Burnout Syndrome Risk in Child and Adolescent Tennis Players and The Role of Adherence to the Mediterranean Diet

This cross-sectional study examined the impact of adherence to Mediterranean diet on burnout syndrome risk in 94 athletes 8–15 years old. Diet pattern and burnout syndrome risk were assessed through the Athlete Burnout Questionnaire and the KIDMED Questionnaire. 55.3% of girls and 16.1% of boys had a high risk of burnout syndrome and the risk increased with age. Of the 78.7% with low adherence to Mediterranean diet, 31.1% showed no risk of burnout syndrome, 33.8% had a moderate risk, and 35.1% high risk. Of the 21.3% with a high adherence to Mediterranean diet, 35% had no risk of burnout syndrome, 45% had a moderate risk, and 20% had a high risk. Participants with moderate/high burnout syndrome risk were more likely to be girls and spend a higher number of hours watching television or playing video games. There is not enough statistical evidence in this study to reject the independence between the level of adherence to the Mediterranean diet and the risk of burnout syndrome in children, except in the case of daily consumption of fresh or cooked vegetables.


Introduction
In the last few decades, the physical, psychological and psychosocial benefits of sports have been established beyond doubt [1][2][3][4][5]. In spite of this, if sport is practiced in an obsessive and uncontrolled way, it can become a risk factor for the development of unhealthy lifestyles [6]. The type of sports discipline, the pressure exerted by the coach or teammates, and the influence of the family [7] can influence the increase in that risk. Activity-related stress, whose associated factors are classified in two general categories-activity demands (physical, psychological, social or organizational aspects of the activity requiring sustained physical/psychological effort or skills and therefore associated with certain physiological/psychological costs) and activity resources (functional in achieving work goals, reducing activity demands and the associated costs, stimulating personal growth, learning and development)-can also play a role in the risk of developing burnout syndrome (BS) [7][8][9]. Certain personality traits such as a high degree of empathy, high degree of altruism, low self-esteem, constancy in action, tendency to emotional over-involvement, locus of external control, unrealistic expectations about work, low self-efficacy, and reduced social skills [10][11][12], as well as an excessive concern for body Due to the presence of some of the same external risk factors described above for a work environment associated with BS and the psychological development stage of the studied population, which hinders their ability to adequately cope with possible stressors, it would be expected that BS is present in young amateur athletes. Given that previous studies have associated BS with negative dietary patterns and intakes in a wide variety of professionals and that the conditions that may be related to the association between BS risk and diet pattern may also be found in amateur youth sports, it would be reasonable to expect similar associations between BS and diet pattern to arise in the context of amateur youth athletic practice.
The Mediterranean Diet (MedDiet) is defined as a diet pattern that has been developed for millennia, as a result of the convergence of many cultures, among the peoples who settled on the shores of the Mediterranean Sea [55][56][57]. It is not only a diet pattern, but it is also a social model, a way of behaving at the table, it being common that meals are shared with family and friends, where the ups and downs of the day are discussed and bonds of kinship and friendship and communication are established [55][56][57]. In this study, the authors have considered the MedDiet as the standard, since it has been the dietary pattern traditionally followed in the location this study took place.
The MedDiet is characterized by an intake rich in nuts, fruits and vegetables, whole grains and pulses (mainly as a source of dietary fiber), fish, low-fat dairy, moderate alcohol intake, limited quantities of red meat and sweets, and with olive oil occupying a privileged place, privative of this diet [58,59]. The consumption of fresh and local products, in terms of seasonality, biodiversity, traditional culinary activities and conviviality, as well as the variety of foods, represents the cornerstone of the MedDiet pattern. Cultural and nutritional aspects, jointly with regular engagement in physical activity, are fundamental parts of this comprehensive Mediterranean lifestyle model [60,61].
Several observational, analytical, and experimental studies have highlighted the potential for health benefits provided by the adherence to the MedDiet pattern [56,57,62,63] through the reduction of risk factors for the majority of non-communicable diseases (type 2 diabetes, metabolic syndrome, CVD, certain types of cancer, cognitive decline, depression and mental disorders).
Favorable health outcomes and a better quality of life cannot be attributed to a single macro or micronutrient or a specific phytochemical, but may be attributed to the synergistic combination of a balanced ratio of ω-6 and ω-3 essential fatty acids, high oleic acid content, and high amounts of dietary fiber, as well as antioxidants, phytonutrients (carotenoids, phytosterols, polyphenols, glycosinolates, sulfur compounds, saponins, polyterpenes) and phytic acid, resulting in antioxidant, anti-inflammatory, and antithrombotic properties, preventing lipoperoxidation, improving lipid profile and endothelial function [63][64][65][66][67][68]. In addition, other non-nutritional aspects of the MedDiet, such as its social, cultural, economic and environmental features, including conviviality, sensory stimulation, socialization, biodiversity, and seasonality, have been recognized as aspects that can reinforce the MedDiet's beneficial effects on wellbeing, quality of life, and health [69,70].
In this context, the specific aim of this study was to investigate the association between adherence to MedDiet and the presence of BS in a group of child and adolescent tennis players between 8 and 15 years of age.
We assessed the hypothesis that a higher adherence to MedDiet may produce a reduction in BS in child and adolescent athletes.

Methods
We conducted an exploratory cross-sectional study in Valencia (Spain) to assess the role of MedDiet pattern adherence and BS risk in a convenience sample of child and adolescent tennis players. The study was carried out in the second half of 2017.

Participants
This study, utilizing a convenience sample of 94 athletes-56 boys (59.6%) and 38 girls (40.4%)-was conducted in the Tennis Club of Valencia. The inclusion criteria for the study included being a member of this club, being between 8 and 15 years old, being free of injury or diseases at the time of the study and training a minimum of 2 h a day. First, we excluded the families that declined to participate, then we applied the criteria established and later we excluded those who did not attend the study interview appointments or who did not complete the questionnaires properly. In a first appointment, before we started the research, we explained the study to the parents of all the child and adolescent tennis players involved, including the tests that would have taken place and that data would have been kept confidential in line with the Spanish data protection law. Afterwards, parents were contacted individually to obtain their permission to include athletes in the study. If the parents gave verbal consent and the child/adolescent met the selection criteria, an appointment was arranged for BS and MedDiet assessments which were carried out by trained dieticians. Informed written consent was obtained from either both parents or whichever parent was present at the time of the assessment visit before athletes completed the questionnaires. During this same appointment, the parents were interviewed with a structured questionnaire to obtain information on their child's age, and their birth and medical history.
The study was conducted in accordance with the Declaration of Helsinki, and the protocol received institutional ethical approval.

MedDiet Assessment
Dietary intake was assessed using the KIDMED test (Mediterranean Diet Quality Index in Children and Adolescents) repeatedly validated in Spain [71,72]. The parents had to answer the KIDMED questionnaire composed of 16 questions (Table 1) for children 8-12 years old while 13-15 years olds completed the questionnaires themselves. This test was used to assess the adherence to the MedDiet in children and adolescents. It consists of 16 items, among which there are 12 questions denoting a positive connotation (olive oil, fruits, vegetables, pulses, cereals, pasta or rice, nuts, fish, dairy products, and yoghurt) and 4 questions denoting a negative connotation (of fast food, baked goods, sweets, and skipping breakfast). Questions denoting a positive connotation are scored with +1, while negative connotation questions are scored with −1. According to the KIDMED index, a score of 0-3 reflects poor adherence to the MedDiet, a score of 4-7 describes adherence that needs improvement, and a score of 8-12 indicates good adherence [72,73]. The test exhibited reasonably good validity for assessing the nutrient intakes of children, and we used an ad hoc computer program developed for this purpose. Trained dietitians updated the nutrient data bank using the available information from the food composition tables for Spain. Table 1. KIDMED test to assess the Mediterranean diet quality scoring [72].

+1
Takes a fruit or fruit juice every day +1 Has a second fruit every day +1 Has fresh or cooked vegetables regularly once a day +1 Has fresh or cooked vegetables more than once a day +1 Consumes fish regularly (at least 2-3 times per week) −1 Goes more than once a week to a fast-food (hamburger) restaurant +1 Likes pulses and eats them more than once a week +1 Consumes pasta or rice almost every day (5 or more times per week) +1 Has cereals or grains (bread, etc.) for breakfast +1 Consumes nuts regularly (at least 2-3 times per week) +1 Uses olive oil at home −1 Skips breakfast +1 Has a dairy product for breakfast (yoghurt, milk, etc.) −1 Has commercially baked goods or pastries for breakfast +1 Takes two yoghurts and/or some cheese (40 g) daily −1 Takes sweets and candy several times every day

Burnout Syndrome Assessment
The Athlete Burnout Questionnaire (ABQ) is the most widely used instrument for the measurement of BS in athletes. It was developed by Raedeke (2009) [74]. We used a reduced model of the Spanish version [75,76], composed of 15 items (plus three alternative statements) that measure three dimensions of BS: physical/emotional exhaustion (PEE), reduced sense of accomplishment (RSA), and devaluation of sport (SD) practice (Table 2). A reduced version enables answering in a shorter time, which decreases the adverse effects of longer questionnaires, such as fatigue or lack of motivation. The ABQ was completed in all cases by the child/adolescent with help from the assessment personnel if needed. 15, which were formulated in the opposite direction; the lower the numerical response, the greater the degree of BS.

Statistical Analysis
KIDMED scores were categorized as poor adherence or need to improve when they were less than or equal to seven and good adherence when they were greater than seven. With regard to risk of BS, we categorized all participants into terciles of risk according to the ABQ score (low: ≤ 19.70, moderate: 19.71 to 24.99, and high: ≥ 25).
Contingency tables were constructed to analyze the relationship between sex, age (categorized in 8-10, 11-12 and 13-15 years) and KIDMED scores (in each item and globally) with the risk of BS that they present. The association between the categorical variables was evaluated by the χ2 test (or Fisher's exact test, as required). In the study of the relationship between the risk of BS and the KIDMED index of children, crude odds ratios (OR) were also calculated with their 95% confidence intervals.
A Student's t-test was run to compare the age of onset of BS according to KIDMED adherence for each risk range of BS. With dichotomous categorical variables, we compared the KIDMED adherence (poor and needs improvement vs. good) with each risk range of BS (low vs. moderate/high) using contingency tables, odds ratios (ORs), and the χ2 test, Student's t-test or Fisher's exact test, as required, to assess statistical significance. Bonferroni corrections were applied to control for multiple comparisons, as appropriate. Non-parametric statistical tests or Shapiro-Wilk tests were used to confirm the assumptions of normality, linearity, homoscedasticity, and independence. Moreover, in order to evaluate the relationship between the risk of BS (low vs. moderate/high) with the sociodemographic characteristics and KIDMED adherence, a multiple logistic regression analysis was performed. The risk of BS (low vs. moderate/high) was considered as a variable response, and the explanatory variables considered were: age (continuous), sex (two categories), level of studies (two categories), mobile use (two categories), television hours and videogames hours (continuous), siblings (two categories), overall KIDMED index, and fresh or cooked vegetables daily (two categories: poor/you need to improve vs. good). Additionally, we calculated the adjusted ORs with 95% CIs. All the p values were two-tailed. Statistically significant associations were considered when p < 0.05.
The data were entered into an Excel spreadsheet, using double-data entry to minimize the risk of errors, and were then transferred to SPSS version 19.0 (IBM SPSS, Statistical Package for Social Sciences, Inc., Chicago, IL, USA). All the p values are two tailed and p < 0.05 was considered statistically significant. Table 3 shows the relationship between gender and age and the BS risk that they present. This analysis revealed that the risks of the girls shifted towards the extremes, with a lower proportion of girls being lower risk, and higher proportion being high risk when compared to their male peers. There is a statistically significant association between sex and the risk of BS (p < 0.001). The main difference is observed in the most extreme, low-and high-risk categories, with the percentages of both sexes in the average level of risk being very similar. The percentage of girls increases as the risk increases, while the opposite happens with boys. Specifically, the percentage of girls at high risk of BS is significantly higher (55.3%) than the percentage of boys (16.1%). On the contrary, the percentage of girls with low BS risk is significantly lower (10.5%) than the percentage of boys with low risk of BS (46.4%). Most of the boys in the study have a low risk of BS, whereas most of the girls in the study have a high risk of BS.

Results
There is no statistically sufficient evidence to reject the independence between the children's age and the type of risk of BS (p = 0.137). However, it is observed that most of the younger children present a low (40%) or medium (42.5%) risk of BS and, on the contrary, the majority of older children present a high risk of BS (44%). Table 4 shows no statistically significant differences between the ages of onset of BS and the KIDMED adherence index of children for the groups with low (p = 0.784) and moderate (p = 0.818) risk of BS, while the high risk group presents a significantly later (p = 0.027) age of onset in children with better adherence to the MedDiet.   Table 5 shows the relationship between the KIDMED adhesion index (in each item of the test and overall) and the type of risk of BS (low vs. moderate/high) presented by children. There is not enough statistical evidence to reject the independence between the KIDMED adhesion index and the type of BS risk of children except in the case of the daily consumption of fresh or cooked vegetables. In this case, the percentage of children with a medium/high risk of BS is significantly higher when their KIDMED index is ≤ 7 (poor or in need of improvement) (89.5%) compared to those who have a good KIDMED index of ≥ 8 (63.5%). It is also observed that the percentage of children with a low risk of BS is higher in the case of children who have a good KIDMED index of ≥ 8 (37.3%) with respect to those who have a poor KIDMED index or who need improvement (≤ 7) (10.5%). In terms of OR, the odds of presenting a medium/high risk of BS is five times higher in children with a KIDMED index of ≤ 7 (poor or in need of improvement) for the intake of fresh or cooked vegetables daily compared to children with a good KIDMED index (≥ 8). In relation to the KIDMED adherence and comparing only the high risk of BS group (3rd tercile) with the reference group (low risk or 1st tercile), there are significant differences between the group with good (≥ 8), and poor/needs improvement (≤ 7) for fresh and cooked vegetables daily (p = 0.004), regular fish (p = 0.042), and pasta or rice almost daily (0.019). In the case of the use of olive oil at home (p = 0.054), freshly cooked vegetables > 1/day (p = 0.059), and sweets and candy (p = 0.053), although it does not reach statistical significance, we can say that it suggests an association, or there is a tendency towards significance. In the participants with a low risk of BS, a greater proportion ate breakfast, consumed more vegetables, pasta and rice, and had a greater adherence to the overall KIDMED index. Table 6 shows the results of the multiple logistic regression model to study the relationship between the risk of BS and the sociodemographic characteristics and adherence to the MedDiet. It is observed that the probability of presenting a moderate/high risk of BS is greater in girls, with the odds of presenting a moderate/high risk of BS being approximately ten times higher in girls than in boys. As the number of hours of television and video games watched by children increases, the probability of their presenting a moderate/high BS risk increases. The odds of having a moderate/high BS risk are multiplied by approximately seven for each one-hour increase in the number of hours of television and video games. In the logistic regression model, the KIDMED score has incorporated vegetable consumption as an explanatory variable given that it was the only KIDMED item to show statistical differences in the previous analysis. The probability of presenting a moderate/high risk of BS is lower in children with a good KIDMED index score for vegetable consumption. The odds of presenting a moderate/high risk of BS is approximately nine times higher in children with a poor KIDMED score for vegetable consumption when compared to children with a good KIDMED score for vegetable consumption.

Discussion
There is no consensus among previous research comparing the prevalence of risk of BS. In our study, the participating child and adolescent athletes were classified according to their risk of suffering from BS. Of these, 31.9% had a low risk, 36.2% had a medium risk and 31.9% had a high risk of suffering from BS. The results agree with Garces de los Fayos (1993) [77], in whose study the sample was composed of 33 tennis players whose ages ranged from 11 to 16 years, and with a prevalence of around 35% of the sample that would have suffered, at least once, a high risk of BS. Likewise, Rotella et al. (1991) [78] [81], whose sample was composed of 84 tennis players, 58 boys and 26 girls, with ages ranging from 13 to 20 years old, referred to similar data. The results of this research show that 4.8% of the study population has high BS values, values similar to those obtained in the studies by Balaguer et al. (2009) [82] in a sample of 225 young tennis players; or in those of Sierra Llamas et al. (2008) [83], and Gustafsson et al. (2007) [34] in athletes of different modalities.
These differences can be explained by considering that individual athletes (e.g., tennis players) may have higher levels of BS than team sports practitioners or that younger athletes tend to have higher levels of BS [84]. In this study, the first assumption is met; however, in this particular case it is the older athletes that present a higher risk of BS.
Several investigations have been conducted aiming to find any association between external (companions, family, coach, public, or club, among others), and internal stressors (personality factors, or specific traits) and different lifestyle variables such as quantity and quality of physical activity [23,[85][86][87], or adherence to a healthy diet [88,89]. Despite previous evidence, unfortunately within sports the possible association between eating behaviors and nutritional intake and BS risk has been given little and non-specific attention. As far as we know, there are no previous studies that have specifically addressed in children and adolescent athletes the association between nutritional intake and the risk of BS.
What foods we choose to consume can be one of the most important factors that influence our mental and physical health [90]. The health benefits of the MedDiet and of physical activity have been documented. Over the last few years, this diet has been recognized for its nutritional composition and effects on human health because it shows an excellent profile of macronutrients and essential micronutrients [65,91]. In this context, some particular deficits of micronutrients or the joint effect of a suboptimal intake [92,93] for several of them could be responsible for the increased risk of psychological disorders, such as BS, among participants who follow low quality dietary patterns.

Study Limitations
Although the sample sizes were a relatively small representation of the population of the study, we believe that our study offers strong internal validity given the low attrition rate. We are confident regarding the good quality of the self-reported information used for the questionnaires.
Parents and coaches were very interested in the study and were extensively trained and supported in the completion of the records. All this can compensate for the limitations that the selection of the sample may impose or for external validity generalization. We should also point out that the data for our study were compared with the potential confounders, with all children having a similar place of residence and socioeconomic status and being followed up over the same time period. We believe that these factors may help to compensate for limitations in generalizability and external validity associated with participant selection. Finally, it could be that the KIDMED index score should be different for child and adolescent athletes, although for the moment, no alternative recommendations have been proposed.

Conclusions
Participants with moderate/high BS risk were more likely to be girls and spend a higher number of hours watching television or playing video games. There is not enough statistical evidence in this study to reject the independence between the level of adherence to the Mediterranean diet and the risk of BS in children, except in the case of daily consumption of fresh or cooked vegetables. An increased risk of BS is likely to be inversely associated with healthier dietary patterns, better nutrient profile, and better eating attitudes. The use of questionnaires that evaluate lifestyles and dietary patterns in child and adolescent athletes can help to add additional information when assessing the risk of BS.
Inadequate intake of some macro and micronutrients could play a role in the development of BS. Those responsible for formulating educational and sports policies, sports doctors, nutritionists and trainers should recommend compliance with high-quality dietary patterns, such as the KIDMED index, thus ensuring adequate nutritional intake that provides an adequate level of different macro and micronutrients for improve the mental and physical health of athletes. More studies are required with representative samples of the target population since inadequate dietary intake and dietary behaviors can adversely affect not only the quality of training, and recovery, but also nutritional status, BS, and general health of the athletes. Finally, the main implications of this study for practice are that coaches and parents should make recommendations on a case-by-case basis after considering a dietary assessment together with a ABQ questionnaire in child and adolescent athletes.