This was a cross-sectional study using 2005–2006 data from the community-based Korean Genome and Epidemiology Study (KoGES, 2005–2006). The KoGES is a cohort study of Korean men and women living in the Ansan (urban) and Anseong (rural) regions of South Korea [18]. Both the semi-quantitative food frequency questionnaire (SQFFQ) and psychosocial well-being index-short form (PWI-SF) data were exclusively available in the second follow-up survey conducted in 2005–2006. The analysis involved 1,656 adults (889 men, 767 women) born between 1955 and 1963. Subjects with a daily energy intake of less than 500 kcal or exceeding 5,000 kcal and those who did not respond to the SQFFQ and PWI-SF questionnaires were excluded (Fig. 1). Approval for this study was obtained from the Institutional Review Board (IRB) of Sungshin Women’s University (IRB No. SSWUIRB-2021-017).

This study utilized a SQFFQ, which comprised 106 items and underwent a revision in 2004. The revision was based on the SQFFQ developed and validated in 2001 by the Korea Disease Control and Prevention Agency (KDCA) [1920]. The SQFFQ was employed to evaluate the participants’ usual dietary intake. The questionnaire featured nine levels of food frequency (rarely, once a month, 2–3 times a month, 1–2 times a week, 3–4 times a week, 5–6 times a week, once a day, twice a day, 3 times a day) and three levels of serving size (less than reference, reference, and more than reference) over the past year. The serving sizes were determined based on the actual consumption amounts reported by Korean adults aged 40–69 yrs, derived from the 24-h recall data of the Korea National Health and Nutrition Examination Survey (KNHANES).

To assess adherence to the Mediterranean diet, we utilized the Mediterranean diet tool based on the Mediterranean diet score (MDS) developed by Panagiotakos et al. [21] and tailored for the community cohort data [22]. The MTDS comprises 11 components: Whole grains, potatoes, fruits, vegetables, legumes, fish, red meat, poultry, dairy products, alcohol, and olive oil. Table 1 details the MTDS components and scoring criteria for this study. Scores for the 11 food group items were aggregated to calculate the MTDS, ranging from 0 to 55, wherein higher scores reflect better adherence to the Mediterranean diet. The consumption of food groups was translated into weekly intake frequency. For the calculation of the MTDS, individuals are assigned points based on their consumption of whole grains, potatoes, fruits, vegetables, legumes, and fish: 0 points for no consumption, 1 point for 0–1 serving/week, 2 points for 1–2 servings/week, 3 points for 2–3 servings/week, 4 points for 3–4.5 servings/week, and 5 points for more than 4.5 servings per week. Conversely, scores were allocated on the inverse scale for the intake of red meat, poultry, and dairy products. Data on alcohol consumption and olive oil were assessed through a dietary habit questionnaire, not the SQFFQ. Daily alcohol intake (mL/day) was calculated by converting the average amount of ethanol consumed per day (g) according to the type and amount of alcoholic beverage consumed. In this study, unlike existing scoring methods that assign the lowest score when alcohol consumption is zero, the criteria were adjusted to evaluate low alcohol consumption including zero as higher scores: 0 points for more than 700 mL/day, 1 point for 600–700 mL/day, 2 points for 500–600 mL/day, 3 points for 400–500 mL/day, 4 points for 300–400 mL/day, and 5 points for less than 300 mL/day. This adjustment was derived based on findings from an earlier domestic study [23] which suggested that moderate alcohol consumption did not demonstrate a discernible benefit in lowering all-cause, cancer, and cardiovascular disease associated mortality. Furthermore, this adjustment is deemed reasonable as certain studies [24] consider regional variations in their calculation of the MDS and, as a result, occasionally exclude alcohol consumption as a criterion. Consumption of olive oil was assessed by asking about the type of oil used in four Korean cooking methods: ‘frying,’ ‘stir-frying,’ ‘Buchim’ (Korean pancakes), and ‘Muchim and Namul’ (Korean salads, cooked and seasoned vegetables). The use of olive oil in all four cooking methods was assigned 5 points. The use of olive oil received 4 points if it was used in three cooking methods, 3 points for two cooking methods, and 2 points for one cooking method. If no olive oil was used, a score of 0 was assigned. Kimchi and pickles were omitted from the vegetable group, while seaweed and mushrooms were included. Items containing juice were excluded from the fruit group.

To assess an individual’s psychosocial stress level, we utilized the PWI-SF, tailored for Korean conditions and proposed by Jang [25]. The PWI-SF facilitates a comparative study of stress and mental health across demographic characteristics. It aims to identify associations between risk factors associated with stress and disease and comprises 18 questions regarding physical and psychosocial conditions experienced or felt in recent weeks. Respondents rate the 18 items on a 4-point scale (0-1-2-3), and the cumulative score measures stress levels. The PWI-SF scores range from 0 to 54, with higher scores indicating elevated stress levels. A score of 27 or more categorizes an individual into the high-risk stress group, 9 to 26 into the potential stress group, and 8 or less into the healthy group.

Demographic characteristics, socioeconomic status, and lifestyle factors of participants, including age, obesity level, daily energy intake, monthly household income, education level, smoking experience, and physical activity, were systematically assessed. Obesity levels, determined by the body mass index (BMI, kg/m²), were classified into 4 groups: underweight (< 18.5 kg/m²), normal weight (18.5 to < 23 kg/m²), overweight (23 to < 25 kg/m²), and obesity (25 kg/m² or higher). Monthly household income was classified into 4 groups: < 1,000,000 won (Korean currency; 1 million Korean won = 765 USD), 1,000,000 to < 2,000,000 won, 2,000,000 to < 4,000,000 won, and 4,000,000 won or higher. Education levels were segmented into 4 groups: elementary school or less, middle school, high school, and college graduate or higher. Smoking status was categorized into 3 groups: never smoked, former smoker, and current smoker. Physical activity, assessed by the metabolic equivalent of tasks (METs) per day, was divided into three intensity levels: light, moderate, and vigorous.

Categorical and continuous data were summarized into frequencies, percentages, means, and standard deviations, respectively. The statistical differences between independent groups were analyzed using the χ² test and one-way analysis of variance, followed by Duncan’s post-hoc test. The distribution of intakes of food groups and nutrients in the MTDS was described with the interquartile range (IQR). The MTDS was divided into tertiles, and food group and nutrient intakes for each quartile were described in the form of the IQR. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by logistic regression to analyze the association between the MTDS (tertiles) and the prevalence of high-level psychosocial stress by gender. All data were processed, and statistical tests were analyzed using SAS 9.4 (SAS Institute, Cary, NC, USA).

The analysis of the general characteristics of the subjects based on the MTDS tertiles is presented in Table 2. The ranges of the MTDS tertile groups for men were T1 (20–33 points), T2 (34–37 points), and T3 (38–39 points), while those for women were T1 (20–33 points), T2 (34–37 points), and T3 (38–48 points). The PWI-SF score was significantly lower at T3 compared to T1 and T2 in men (P = 0.014), and it was significantly lower at T2 and T3 compared to T1 in women (P = 0.027). There was no statistically significant difference in the general characteristics of subjects between the tertile groups for both men and women. However, women in the T3 group exhibited a higher educational level compared to the other groups (P = 0.001).

The intake of food groups according to the MTDS tertiles is presented in Table 3. The results of the energy-adjusted logistic regression analysis revealed that the consumption of whole grains, potatoes, fruits, vegetables, legumes, and fish increased with higher MTDS, while the consumption of red meat and dairy products decreased in both men and women (P for trend < 0.05). Additionally, poultry and alcohol consumption decreased as MTDS decreased in men (P for trend < 0.05), whereas P for the trend was not obtained, given the absence of a difference in the median value of intake between the tertile groups in women.

A summary of the energy and nutrient intakes including carbohydrates, proteins, fats, fiber, calcium, phosphorus, iron, potassium, natrium, vitamin A, retinol, β-carotene, vitamin E, thiamin, riboflavin, niacin, vitamin B6, vitamin C, zinc, folate, ash, and cholesterol according to the tertiles of the MTDS is presented in Table 4. As MTDS increased, energy and nutrient intakes excluding protein, calcium, phosphorus and riboflavin increased in men (P for trend < 0.05) and energy and nutrient intakes excluding fat and retinol increased in women (P for trend < 0.05).

Table 5 presents the adjusted OR and 95% CIs for high psychosocial stress from the multiple logistic regression analyses by MTDS tertiles. A higher MTDS was significantly associated with a lower prevalence of high psychosocial stress and showed a declining trend (P for trend = 0.016 for men and 0.026 for women). Men in the highest MTDS tertile had a 41% lower OR of high psychosocial stress compared with those in the lowest tertile (OR, 0.59; 95% CI, 0.38–0.91). Similarly, women in the highest tertile of the MTDS had a 39% lower OR of high psychosocial stress compared with those in the lowest tertile (OR, 0.61; 95% CI, 0.40–0.95).