| Home | E-Submission | Sitemap | Contact Us |  
top_img
J Prev Med Public Health > Volume 47(3); 2014 > Article
Hong, Bae, Kim, and Park: Variation in Meal-skipping Rates of Korean Adolescents According to Socio-economic Status: Results of the Korea Youth Risk Behavior Web-based Survey

ABSTRACT

Objectives

To identify and evaluate the trend of meal-skipping rates among Korean adolescents with their contributing causes and the influence of household income level on meal skipping.

Methods

Using 2008, 2010, and 2012 data from the Korea Youth Risk Behavior Web-based Survey of 222 662 students, a cross-sectional study with subgroup analysis was performed. We calculated odds ratios for skipping each meal 5 or more times in a week by household socio-economic status using a multiple logistic regression model. The secular change in the meal-skipping rates by the students' family affluence scale was analyzed by comparing the meal-skipping students within each subgroup and odds ratios for the same event over time.

Results

Through 2008 to 2012, most of the meal-skipping rates generally showed a continuous increase or were almost unchanged in both sexes, except for breakfast skipping in several subgroups. Students in low-income households not living with both parents had the highest meal-skipping rates and odds ratios for frequent meal skipping. In a time-series subgroup analysis, the overall odds ratios for the same event increased during 2008 to 2012, with a slight reduction in the gap between low and higher income levels with regard to meal skipping during 2010 to 2012.

Conclusions

Household socio-economic status and several other factors had a significant influence on Korean adolescent meal-skipping rates. Although the gap in eating behavior associated with household socio-economic differences is currently decreasing, further study and appropriate interventions are needed.

INTRODUCTION

Nutrition and eating habits are important factors affecting the health of a population. Among related indicators, the meal-skipping rate is a simple but meaningful measure for evaluating the eating habits of a particular group. Indeed, many researchers have investigated the number of food consumption periods per day or meal skipping to understand the dietary patterns of a population [1,2,3,4]. These studies have usually demonstrated a relationship between a higher household socio-economic status (SES) and a higher total number of eating occasions or maintenance of good eating habits in children and adolescents [1,4]. In addition, these outcomes were connected to a high intake of nutritious food and overall good health.
Given the important role of dietary intake and nutrition during adolescence in the prevention of disease and the prediction of health status in adulthood, numerous national policies and guidelines on food intake have been established [5] in many countries. Many studies have demonstrated a positive association between low household SES or a high poverty ratio and undesirable dietary behaviors such as meal skipping or poor nutritional intake [6,7,8,9,10,11]. On the other hand, however, a number of long-term studies regarding the trend of change in eating habits over time within a specific society have generally demonstrated that the number of eating episodes and the quality of food eaten have shown improvement over time, with a decrease in the dietary behavioral gap among different social or economic classes [2,3,7,9].
Evidence also suggests that eating regularly and not skipping meals have a positive effect on overall quality of life and achievement, including academic performance, energy, and emotional functioning [12,13]. This is an especially important matter for those who are more vulnerable in society, such as adolescents from low-income families, because meal skipping can itself act as a social signal of impoverishment, as well as have secondary adverse effects or even an association with the inheritance of poverty. On this matter, some research has focused specifically on the meal skipping of adolescents from low-income households, searching for policy alternatives as well as analyzing related other problems [14,15,16].
Proper nutrition has a tremendous impact on a child's well-being, making it a fundamental human right [14]. Therefore, many studies of meal skipping among Korean children and adolescents have been conducted [17,18,19]. To address the problem of individual food insecurity among children in Korea, the Korean government has conducted school meal programs during school vacations since 2000, and has provided financial support to the children of low-income households for school meals. Beyond these two major national school feeding support systems, a free school lunch service was established in 2011, aiming for nationwide implementation. By gradually adding educational institutions receiving support, it has already been expanded to include almost all middle and high schools.
In this investigation, we focused on and analyzed the pattern and associated factors of Korean adolescent rates of skipping each of the three daily meals, especially the differences related to variation in household SES. We also conducted a subgroup analysis in order to compare these differences before and after the expansion of free meal services for lunch in Korea, and to evaluate the effectiveness of this policy.

METHODS

Study Design and Participants

We analyzed three years of survey data through a three-time cross-sectional study. The data for this study were obtained from the Korea Youth Risk Behavior Web-based Survey 2008, 2010, and 2012 (KYRBWS-IV/VI/VIII) [20]. The KYRBWS is a nationwide web-based survey regarding health behaviors of Korean adolescents (middle and high school students) conducted jointly by the Korea Centers for Disease Control and Prevention, the Ministry of Health and Welfare, and the Ministry of Education, Science, and Technology.
The study samples included 79 099 students in 2008, 74 980 students in 2010, and 76 980 students in 2012 from 400 middle schools and 400 high schools, selected by using a two-stage stratified cluster sampling method from all middle and high school students in Korea. A population was stratified by 43 regional and school-type variables, and then the sample was selected through two stages of school (the first sampling unit) and class (the second sampling unit). The total number of respondents was 222 662, consisting of 75 238 individuals in 2008, 73 238 individuals in 2010, and 74 186 individuals in 2012. The response rates of each year were 95.1% (75 238/79 099) in 2008, 97.7% (73 238/74 980) in 2010, and 96.4% (74 186/76 980) in 2012.

Measurement of Variables

The dependent variable of our research was the rate of skipping each meal in Korean adolescents. The nine independent variables were as follows: student's grade level, body mass index (BMI), family affluence scale as a representation of household income level, average spending money per week, household members, perceived body image, effort towards changing body weight during the last month, stress awareness, and research year. In addition to the demographic factors and variables of direct interest to this study, other variables were mostly selected based on the factors reported in previous research to be associated with adolescents' eating habits [14,15,21]. The average amount of spending money per week was also chosen as one of the independent variables in order to examine the effect of being able to afford consuming meal substitutes, such as fast foods or other snacks. All variables were measured through responses to multiple-choice or combination questions, except BMI, which was calculated from reported height and weight.
The data was analyzed separately for males and females except for the subgroup analysis, because meal skipping reveals sex-dependent differences in itself, and is often closely related to the variables that are dependent upon sex, such as efforts to change body weight or self-perception of body image [15,22,23,24]. For most of the variables, we undertook recategorization of choices or created new variables. Groups were categorized according to BMI (obesity, overweight, normal body weight, and low body weight groups), according to the standard growth chart for adolescents published by the Korea Centers for Disease Control and Prevention in 2007 [25]. As an indicator of the SES of the household, the Family Affluence Scale (FAS) was used, which was first developed from an international cooperative study of the World Health Organization and then adopted by the KYRBWS [20]. The FAS is a four-item questionnaire that helps students report their family income objectively: It evaluates the sum of scores regarding whether the family owns a car, whether the student has his/her own bedroom, the number of family vacations during the past 12 months, and the number of computers the family owns [26,27,28,29]. Comparison with other socioeconomic indicators, such as father's education level or student's perceived SES, will be discussed later. In addition, the responses regarding average spending money per week were divided into a total of six groups, by combining several groups together that were too small for analysis. The variable of family members, which consists of selected combination-type responses, was transformed to a binary variable with reclassified groups divided by whether a student lives with both parents or not. Those who did not attempt to control body weight in any way and those who attempted to maintain body weight were combined into one group, while stress awareness, which had been measured on a five-point scale, was reclassified into three groups.
As a dependent variable, the rates of skipping each meal were defined as a proportion of students who ate their breakfast, lunch, or dinner less than three times during the previous seven days (skipping a meal five or more times per week) in each subgroup of all independent variables, according to the equation in the guidelines of the KYRBWS 2012 for using primitive data [20]. The denominator would be the total numbers of students in each specific subgroup, for example, total female students, or male students, who responded that their family affluence scale was high.

Statistical Analysis

For all categorical variables, data are presented as the number of cases and their respective proportions. The statistical analysis was done on weighted data using the SURVEYFREQ and SURVEYLOGISTIC procedures, so that the population and the sample were consistent with each other in demographic composition. In order to see the differences in meal-skipping rates among different responding subgroups of the variables for each single factor, a chi-squared test was performed. Logistic regression analysis was performed with the trend test to determine the odds ratios of skipping each meal five or more times per week due to the change in each variable while controlling for the effect of the other variables. In addition, particularly for the purpose of examining the patterns of change in the effect of household economic conditions over time on a student's meal-skipping rate, a subgroup analysis according to family affluence scale was conducted for each year, as a variable of interest among the eight variables originally selected. The results have also been presented as an every-other-year change in meal-skipping rates and odds ratios for skipping each meal. Trend tests for these changing patterns were also conducted. In every logistic regression analysis including all dependent and independent variables, we confirmed that there was no multicollinearity between these variables using the VIF (variance inflation factor). A p-value of <0.05 was considered to indicate significance and all statistical analyses were performed using SAS version 9.2 (SAS Inc., Cary, NC, USA).

RESULTS

Surveys were conducted in 2008, 2010, and 2012 in order to target the entire sample group consisting of 115 890 male and 106 772 female students.
Breakfast-skipping rates were lowest in 2012, but lunch- and dinner-skipping rates increased continuously or remained nearly unchanged during 2010 and 2012 for both sexes. This prevalence of frequent meal skipping increased from 5.7% to 12.8% in males and 3.9% to 8.8% in females for lunch, and from 6.4% to 12.6% in males and 8.1% to 9.8% in females for dinner. Meanwhile, the breakfast-skipping rates decreased steadily from 25.1% to 24.9% in males and 26.7% to 24.6% in females. We also observed that the group of students with low household SES had the highest meal-skipping rates among the entire research population, and paradoxically, students who received the largest amount of spending money also had the highest meal-skipping rate in both sexes. The same result was seen in the group of students reporting that at least one of their parents did not live with them, and those that felt a lot of stress in daily living (Table 1).
The odds ratios for skipping each meal five times or more per week in each subgroup were calculated for all independent variables after adjusting for the effect of other variables. In the case of lunch and dinner in both sexes, the overall risk of meal skipping increased substantially in 2010 and 2012 compared with 2008 (p=0.004 for trend in dinner skipping among females; p<0.001 for all other trend comparisons). The group of students whose household SES was low, those who did not live with both parents, and those whose average spending money per week was high presented the highest risk of meal skipping. These results were statistically significant and were true of every meal for both sexes.
In the case of breakfast, students of both sexes showed the lowest risk of frequent meal skipping when they had a lean body image, but this was only statistically significant in males. For lunch and dinner, in contrast, students who perceived themselves as fat recorded the lowest risk of frequent meal skipping, and almost all cases were statistically significant. The group with the highest risk for meal skipping was students who replied they had an average body type in both sexes, for every meal. On the other hand, when we focused on actual BMI, the low bodyweight group showed a tendency toward having the highest risk of meal skipping in both sexes, but these differences were not significant except for breakfast skipping in male students. Also, the risk for meal skipping showed a certain correlation with efforts to change bodyweight for boys, meaning that, for every meal, there was the highest risk for frequent meal skipping in those who were trying to lose weight and the lowest risk for frequent meal skipping in those who were trying to gain weight. However, among the girls, for every meal, the students who did not make any effort to change their weight demonstrated the lowest risk for meal skipping. Whereas the girls who were trying to lose weight recorded the highest breakfast-skipping risk and this risk showed an in turn showed a correlation with an attempt to change bodyweight, girls showed a contradictory tendency for lunch and dinner, meaning that the highest meal-skipping risk was recorded in those who were trying to gain weight, and it was statistically significant. Those with a stress levels that were slightly above the average did not have a significantly different risk of frequent lunch or dinner skipping in both sexes (Table 2).
Among these variables, we focused on the effect of household SES on meal-skipping rates, specifically the variation in meal skipping associated with SES. For this purpose, we conducted a time-series comparison of skipping rates of each meal within high, middle, and lower family affluence scale classes. The results from simply measuring the meal-skipping rate of five or more times per week according to family affluence scale each year showed that the low SES class had a significantly higher meal-skipping rate compared to the middle and high SES classes, for every meal and both sexes. The overall rates of skipping lunch and dinner were on the rise in both sexes, but between 2010 and 2012, particularly in the low family affluence scale class, the incremental slope showed definite slowing aspects and the rates themselves decreased. In the case of breakfast, on the other hand, the meal-skipping rates in all family affluence scale classes and both sexes were lower in 2012 compared to 2010 (Table 3). Although low-SES households had the highest meal-skipping rates under most conditions, such a reduction in the meal-skipping rate itself or a relatively large reduction in the incremental width in a group of low-SES household students can be interpreted as a positive sign of an improving gap in eating behavior by household income level.
When considering and adjusting other independent variables that may have an influence on meal skipping (i.e., grade level, BMI, average spending money per week, family members, perceived body image, effort towards changing bodyweight during the past month, and stress awareness), the odds ratios for skipping a meal five or more times per week showed a decrease in many cases over the study period. In some cases that showed increased odds ratios for meal skipping, the incremental change was much smaller in the low-SES class than in the high-SES class. The risk of breakfast skipping uniquely showed a gradual decrease in female students, regardless of SES status. In contrast, a relatively obvious increase in the risks of dinner skipping in male students of all SES classes, which reached doubling or even more, was observed. However, the increase in dinner skipping was less remarkable in female students. For the students in the low-SES class, only the boys revealed a discernable increasing risk of meal skipping for lunch (Table 4).

DISCUSSION

In this study, we measured the overall meal-skipping rates of Korean adolescents and differences between them associated with various factors in boys and girls, with a specific focus on household SES classes. Our observations included the changing patterns in meal-skipping through 2010 and 2012, compared to 2008. For some variables, the pattern of skipping breakfast demonstrated different features from that of lunch and dinner. In most cases, for both sexes, we could infer that breakfast skipping was a voluntary choice, as our results showed that the group of students who perceived themselves as thin had a lower breakfast-skipping rate than those who perceived themselves as fat, and the groups of students trying to lose weight showed the highest meal-skipping rate. On the other hand, the pattern of skipping lunch and dinner showed the opposite characteristics, meaning that students who perceived themselves as thin or had low bodyweight recorded an even higher skipping rate, and girls trying to gain weight recorded the highest meal-skipping rate. This has two possible implications: One is that unintended lunch- or dinner-skipping itself functions as an independent variable and influences each individual's body type or BMI, rather than being an intended action resulting from one's perceived body type or effort to change bodyweight; the other is that there could be a tendency in the low bodyweight group and the group trying to gain weight to skip lunch or dinner and substitute other snacks in place of their regular meals. Meanwhile, for boys, when classified by the effort to change bodyweight, all groups showed the tendency to skip meals in accord with their weight changing goals for every meal. This suggests that, for the male students, lunch- and dinner skipping were also intentional and voluntary like breakfast skipping.
We found that, the greater the allowance the students received, the higher the rate of meal skipping and the risk of meal skipping. Particularly in the group of students with the greatest average spending money per week, which exceeded 60 000 Korean wons (about 50 US dollars), the odds ratios for skipping meals five times or more per week markedly increased in both sexes. This could be inferred to be related to the relatively large allowance, perhaps in dual-income families, but further research would be needed to confirm this. Regarding stress awareness, girls reported gradually increasing meal-skipping rates for all three meals as their stress from daily life increased. However, for the boys, only the breakfast-skipping rate increased as their stress increased; lunch- and dinner-skipping rates were lower in boys who experienced a bit of stress than those who felt almost none.
Students who belonged to the low household SES class had higher meal-skipping rates and higher odds ratios for frequent meal skipping compared to those of the other SES classes, which was particularly clear in the male students, whose differential between the low class and other classes was about 1.8 times. This indicates an apparent gap in regular eating behavior between the low household income level group and other groups. In addition, the students who reported not living with both parents had the highest skipping rates in both sexes for every meal, and it was more remarkable in breakfast skipping than lunch- or dinner skipping. Therefore, we propose that government-led programs be implemented to support these disadvantaged groups. For example, establishment of a public free meal service center, or implementation of programs such as food stamps or the School Breakfast Program, which have been implemented in the US, as well as programs for educating and encouraging healthy eating habits for disadvantaged adolescents, would be advantageous [30,31].
Our results show that, though the overall meal-skipping rates and odds ratios for frequent meal skipping for lunch and dinner increased over time in both sexes, the incremental rates decreased between 2010 and 2012, particularly in the low-income group, suggesting that the existing difference in basic health behaviors or eating patterns between low SES households and others, whether it is a problem of accessibility or a problem of personal inclination, had a tendency to decrease over time. This data constitutes an improvement from a public health perspective and may be partially associated with the implementation of free lunch services in schools, which was implemented and has been extended to nearly all Korean middle and high schools throughout 2011 and 2012. On the other hand, unlike the decrease in the SES gap in lunch- and dinner-skipping, there is almost no change in the differences in breakfast-skipping rates between the low SES class and others from 2008 to 2012. The meal-skipping rates and odds ratios for breakfast generally decreased a little but did not change much between 2010 and 2012 in both sexes, and all SES classes. Nevertheless, ongoing high or increasing meal-skipping rates for any of the three meals in adolescents should be considered a significant problem.
In addition to household SES, whether or not the student lived with both parents also appears to be an important factor in adolescent eating habits and meal-skipping rates. Further consideration regarding an interaction between these two factors, as well as the high meal-skipping rates of those in the low bodyweight and high-stress groups, followed by appropriate management, may be fruitful.
Some studies in Korea have investigated the various factors that are associated with adolescents' meal skipping, such as accurate knowledge of nutrition, type of residential district, parental occupation, and health behaviors like cigarette smoking, along with the variables included in the present study such as household SES [13,15,17,18,19,23]. Several other studies have emphasized the factors associated with social support like care from the family, rather than household statuses [14,15,32]. Those factors turned out to be an important variable in our research as well; as represented by "family members" variable, which meant 'single- or two-parent household' in our study, it could be one explanation of why students were more likely to skip breakfast than any other meal, in that breakfast is usually and mainly served at home. Moreover, the influences of all these factors taken together may explain meal-skipping rates which remain high in spite of the expansion of free school meal services.
To conclude, the reasons behind continuously high and increasing meal-skipping rates should be analyzed, considering all the above mentioned factors, followed by appropriate interventions. Special consideration should be taken for high-risk groups such as students who are attempting to lose weight or belong to the low bodyweight group, children of single-parent or low-income families, socially alienated or neglected students, and so on, who have been identified through previous research or our study. A range of interventions should be attempted, such as providing education about nutrition, managing stress, helping students to establish a healthy self-image, and managing or encouraging schools and students to engage in a school food service program that aims to ingrain healthy eating habits into adolescents [14,15,22,23,24]. In addition, we should consider a range of approaches for improving our current school food service and its implementation. First, we could focus on several gaps in our program. This survey was conducted when school was in session, but more students will be at risk of frequent meal skipping during holidays when regular school lunch service is not provided. The Korean government has made an effort to cope with this problem, by running a boxed lunch delivery program with various organizations and institutions during vacations. Like this, other supplementary services can be provided to fill the gaps in meal services of existing programs. Second, satisfaction with the existing services, on the basis of frequent surveys or interviews, could be improved, using a number of different policies and interventions [33,34]. Efforts to expand school meal programs and providing better meals at school while improving hygiene and service, combined with expertise and programs that incorporate the social welfare system, would result in achieving greater efficacy of school meal programs [33,34,35,36].
In this study, we used data from 2008, 2010, and 2012 to observe the change in Korean student meal-skipping rates over time. In analyses of subgroups divided by household SES, we identified the differences in meal-skipping rates by SES. However, this was a cross-sectional study, making it difficult to establish definite causality between dependent and independent variables. Furthermore, there are calculations other than those used in this study that could have been used to determine the meal-skipping rate, which could have resulted in different findings than those reported here. Among several indices for SES of household, we used the FAS score, which is known for strong associations with health outcomes that are related to family culture and behavior, but less so for some behaviors where peer norms are a potentially powerful influence. The FAS also has its own limitations, in that it does not reflect changing patterns in family consumption and lifestyle today [26,27]. The results from using different socioeconomic indicators other than the FAS, such as father's education level or student's perceived SES, did not reveal remarkable differences from the results of this study. Additionally, with regard to policy implementation, it is worth noting that there is a difference between an interpretation based on statistical significance, which is subject to a large sample size, and the meaning of results based on effect size, which should be considered in policy-making processes.
In conclusion, in the context of increased overall meal-skipping rates, household SES continues to play an important role in Korean adolescent eating habits related to appropriate meal intake. Although we found that the gap in eating behavior between households of differing SES showed a small reduction over time, there were also many other factors that perturb healthy eating habits, such as not living in a two-parent household or excessive stress. These factors should be further investigated and appropriate government-led interventions enacted.

CONFLICT OF INTEREST

The authors have no conflicts of interest with the material presented in this paper.

REFERENCES

1. Abudayya AH, Stigum H, Shi Z, Abed Y, Holmboe-Ottesen G. Sociodemographic correlates of food habits among school adolescents (12-15 year) in North Gaza Strip. BMC Public Health 2009;9: 185. 19527503.
crossref pmid pmc
2. Kant AK, Graubard BI. Secular trends in patterns of self-reported food consumption of adult Americans: NHANES 1971-1975 to NHANES 1999-2002. Am J Clin Nutr 2006;84(5):1215-1223 17093177.
pmid pmc
3. Kant AK, Graubard BI. 20-Year trends in dietary and meal behaviors were similar in U.S. children and adolescents of different race/ethnicity. J Nutr 2011;141(10):1880-1888 21865567.
crossref pmid pmc
4. Kant AK, Graubard BI. Family income and education were related with 30-year time trends in dietary and meal behaviors of American children and adolescents. J Nutr 2013;143(5):690-700 23514763.
crossref pmid pmc
5. Kemm JR. Eating patterns in childhood and adult health. Nutr Health 1987;4(4):205-215 3554044.
crossref pmid
6. Fahlman MM, McCaughtry N, Martin J, Shen B. Racial and socioeconomic disparities in nutrition behaviors: targeted interventions needed. J Nutr Educ Behav 2010;42(1):10-16 19910257.
crossref
7. Hulshof KF, Brussaard JH, Kruizinga AG, Telman J, Löwik MR. Socio-economic status, dietary intake and 10 y trends: the Dutch National Food Consumption Survey. Eur J Clin Nutr 2003;57(1):128-137 12548307.
crossref pmid
8. Hulshof KF, Lowik MR, Kok FJ, Wedel M, Brants HA, Hermus RJ, et al. Diet and other life-style factors in high and low socio-economic groups (Dutch Nutrition Surveillance System). Eur J Clin Nutr 1991;45(9):441-450 1959516.
pmid
9. Kant AK, Graubard BI. Secular trends in the association of socio-economic position with self-reported dietary attributes and biomarkers in the US population: National Health and Nutrition Examination Survey (NHANES) 1971-1975 to NHANES 1999-2002. Public Health Nutr 2007;10(2):158-167 17261225.
crossref pmid
10. Wang Y, Jahns L, Tussing-Humphreys L, Xie B, Rockett H, Liang H, et al. Dietary intake patterns of low-income urban african-american adolescents. J Am Diet Assoc 2010;110(9):1340-1345 20800126.
crossref pmid pmc
11. Munoz KA, Krebs-Smith SM, Ballard-Barbash R, Cleveland LE. Food intakes of US children and adolescents compared with recommendations. Pediatrics 1997;100(3 Pt 1):323-329 9282700.
crossref pmid
12. Kim HY, Frongillo EA, Han SS, Oh SY, Kim WK, Jang YA, et al. Academic performance of Korean children is associated with dietary behaviours and physical status. Asia Pac J Clin Nutr 2003;12(2):186-192 12810409.
pmid
13. Huang CJ, Hu HT, Fan YC, Liao YM, Tsai PS. Associations of breakfast skipping with obesity and health-related quality of life: evidence from a national survey in Taiwan. Int J Obes (Lond) 2010;34(4):720-725 20065977.
crossref pmid
14. Bae HO, Kim M, Hong SM. Meal skipping children in low-income families and community practice implications. Nutr Res Pract 2008;2(2):100-106 20126373.
crossref pmid pmc
15. June KJ, Kim JY, Park S, Lee JY. Breakfast skipping and related factors in children in poverty. J Korean Acad Community Health Nurs 2011;22(2):204-211 (Korean).
crossref
16. Kim HK, Yoon CH. Development of health promotion program for low income children in underserved area. Seoul: Korea Association of Health Promotion; 2007. p. 3-4 p. 111. p. 198-199 (Korean).

17. Chung SJ, Lee Y, Kwon S. Factors associated with breakfast skipping in elementary school children in Korea. Korean J Community Nutr 2004;9(1):3-11 (Korean).

18. Park HA, Kang JH, Kim KW, Cho YG, Hur YI, Kim OH. Breakfast skipping, related factors, and nutrients intake of 5th grade students. Korean J Fam Med 2011;32(1):11-20 (Korean).
crossref
19. Bang YM, Lee MS, Na BJ, Kim KY. A study on the related factors of skipping breakfast in elementary students. Korean J Health Educ Promot 2006;23(3):17-35 (Korean).

20. Ministry of Education, Science and Technology. Ministry of Health and Welfare. Korea Center for Disease Control and Prevention. Guidelines of using a primitive data of Korea Youth Risk Behavior Web-based Survey indices definition 2012. Seoul: Ministry of Education, Science and Technology; 2012. p. 107-108 p. 133 (Korean).

21. Choi IS, Ro HK. A comparison study on middle school students of Gwangju and Jeonnam in terms of dietary behavior, body perception and weight control concerns according to BMI. J Korean Soc Food Sci Nutr 2010;39(3):383-391 (Korean).
crossref
22. Kim YK, Shin WS. A comparison study on perception of body image and dietary habits of high school students between urban and rural areas. Korean J Community Nutr 2008;13(2):153-163 (Korean).

23. Kang HW, Lee SS. A study on weight control and dietary life among middle school students: nutrition knowledge, dietary behavior and food frequency. Korean J Nutr 2006;39(8):817-831 (Korean).

24. Eom HS, Jeong MJ, Kim SB. A study on nutrition knowledge, dietary attitude, food habit of middle school students in Chonbuk area. Korean J Community Nutr 2005;10(5):574-581 (Korean).

25. Korea Centers for Disease Control and Prevention. Korean Pediatric Society. 2007 KCDC growth charts for the Korea. Seoul: Korea Centers for Disease Control and Prevention; 2007. p. 1-60 (Korean).

26. Currie C, Molcho M, Boyce W, Holstein B, Torsheim T, Richter M. Researching health inequalities in adolescents: the development of the Health Behaviour in School-Aged Children (HBSC) family affluence scale. Soc Sci Med 2008;66(6):1429-1436 18179852.
crossref pmid
27. Boyce W, Torsheim T, Currie C, Zambon A. The family affluence scale as a measure of national wealth: validation of an adolescent self-report measure. Soc Indic Res 2006;78(3):473-487.
crossref
28. Currie C, NicGabhainn S, Godeau E, Roberts C, Smith R, Currie D, et al. Inequalities in young people's health: HBSC international report from the 2005/2006 Survey. Copenhagen: World Health Organization; 2008. p. 13-15.

29. Currie CE, Elton RA, Todd J, Platt S. Indicators of socioeconomic status for adolescents: the WHO Health Behaviour in School-aged Children Survey. Health Educ Res 1997;12(3):385-397 10174221.
crossref pmid
30. Hanson KL, Olson CM. School meals participation and weekday dietary quality were associated after controlling for weekend eating among school children aged 6 to 17 years. J Nutr 2013;143(5):714-721 23486981.
crossref pmid
31. Bhattacharya J, Currie J, Haider SJ. Breakfast of champions? The school breakfast program and the nutrition of children and families. J Hum Resour 2006;41(3):445-466.
crossref
32. Hur MH. The public school lunch program and its contribution to the alleviation of problem behaviors. Int J Bus Soc Sci 2012;3(18S):142-150.

33. Chung KH. Lunch program for needy children: current status and policy recommendations. Seoul: Korea Institute for Health and Social Affairs; 2005. p. 5-14 (Korean).

34. Yoon JH. Opinions on food service program for low-income children when schools are not in session. Seoul: Korea Institute for Health and Social Affairs; 2005. p. 37-48 (Korean).

35. Lee KA, Lyu ES. Comparison of student satisfaction with foodservice of middle school by place for meal service in Busan area. J Korean Soc Food Sci Nutr 2005;34(2):209-218 (Korean).
crossref
36. Song HJ, Moon HK. Comparing school lunch program served at dining room with program at classroom for sanitation and contentment at one middle school. Korean J Community Nutr 2010;15(3):369-378 (Korean).

Table 1.
Overall distribution of study population and rates of skipping each meal
Male (n = 115 890)
Female (n = 106 772)
Total Breakfast-skip Lunch-skip Dinner-skip Total Breakfast-skip Lunch-skip Dinner-skip
Grade
 Middle school     1st year 19 776 4390 (21.7) 2623 (12.7) 2633 (12.6) 18 091 4171 (22.5) 1618 (8.6) 1795 (9.8)
2nd year 19 952 4649 (23.0) 2231 (10.7) 2304 (11.1) 17 987 4462 (24.5) 1383 (7.1) 1804 (9.6)
3rd year 20 242 5157 (25.3) 2003 (9.6) 2088 (10.1) 17 762 4556 (25.4) 1122 (6.0) 1743 (9.5)
 High school     1st year 19 509 4726 (24.6) 1541 (7.7) 1612 (8.0) 17 683 4635 (26.2) 1054 (5.5) 1568 (8.5)
2nd year 18 561 4927 (26.6) 1472 (7.7) 1588 (8.2) 18 253 4960 (27.6) 1071 (5.6) 1745 (9.6)
3rd year 17 850 5187 (29.8) 1415 (8.1) 1511 (8.5) 16 996 4644 (27.7) 1022 (5.8) 1505 (8.6)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 0.004
Year
 2008 39 278 9679 (25.1) 2290 (5.7) 2604 (6.4) 35 960 9595 (26.7) 1584 (3.9) 3071 (8.1)
 2010 38 391 9794 (25.5) 3950 (9.9) 4196 (10.4) 34 847 8940 (25.6) 2453 (6.7) 3530 (9.9)
 2012 38 221 9563 (24.9) 5045 (12.8) 4936 (12.6) 35 965 8893 (24.6) 3233 (8.8) 3559 (9.8)
p-value 0.57 <0.001 <0.001 0.009 <0.001 <0.001
BMI1
 Underweight 7362 2010 (27.3) 738 (9.7) 752 (9.8) 7184 2016 (28.0) 513 (6.6) 617 (8.5)
 Normal 89 330 21 954 (24.7) 8428 (9.1) 8681 (9.3) 84 921 21 322 (25.0) 5452 (6.0) 7871 (9.0)
 Overweight or obesity 15 533 3869 (24.8) 1476 (9.1) 1637 (10.1) 11 411 3051 (26.7) 830 (6.9) 1117 (9.5)
p-value 0.001 0.32 0.03 <0.001 0.005 0.13
Family affluence scale
 High 38 197 8169 (21.3) 3548 (8.9) 3650 (9.1) 33 674 7400 (22.1) 2187 (6.3) 2874 (8.5)
 Middle 63 049 15 902 (25.4) 5739 (8.8) 6011 (9.2) 58 924 15 219 (25.8) 3805 (6.0) 5492 (9.0)
 Low 14 644 4965 (34.7) 1998 (13.4) 2075 (13.9) 14 174 4809 (34.3) 1278 (8.5) 1794 (12.4)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Average spending money per week (KRW)
 0-10 000 38 841 8504 (21.6) 3686 (9.1) 3717 (9.1) 36 558 7997 (21.6) 2319 (6.0) 3032 (8.1)
 10 000-20 000 29 762 7281 (25.0) 2698 (8.7) 2841 (9.2) 27 772 7082 (25.2) 1858 (6.2) 2525 (8.7)
 20 000-30 000 15 741 4125 (26.4) 1530 (9.5) 1585 (9.7) 14 905 3985 (26.8) 981 (6.2) 1444 (9.6)
 30 000-40 000 10 945 2938 (27.0) 1059 (9.6) 1123 (10.0) 10 409 2932 (28.2) 711 (6.2) 1047 (9.7)
 40 000-60 000 10 971 3018 (27.2) 1083 (8.9) 1135 (9.4) 10 476 3106 (29.7) 735 (6.8) 1148 (10.7)
 60 000 or more 9630 3170 (33.0) 1229 (12.6) 1335 (13.7) 6652 2326 (35.9) 666 (9.7) 964 (14.3)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Family members
 Living with both parents2 95 188 21 737 (22.9) 8659 (8.8) 8892 (9.0) 88 021 20 583 (23.4) 5587 (6.0) 7670 (8.5)
 Do not live with at least one parent 20 702 7299 (36.2) 2626 (12.5) 2844 (13.4) 18 751 6845 (36.9) 1683 (8.6) 2490 (13.5)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Perceived body image
 A little or very lean 40 571 10 029 (24.8) 3832 (9.2) 3952 (9.3) 22 847 5753 (25.0) 1543 (6.4) 1941 (8.4)
 A little or very fat 38 853 9824 (25.4) 3699 (9.2) 3928 (9.7) 45 363 11 869 (26.1) 3099 (6.4) 4471 (9.7)
 Fair 36 466 9183 (25.4) 3754 (9.9) 3856 (10.3) 38 562 9806 (25.5) 2628 (6.4) 3748 (9.3)
p-value 0.16 0.001 0.001 0.04 0.99 <0.001
Effort for changing BW during the last 1 mo
 Effort for gaining weight 12 461 3001 (23.9) 1166 (9.1) 1184 (9.2) 1810 480 (27.3) 219 (11.9) 252 (14.5)
 Effort for losing weight 26 591 7050 (26.8) 2950 (10.8) 3308 (12.0) 45 143 12 187 (26.9) 3372 (7.0) 5374 (11.5)
 No effort 56 768 13 936 (24.6) 6068 (10.3) 6042 (10.2) 44 743 10 870 (24.3) 3087 (6.6) 3615 (8.0)
p-value <0.001 0.002 <0.001 <0.001 <0.001 <0.001
Stress awareness
 A lot of stress 42 747 12 197 (28.7) 4482 (10.3) 4810 (11.0) 53 788 15 786 (29.4) 4016 (7.1) 5851 (10.7)
 A little bit of stress 50 252 11 755 (23.4) 4520 (8.6) 4627 (8.8) 40 517 9164 (22.6) 2473 (5.8) 3322 (8.0)
 Almost no stress 22 891 5084 (22.3) 2283 (9.6) 2299 (9.5) 12 467 2478 (19.4) 781 (5.7) 987 (7.5)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Total 115 890 29 036 (25.2) 11 285 (9.4) 11 736 (9.7) 106 772 27 428 (25.6) 7270 (6.4) 10 160 (9.3)

BMI, body mass index; KRW, Korean won; BW, body weight.

Values are presented as the number of individuals (percent of the total).

1 Underweight defined as one’s BMI <5th percentile of same age group’s distribution; Overweight as 85th percentile≤BMI <95th percentile of same age group’s distribution; Obesity defined as BMI ≥95th percentile of same age group’s distribution.

2 Including stepfather and stepmother.

Table 2.
Odds ratios for skipping each meal 5 or more times per week
Male
Female
Breakfast Lunch Dinner Breakfast Lunch Dinner
Grade
 Middle school     1st year 1.00 1.00 1.00 1.00 1.00 1.00
2nd year 1.05 (0.98, 1.12) 0.79 (0.73, 0.86) 0.83 (0.77, 0.90) 1.07 (1.00, 1.16) 0.76 (0.68, 0.86) 0.91 (0.82, 1.01)
3rd year 1.13 (1.06, 1.20) 0.67 (0.61, 0.73) 0.72 (0.66, 0.79) 1.04 (0.97, 1.12) 0.62 (0.55, 0.70) 0.86 (0.78, 0.94)
 High school     1st year 0.99 (0.91, 1.07) 0.49 (0.44, 0.54) 0.50 (0.46, 0.56) 1.01 (0.93, 1.09) 0.53 (0.47, 0.61) 0.73 (0.66, 0.82)
2nd year 1.07 (0.99, 1.16) 0.45 (0.41, 0.50) 0.48 (0.44, 0.53) 1.02 (0.94, 1.10) 0.51 (0.45, 0.59) 0.79 (0.71, 0.88)
3rd year 1.15 (1.07, 1.25) 0.45 (0.40, 0.49) 0.47 (0.42, 0.52) 0.98 (0.90, 1.06) 0.51 (0.44, 0.59) 0.72 (0.64, 0.81)
p-value <0.001 <0.001 <0.001 0.27 <0.001 <0.001
Year
 2008 1.00 1.00 1.00 1.00 1.00 1.00
 2010 1.10 (1.03, 1.19) 1.91 (1.72, 2.14) 1.77 (1.59, 1.97) 0.95 (0.89, 1.03) 1.73 (1.51, 1.99) 1.13 (1.02, 1.25)
 2012 1.09 (1.02, 1.17) 2.73 (2.47, 3.03) 2.33 (2.11, 2.57) 0.92 (0.86, 0.99) 2.50 (2.21, 2.84) 1.17 (1.07, 1.28)
p-value 0.02 <0.001 <0.001 0.06 <0.001 0.004
BMI1
 Underweight 1.23 (1.11, 1.37) 1.08 (0.93, 1.25) 1.01 (0.87, 1.17) 1.11 (0.99, 1.24) 1.06 (0.88, 1.27) 1.11 (0.95, 1.31)
 Normal 1.05 (0.99, 1.12) 1.00 (0.91, 1.10) 0.93 (0.85, 1.02) 0.96 (0.90, 1.02) 0.94 (0.85, 1.05) 1.06 (0.97, 1.17)
 Overweight or obesity 1.00 1.00 1.00 1.00 1.00 1.00
p-value 0.001 0.42 0.11 0.001 0.14 0.36
Family affluence scale
 High 1.00 1.00 1.00 1.00 1.00 1.00
 Middle 1.25 (1.20, 1.31) 1.15 (1.09, 1.23) 1.17 (1.10, 1.24) 1.19 (1.14, 1.24) 1.06 (0.98, 1.14) 1.13 (1.06, 1.20)
 Low 1.78 (1.68, 1.89) 1.75 (1.61, 1.90) 1.73 (1.59, 1.89) 1.64 (1.53, 1.74) 1.41 (1.26, 1.59) 1.43 (1.30, 1.58)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Average spending money per week (KRW)
 0-10 000 1.00 1.00 1.00 1.00 1.00 1.00
 10 000-20 000 1.22 (1.16, 1.28) 1.09 (1.02, 1.18) 1.16 (1.09, 1.25) 1.22 (1.17, 1.28) 1.13 (1.04, 1.23) 1.11 (1.03, 1.19)
 20 000-30 000 1.35 (1.28, 1.44) 1.25 (1.14, 1.37) 1.27 (1.16, 1.40) 1.33 (1.25, 1.41) 1.19 (1.07, 1.32) 1.27 (1.17, 1.39)
 30 000-40 000 1.31 (1.22, 1.40) 1.30 (1.17, 1.44) 1.35 (1.22, 1.48) 1.48 (1.39, 1.58) 1.23 (1.09, 1.37) 1.30 (1.18, 1.43)
 40 000-60 000 1.34 (1.24, 1.44) 1.24 (1.12, 1.37) 1.34 (1.21, 1.48) 1.51 (1.42, 1.61) 1.36 (1.20, 1.53) 1.38 (1.25, 1.53)
 60 000 or more 1.78 (1.65, 1.91) 1.82 (1.64, 2.02) 1.93 (1.74, 2.14) 2.00 (1.85, 2.16) 1.76 (1.55, 1.99) 1.80 (1.60, 2.02)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Family members
 Living with both parents2 1.00 1.00 1.00 1.00 1.00 1.00
 Do not live with at least one parent 1.56 (1.48, 1.65) 1.16 (1.07, 1.26) 1.24 (1.15, 1.35) 1.57 (1.49, 1.65) 1.23 (1.12, 1.34) 1.33 (1.23, 1.43)
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Perceived body image
 A little or very lean 0.93 (0.89, 0.98) 0.90 (0.84, 0.97) 0.90 (0.83, 0.96) 0.95 (0.89, 1.01) 0.90 (0.81, 0.99) 0.92 (0.84, 1.00)
 A little or very fat 0.98 (0.93, 1.03) 0.83 (0.77, 0.89) 0.79 (0.73, 0.86) 0.97 (0.92, 1.01) 0.89 (0.82, 0.96) 0.92 (0.86, 0.99)
 Fair 1.00 1.00 1.00 1.00 1.00 1.00
p-value 0.03 <0.001 <0.001 <0.001 <0.001 <0.001
Effort for changing BW during the last 1 mo
 Effort for gaining weight 0.92 (0.87, 0.98) 0.99 (0.90, 1.09) 0.98 (0.89, 1.08) 1.01 (0.86, 1.19) 1.61 (1.29, 2.01) 1.57 (1.27, 1.94)
 Effort for losing weight 1.13 (1.08, 1.18) 1.25 (1.17, 1.34) 1.40 (1.31, 1.50) 1.10 (1.06, 1.14) 1.17 (1.10, 1.26) 1.54 (1.44, 1.65)
 No effort 1.00 1.00 1.00 1.00 1.00 1.00
p-value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
Stress awareness
 A lot of stress 1.29 (1.22, 1.36) 1.11 (1.04, 1.20) 1.17 (1.09, 1.26) 1.62 (1.52, 1.73) 1.32 (1.18, 1.46) 1.42 (1.30, 1.56)
 A little bit of stress 1.03 (0.98, 1.08) 0.95 (0.89, 1.02) 0.96 (0.90, 1.03) 1.19 (1.11, 1.28) 1.10 (0.99, 1.24) 1.10 (1.00, 1.21)
 Almost no stress 1.00 1.00 1.00 1.00 1.00 1.00
p for trend <0.001 <0.001 <0.001 <0.001 <0.001 <0.001

Values are presented as odds ratio (95% confidence interval).

BMI, body mass index; KRW, Korean won; BW, body weight.

1 Underweight defined as one’s BMI <5th percentile of same age group’s distribution; Overweight as 85th percentile≤ BMI <95th percentile of same age group’s distribution; Obesity defined as BMI ≥95th percentile of same age group’s distribution.

2 Including stepfather and stepmother.

Table 3.
Skipping rates of each meal according to family affluence scale, varying by year
Male
Female
Total Breakfast-skip Lunch-skip Dinner-skip Total Breakfast-skip Lunch-skip Dinner-ski
High 2008 11 507 2419 (21.2) 609 (5.1) 706 (5.7) 10 017 2270 (22.7) 346 (3.1) 666 (6.2)
2010 13 313 2913 (21.6) 1240 (9.0) 1309 (9.4) 11 480 2554 (22.6) 751 (6.5) 1041 (9.4)
2012 13 377 2837 (20.9) 1699 (12.3) 1635 (12.0) 12 177 2576 (21.0) 1090 (9.0) 1167 (9.7)
p-value 0.61 <0.001 <0.001 0.04 <0.001 <0.001
Middle 2008 22 339 5473 (25.0) 1171 (5.2) 1344 (5.9) 20 458 5453 (26.7) 869 (3.8) 1738 (8.1)
2010 20 426 5240 (25.9) 1987 (9.3) 2133 (9.9) 19 074 4909 (25.5) 1263 (6.1) 1894 (9.5)
2012 20 284 5189 (25.5) 2581 (12.3) 2534 (12.1) 19 392 4857 (25.2) 1673 (8.4) 1860 (9.5)
p-value 0.52 <0.001 <0.001 0.10 <0.001 0.001
Low 2008 5432 1787 (33.9) 510 (8.9) 554 (9.7) 5485 1872 (34.7) 369 (6.3) 667 (12.2)
2010 4652 1641 (36.1) 723 (15.5) 754 (16.1) 4293 1477 (35.3) 439 (9.8) 595 (13.7)
2012 4560 1537 (34.2) 765 (16.7) 767 (16.6) 4396 1460 (32.8) 470 (9.9) 532 (11.4)
p-value 0.18 <0.001 <0.001 0.22 <0.001 0.04
Total 2008 39 278 9679 (25.1) 2290 (5.7) 2604 (6.4) 35 960 9595 (26.7) 1584 (3.9) 3071 (8.1)
2010 38 391 9794 (25.5) 3950 (9.9) 4196 (10.4) 34 847 8940 (25.6) 2453 (6.7) 3530 (9.9)
2012 38 221 9563 (24.9) 5045 (12.8) 4936 (12.6) 35 965 8893 (24.6) 3233 (8.8) 3559 (9.8)

Values are presented as the number of individuals (percent of the total).

Table 4.
Odds ratios for skipping each meal 5 or more times per week according to family affluence scale, varying by year1
Male
Female
Breakfast Lunch Dinner Breakfast Lunch Dinner
High 2008 1.00 1.00 1.00 1.00 1.00 1.00
2010 1.11 (0.99, 1.23) 2.01 (1.70, 2.38) 1.78 (1.52, 2.09) 0.99 (0.89, 1.11) 2.12 (1.70, 2.65) 1.42 (1.21, 1.66)
2012 1.08 (0.97, 1.20) 3.02 (2.57, 3.55) 2.50 (2.15, 2.90) 0.94 (0.84, 1.04) 3.30 (2.69, 4.06) 1.57 (1.36, 1.82)
p for trend 0.21 <0.001 <0.001 0.28 <0.001 <0.001
Middle 2008 1.00 1.00 1.00 1.00 1.00 1.00
2010 1.07 (0.98, 1.18) 1.78 (1.56, 2.03) 1.69 (1.48, 1.92) 0.93 (0.85, 1.01) 1.57 (1.33, 1.84) 1.01 (0.90, 1.14)
2012 1.07 (0.98, 1.16) 2.51 (2.22, 2.84) 2.18 (1.93, 2.46) 0.92 (0.85, 1.00) 2.27 (1.95, 2.64) 1.03 (0.92, 1.15)
p for trend 0.25 <0.001 <0.001 0.12 <0.001 0.84
Low 2008 1.00 1.00 1.00 1.00 1.00 1.00
2010 1.14 (0.99, 1.31) 1.77 (1.43, 2.19) 1.67 (1.35, 2.06) 1.00 (0.86, 1.15) 1.62 (1.26, 2.09) 1.14 (0.94, 1.38)
2012 1.06 (0.93, 1.21) 2.10 (1.70, 2.58) 1.87 (1.52, 2.29) 0.90 (0.79, 1.03) 1.69 (1.34, 2.12) 0.93 (0.78, 1.11)
p for trend 0.18 <0.001 <0.001 0.19 <0.001 0.06

Values are presented as odds ratio (95% confidence interval).

1 Adjusted for grade level, body mass index, average spending money per week, household members, perceived self-body image, effort towards changing body weight during the last month, and amount of stress.

Editorial Office
Graduate School of Public Health, Seoul National University
1 Gwanak-ro, Gwanak-gu, Seoul 08826, Korea
Tel : +82-2-740-8328   Fax : +82-2-764-8328   E-mail: jpmphe@gmail.com
About |  Browse Articles |  Current Issue |  For Authors and Reviewers
Copyright © 2019 by Korean Society for Preventive Medicine. All rights reserved.                 powerd by m2community