Introduction

The rise in the prevalence of atopic diseases, mainly asthma, allergic rhinitis, and eczema, in children and adolescents during the last decades, generated the need for further study of the various lifestyle and environmental factors implicated in their etiology.¹ A critical lifestyle factor is nutrition, which has already been identified as a contributor to the burden of various chronic diseases, including atopic diseases in children. It has been hypothesized that the influence of nutrition on atopic disease incidence may be due to the antioxidant intake, as well as the balance of n-6, n-3 polyunsaturated fatty acids (PUFA), and the intake of vitamin D.^2–6 One of the most extensively studied nutritional patterns is the Mediterranean Diet (MedDiet). MedDiet can be defined as a cultural lifestyle model for a healthy and sustainable diet, identified in the upper Mediterranean region in the mid-1950s and 1960s.⁷ Extensive research has shown that greater adherence to the MedDiet, in general the Mediterranean lifestyle, has been associated with lower incidence of chronic diseases, such as cardiovascular, neurophysiological, and pulmonary diseases, and some types of cancer, mainly in adult population living in the Mediterranean Sea region.⁸ The high content of vegetables, salads, fresh fruits, fish and cereals, and the considerable use of olive oil in the MedDiet guarantee an efficient intake of β-carotene, vitamin C, tocopherols, α-linolenic acid, various important minerals, and several bioactive compounds, such as polyphenols and antioxidants. Many of these aspects of MedDiet have been found to be associated with asthma and atopic disease prevalence in several studies conducted, but with conflicting results.^8–10

Thus, in the context of the Global Asthma Network (GAN) study,¹¹ the association between the level of adherence to the MedDiet and prevalence of asthma, allergic rhinitis, and eczema was evaluated in Greek adolescents. Furthermore, we expanded our research hypothesis by ecologically evaluating the association between the prevalence of atopic diseases globally, as recorded in the GAN study,¹¹ with the consumption of the major food groups or components of the MedDiet in the participating countries.

Materials and Methods

Design

GAN study is an international epidemiological project aimed to monitor the worldwide prevalence, severity, management, and risk factors of asthma and other atopic diseases, such as allergic rhinitis and eczema, in two life periods of childhood, that is, during 6–7 and 13–14 years old.¹¹ The Greek part of the GAN study included adolescents aged 13–14 years and followed the same protocol of the study.¹²

Setting

The sampling of this study took place in the greater Athens metropolitan area, Greece, from February to March 2020, in high school settings. In particular, 20 high schools were enrolled by convenience sampling, from a list that was provided by the Secondary Education Office of the Ministry of Education. All children in the first and second grades of each school were asked to participate. Schools for children with special educational needs or disabilities were excluded.

Sample

In total, 2560 adolescents were asked to participate. Of them, 1934 adolescents (921 boys [47.5%] and 1013 girls [52.5%]), with mean age (Standard Deviation [SD]) 12.7 (0.6) years agreed to participate (participation rate 76%).

Statistical power analysis

The number of enrolled children was adequate to achieve a statistical power of 90% or greater to evaluate standardized differences between MedDiet adherence score and the presence of atopic disease, at a probability level <0.05 (two-sided hypothesis).

Measurements

A physical examination was performed at the schools by trained investigators of the study. Current asthma was defined as the positive response to the question “Have you had wheezing or whistling in the chest in the past 12 months?” Similarly, current rhinitis was defined as the positive answer to the question “In the past 12 months, have you had a problem with sneezing or a runny or blocked nose when you did not have a cold or flu?” and current eczema as the positive answer to the question “Have you had this itchy rash at any time in the past 12 months?”. Moreover, adolescents were asked if they had any siblings, and if so, how many.

The protocol of the GAN study also included an interview, after the physical examination, based on a standardized, validated questionnaire that included questions regarding dietary habits, physical activity, and information about the living environment of the participating adolescents.¹³ Especially, the consumption frequency of various food groups or food items (i.e., meat and meat products, seafood, fruits, cooked [green and roots] and raw vegetables, salads [green and raw], pulses [peas, beans, lentils], legumes, cereals, and dairy, as well as lollies, candies, sweets, and fizzy or soft drinks), during the past 12 months was recorded. Adherence to the Mediterranean diet was evaluated using the KIDMED index (Mediterranean Diet Quality Index for children and adolescents). The index is derived from 16 components, which summarize the inherent characteristics of the Mediterranean diet. The theoretical score ranges from 0–12 (a score <3 reflects poor adherence to the Mediterranean diet, whereas scores between 4–7 and 8–12 indicate average and good adherence to the Mediterranean diet, respectively).¹⁴ Physical activity status of the adolescents was assessed through detailed questions, evaluating the number of occasions per week spent on engaging in vigorous physical activities (i.e., up to the point of breathing hard), the total hours in a day spent watching television or film or videos, and the number of hours per day spent on computer and internet activities (gaming, chatting, Facebook, YouTube, etc.). Adolescents were then classified into as adopting an active or sedentary lifestyle. Height and weight were measured through standard procedures, and body mass index was calculated to classify adolescents as normal weight, overweight, and obese, using the International Obesity Task Force (IOTF) classification.¹⁵

Moreover, adolescents’ parents (or guardians) were asked (by completing a part of the questionnaire at home) to report if they had a history of atopic diseases (asthma, eczema, or allergic rhinitis), if there was visible moisture or mold spots on the walls or ceiling of their homes, as well as their education level (compulsory education up to 9 years, secondary or tertiary [university or college or postgraduate studies]), and whether they were currently smoking.

Global data

The published information regarding the global prevalence of atopic diseases from the entire GAN study participating centers was retrieved by Asher et al.,¹⁶ whereas data on consumption of fruits, vegetables, meat, seafood, and dairy products (per capita, and in kilogram per person), as well as energy intake, were retrieved through the Food and Agriculture Organization of United Nations databases.¹⁷

Bioethics

The study was designed according to the principles of the declaration of Helsinki (1989) and was approved by the Ethics Committee of the National and Kapodistrian University of Athens (Decision number: 214/13-12-19). Permission for the accomplishment of the study was issued by the Ministry of Educational Affairs (Decision number: 10053/24-01-2020).

Statistical analysis

Continuous variables were presented using mean values and SD, and categorical variables were presented using absolute and relative frequencies. Pearson’s chi-squared and Fisher’s exact tests, when appropriate, as well as Analysis of Variance (ANOVA), were applied to examine for univariate differences among the several atopic, anthropometric, and lifestyle characteristics of adolescents and the level of adherence to the MedDiet (poor, moderate, and good), respectively. Logistic regression models (adjusted for several well-known confounders based on the related literature) were applied to assess the relationship between the level of adherence to the MedDiet with the atopic outcomes. Results were presented as odds ratios and corresponding 95% confidence intervals. Deviance residuals were calculated to evaluate the model’s goodness-of-fit. To further explore the discriminating ability of the nutrition items in the classification of the adolescents according to atopic disease outcomes, hierarchical discriminant analysis was applied. The analysis produced a discriminant function (i.e., Fisher’s discriminant function) as a linear combination of the predictor variables which delivered the best discrimination for the outcome variable. Wilk’s lambda was also calculated to reveal which food group had better classification ability. Wilk’s lambda is a multivariate test statistic, with values ranging between 0 and 1; values close to 0 indicate that the investigated factor has good classification ability, while values close to 1 indicates that the factor has poor ability to classify cases in each group. Spearman rho ranked correlation coefficient was used to evaluate the relationships between atopic diseases’ prevalence in the entire GAN study participating centers and specific food consumption. All reported probability values (P) were based on two-sided hypothesis testing and compared to a significant level of 5%. STATA 17 software was used for all the statistical calculations (STATA Corp., College Station, Texas, USA).

Results

Prevalence of atopic diseases

From the total study sample of 1934 adolescents, 6.9% reported symptoms of current asthma (at least one episode of wheezing or whistling in the chest in the past 12 months), 25.3% reported symptoms of current rhinitis (at least one episode of sneezing or runny nose without cold symptoms in the past 12 months), and 8.9% reported symptoms of current eczema (at least the appearance of an itchy rash in the past 12 months).

Mediterranean diet and atopic diseases

The mean KIDMED score was 4.8 (2.1) for boys and 5.1 (2.1) for girls (P = 0.002). Majoity of the participating adolescents were classified as having “average” adherence (i.e., 63%), almost 1/4 as “poor” adherence (i.e., 24%), and only 13% as having “good” adherence to the Mediterranean dietary pattern. In Table 1, data regarding the prevalence of atopic diseases, as well as various demographic and lifestyle characteristics, by the level of adherence to the Mediterranean diet, are presented. Significantly higher rates of current rhinitis were observed in adolescents with poor adherence to the MedDiet, while adolescents with good adherence to the MedDiet were significantly more active and were originating from families with parents of higher educational level (P < 0.05).

However, residual confounding may exist, due to the observational nature of the study. Thus, in Table 2, the results from nested, multiadjusted logistic regression models evaluating the association between the level of adherence to the MedDiet and the prevalence of current asthma, allergic rhinitis, and rash symptoms as outcomes are presented. Compared to poor adherence, adolescents with moderate and good adherence to the MedDiet had 34 and 60% lower odds of having any asthma symptoms, respectively, and 20 and 41% lower odds of having any allergic rhinitis symptoms, respectively, adjusted for several confounders (Model 3, Table 2). As far as current allergic rash symptoms are concerned, no significant associations with any level of adherence to the MedDiet were observed. Regarding the other factors included in the models, almost none of them were associated with the presence of atopic symptoms in adolescents.

Decomposition of the Mediterranean diet in relation to atopic diseases

To further explore which of the studied food groups were associated more with the classification of adolescents as having atopic diseases, hierarchical, discriminant analysis was applied. Based on the values of Wilk’s lambda which was derived from the discriminant analysis, it was revealed that cooked and raw vegetables’ consumption (λ = 0.881), followed by consumption of fruits (λ = 0.957) and nuts (λ = 0.989), were the factors which had the higher classification ability among the food groups studied.

Global prevalence of atopic diseases and foods consumption

To evaluate the global prevalence of atopic diseases in relation to food consumption, information on the prevalence of asthma, allergic rhinitis, and rash symptoms, for each of the 27 centers in 14 countries that participated in GAN study, were used.¹⁶ Food consumption was retrieved from the Food and Agriculture Organization of the United Nations (FAOSTAT) databases.¹⁷ Overall, 74,361 adolescents participated in GAN Phase I study. The prevalence of asthma varied from around 4% (Nigeria, India, Kosovo and Equator) to above 20% (Costa Rica, New Zeeland), allergic rhinitis varied from around 22% (Nicaragua, Honduras, Nigeria) to above 50% (Taiwan, Argentina, Brazil), and that of allergic rash varied from below 4% (Equator, Mexico) to above 20% (Spain). Inverse correlations were observed between fruits and vegetables consumption and asthma rates across participating countries (rho = 0.06 for fruits and 0.19 for vegetables). Moreover, a positive correlation was observed between saturated fatty acids intake and allergic rash (eczema) rates (rho 0.26). No other correlations were observed between other food groups (i.e., meat, seafood) consumed and atopic diseases rates in GAN study participating centers.

Discussion

Under the context of the Greek branch of the GAN study, the hypothesis of whether adherence to the MedDiet is associated with a lower prevalence of asthma, allergic rhinitis, and eczema symptoms was evaluated in adolescents (i.e., in 13- to 14-year olds). A strong, inverse, and independent (from physical activity and obesity status) relationship between the level of adherence to the MedDiet and prevalence of asthma and allergic rhinitis symptoms was observed, while no significant associations were found for eczema. Most importantly, adolescents who had good adherence to the MedDiet had almost 50% lower odds of having current asthma and rhinitis compared to the ones who exhibited poor adherence. The food group with the most significant discriminating capability between good and moderate and/or poor MedDiet adherents was cooked and raw vegetables, indicating that vegetable consumption could be considered as a marker for the evaluation of the level of adherence to the MedDiet. Furthermore, this food group should be the first one to be promoted when public health interventions related to atopic diseases prevention and nutrition are about to be designed. Taking into consideration that atopic diseases are widespread among adolescents in Greece, although lesser than in other Northern European countries, the presented evidence has important public health implications, because they suggest a strong, nonpharmacological, specific means for preventing atopic diseases in adolescence.

Evidence from the past two decades suggests that children and adolescents with a good level of adherence to the MedDiet in early life are protected against the development of asthma and atopy. In one of the first studies on this topic, among 690 Greek children aged 7–18 years from Crete, Chatzi et al. revealed that good adherence to the MedDiet had a protective effect on allergic rhinitis, asthma-like symptoms, and atopy.¹⁸ These findings were in line with those from PANACEA study conducted in Greece,¹⁹ as well as other studies, mainly from Spain and other Mediterranean countries.^20–22 Results from a recent birth cohort study in children originating from Paris reported a protective association of MedDiet not only to allergic morbidity but also to the lung function, as demonstrated through improved spirometry, in children aged eight.²³ However, there are studies which have not replicated the favorable association between MedDiet and atopic disease prevalence. For example, in a case–control study of 287 asthmatic Peruvian children, adolescents aged 9–19 years, and 96 nonasthmatic counterparts, no association was observed between MedDiet and allergic sensitization.²⁴

Our study provided evidence about the protective association between the level of adherence to the MedDiet and asthma and allergic rhinitis. MedDiet is a dietary pattern rich in antioxidants, carbohydrates, fiber, and mono- and n-3 PUFA which through their regulating role in the inflammatory pathway could suppress the development of atopic diseases. It is well known that Immunoglobulin E (IgE), a major allergic mediator, is released from B-lymphocytes and is influenced by cytokines produced by T-helper lymphocytes. Interleukin (IL)-4 produced by Type 2 T-helper (Th2) cells, promotes the synthesis of IgE from B-lymphocytes, whereas IFN-γ released from Type 1 T-helper (Th1) cells inhibits it. Moreover, prostaglandin 2 (PGE2) modulates the formation of cytokines by inhibiting IFN-γ release; thus, high levels of PGE2 result in a rise in IgE concentration and subsequently exaggeration of atopic phenotype in sensitized subjects.²⁵ Therefore, a high dietary intake of n-3 PUFA reduces the production of arachidonic acid, the precursor of PGE2, which leads to a decrease in the secretion of IgE. Antioxidants (e.g., vitamins, selenium, β-carotene, flavonoids) which are mainly found in fruits and vegetables play an important anti-inflammatory role by interfering in critical pathways in these pathophysiological mechanisms.²⁶ Specifically, vitamin C provides antioxidant capacity by inactivating oxygen free radicals and preventing macrophage secretions. Vitamin E and selenium own a major role in protecting cells from oxidative stress 6, which is regarded as one of the critical factors involved in the chronic inflammation of the airways in asthma. Vitamin D inhibits the release of IgE mediated by mast cells and the stimulator of the skin mast cells IL-33, thereby reducing the level of allergic response to various atopic stimulants.²⁷ However, in a very recent systematic review, it was reported that the protective role of the MedDiet against childhood asthma seems prominent, but also the analysis of the retrieved studies may imply that the MedDiet probably does not affect the development of allergies.²⁸ It is more than evident that randomized clinical trials are needed to further explore the role of MedDiet and its constituents on atopic diseases.

Limitations

The GAN study, as a cross-sectional study, suffers from the inherited limitations of this type of observational studies, like recall bias and inability to assess causal relationships. However, effort was made not to overinterpret the study results. Moreover, reverse causality may also exist (i.e., adolescents with atopic disease symptoms may have adopted healthier dietary habits as prompted by their physicians). In addition, accounting for residual confounding was feasible only for physical activity status and body weight level of participating adolescents, and not for other activities and lifestyle behaviors.

Conclusion

Our study provides evidence for a strong and inverse association between the level of adherence to the MedDiet and the occurrence of asthma and allergic rhinitis symptoms. Moreover, it signifies the importance of contribution of fruits and vegetable consumption in the MedDiet and the beneficial association with atopic diseases. The same inverse relation was observed in the study of the associations between the global atopic diseases prevalence as recorded in the GAN study and asthma occurrence in the corresponding countries. Thus, the promotion of MedDiet and especially the promotion of fruits and vegetable consumption could be an efficient lifestyle intervention that can contribute to the reduction of the burden of these atopic diseases in adolescents worldwide. Furthermore, public health workers, as well as clinicians implicated in the healthcare activities of atopic adolescents, should always consider MedDiet and especially fruits and vegetable consumption as an important nonpharmacological means of atopic disease prevention.