Obesity is a disease which has long been regarded as an important social problem in Korea. Among them, child obesity is receiving more attention because it causes children to be highly likely to develop chronic metabolic diseases such as hypertension, hypertriglyceridemia, hyperinsulinemia, nonalcoholic fatty liver disease, diabetes, and et Cetra [12]. Recently, Jeju Island had the highest student obesity prevalence, with 14.8% in the national elementary, middle and high school students' health examination data as conducted by the Korea Educational Development Institute in 2012 [3]. The prevalence of the national student obesity was 12.3% in 2012. Jeju Island showed the highest obesity rate for 3 consecutive years. From 2012, it showed 15.9% in 2013 [4] and 16.7% in 2014 [5]. It became a big issue in the community of Jeju Island. This study was designed to find out why Jeju Island had a higher obesity rate than did other regions.

This study used original data from the 82,581 sample data of nationwide elementary, middle, and high school health examinations, conducted by the Ministry of Education in 2014. It selected 33,488 elementary school students' data from this data set.

This study included 28,026 elementary students who belonged to the normal and obesity groups. It excluded underweight and overweight groups among the 33,488 elementary students in 15 provinces, nationwide, based on body mass index. This classification is based on the percentile of body mass index by sex and age. The underweight group is less than the 5th percentile. The normal group is more than the 5th and less than the 85th percentile. The overweight group is more than the 85th and less than the 95th percentile. The obesity group is more than the 95th percentile or weighs more than 25 kg/m² regardless of percentile.

This study selected twelve questions which were known to be factors related to obesity; these were selected from among the health survey questionnaire of the original data. Twelve questions were composed of 7 eating habits (Ramen intake rate ≥3 times a week, consuming soft drinks rate ≥3 times a week, fast food intake rate ≥3 times a week, skipped breakfast rate, meat intake rate ≥3 times a week, fruit intake rate ≥3 times a week, and vegetable intake rate (not including Kimchi) ≥3 times a week), 1 exercise habit (intense physical activity practice rate ≥3 times a week), 2 lifestyle habits (sleeping time ≤7 hours and using computer ≥2 hours a day) and 2 mental health areas (bullying experience within the past year and experience of runaway impulse) (Table 1).

From our analysis of 28,026 elementary school children with obesity or normal weight in the 15 provinces, we analyzed 12 factors related to eating habits, exercise habits, lifestyle, and mental health. The differences related to 12 factors between the obese and normal weight children were researched. Finally, Jeju was compared with Seoul, which has the lowest obesity prevalence in school age children.

Statistical analysis was performed with PASW Statistics ver. 18.0 (IBM Co., Armonk, NY, USA). The chi square was used to clarify factors related to child obesity, as well as to compare Jeju Island to Seoul.

This study was designed to analyze why Jeju Island has had the highest obesity rate since 2012. According to the study of Seo [6], the obesity prevalence of school age children on Jeju Island in 2002 was 9.9%. But in 2014, it was 16.7% [5]. Therefore, we tried to discover what made the school age children of Jeju Island become obese. This study utilized the raw data of the health examination and survey conducted by the Ministry of Education in 2014. Among the several questions in the survey, we selected 12 factors known to be related to obesity (Table 1). First, we compared the normal group with the obesity group to confirm the correlation between the 12 factors and obesity (Table 2). Then, we compared Jeju Island to Seoul (Table 3).

For the eating habits, the present study showed that the frequency of consuming soft drinks, fast food intake, and skipping breakfast are associated with obesity. However, the frequency of Ramen intake is not associated with obesity (Table 2). Jee and Kim [7] showed the same result about the frequency of Ramen intake. Camargo and Marín-León [8] reported the frequency of fast food intake is related to obesity, but consumption of soft drinks and Ramen are not. And Janssen et al. [9] also demonstrated that there was no significant finding between soft drink intake and being overweight in 34 western countries. Maddah and Nikooyeh [10] discovered that obesity may develop easily in those who skipped breakfast. Szajewska and Ruszczynski [11] proved the importance of breakfast consumption in children and adolescents in Europe. They showed eating breakfast helps individuals not become obese.

For the exercise habit, the present study showed that the rate of practicing an intense physical activity in the obesity group was not significantly different from that in the normal group (Table 2). But other studies showed different results. Kim et al. [12] showed that physical activity is significantly associated with the reduction of obesity, by increasing heart rate and respiratory rate. Janssen et al.'s results [9] demonstrated a negative relationship between physical activity and body mass index (BMI) classification.

For the lifestyle habit, the present study showed that having a sleeping time of less than 7 hours and using computers more than 2 hours are strongly associated with the risk of obesity (Table 2) Chronic partial sleep loss may increase the risk of obesity [13]. Li et al. [14] showed that children who had screen time of more than 2 hours are more likely to become obese. Park et al. [15] also presented the idea that watching TV more than 2 hours can make a child obese.

For the mental health area, having a bullying experience within the past year, and having the experience of a runaway impulse, are associated with the risk of obesity (Table 2). However, Peltzer and Pengpid [16] showed that a bullying experience within the past 30 days did not affect children's weight gain. In addition, Janssen et al. [17] pointed out an important part of the bullying experience. Overweight children are significantly more likely to be victims of bullying than are normal weight children.

In our cross-sectional analysis between the normal group and the obesity group, we confirmed that seven non-healthy behaviors (soft drink, fast food, skipped breakfast, sleep ≤7 hours a day, computer ≥2 hours a day, bullying experience, and runaway impulse) and two healthy behaviors (fruit intake, vegetable intake) had a significant relationship with obesity (Table 2). Compared to Seoul, only the Ramen intake rate and meat intake rate of Jeju Island were significantly higher (Table 3). Among the seven confirmed obesity factors, five factors (soft drink, fast food, skipped breakfast, computer ≥2 hours a day, bully experience) have no significant differences between them (Table 3). In addition, the two confirmed obesity factors (sleep ≤7 hours a day, runaway impulse) are significantly higher in Seoul (Table 3). Finally, the authors concluded that the main cause of a high obesity rate in Jeju Island was the low level of a healthy food intake rate such as the children's fruit and vegetable intake. It is ironic that Jeju Island has a lower fruit intake rate than Seoul, considering that Jeju Island is the place to sell various fruits especially mandarins to Seoul.

This study has several limitations. First is that all of the results, except for the BMI, were obtained through a survey with a questionnaire. In the case of the first grade students in elementary school, their guardian responded to the survey. In the case of 4th grade students in elementary school, students responded to the survey with their guardian. Therefore, the result of a survey might not reflect the actual eating habits, physical activity, lifestyle and mental health of the student. Second, the present study did not include other obesity factors such as socioeconomic status, educational status of a parent, and the distance of commuting because this study utilized the raw data which was designed not for analyzing obesity but for checking overall health. The third is that the present study cannot discover the impact of compounded factors which could occur between the twelve obesity factors. Regardless of these limitations, the present study has several strengths. First, this study analyzed 28,206 elementary students in Korea nationwide. Second, the present study is designed to focus on specific local health issues in Jeju Island. Third, the result of study can be applied directly to solve the problem of obesity in elementary school students in the community of Jeju Island.

In conclusion, our obesity factor analysis showed that high soft drinks and fast food consumption rate, skipped breakfast, lack of sleep, excessive computer use, runaway urge, bullying experience rate and lower fruit, vegetable were the causes of obesity. As a result of factor analysis, the low rate of fruit and vegetable intake seemed to be the main obesity cause of the Jeju Island high obesity rate in children. It showed no significant difference, but high Ramen intake rate and insufficient physical activity were more prominent in elementary school students in Jeju than in Seoul. These results can be used as educational material for the Jeju elementary school health care project to lower the obesity rate of Jeju Island.