The Prevalence of Breakfast Skipping and its Association with Lifestyle Factors and Weight in 11-15 years Adolescents from Selected Lebanese Regions

Today, overweight and obesity rates are on the rise worldwide reaching an epidemic level in the twenty-first century [1]. The IOTF (International Obesity Task Force) indicated that 1 in 10 school-aged children are overweight or obese [2]. According to the World Health Organization (WHO), high body mass index (BMI) during adolescence predicts elevated adult mortality and cardiovascular disease rates, even if the excess body weight is lost [3].


Introduction
Today, overweight and obesity rates are on the rise worldwide reaching an epidemic level in the twenty-first century [1]. The IOTF (International Obesity Task Force) indicated that 1 in 10 school-aged children are overweight or obese [2]. According to the World Health Organization (WHO), high body mass index (BMI) during adolescence predicts elevated adult mortality and cardiovascular disease rates, even if the excess body weight is lost [3].
Obesity usually results from a combination of factors including physical inactivity, unhealthy diet, and certain eating habits including breakfast skipping [4]. The prevalence of breakfast skipping among children and adolescents ranges from 10 to 30% in the United States and Europe [5]. Several studies have demonstrated that breakfast consumption may influence appetite [6], dietary intake and composition [7] which may lead to weight gain [8,9] and increase the risk of chronic diseases [10]. Nonetheless, most of the existing studies remain unclear about whether there is a strong relationship between omitting breakfast and weight gain [11,12].
In Lebanon, several studies have shown a high prevalence of obesity in different age groups [13,14] and it was higher among adolescent boys than girls [15]. Moreover, in a large sample of Lebanese adolescents, Salameh et al. found a positive association between obesity and the frequency of dieting and taking diet pills [16].
An interesting hypothesis is that skipping breakfast in adolescents living in Lebanon may be associated with higher BMI-for-age, dieting practices, lower physical activity level, higher screen time and lower socio-economic status (SES).
The aim of the present study is to determine the prevalence of skipping breakfast and associated risk factors among a random sample of adolescents in public and private Lebanese schools from selected Lebanese regions.
Results of this study would be necessary to provide adequate preventive measures in order to decrease the impact of breakfast skipping on weight and healthy eating practices of adolescents living in Lebanon.
The major strength to this study is that, it would be the first comprehensive study to simultaneously assess the prevalence of breakfast skipping in a sample of Lebanese adolescents in combination with dietary assessment, age, gender, SES and BMI. However, the study is not without limitations; the study findings are not be generalizable beyond adolescents attending Lebanese private and public schools in all Lebanese regions, and because the study is cross-sectional in nature, causality cannot be inferred.
BMI-for-age. Data were collected between November 2012 and May 2013.
According to the Ministry of higher education, 203 private and public middle schools are spread in these regions, including 25348 students registered for the academic year 2011-2012 [17]. Sample selection took into consideration the different proportions of private and public schools, the different sizes of schools in terms of students' enrollment and the differences between districts. In Jbeil and Kesrouan regions, 2% of the schools' adolescents and 1.3% in Metn District were contacted for participation.
In total 404 adolescents agreed to participate and were selected from 34 Lebanese middle schools out of which 12 schools were public and 22 were private. Adolescents aged less than 11 years or more than 15 years were excluded from the sampling.

Screening tests
Questionnaire: Students were asked to fill a pre-tested questionnaire offered in two languages (French or English) (Appendix 2) and adapted for the purpose of the study. Data collection and anthropometric measurements were held by a licensed dietitian trained to assess in the field.
The questionnaire included questions about age, gender, SES (parental income and private or public school) and parental presence. Several questions were formulated to assess skipping meals, particularly breakfast. Physical activity, screen time and some lifestyle habits (cigarette smoking, sleep duration, alcohol consumption), eating disorders, and some dietary habits were also assessed. Table 1 displays a brief description of the variables used in the current analysis.
Anthropometric measurements: Anthropometric measurements were taken using standardized techniques and calibrated equipment. Subjects were weighed to the nearest 0.1 kg in the morning wearing basic school uniform. Using a stadiometer, height was measured without shoes and recorded to the nearest 0.5 cm. WHO 2007 growth charts were used to monitor growth for children and adolescent aged between 5 and 19 years old that defined the development using z-scores cutoff points [18]. Normal children have a z-score between -1 and 1. A z-score above or equal 1 SD classifies the child as overweight; a z-score above or equal 2 SD mark that the child is obese. As for a z-score below -1 or equal is classified as marginally underweight, below or equal -2 the children is moderately underweight and below or equal -3 he is severely underweight. Z-scores were calculated using the Anthro Plus software [18].
Assessing skipping breakfast: Due to the absence of a standard definition, skipping breakfast was assessed by several questions. First, an open-ended question "In the last 7 days, on how many days did you eat breakfast?" was asked with responses ranging from 0 to 7 days. Thus, three definitions of breakfast skipping were created: Missing breakfast at least once, at least three times, or at least six times in the past seven days. Furthermore, skipping breakfast was also assessed by "where do you usually eat breakfast?" Skippers were those who answered "I never eat breakfast". Energy intake during breakfast, derived from the 24-hour recall, was another tool to assess skipping breakfast. Two definitions of skipping breakfast were used. An adolescent is considered a skipper, if on the day before collecting the data; ate nothing (0 Kcal) before 10 a.m. or ate nothing (0 Kcal) before 12 p.m.

Dietary intake:
A 24-hour recall was used to assess dietary intake and was collected for the day before the adolescents' interview, then assessed by a free automated self-administered 24-hour recall (ASA24) software [19].
Assessing other variables: Physical activity and screen time were assessed using the 2011 Middle School Youth Risk Behavior Survey questionnaire developed by the CDC [20]. For assessment of eating disorders, the SCOFF questionnaire validated among adolescents has been used [21,22]. An adolescent having a score equal or above 2 was considered to be at risk for eating disorders.

Statistical analysis
Statistical analysis was executed using the Statistical Analysis for Social Sciences (SPSS, version 16.0) and the level of significance was set at p<0.05.
Frequencies and descriptive variables were conducted stratified either by gender or by skipping breakfast. Means of age and anthropometric measurements were calculated through an independent sample t-test and ANOVA regression.
The z-scores drawn from the WHO 2007 reference were coded according to different cutoffs and deducing the prevalence of overweight and obesity.
Linear regression was performed for variables with more than 2 categories with the calculation for the significance.
The association between dietary intake of energy and macronutrients was then assessed through Pearson correlation with the calculation of the significance.

Results and Discussion
The main characteristic of the sample for this study is illustrated in Tables 2 and 3. The mean age of the study, indicated in Table 2, is 13.7 ± 1.010. Half of the participants have a normal BMI-for-age, while 25% were overweight and 20% were obese. Boys were more obese than girls; whereas, girls were overweight more than boys (Table 2).
Various definitions were used to assess skipping breakfast and the prevalence varied between 8.4% according to the location till 42.8% for skipping one or more days per week ( Figure 1). This percentage is comparable to the findings in previous studies. Tin et al. found that 5.2% of Hong Kong children were skippers when they used the location to define breakfast. Whereas in the same country, when Cheng et al. defined breakfast skippers as missing to drink and to eat before morning classes at least one school day during the past week, they found a prevalence of 30.5% skippers [23]. Dialektakou et al. used 24 different definitions for breakfast and the prevalence ranged between 3.6% never eats in the morning and 74.7% eats on average 0-6 mornings/week [24,25]. Thus, care is needed while comparing breakfast skipping with other studies and while assessing the association with different health outcomes [25].
In this study, almost all various definitions agreed that girls skip more breakfast than boys with p<0.05 but without any significance when defining breakfast by calories similar to other studies ( with p>0.05. Thus, breakfast skipping could not be a useful marker in eating disorders for our study. Furthermore, the intake of fruits, vegetables, pastries and sweets were similar for adolescent that were skippers and non-skippers. This was not consistent in the literature [29]. This inconsistency might be explained by the changes in dietary behaviors among adolescents apart from breakfast eating.

SES:
Socio-economic variables did not affect skippers as mentioned in the literature [29][30][31]. This could be due to the similarity in SES for the adolescent that tend to skip or not breakfast with majority of them had middle income. Shaw describes skipping breakfast as a matter of personal choice rather than influenced by SES [32]. However, the type of school has affected the consumption of breakfast where adolescents that attend public schools tend to skip breakfast more than adolescents studying in private schools with p<0.05 (Table 3). These results might be explained by less quality of education regarding healthy lifestyle in public schools. Furthermore, the frequency of parents preparing food for adolescents and specifically breakfast were strongly associated to breakfast skipping with p<0.001 (Table 3). The result from the present

Skipping breakfast in our sample is associated with many factors Dietary behaviors and lifestyle factors:
The majority of our sample tends to eat three to five meals per day. As for eating lunch, breakfast skippers tend to skip lunch more than eaters (17.6% and 8.7% consecutively) (Table 3). For milk consumption, the majority of breakfast skippers never drink milk (82.4%) (Table 3). Furthermore, there is an association between skipping breakfast and sleeping duration where skippers tend to sleep less than six hours per day (29.4%) more than eaters (12.2%) with p<0.05 (Table 5). Thus, similar to our findings, healthy behaviors were reported with the daily intake of breakfast [27].
Contradictory results were however found in our study when comparing with previous research. First, physical activity and television viewing were not associated to skipping breakfast. This result might be due to the absence of a universally acceptable method for assessing physical activity and sedentary behaviors in adolescents [28]. Second, eating disorders were not correlated in our sample to skipping breakfast   study supports the literature for this positive association [33]. Thus, parents could play an important role in adolescents' dietary behavior.
Linear regression model non-adjusted or adjusted did not found any association between BMI-for-age and skipping breakfast, even though various definitions were taken into consideration and different confounders were examined as well with p>0.05 (Tables 6 and 7).
Many studies in the literature have found a correlation between breakfast intake and weight [5,8]. Several studies have found also no association between breakfast skipping and weight changes [7,11,25,34]. This inconsistency in results might be explained by the absence of a standardized definition for skipping breakfast. Finally, fat intake and carbohydrate intake were positively correlated to BMI-for-age with p<0.05, but no association was found while adjusting to total energy p=0.091 (Table 8). However, comparing the definition of breakfast by calorie intake to location only 9 adolescents were found to be common (Table 9) with a mean BMI-for-age 1.55 ± 0.78 (Table 10).

Strength and limitations
Other than the absence of standard definition, the major limitation of this study is that it is cross-sectional so the causality could not be drawn. Underreporting energy intake and breakfast for adolescents might be another barrier for studying the relationship between breakfast and BMI. Overweight and obese children tend to underreport food intakes than normal weight [35], so they may also underreport breakfast consumption too [36].
On the other hand, the usage of one single 24-hour recall might not reflect the real intake of the adolescents' population. Multiple 24hour recalls for the individual are required. Moreover, 24-hour recall reliance on memory might be another limitation besides errors in data collection and the absence of a Lebanese food database.
Finally, this study is the first national study examining breakfast skipping in relation with weight in Lebanon; a few Lebanese studies have studied skipping breakfast apart and overweight and obesity without searching for a relation between them. The strength of the current study relies in the questionnaire's assessment of a multitude of covariates. Not to mention that, this study assessed skipping breakfast using different definitions that allows for more accurate comparison. However, those results could only be generated to the region where this study was conducted; due to small sampling size that does not represent the entire Lebanese adolescent population.

Conclusion
Prevalence of breakfast skipping and on the other hand overweight and obesity among adolescents has been on the rise in the last few decades. The relation between these two phenomena however remains complex. Due to the importance of regularity in breakfast consumption on nutritional quality of the diet and the improvement in academic performance and psychosocial functioning, the purpose of the current study was to estimate the prevalence of skipping breakfast among 11-15 years old adolescents living in Lebanon and identifying whether there is an association between breakfast patterns and BMI-for-age taking into consideration confounders [37].
According to this study, the prevalence of overweight and obesity were high attaining 25% overweight and 19% obesity among adolescents. In contrast to the absence of a specific definition of breakfast skipping, several definitions were used to describe breakfast skipping and its prevalence varied widely between these definitions from 8.4% till 42.8%.
In an attempt to identify the correlation between breakfast and weight, this analysis found no association between the components despite taking into consideration the various definitions and confounders.
As for dietary intake, fat and carbohydrate were positively associated with BMI-for-age (p<0.05) contrary to energy intake where there was On average skipping breakfast more than 6 days per week Don't know 6 (1.6%) 0 (0%) 6 (3.2%) Skipping breakfast was correlated with the school type, parental preparation of food specifically breakfast, sleeping duration, a lower consumption of milk, skipping lunch more often and eating fewer meals per day.
This study could not conclude a relation between BMI-for-age and breakfast in the Lebanese adolescents' population. Nevertheless, admitting that skipping breakfast has no effect on the outcome of health and specifically BMI is inappropriate because causality cannot be settled with cross-sectional studies. However, this research represents a guide for further research that will focus on skipping breakfast regarding weight changes while taking into consideration other variables.
Finally, this research emphasizes the importance and need of more in-to-depth cross-sectional or preferably longitudinal studies to be conducted by public health agencies. These new researches must be      based on large representative samples of the entire Lebanese population in all districts addressing the high prevalence and causes of overweight and obesity [38] and their possible relation to breakfast consumption in Lebanon [39].

Conflict of Interest
The following is a research article in which all participating authors meet the uniform requirements of the Occupational Medicine and Health Affairs criteria for authorship. This work was approved by the ethics committee at the Holy Spirit University, it was not sponsored by any organization, and there were no conflict of interest regarding this work.