ferrer-Svoboda, C, Torres-Fernàndez, T, et al. (2019) Determinants of adherence to healthy eating patterns in a population of children and adolescents: evidence on the Mediterranean diet in the city of Mataró (Catalonia, Spain). Nutrients 11, 854.CrossRefGoogle Scholar
Roccaldo, R, Censi, L, D’Addezio, L, et al. (2014) Adherence to the Mediterranean diet in Italian school children (The ZOOM8 Study). Int J Food Sci Nutr 65, 621628.CrossRefGoogle Scholar
Santomauro, F, Lorini, C, Tanini, T, et al. (2014) Adherence to Mediterranean diet in a sample of Tuscan adolescents. Nutrition 30, 13791383.CrossRefGoogle Scholar
Tambalis, KD, Panagiotakos, DB, Moraiti, I, et al. (2018) Poor dietary habits in Greek schoolchildren are strongly associated with screen time: results from the EYZHN (National Action for Children’s Health) Program. Eur J Clin Nutr 72, 572580.CrossRefGoogle ScholarPubMed
Agostinis-Sobrinho, C, Santos, R, Rosário, R, et al. (2018) Optimal adherence to a Mediterranean diet may not overcome the deleterious effects of low physical fitness on cardiovascular disease risk in adolescents: a cross-sectional pooled analysis. Nutrients 10, 815.CrossRefGoogle Scholar
Chacón-Cuberos, R, Zurita-Ortega, F, Martínez-Martínez, A, et al. (2018) Adherence to the Mediterranean diet is related to healthy habits, learning processes, and academic achievement in adolescents: a cross-sectional study. Nutrients 10, 1566.CrossRefGoogle Scholar
Obradovic Salcin, L, Karin, Z, Miljanovic Damjanovic, V, et al. (2019) Physical activity, body mass, and adherence to the Mediterranean diet in preschool children: a cross-sectional analysis in the Split-Dalmatia County (Croatia). Int J Environ Res Public Health 16, 3237.CrossRefGoogle Scholar
Pino-Ortega, J, De La Cruz-Sánchez, E & Martínez-Santos, R (2010) Health-related fitness in school children: compliance with physical activity recommendations and its relationship with body mass index and diet quality. Arch Latinoam Nutr 60, 374379.Google ScholarPubMed
Cabanas-Sánchez, V, Martínez-Gómez, D, Izquierdo-Gómez, R, et al. (2018) Association between clustering of lifestyle behaviors and health-related physical fitness in youth: the UP&DOWN study. J Pediatr 199, 4148.e1.CrossRefGoogle ScholarPubMed
Manzano-Carrasco, S, Felipe, JL, Sanchez-Sanchez, J, et al. (2020) Weight status, adherence to the Mediterranean diet, and physical fitness in Spanish children and adolescents: the Active Health Study. Nutrients 12, 1680.CrossRefGoogle Scholar
Bawaked, RA, Gomez, SF, Homs, C, et al. (2018) Association of eating behaviors, lifestyle, and maternal education with adherence to the Mediterranean diet in Spanish children. Appetite 130, 279285.CrossRefGoogle Scholar
Bibiloni, MM, Pich, J, Córdova, A, et al. (2012) Association between sedentary behaviour and socioeconomic factors, diet and lifestyle among the Balearic Islands adolescents. BMC Public Health 12, 718.CrossRefGoogle ScholarPubMed
Farajian, P, Risvas, G, Karasouli, K, et al. (2011) Very high childhood obesity prevalence and low adherence rates to the Mediterranean diet in Greek children: the GRECO study. Atherosclerosis 217, 525530.CrossRefGoogle Scholar
Fauquet, J, Sofi, F, López-Guimerà, G, et al. (2016) Adherencia a la dieta mediterránea en adolescentes catalanes: factores socioeconómicos y de estilo de vida (Mediterranean diet adherence among Catalonian adolescents: socio-economic and lifestyle factors). Nutr Hosp 3333, 12831290.Google Scholar
Galan-Lopez, P, Ries, F, Gisladottir, T, et al. (2018) Healthy lifestyle: relationship between Mediterranean diet, body composition and physical fitness in 13 to 16-years old Icelandic students. Int J Environ Res Public Health 15, 2632.CrossRefGoogle ScholarPubMed
Galan-Lopez, P, Sánchez-Oliver, AJ, Ries, F, et al. (2019) Mediterranean diet, physical fitness and body composition in Sevillian adolescents: a healthy lifestyle. Nutrients 11, 2009.CrossRefGoogle ScholarPubMed
Petrie, HJ, Stover, EA & Horswill, CA (2004) Nutritional concerns for the child and adolescent competitor. Nutrition 20, 620631.CrossRefGoogle ScholarPubMed
Kelishadi, R, Ardalan, G, Gheiratmand, R, et al. (2007) Association of physical activity and dietary behaviours in relation to the body mass index in a national sample of Iranian children and adolescents: CASPIAN Study. Bull World Health Organ 85, 1926.CrossRefGoogle Scholar
López-Gil, JF, Oriol-Granado, X, Izquierdo, M, et al. (2020) Healthy lifestyle behaviors and their association with self-regulation in Chilean children. Int J Environ Res Public Health 17, 5676.CrossRefGoogle ScholarPubMed
Leech, RM, McNaughton, SA & Timperio, A (2014) The clustering of diet, physical activity and sedentary behavior in children and adolescents: a review. Int J Behav Nutr Phys Act 11, 4.CrossRefGoogle ScholarPubMed
O’Donoghue, G, Kennedy, A, Puggina, A, et al. (2018) Socio-economic determinants of physical activity across the life course: a ‘DEterminants of DIet and Physical ACtivity’ (DEDIPAC) umbrella literature review. PLOS ONE 13, e0190737.CrossRefGoogle ScholarPubMed
Shi, X, Tubb, L, Fingers, ST, et al. (2013) Associations of physical activity and dietary behaviors with children’s health and academic problems. J Sch Health 83, 17.CrossRefGoogle ScholarPubMed
Avery, A, Anderson, C & McCullough, F (2017) Associations between children’s diet quality and watching television during meal or snack consumption: a systematic review. Matern Child Nutr 13, e12428.CrossRefGoogle ScholarPubMed
Lobstein, T & Dibb, S (2005) Evidence of a possible link between obesogenic food advertising and child overweight. Obes Rev 6, 203208.CrossRefGoogle ScholarPubMed
Rey-López, JP, Vicente-Rodríguez, G, Biosca, M, et al. (2008) Sedentary behaviour and obesity development in children and adolescents. Nutr Metab Cardiovasc Dis 18, 242251.CrossRefGoogle ScholarPubMed
Peters, JL, Sutton, AJ, Jones, DR, et al. (2006) Comparison of two methods to detect publication bias in meta-analysis. J Am Med Assoc 295, 676680.CrossRefGoogle ScholarPubMed
García-Hermoso, A, Saavedra, JM, Ramírez-Vélez, R, et al. (2017) Reallocating sedentary time to moderate-to-vigorous physical activity but not to light-intensity physical activity is effective to reduce adiposity among youths: a systematic review and meta-analysis. Obes Rev 18, 10881095.CrossRefGoogle Scholar
Figure 0

Fig. 1. Study selection.

Figure 1

Table 1. Characteristics of the studies

Figure 2

Fig. 2. Forest plot showing the correlation of Mediterranean diet adherence and physical activity for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Figure 3

Fig. 3. Forest plot showing the correlation of Mediterranean diet adherence and sedentary behaviours for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Figure 4

Fig. 4. Forest plot showing the correlation of Mediterranean diet adherence and cardiorespiratory fitness for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Figure 5

Fig. 5. Forest plot showing the correlation of Mediterranean diet adherence and muscular fitness for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Figure 6

Fig. 6. Forest plot showing the correlation of Mediterranean diet adherence and speed–agility (lower values represent higher performance) for each study. Squares represent pooled effect size for each subgroup analysis, and the diamond represents the overall pooled effect size.

Figure 7

Table 2. Subgroup analysis of the association between adherence to the Mediterranean diet and physical activity, sedentary behaviour and physical fitness components according to sex and age(Correlation coefficients (r) and 95 % confidence intervals)

Supplementary material: File

García-Hermoso et al. supplementary material

García-Hermoso et al. supplementary material
Download García-Hermoso et al. supplementary material(File)
File 1.9 MB