Plant-based diets and incident metabolic syndrome: Results from a South Korean prospective cohort study

Background Prior studies have shown that plant-based diets are associated with lower risk of cardiovascular risk factors and incident cardiovascular disease, but risks differed by quality of plant-based diets. No prospective studies have evaluated the associations between different types of plant-based diets and incident metabolic syndrome (MetS) and components of MetS. Furthermore, limited evidence exists in Asian populations who have habitually consumed a diet rich in plant foods for a long period of time. Methods and findings Analyses were based on a community-based cohort of 5,646 men and women (40–69 years of age at baseline) living in Ansan and Ansung, South Korea (2001–2016) without MetS and related chronic diseases at baseline. Dietary intake was assessed using a validated food frequency questionnaire. Using the responses in the questionnaire, we calculated 4 plant-based diet indices (overall plant-based diet index [PDI], healthful plant-based diet index [hPDI], unhealthful plant-based diet index [uPDI], and pro-vegetarian diet index). Higher PDI score represented greater consumption of all types of plant foods regardless of healthiness. Higher hPDI score represented greater consumption of healthy plant foods (whole grains, fruits, vegetables, nuts, legumes, tea and coffee) and lower consumption of less-healthy plant foods (refined grains, potatoes, sugar-sweetened beverages, sweets, salty foods). Higher uPDI represented lower consumption of healthy plant foods and greater consumption of less-healthy plant foods. Similar to PDI, higher pro-vegetarian diet score represented greater consumption of plant foods but included only selected plant foods (grains, fruits, vegetables, nuts, legumes, potatoes). Higher scores in all plant-based diet indices represented lower consumption of animal foods (animal fat, dairy, eggs, fish/seafood, meat). Over a median follow-up of 8 years, 2,583 participants developed incident MetS. Individuals in the highest versus lowest quintile of uPDI had 50% higher risk of developing incident MetS, adjusting for demographic characteristics and lifestyle factors (hazard ratio [HR]: 1.50, 95% CI 1.31–1.71, P-trend < 0.001). When we further adjusted for body mass index (BMI), those in the highest quintile of uPDI had 24%–46% higher risk of 4 out of 5 individual components of MetS (abdominal obesity, hypertriglyceridemia, low high-density lipoprotein [HDL], and elevated blood pressure) (P-trend for all tests ≤ 0.001). Greater adherence to PDI was associated with lower risk of elevated fasting glucose (HR: 0.80, 95% CI 0.70–0.92, P-trend = 0.003). No consistent associations were observed for other plant-based diet indices and MetS. Limitations of the study may include potential measurement error in self-reported dietary intake, inability to classify a few plant foods as healthy and less-healthy, lack of data on vegetable oil intake, and possibility of residual confounding. Conclusions In this study, we observed that greater adherence to diets consisting of a high intake of refined carbohydrates, sugars, and salty foods in the framework of plant-based diets was associated with an elevated risk of MetS. These results suggest that considering the quality of plant foods is important for prevention of MetS in a population that habitually consumes plant foods.


We submit this research plan to seek approval from the Institutional Review Board.
Date: December/9/2019 Principle investigator: Jihye Kim (signature)  ❍ Dietary factors related to metabolic syndrome -In previous studies, metabolic syndrome is known to be closely related to dietary intake. Recently, it has been known that it is closely related to food, food group intake, and dietary pattern.
-Whole grains : In a cross-sectional study of adults over the age of 18, reported that upper category of whole grain intake had 39% lower prevalence of hypertriglyceridemia, 16% lower prevalence of hypertension and 32% lower prevalence of metabolic syndrome than lowest category of whole grain intake. Higher consumption of refined grains was associated with 23% higher prevalence of having hypercholestrolemia, 2 fold higher prevalence of hypertriglyceridemia, 69% higher prevalence of hypertension and 2.2 fold higher prevalence of metabolic syndrome than lowest category of refined grain intake (Eur J Clin Nutr 2005).
-Legume: Subjects in the highest quartile of legume intake had 75% lower prevalence of having metabolic syndrome compared with those in the lowest quartile of legume intake. Subjects in the highest versus lowest quartile of legume intake have significantly reduced risk of metabolic syndrome (Arch Iran Med 2012).
-Fish: A cross-sectional study of Iranian women reported that subjects in the highest tertile of fish intake were 65% less likely to have the metabolic syndrome than those in the lowest tertile (Eur J Clin Nutr 2014).
-Red meat: In a study of Iranian women between the ages of 40 and 60, individuals in the top quintile of red meat intake had 2.3 fold greater prevalence of having metabolic syndrome compared with those in the bottom quintile (J Nutr 2009).
-Milk and dairy product: In the French general population, higher total dairy and cheese intake were associated with decreased waist circumference and triglycerides during the 9-year follow-up (J Am Coll Nutr 2011). Using data from the fifth Korean National Health and Nutrition Examination Survey, researchers found that higher consumption of milk or yogurt was significantly associated with a 29% lower risk of the metabolic syndrome (J Hum Nutr Diet 2013). A recent cohort study, which conducted in 26,445 healthy men and women followed up during 12 years, reported that highest versus lowest intake category of fermented milk was associated with 15% decreased incidence of cardiovascular disease (Eur J Epidemiol 2011).
-Fruit and vegetable: In a study of adults between the ages of 40 and 60, reported subjects in the highest quintile of fruit intake had a 34% lower and those in the highest quintile of vegetables intake had a 30% lower chance of having the metabolic syndrome than did those in the lowest quintiles (Am J Clin Nutr 2006).
-Sugar sweetened beverage: In precious cross sectional study, which was conducted tin Mexican aged between 20 and 70 years reported that subjects consuming more than two servings of sweetened beverages daily were having 2 times greater risk of metabolic syndrome than those who did not consume sweetened beverages (Public Health Nutr 2010).
-Alcohol: The cohort study of Italian men whose aged over 65 years reported that high alcohol intake significantly increases the risk of developing metabolic syndrome compared to those who consume low alcohol intake (Eur J Clin Nutr 2010).
-Refined grain: Data from the Fourth Korea National Health and Nutrition Examination Survey reported that women in the highest quintiles of refined grains and white rice intakes were more likely to have metabolic syndrome than women in the lowest quintiles (J Acad Nutr Diet 2014 -Several cross-sectional studies have been conducted between metabolic diseases and dietary risk factors, but most of them are conducted in limited population subgroups. In this study, we aim to provide appropriate dietary guidelines for the prevention and management of metabolic diseases in Koreans using large-scale population-based data (cohort) represented by Koreans. We aim to identify dietary risk factors associated with the development of metabolic diseases because some of these data have been followed up for a long term.

Expected results
❍ Based on the large-scale population survey data, we believe that it is possible to suggest the future direction on related chronic diseases if we identify the associated dietary risk factors in Korean adults such as metabolic syndrome and related metabolic diseases (obesity, high blood pressure, diabetes, cardiovascular disease and cancer). ❍ The findings can be used as a basis for establishing and informing national welfare policies for the implementation of a healthy society.

Criteria for Study Discontinuation
Not applicable

Sample size of study population
This study will identify dietary factors related to metabolic diseases in Korean adults. We made assumptions about diabetes because diabetes has the lowest incidence in this data set. A previous study which showed a 12% reduction in the risk of diabetes among those with higher versus lowest quality of diet reported that the number of samples required at power 80 % was approximately 220,000 (PLoS Med, 2016). The number of participants is similar to the total number of cohort participants of 210,000, requiring 210,000 participants.

Privacy policy
Privacy data from cohort research are not provided to the researchers who receive cohort data. The information provided to the researchers will be managed by a unique identification ID and the data obtained from this study will be accessible only to the researchers participating in the study. After the end of the study, the results of this study will not be discarded, but personal identification ID and healthrelated information will be discarded permanently after the end of the study.

Study Method 2.3.1 Study period
One year after IRB permission mm Hg or antihypertensive medication use ❍ Ascertainment of overweight and obesity: overweight is defined as 23≤body mass index (BMI)<25 and obesity is defined as BMI≥25 kg/m 2 ❍ Ascertainment of cancer: Incidence of gastric cancer, colon cancer, breast cancer, and thyroid cancer are examined. ❍ Measurement of biochemical indicators: blood glucose, triglyceride, cholesterol, blood pressure, insulin, and hs-CRP were meausred ❍ Covariates: demographic variables and lifestyle factors such as age, sex, education, smoking status, alcohol drinking, physical activity, nutrient intake will be used as covariates ❍ Anthropometric measurements will be used as covariate to evaluate the relationship between dietary factors and metabolic diseases

Data collection
Not applicable 2.3.9 Statistical analysis All data are analyzed using SAS version 9.4 (SAS institute, Cary, NC, USA). P values < 0.05 are considered as statistically significant. To evaluate the prospective associations between dietary factors and incidence of metabolic diseases such as metabolic syndrome, hypertension, diabetes, we will use Cox proportional hazards models. To investigate the cross-sectional association, we will use logistic regression analysis.

Concerns about side effects and safety
Not applicable

Data protection
The information provided to the researchers will be managed by a unique identification ID and the data obtained from this study will be accessible only to the researchers participating in the study. After the end of the study, the results of this study will not be discarded, but personal identification ID and healthrelated information will be discarded permanently after the end of the study.

Compensation policies for side effects and safety
Not applicable