In recent years, growth of the commercial food industry in combination with increasing incomes has led to increased consumption of high-energy foods and increased daily energy intake, particularly in developing countries(Reference Kant and Graubard1,Reference Galisteo, Duarte and Zarzuelo2) . This shift in dietary habits has resulted in increased obesity and obesity-related diseases such as the metabolic syndrome (MS) and CVD(Reference Kang, Shim and Lee3). According to data from the 2003–2018 National Health and Nutrition Examination Survey conducted in the USA, the reported prevalence of the MS increased from 31·2 % (95 % CI 31·2, 33·2 %) in 2003–2006 to 38·1 % (95 % CI 34·5, 41·9 %) in 2015–2018(Reference Aguilar, Bhuket and Torres4). According to the Korean National Health and Nutrition Examination Survey (KNHANES) in 2014–2017, the prevalence of the MS among adults aged 20 years and older in Korea is 30·9 %(Reference Kim, Ju and Park5). The WHO estimates that the global prevalence of the MS is 300 million individuals in 2025(Reference Borch-Johnsen6).
The concept of the MS was introduced by Reaven in 1988(Reference Reaven7). Reaven described ‘syndrome X’ as insulin resistance clusters with glucose intolerance, dyslipidaemia and hypertension that was associated with increased association for CVD(Reference Reaven8). As the awareness and prevalence of these abnormalities increased, the condition was classified as the MS(Reference Kereiakes and Willerson9). The MS is defined as a cluster of clinical cardiovascular risk factors including dyslipidaemia, impaired glucose tolerance, hypertension, physical inactivity and abdominal obesity(Reference Grundy10,Reference Isomaa, Almgren and Tuomi11) . The MS involves at least three risk factors related to waist circumference, blood pressure and serum concentrations of glucose, TAG and HDL-cholesterol under fasting conditions(Reference Renaldi, Pramono and Sinorita12). The pathogenesis of the MS is complex and incompletely understood; however, obesity, a sedentary lifestyle, and dietary and genetic factors contribute to the MS(Reference Hu, Qiao and Tuomilehto13). The presence of the MS increases the likelihood for complications in patients with type 2 diabetes mellitus, CVD and other chronic conditions(Reference Costa, Canani and Lisboa14,Reference Kilpatrick, Rigby and Atkin15) .
The MS is associated with significant social and financial burden. The National Health Service in the UK has estimated the social cost of obesity to be £5·1 billion annually(Reference Datta16). In addition, working-age individuals with the MS have significantly higher medical costs compared to those without the MS ($626 in additional expenses per member per month)(Reference Fitch, Pyenson and Iwasaki17). A primary objective in the management of the MS is to mitigate modifiable risk factors. Many previous studies have reported that lifestyle factors such as maintaining healthy body weight, exercise and healthy dietary intake may prevent the MS(Reference Anderson18–Reference Riccardi, Giacco and Rivellese24). Furthermore, dietary interventions have been shown to improve diet quality, body weight and nutrition-related biomarkers. Healthy dietary habits reduce the likelihood for the MS, and several studies have shown that dietary counsellor programmes among individuals with the MS significantly decrease the MS components(Reference Poppitt, Keogh and Prentice23,Reference Riccardi, Giacco and Rivellese24) . These findings highlight the importance of diet in metabolic health.
Despite this importance, many food choice decisions are dependent on momentary choice with limited consideration and information processing(Reference Bialkova, Grunert and Juhl25). Given these points, healthcare policy authors have faced public health policy when the purpose is to improve consumer’s healthier choice in order to reduce burden associated with nutrition-related chronic diseases(Reference Egnell, Crosetto and D’almeida26). As the consumption of processed foods has increased, the need for nutritional information has become an important public health issue(Reference Egnell, Crosetto and D’almeida26). In recent years, governments have promoted nutrition labels to increase public knowledge of nutritional information. Nutrition labelling defined any visual information on food ingredients on the package or container of food(Reference Gezmen-Karadağ and Türközü27).
In South Korea, nutrition labelling of foods was mandated in 1995, to help consumers make informed choices about their food purchases(Reference Park28). Since then, consumers have been able to review and consider the nutrition value of food distributions(Reference Park28,Reference Barreiro-Hurlé, Gracia and De-Magistris29) . Nutrition labelling also safeguards consumers from unreliable advertisement by providing exact nutrition information(Reference Kim, Kweon and Kim30). Nutrition labelling included information food size, energy, carbohydrate, protein, fat, sugar, trans-fatty acids, cholesterol and Na(Reference Oh and Kim31). Previous study showed the effects of nutrition labelling use intake in lower energy, fat, cholesterol and sugar than non-user(Reference Oh and Kim31). Nutrition labelling is intended to motivate consumers to choose more healthy food options and ultimately reduce association for the MS(Reference Barreiro-Hurlé, Gracia and De-Magistris29). Therefore, label-dependent consumption would be more effective in health self-management among relatively vulnerable participants than the general population(Reference Oh and Kim31,Reference Han, Kim and Kim32) . Despite increased MS prevalence and the expansion of nutrition labelling in South Korea, few studies have investigated their relationship. Numerous studies suggest a positive correlation between nutrition consumption and the MS; few have examined the effects of nutrition labelling and/or its risk factors. The introduction of nutrition labelling of foods in South Korea was expected to provide food-related health information to the South Korean population, the MS patients included, to help them better manage their health. The present study assessed the association between nutrition labelling awareness and the MS.
Experimental methods
Study population
The data were obtained from the KNHANES 2016–2018. The KNHANES are cross-sectional surveys that have been conducted annually since 1998 by the Korea Centers for Disease Control. The Korea Centers for Disease Control randomly selects households and invites them to complete the voluntarily survey. Stratified multistage probability sampling design was used with selection made from sampling units based on geographical area, sex and age groups from household registries.
A total of 24 269 participants completed the 2016–2018 KNHANES. To reduce the uncertainty associated with incomplete surveys, we excluded participants with incomplete data for the MS criteria including high blood pressure, elevated fasting blood glucose, hypertriacylglycerolaemia, low HDL-cholesterol and abdominal obesity. Participants who did not complete the nutrition awareness question were also excluded. The present study included individuals aged 20 years and older. After exclusions, 14 490 participants were included (Fig. 1).
Flow chart of study population. KNHANES, Korean National Health and Nutrition Examination Survey.
Venn diagrams illustrating the concept of nutrition labelling awareness.
Variables
The present study used the definition of the MS adopted by the National Cholesterol Education Program Adult Treatment Panel III and the Korean Journal of Obesity criteria(Reference Takahashi, Shimomura and Proks33,Reference Kim, Jeong and Kim34) . Blood pressure was measured using a mercury sphygmomanometer in a seated position after 10 min of rest. Two measurements were made for all participants at 5- min intervals. An average of the two measurements was used in the analyses. Waist circumference was measured at the midpoint between the bottom of the rib cage and the top of the lateral border of the iliac crest on full expiration. Blood samples were collected from participants in the morning after overnight fasting and analysed at a centralised national laboratory. Participants with three or more of the following criteria were classified as having the MS: high blood pressure (≥130/85 mmHg or using medication to control blood pressure), elevated fasting blood glucose (≥5·55 mmol/l), hypertriacylglycerolaemia (≥8·325 mmol/l), low HDL-cholesterol (≤2·22 mmol/l in men and ≤2·775 mmol/l in women) and abdominal obesity (waist circumference of ≥90 cm for men and ≥85 cm for women).
The main independent variable was the use of nutrition labelling. We defined based on response for three phases question in KNHANES(Reference Kim, Kweon and Kim30,Reference Han, Kim and Kim32,Reference Kim35) . First, ‘Do you know the nutrition labelling?’. Second, ‘Do you read the nutrition label when buying or choosing processed foods?’. Third, ‘Do you have the effect of your decision on buying food on nutrition labelling?’. These questions were answered with ‘yes or no’. If participants responded as ‘yes’ for question, they responded for the next phase. The aware + consider group responded ‘yes’ three times in a row. The aware + not consider group answered as ‘yes’ for first question, then they answered as the next phase either ‘yes’ for second and ‘no’ for third question or ‘no’ for second question. The not aware group categorised who responded ‘no’ for first question. Awareness of nutrition information was divided into three categories: the aware + consider group checked nutrition facts and made labelling-dependent purchase decisions, the aware + not consider group checked nutrition facts but did not make labelling-dependent purchase decisions or were aware of nutrition facts but did not check them when making food purchase decisions and the not aware group was unaware of nutrition labelling(Reference Yoo, Han and Chung36) (Figs. 2 and 3).
Flow chart of nutrition labelling.
Several demographic characteristics were collected. Participants were categorised by age (<40, 40–49, 50–59, ≥60 years), household income quartile, educational level (elementary school graduation, junior high school, high school graduation, above university graduation), marital status (married, unmarried, divorced, widowed), employment status (unemployed, white-collar, blue-collar), public health insurance coverage (national health insurance, medical aid), supplemental private health insurance (yes, no) and smoking status (non-smoker, ex-smoker, current smoker). High-risk drinking (yes, no) was defined as the consumption of more than seven (male) or five (female) alcoholic drinks on a single occasion at least twice per week. The cut-off for weekly aerobic exercise (yes, no) was more than 2 h 30 min of moderate physical activity, 1 h 15 min of vigorous physical activity or a combination of both for >2 h 30 min, where 1 min of moderate physical activity was equal to 1 min of total physical activity time and 1 min of vigorous physical activity was equal to 2 min. Nutritional characteristics included energy intake (calculated as the average number of kJ consumed per d) and macronutrient intake (carbohydrate, protein, fat; calculated as the average number of g consumed per d).
Statistical analyses
All statistical analyses were performed using SAS version 9.4. Categorical variables were reported as frequencies and percentages, and descriptive statistics were generated using χ 2 tests. Continuous variables are reported as means and standard deviations, and study groups were compared using t tests. Logistic regression analyses were used to determine the OR and 95 % CI of the MS associated with nutrition labelling awareness while controlling for potential covariates including sex, age, household income, educational level, marital status, public health insurance coverage, supplemental private health insurance, smoking status, high-risk drinking, aerobic exercise, energy intake and macronutrient intake. Subgroup analyses were conducted using multivariable logistic regression analyses to investigate the association between the MS and sex, age, education level, household income, marital status and employment status. In the subgroup analyses, the Cochran–Armitage test was used to assess the association between the MS and each variable and the awareness of nutrition labelling. P values < 0·05 were considered statistically significant.
Results
General characteristics of the study population
A total of 14 490 participants were included, of which 4001 (27·6 %) had the MS and 10 489 (72·4 %) did not. The demographic and health characteristics of the study population are summarised in Table 1. In total, 3815 (26·3 %) participants were classified as aware + consider, 7001 (48·3 %) were classified as aware + not consider and 3674 (25·4 %) were not aware of nutrition labelling. Statistically significant differences were observed between nutrition labelling awareness and having the MS (aware + consider 18·2 % among participants with v. 81·8 % among those without the MS, aware + not consider 25·8 v. 74·2 %, not aware 40·8 v. 59·2 %; all P < 0·0001). In addition, there were differences in sex, age, household income, educational level, marital status, employment status, public health insurance, supplemental private health insurance, smoking status, high-risk drinking, aerobic exercise, energy intake and macronutrient intake between participants with and without the MS.
General characteristics of the study population
(Numbers and percentages; mean values and standard deviations)
Univariable and multivariable logistic regression analyses of the relationship between awareness of nutrition labelling and the metabolic syndrome
Table 2 presents findings from univariate and multivariate logistic regression analyses of the relationships between nutrition information awareness and the MS. In univariable logistic regression analyses, the aware + not consider (OR 1·56, 95 % CI 1·42, 1·72) and not aware (OR 3·10, 95 % CI 2·79, 3·44) groups had increased likelihood for the MS compared with the aware + consider group. In covariate analyses, male and older participants had increased likelihood for the MS compared with female and young participants. In addition, higher association for the MS was lower among participants with higher education levels compared to those with less education. Participants with high education level, other marital status (unmarried, divorced, widowed) or white-collar jobs were at reduced likelihood for the MS compared with those with low education level, married or unemployed. The MS was higher among participants on medical aid or not with private health insurance compared to those with national health insurance or with supplemental private health insurance. Smokers, high-risk drinkers and those that do not exercise regularly had a higher prevalence of the MS. In addition, energy, protein and fat intake were inversely correlated with the MS. In multivariable logistic regression analyses, participants in the not aware group had increased likelihood for the MS (OR 1·23, 95 % CI 1·08, 1·39); however, no statistically significant difference was observed in the aware + not consider group. Female, younger age and high education level were associated with lower MS. The MS was lower among participants with blue-collar jobs compared with those who were unemployed. Participants on medical aid had higher rates of the MS compared with those on the national health insurance plan. Participants with higher-risk health behaviours including smoking, high-risk drinking or no aerobic exercise were at increased likelihood for the MS.
Univariable and multivariable logistic regression analyses of the relationship between nutrition labelling awareness and the metabolic syndrome
(Odds ratios and 95 % confidence intervals)
Subgroup analyses of nutrition labelling awareness and the metabolic syndrome stratified by sex, age, household income, married status and employment status
The results of subgroup analyses on nutrition labelling awareness and the MS stratified by sex, age, household income, married status and employment status are summarised in Table 3. Among female participants, those in the aware + not consider and not aware groups had a statistically significant higher association for the MS compared with those in the aware + consider group. Among participants aged 60 years and older, those in the aware + not consider (OR 1·48, 95 % CI 1·19, 1·83) and not aware (OR 1·76, 95 % CI 1·43, 2·17) groups had a statistically significant higher likelihood for the MS compared with those in the aware + consider group. Among participants in the fourth quartile for household income, likelihood for the MS was significantly higher in the aware + not consider (OR 1·34, 95 % CI 1·09, 1·65) and not aware groups (OR 1·77, 95 % CI 1·37, 2·29). Only married participants in the aware + not consider (OR 1·19, 95 % CI 1·05, 1·34) and not aware (OR 1·51, 95 % CI 1·31, 1·75) groups had significantly higher association for the MS compared with the aware + consider group. Participants in the not aware group had increased likelihood for the MS compared with the aware + consider groups regardless of employment status (unemployed, OR 1·70, 95 % CI 1·40, 2·06; white-collar job, OR 1·28, 95 % CI 1·01, 1·63; blue-collar job, OR 1·32, 95 % CI 1·03, 1·69), with the trend being statistically significant (P for trend < 0·0001).
Subgroup analyses of nutrition labelling awareness and the metabolic syndrome by sex, age, household income, marital status and employment status
(Odds ratios and 95 % confidence intervals)
Discussion
The present study identified an association between nutrition information awareness and the MS after adjusting for potential confounding among a nationally representative sample of adults aged 20 years and older in Korea.
Healthy behaviours are associated with a reduced incidence of the MS(Reference Dalle Grave, Calugi and Centis37,Reference Chen, Pang and Li38) . Previous studies have emphasised the importance of dietary behaviour in weight management and nutrition(Reference Sonnenberg, Pencina and Kimokoti39), with some suggesting that modifying dietary habits may be more effective than exercise in promoting metabolic health(Reference Barreiro-Hurlé, Gracia and De-Magistris29). Current nutritional guidelines encourage the intake of healthy foods including fruits and vegetables and recommend reducing the consumption of energy-dense foods including sugars, carbohydrates and processed foods(Reference Post, Mainous and Diaz40). The National Cholesterol Education Program Adult Treatment Panel III and the American Heart Association proposed dietary recommendations for preventing and treating the MS, which outlines the appropriate macronutrient distribution of carbohydrates, protein and fat and recommends limiting intake of trans-fat and refined sugar, and increasing consumption of fruits, vegetables and whole grains(Reference Feldeisen and Tucker41,Reference Crujeiras, Zulet and Abete42) . The WHO and Agriculture Organization of the United Nations have suggested that the increased consumption of energy-dense foods and fast foods has contributed to the increasing prevalence of obesity causing to the MS(43). Energy-dense foods and fast foods have been associated with increased energy intakes and poor diet quality(Reference Esmaillzadeh and Azadbakht44). Unhealthy diets may contribute to insulin resistance by their higher levels of saturated fats(Reference Vessby, Uusitupa and Hermansen45). Reducing imbalance diet might then become a viable therapeutic option not only for obesity but also for type 2 diabetes and the MS(46). Energy-dense and fast foods have been associated with increased energy intakes and poor diet quality(Reference Esmaillzadeh and Azadbakht44). Unhealthy diet may contribute to insulin resistance by their higher levels of saturated fats, which have been shown to be related to impaired insulin sensitivity(Reference Vessby, Uusitupa and Hermansen45,46) . Reducing imbalance diet might then become a viable therapeutic option not only for obesity but also for type 2 diabetes and the MS(46).
Because of these problems, public healthcare experts seek primarily to improve nutrition quality worldwide. In South Korea, nutrition labelling provides information on nutritional content including energy, carbohydrate, protein, fat, sugar and cholesterol since 1995(Reference Kwon, Yoon and Kim47). Research suggests that nutrition information on packaged food may influence food purchasing decisions and promote healthier dietary habits(Reference Han48). Previous studies have also looked into the association between the use of nutrition labels and the actual pattern of nutrient intake, some of which have shown that those who read nutrition labels tended to consume less energy and fats, and more fruits, vegetables(Reference Gezmen-Karadağ and Türközü27). We hypothesised that nutrition labelling-dependent purchasing decisions significantly affect the MS and explored likelihood between nutrition labelling and the MS.
The present study showed the MS was higher in participants who are not aware of nutrition labelling than who are taking nutrition labelling-dependent purchasing decisions. Consistent with the present study, nutrition labelling is positively associated with patient self-management of chronic diseases(Reference Han, Kim and Kim32,Reference Yoo, Han and Chung36) . Our subgroup analysis showed positive relationship between nutrition labelling and socio-demographic characteristics. The present study showed the MS in males is higher than in females, with 32·4 % male with the MS and 24·2 % female with the MS. The previous study also confirmed that men have a higher relation factor level compared with women. Females were more likely than males to comply with dietary recommendations and choose foods than align with dietary guidelines(Reference Campos, Doxey and Hammond49). In addition, females perceive nutrition as more important, use food labels more often and are more likely to be influenced by nutrition labelling in their food purchasing decisions compared with males(Reference Rimpeekool, Kirk and Yiengprugsawan50). In the present study, sex subgroup analyses showed analogous results with higher nutrition labelling awareness among females. The present study showed a promotion in the MS with advancing age, with significantly higher association of this condition among individuals when compared with younger groups(Reference Kim51). In fact, age is known to increase the MS since several age-related physiological changes facilitate the development of insulin resistance(Reference Kim51,Reference Oh, No and Kim52) . In particular, postmenopausal women are vulnerable to metabolic changes due to oestrogen deficiency that increase total cholesterol and lipoprotein level(Reference Tardivo, Nahas-Neto and Nahas53). Unhealthy diet contributes to increases of atherosclerosis and development of cardiovascular risk, especially at 50–69 years of age(46,Reference Tardivo, Nahas-Neto and Nahas53) . Ageing is accompanied by an increased need in several nutrients, such as vitamins and minerals, whereas the overall energy requirements decline(Reference Cano-Ibáñez, Bueno-Cavanillas and Martínez-González54). Therefore, nutrition labelling could be more crucial for elderly people due to hormone and metabolic changes(Reference Kim, Kweon and Kim30). However, regarding nutrition education experience, 94·8 % of older adults do not have an educational background in nutrition(Reference Han and Yang55). In addition, older adults have less awareness of the importance of the nutrients on food labels and therefore may be less likely to use nutrition labels in their food purchasing decisions(Reference Evenson and Lee56). For this reason, subgroup analysis showed that the aware + not consider and not aware groups have higher association MS than the aware + consider group in older adults. Therefore, more public health promotion of nutrition labelling should be provided for elderly populations(Reference Kim, Jeong and Kim34,Reference Yoo, Han and Chung36) . The present study identified high-income level was a significant protective effect against the MS. Based on the previous study, high-income earner paid more attention towards healthy choices(Reference Bernstein57,Reference Sidorenkov, Nilssen and Grjibovski58) . High-income households are more likely to consider the nutritional information, including energy, fat and cholesterol, on food labels than lower-income households(Reference Kim, Nayga and Capps59,Reference Lin, Lee and Yen60) . Moreover, high-income earner has more healthy lifestyle than low-income earner such as leisure time physical activity and balance diet(Reference Bernstein57). In the present study, education level influenced food label use among participants with the MS. Consumers with higher education may better understand the nutrition label information and the effects of nutrients on health, therefore informing their food purchasing decisions(Reference Kim, Nayga and Capps59). As a result, those with higher education may be more likely to adapt their dietary habits to promote metabolic health. These results should motivate healthcare professionals to consider the positive effects of that nutrition labelling-dependent purchasing decisions among specific population(Reference Han, Kim and Kim32,Reference Yoo, Han and Chung36) .
The present study had several limitations. First, the cross-sectional design of KNHANES precluded the ability to assess likelihood between nutrition labelling awareness and the MS. Second, the KNHANES are mostly self-reported surveys. Thus, questions on the socio-economic status, health behaviour, awareness of nutrition, and body size perception may be subject to recall bias. Third, the present study did not account for all health-related factors as part of the MS history. Despite these limitations, the study had several strengths. First, we identified several factors that contribute to the MS in relation to nutrition labelling. This provides an opportunity to develop and appropriate nutritional and educational interventions to manage the MS. Second, we used data from a nationally representative sample.
In conclusion, we found that nutrition labelling awareness is associated with the MS among a nationally representative sample of adults in Korea. Nutrition labelling-dependent purchasing decisions were associated positively with the MS. This finding could be used as evidence-based data for promoting the use of nutrition labelling to manage the MS in South Korea.
Acknowledgements
By submitting a written statement and data utilisation plan, the KNHANES data are openly available at https://knhanes.cdc.go.kr/knhanes/index.do.
This work was supported by the Soonchunhyang University Research Fund.
H. H. C. and M. I. H. conceptualised the research. H. H. C. and D. W. L. conducted the formal analysis. H. H. C. and D. W. L. investigated the data. D. W. L. designed the methodology. M. I. H. managed project administration and supervision. H. H. C. wrote the paper. All authors reviewed and revised the manuscript and approved the manuscript for publication.
There are no conflicts of interest.
