Study population. We used health check-up data of 15,381 Seoul National University students, including undergraduate, graduate, and doctoral students, from 2016 through 2018. All students were eligible to voluntarily have health check-up for free once a year, and the health screening included a self-reported questionnaire (health behavior), anthropometric measurements (height, weight, and waist circumference), and laboratory tests34. We excluded participants who (1) were under 18 years or over 40 years of age (n = 137), (2) declined to participate in the study (n = 638), (3) were taking medication for hypertension, diabetes mellitus, or dyslipidemia (n = 45), (4) were foreigners (n = 1,094), (5) were pregnant (n = 19), or (6) had incomplete information (n = 1,146). Ultimately, a total of 12,302 participants were included in this study. All individuals provided informed consent to participate in the study, which was approved by the Institutional Review Board of Seoul National University (IRB number, C-1304-062-481), and have been performed in accordance with the Declaration of Helsinki.
Anthropometric and laboratory measurements. Weight and height were measured with participants in light clothing on the day of health check-up. Body mass index (BMI) was calculated as weight (kg) divided by the square of the height (m2). Waist circumference was measured at the midpoint between the last rib and the top of the iliac crest. Blood pressure (BP) measurements were taken with the participant in a sitting position using an automatic BP measurement system after a rest period of at least 5 minutes. Blood samples were taken after at least 12 hours of fasting. All experiments were performed in accordance with relevant guidelines and regulations.
Definition of metabolic syndrome. The definition of metabolic syndrome was taken from the modified National Cholesterol Education Program (NCEP) Adult Treatment Panel III guidelines35. Subjects who met three or more of the following criteria were diagnosed with metabolic syndrome: (1) central obesity (waist circumference ≥ 90 cm for men or ≥ 85 cm for women); (2) hypertriglyceridemia with fasting plasma triglyceride levels ≥ 150 mg/dL; (3) decreased levels of high density lipoprotein cholesterol (HDL-C) with HDL-C levels < 40 mg/dL for men and < 50 mg/dL for women; (4) hypertension with systolic or diastolic BP ≥ 130/85 mmHg; and (5) hyperglycemia with fasting plasma glucose ≥ 100 mg/dL.
Assessment of breakfast eating, meal patterns and dietary quality. Using a self-administered questionnaire, participants were asked how often, on average, they consumed breakfast during the past year: “How often do you eat breakfast?” According to the answers, the participants were then categorized into three groups: having breakfast for 7 days per week (non-skipper), 4–6 days per week (skipper for 1–3 days), and 0–3 days per week (skipper for 4–7 days).
Using the question “How often do you overeat or binge?”, we divided participants into three groups based on frequency of binge eating per week: less than once per week (e.g., 2–3 times per month), 1–2 times per week, and ≥ 3 times per week.
To assess meal frequency per day, we categorized the participants into two groups, using the question “How many meals do you usually eat a day?”: regular (3 meals per day) and irregular (< 3 meals per day).
Overall meal pattern was defined by combination of three variables: breakfast frequency per week (> 3 days per week or ≤ 3 days per week), binge eating per week (< 3 times per week or ≥ 3 times per week), and meal frequency per day (3 meals per day or < 3 meals per day on average). An “unhealthy” meal pattern was defined if an individual had all the bad eating patterns; “fair” was defined if an individual had 1 or 2 bad eating patterns; and “healthy” was defined if a participant had no bad eating patterns. For example, someone had an unhealthy meal pattern if they had 0–3 days with breakfast per week, ≥ 3 times of binge eating per week, and < 3 meals per day.
We also evaluated dietary intake of various foods by asking consumption frequency of fruits, vegetables, milk and dairy products, high-fat meat, processed meat, and sugared beverages (< once a week, 1–4 times a week, ≥ 5 times a week).
Other variables. Alcohol consumption was classified into non-drinker, moderate drinker, and heavy drinker based on self-questionnaires. Moderate drinking was defined as consuming 14 or less standard drinks per week in men and 7 or less standard drinks per week in women, and heavy drinking was defined as more than these amounts (1 standard drink = 12 g of alcohol)36. Smoking status was categorized into never, ex-smoker, and current smoker based on self-questionnaires. Physical activity levels were evaluated using the International Physical Activity Questionnaire and total physical activity levels (metabolic equivalent [MET]-min/week) were categorized into low (< 600 MET-min/week), moderate (600–2999 MET-min/week), and high (≥ 3000 MET-min/week) physical activity37.
Statistical analysis. Data were expressed as the mean with standard deviations (SD) for continuous variables or as number with percentages for categorical variables. χ2 and the analysis of variance for categorical and continuous variables were used to compare general characteristics of the study population by breakfast consumption status. Logistic regression analyses were performed to estimate odds ratios (ORs) and 95% confidence intervals (CIs) of metabolic syndrome, with adjustment for various lifestyle and dietary factors. Model 1 was adjusted for age and sex, and model 2 was adjusted for alcohol consumption, smoking status, physical activity, and BMI in addition to the covariates of model 1. Model 3 was adjusted for dietary intake of fruits, vegetables, milk and dairy products, high-fat meat, processed meat, and sugared beverages in addition to the covariates of model 2. All statistical analyses were performed using Stata ver. 16.0 for Windows (Stata Corp., College Station, TX, USA). P < 0.05 was considered statistically significant.