The association between dietary pattern and depression in middle-aged Korean adults

Abstract

BACKGROUND/OBJECTIVES

The association between dietary patterns and depression has been reported but the results have been inconsistent. This study was conducted to investigate the association between dietary patterns and depression in middle-aged Korean adults.

SUBJECTS/METHODS

The participants were selected from a community-based cohort, a subset of the Korean Genome and Epidemiology Study. Depression was assessed using the Korean version of the Beck Depression Inventory (BDI) and those with a BDI score ≥ 16 were defined as having depression. The subjects' food intakes over the year preceding the survey were estimated by using a validated semi-quantitative food frequency questionnaire. Dietary patterns were identified by using factor analysis. Multiple logistic regression was used to assess the association of dietary pattern with depression.

RESULTS

Among 3,388 participants, 448 (13.2%) were identified as having depression. We identified two major dietary patterns: ‘Healthy’ dietary pattern was characterized by high intakes of vegetables, soybeans, mushroom, seaweeds, white fish, shellfish and fruits and a low intake of white rice. ‘Unhealthy’ dietary pattern was characterized by high intakes of white rice, meats, ramen, noodles, bread and coffee and a low intake of rice with other grains. Compared with subjects in the lowest quartiles, those in the highest quartiles of the healthy dietary pattern had a significantly lower odds ratio (OR = 0.59, 95% CI: 0.42–0.82, P for trend = 0.0037) after adjusting for potential confounders. In contrast, the unhealthy dietary pattern was negatively associated with depression (OR = 1.65, 95% CI: 1.19–2.28, P for trend = 0.0021).

CONCLUSION

This results suggest that a healthy dietary pattern (rich in vegetables, soybeans, mushroom, seaweeds, white fish, shellfish, and fruits) is associated with low risk of depression. Whereas an unhealthy dietary pattern (rich in white rice, meats, ramen, noodles, bread, and coffee) is associated with a high risk of depression in middle-aged Korean adults.

INTRODUCTION

Depression is a psychiatric disorder and the burden of that disease is greater than any other single disease [1]. The World Health Organization has predicted that 300 million people have depressive symptoms [2]. In the 2014 Korea National Health and Nutrition Examination Survey, 6.7% of adults aged > 19 years had depression, as defined as a score on the Patient Health Questionnaire (PHQ)-9 of 10 or higher [3]. Depression is associated with a variety of social problems and is closely related to suicide, thus appropriate prevention strategies are needed [4].

To alleviate this health issue, studies have proposed that several nutrients, such as omega-3 fatty acid, folate, vitamin B₆, and vitamin B₁₂, may help prevent depressive symptoms [5,6,7,8]. However, since the effect of an individual nutrient on a disease is limited, a dietary pattern analysis method that evaluates the effect of the entire diet has been proposed [9,10]. In a review paper, dietary patterns including the Mediterranean diet, the prudent diet pattern, and the pro-vegetarian diet pattern, were reported to reduce the risk of depression [11]. In an assessment of the individual dietary patterns and incidence of depression in middle-aged subjects, the Mediterranean diet was shown to be inversely associated with depression of middle-aged subjects in the PREDIMED randomized trial [12]. Also, the whole-food pattern (high intake of vegetables, fruits, and fish) was shown to be associated with decreased depression risk in middle-aged subjects [13]. On the other hand, dietary pattern and depression did not show any association in a large-scale cohort study of middle-aged US women [14]. To the best of our knowledge, there is only one study that conducted to examine the association between dietary pattern and depression in Korean adolescents [15]. Therefore, the objective of this study was to examine the association between dietary patterns and depression in Korean middle-aged adults.

SUBJECTS AND METHODS

Participants

The participants were selected from a community-based cohort of Korean Genome and Epidemiology Study; the design of that study has been described in detail previously [16]. In brief, Korean adults aged 40–69 years who lived in rural (Ansung) and urban (Ansan) areas were recruited from 2001 to 2002. Among the 10,030 participants, we selected 3,388 men and women who had completed the dietary questionnaire, and underwent screening for depression. A total of 6,642 participants were excluded for following reasons: did not attend first follow up examination (n = 2,515); no dietary data (n = 33); no depression screening data (n = 4,065); history of mental disorder (n = 7); outside of the recommended calorie cut-off range (< 500 and < 800 kcal or > 3,500 and > 4,000 kcal in women and men, respectively) (n = 22) [17]. Informed consent was obtained from all study participants. The data was provided National Biobank of Korea, the Centers for Disease Control and Prevention, Republic of Korea (4845-301, 4851-302, and -307). The study protocol was approved by the Gachon University Institutional Review Board (1044396-201604-HR-028-01).

Screening for depression

For depression screening, the Beck Depression Inventory (BDI) was administered. The presence of depression was determined based on the BDI score: normal subjects (BDI score range: 0–15) and depression subjects (BDI score range: 16–63). The validity of the BDI has been previously verified and a BDI score of 16 is the optimal cut-off score for Koreans [18].

Dietary assessment

The subjects' food intakes over the year preceding the survey were assessed by using a validated semi-quantitative food frequency questionnaire (SQFFQ) containing a list of 106 food items. The intake frequency of each food items was presented in nine categories: never or seldom, 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, and three or more times a day. The intake amount of each food items was classified into three categories: small, medium (1 serving), or large. The SQFFQ was previously validated by using 12-day diet record data of 124 subjects [19]. Nutrient intakes were calculated for each subject by using the seventh edition food composition tables provided by the Korean Nutrition Society [20].

Measurement of covariates

Information on various covariates such as general characteristics and lifestyle data were collected by using an interviewer-administered questionnaire. The following is the list of covariates: age, marital status (married or other), exercise (yes or no), alcohol drinking (no or ex/current drinker), smoking (no or ex/current smoker), educational level (graduated elementary school, middle school, high school, or college or higher degree), family history of mental disorder (yes or no), sleep hours (≤ 5 hours, 5–6 hours, 6–7 hours, or ≥ 7 hours), chronic disease status such as diabetes, hypertension, and cardiovascular diseases (yes or no).

Height and weight were measured to the nearest 0.1 cm and 0.1 kg, respectively, by trained staff using a scale and a wall-mounted extensometer. Body mass index (BMI) was calculated as the weight in kilograms (kg)/height in meters (m) squared.

Statistical analysis

The characteristics of subjects are expressed as a percentage (categorical variables) or as mean and standard deviation values (continuous variables). Differences between groups were tested using the t-test for continuous variables and the chi-square test for categorical variables.

To reduce the complexity of the data, the 106 food items were categorized into 36 groups. In general, the food grouping was based on food and nutrient composition similarity (Table 1). Dietary patterns were derived using factor analysis (principal component) based on the 36 food groups of SQFFQ using PROC FACTOR. The factors were rotated via an orthogonal transformation to simplify the structure and to obtain greater interpretability. To determine the number of factors, we considered eigenvalues (> 2), scree test plots, and factor interpretability. The factor score for each pattern was calculated by summing intake of food groups weighted by factor loadings. Each subjects was then assigned a score for each of the identified patterns. Sujbects were divided into quartiles based on the factor score of each dietary pattern. We computed odds ratio (OR) values and their 95% confidence intervals (CI) using logistic regression. Multivariate models were adjusted for age, BMI, education level, smoking, alcohol drinking, exercise, sleep hours, chronic diseases status, marital status, family history of mental disorder, and total energy intake. Data were analyzed using SAS 9.4 (SAS Institute Inc. Cary, NC, USA) and statistical significance was accepted at P < 0.05.

Table 1. Food grouping used in dietary pattern analysis¹⁾.

¹⁾Food items were from the semi-quantitative food frequency questionnaire

RESULTS

Dietary patterns of subjects

We identified two major dietary patterns by using factor analysis: healthy and unhealthy dietary patterns. The factor loadings associated with each of the 36 food groups for both of the dietary patterns in male and female subjects are presented in Table 2. A positive loading score means that the food group is positively associated with the dietary pattern whereas a negative loading score indicates an inverse association with the dietary pattern. On that basis, the healthy dietary pattern is characterized by high intakes of vegetables, soybeans, mushroom, seaweeds, white fish, shellfish and fruits and a low intake of refined rice. Similarly, the unhealthy dietary pattern was characterized by high factor loadings in refined rice, meats, ramen, other noodles, bread, and coffee and low factor loadings in rice with grains. Both male and female had similar dietary patterns.

Table 2. Factor loading matrix for the two dietary patterns identified from the responses to the Food Frequency Questionnaire.

¹⁾Food groups with absolute values < 0.20 are not shown for simplicity.

Prevalence of depression and general characteristics

A summary of the sociodemographic characteristics, health behavior, marital status, sleep hours, and BMI of normal and depression subjects are shown in Table 3. Among the 3,388 subjects, 448 (13.2%) were identified as having depression. There was a significantly greater age, more ‘others’ marital status, lower education level, less sleep hours, less exercise, and more having chronic diseases in subjects with depression than in normal subjects. In male, there were significant differences in age, marital status, education, and sleep hours between depression and normal subjects. In female, age, marital status, education, sleep hours, and chronic disease status were significantly different between the normal and depression groups.

Table 3. Characteristics of subjects with and without depression.

¹⁾Chronic diseases such as diabetes, hypertension, hyperlipidemia, congestive heart failure, coronary artery disease, and myocardial infarction

BMI: body mass index

Dietary patterns and depression

Table 4 presents the OR and 95% CI of depression across quartiles of the healthy and unhealthy dietary patterns. Compared with the subjects in the lowest quartile of the healthy dietary pattern group, those in the highest quartile had a significantly lower odds ratio (OR = 0.59, 95% CI: 0.42–0.82, P for trend = 0.0037 for all subjects, OR = 0.59, 95% CI: 0.35–0.98, P for trend = 0.0359 for male, OR = 0.64, 95% CI: 0.42–0.99, P for trend = 0.1179 for female) after adjustment for age, BMI, exercise, smoking, alcohol drinking, marital status, education, sleep hours, family history of mental disorder, total energy intake and chronic disease status. In addition, the OR for the presence of depression in the highest quartile of the unhealthy dietary pattern group was significantly high compared to that in the lowest quartile as a reference (OR = 1.65, 95% CI: 1.19–2.28, P for trend = 0.0021 for all subjects, OR = 1.70, 95% CI: 1.03–2.80, P for trend = 0.0621 for male, OR = 1.50, 95% CI: 1.01–2.24, P for trend = 0.0104 for female).

Table 4. Association of dietary pattern with the risk of depression by multivariate-adjusted regression analysis.

Model 1: adjusted for sex (total) and age, Model 2: adjusted for sex (total), age, body mass index, exercise, smoking, alcohol drinking, marital status, education, sleep hours, family history of mental disorder, chronic diseases status, and total energy intake

Q, quartile; CI, confidence interval.

DISCUSSION

We conducted this study to determine whether dietary patterns derived from food group intake are associated with the presence of depression in middle-aged Korean adults. In this study, a ‘healthy’ dietary pattern (high intakes of vegetables, soybeans, mushroom, seaweeds, white fish, shellfish, and fruits and a low intake of white rice) showed a lower risk of depression whereas the ‘unhealthy’ dietary pattern (high intakes of white rice, meats, ramen, noodles, bread, and coffee and a low intake of rice with grains) increased the risk of depression after controlling for various social, health, and dietary confounders. To the best of our knowledge, this is the first study to examine the association between dietary pattern and depression in middle-aged Korea adults.

Our findings are similar to the results of a study of Korean adolescent girls, which indicated that a high intake of instant and processed foods increased the risk of depression and that high intakes of green vegetables, fruits, and beans was negatively associated with the risk of depression [15].

The healthy Japanese dietary pattern, which is characterized by a high intakes of vegetables, fruit, mushrooms, and soy products, has been associated with a low incidence of depressive symptom in Japanese adults [21]. Also, maintaining a dietary pattern rich in vegetables, mushrooms, seaweeds, soybean products, green tea, potatoes, and fish was shown to lower the risk of depressive symptoms among Japanese employees [22].

Western studies have also reported associations between dietary patterns and depression. Typically, the Mediterranean dietary pattern, which is characterized by a high intakes of vegetables, nuts, legumes, fruit, and fish, and a moderate alcohol consumption, was reported to lower the incidence of depression in the Spanish SUN cohort [23]. Also, in an Australian prospective study, Mediterranean-like and fruit dietary pattern was associated with a lower risk of depression in middle-aged women [24]. A three-year intervention study revealed that a Mediterranean diet with nuts has a beneficial effect on depression risk in diabetes patients [12]. Similarly, a healthy dietary pattern characterized by a high intake of various vegetables and nuts was associated with a low prevalence of depression in an Iranian population [25]. The whole-food dietary pattern (high intake of vegetables, fruits, and fish) was associated with a lower risk of depression and a processed food dietary pattern increased the risk of depression among middle-aged Whitehall II cohort participants [13].

The Korean healthy food pattern represents the traditional Korean diet, which includes plenty of vegetables, a high intake of legumes and fish, and a low intake of red meat. Also, the Korean diet generally includes banchan (small side dishes) mainly consisting of vegetables with various jang (fermented soy products or pepper paste), garlic, pepper powder, sesame oil, and perilla oil [26]. Our study revealed that a healthy dietary pattern of middle-aged Koreans was characterized by high intakes of vegetables, soybeans, mushroom, seaweeds, white fish, shellfish and fruits. Therefore, the healthy dietary pattern is rich in dietary fiber, omega-3 fatty acid, and various antioxidant nutrients. A systematic review reported that dietary n-3 PUFA and fish consumption are associated with a low risk of depression [27]. And it has been reported that fish or seafood intake, as well as omega-3 fatty acid intake, could prevent depression in a Korean population [28,29,30]. Taylor et al. [31] reported that a fiber and n-3 PUFA rich diet may reduce the risk of depression, anxiety, and stress. Also, dietary fiber intake from vegetables and fruits significantly reduced the depressive symptoms among Japanese workers [32]. Dietary fibers are transferred to the cecum and large bowel for anaerobic fermentation by gut microbiota into short chain fatty acids, such as acetic, propionic, and caproic acids, which has been reported to partially contribute to the prevention of depression [33].

Because this was a cross-sectional study, it is difficult to determine causal relationships between a healthy dietary pattern and a reduced risk of depression. Nonetheless, the results of this study may help to prevent middle-aged depression by reducing these unhealthy eating patterns and increasing healthy eating patterns. In conclusion, our findings suggest that a healthy dietary pattern is associated with a lower risk of depression whereas an unhealthy dietary pattern is associated with a high risk of depression in middle-aged Korean adults.