Food and Chemical Toxicology 74 (2014) 177–183

Contents lists available at ScienceDirect

Food and Chemical Toxicology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / f o o d c h e m t o x

Dietary patterns and pulmonary function in Korean women: Findings from the Korea National Health and Nutrition Examination Survey 2007–2011 Yoonsu Cho a,b, Hye-Kyung Chung c, Seung-Sup Kim b, Min-Jeong Shin a,b,d,* a

Department of Food and Nutrition, Korea University, Seoul 136-703, Republic of Korea Department of Public Health Sciences, Graduate School, Korea University, Seoul 136-703, Republic of Korea c Severance Institute for Vascular and Metabolic Research, Yonsei University College of Medicine, Seoul 120-749, Republic of Korea d Korea University Guro Hospital, Korea University, Seoul 152-703, Republic of Korea b

A R T I C L E

I N F O

Article history: Received 24 April 2014 Accepted 22 September 2014 Available online 5 October 2014 Keywords: Dietary patterns Pulmonary function FEV1 FVC Vitamin A Dietary fat

A B S T R A C T

In the present study, we evaluated the association between dietary patterns and pulmonary functions in Korean women older than 40 years. This study analyzed the data from the Korea National Health and Nutrition Examination Survey IV and V (2007–2010). In total, 7615 women were included in the analysis. Using principal component analysis, two dietary patterns were identified, namely a balanced diet pattern (vegetables, fish, meat, seaweed, and mushrooms) and a refined diet (snacks, bread, milk, dairy products, and fast food). The refined diet pattern was positively associated with energy from fat but negatively associated with vitamin A, β-carotene, niacin, and fiber. After adjusting for potential confounders, the refined diet pattern was negatively associated with levels of predicted forced vital capacity (odds ratio (OR): 0.84, 95% confidence intervals (CIs): 0.70, 0.99) and predicted forced expiratory volume in 1 second (OR: 0.79, 95% CIs: 0.66, 0.93). In conclusion, the refined diet pattern was associated with decreased pulmonary function in Korean women. This information may be useful toward the development of nutritional guidelines for improving pulmonary function in Korean women. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Chronic respiratory diseases constitute a serious public health problem in all countries throughout the world (World Health Organization, 2007), of which the high prevalence and associated medical costs impose large socioeconomic burdens. Among the diseases that are characterized by airflow obstruction and decline in pulmonary function, chronic obstructive pulmonary disease (COPD) and asthma are the most common (Bommart et al., 2014) and especially true in Asian countries due to high rates of smoking, air pollution, and occupational exposure to dusts (Oh et al., 2011). According to data from the 2011 Korea National Health and Nutrition Examination Survey (KNHANES), 12.5% of Koreans reported a

Abbreviations: COPD, chronic obstructive respiratory disease; KNHANES, Korea National Health and Nutrition Examination Survey; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; BMI, body mass index; OR, odds ratio; CI, confidence interval. * Corresponding author. Department of Food and Nutrition, Korea University, Seoul 136-703, Republic of Korea; Department of Public Health Sciences, Graduate School, Korea University, Seoul 136-703, Republic of Korea; Korea University Guro Hospital, Korea University, Seoul 152-703, Republic of Korea. Tel.: +82 2 940 2857; fax: +82 2 940 2850. E-mail address: [email protected] (M.-J. Shin). http://dx.doi.org/10.1016/j.fct.2014.09.014 0278-6915/© 2014 Elsevier Ltd. All rights reserved.

current diagnosis of COPD (Korea Ministry of Health and Welfare, 2012). Pulmonary function is an important predictor of mortality in the general population as well as in the populations with respiratory disease (Hole et al., 1996; Thomason and Strachan, 2000). Several previous studies have reported the relationship between dietary factors, especially antioxidant nutrients and omega-3 fatty acid, and measures of pulmonary function (Kelly, 2005; Romieu, 2005; Spector and Surette, 2003). However, most of the previous studies have focused on the effects of single nutrients or food items on pulmonary function. Considering that nutrients and foods are consumed together rather than in isolation, characterizing dietary patterns within a population and examining whether dietary patterns are related to pulmonary function would be an appropriate approach to analyzing this relationship. Interestingly, recent reports suggested that sexspecific differences exist in the prevalence of chronic respiratory diseases (Po et al., 2011; Han et al., 2007; National Emphysema Treatment Trial Research Group, 2007). It is indicated that women are exposed to the risk of chronic bronchitis and COPD due to household biomass fuel (Po et al., 2011). Women with COPD showed more dyspnea and a lower subjective health status than those in men (Han et al., 2007). Moreover, the pathological nature of respiratory diseases may differ between men and women (Varkey, 2004; National Emphysema Treatment Trial Research Group, 2007), yet further

178

Y. Cho et al./Food and Chemical Toxicology 74 (2014) 177–183

investigation is required. In addition, limited studies have examined the associations between dietary patterns and pulmonary disease mainly conducted in western population. For example, previous observational studies suggested that adults following a “prudent” diet (high in fruit, vegetables, fish, and whole grains) are positively associated to forced expiratory volume in 1 second (FEV1), with reduction in the OR for COPD (Shaheen et al., 2010), and asthma (Varraso et al., 2009). Compared to Western people, Asian population including Koreans has a unique diet represented as high in refined carbohydrate (noodle, and rice cake), soyfoods, and fermented vegetables; therefore, it is interesting to characterize the dietary pattern predicting pulmonary function especially in this population. In the present study, we aimed to construct dietary patterns in Korean women using factor analysis. Then, we sought to examine the associations between dietary patterns and pulmonary function (i.e. FEV1 and forced vital capacity (FVC)) among Korean women over 40 years. 2. Subjects and methods 2.1. Study population This study was based on data from the KNHANES IV and V, 2007– 2010. Details of the KNHANES are available (Kweon et al., 2014). The KNHANES is a cross-sectional survey conducted nationwide by the Ministry of Health and Welfare. The KNHANES is composed of the following three sections: a health interview, health examination, and nutrition survey. A nationally representative sample was chosen from the Korean population using household records that were provided by the 2005 Population and Housing Census in Korea. Twenty households were selected from each survey section using a stratified, multistage probability cluster sampling method that considers each participant’s geographical area, age, and sex. In the KNHANES IV (2007–2009) and V (2010–2011), 42,347 individuals participated in the examination (response rate: higher than 80% for age ≥ 1 year). We limited our analyses to women older than 40 years because lung function tests were only implemented in those older than 40. We excluded subjects with missing data for pulmonary function levels and dietary intake. Finally, 7615 women were included in the statistical analysis. The institutional review board of the Centers for Disease Control and Prevention in Korea approved the KNHANES. All the participants in the survey provided informed written consent. 2.2. General characteristics of the subjects We obtained data from KNHANES IV and V, including demographic, anthropometric, and biochemical measurement data. Trained experts obtained anthropometric measurements by following standardized protocols. The body weights and heights of the subjects were measured to the nearest 0.1 kg and 0.1 cm, respectively. Body mass index (BMI) was calculated as weight (kg)/height squared (m2). Demographic variables that were potential confounders included age, smoking status, secondhand smoking status, physical activity, education, monthly income, occupation, and residence area. Smoking status was defined based on the questionnaire: “Do you currently smoke?” Subjects who answered “everyday” or “sometimes” were regarded as current smokers and subjects who answered “smoked in the past but not currently smoking” were regarded as past smokers. The mean number of cigarette packs smoked per day was also determined. Secondhand smokers were defined as subjects exposed to indirect cigarette smoke at their workplace or home. Physical exercise was divided into two categories, either as “practice” or as “do not practice,” according to whether or not the individual participated in any of the following at least 5 days within each week: intense physical activity for at least 20 minutes, moderate physical activity for at least 30 minutes, or walking for at least

30 minutes. Education level was divided into four categories as elementary school, middle school, high school, or university, according to the subject’s highest achieved level. Monthly income was divided into quartiles and reported in the South Korean currency won as follows: lowest (≤1 million won), lower middle (1 million won ≤ 2 million won), upper middle (2 million won ≤ 3 million won), or highest (>3 million won). The area of residence was classified as urban or rural. Occupation type was categorized into four groups according to a previous study (Lee et al., 2013), with slight modifications, as an office worker (administrator, professional, salesperson, or service worker), manual worker (agriculture, fishing, or simple laborer), technician, or unemployed. 2.3. Dietary assessment Dietary data were obtained from the nutrition survey in the KNHANES. To identify dietary patterns, all food items in the food frequency questionnaire were categorized into 18 groups based on the Korean nutrient database (The Korean Nutrition Society, 2010; Fig. 1). The food frequency questionnaire reflects how often subjects consumed each particular food over the prior 12-month period according to a 10-point scale (9 = 3 times per day, 8 = 2 times per day, 7 = once per day, 6 = 4–6 times per week, 5 = 2–3 times per week, 4 = once per week, 3 = 2–3 times per month, 2 = once per month, 1 = 6–11 times per year, and 0 = almost never). The content and amount of food consumed were obtained by the 1-day 24-hour recall method; the nutrient intake was then analyzed using the database from the food composition table published by the Rural Development Administration (2006). Nutrient intake data included total energy (kcal), carbohydrate (g), protein (g), fat (g), fiber (g), calcium (mg), phosphorous (mg), iron (mg), sodium (mg), potassium (mg), vitamin A (μg RE), vitamin B1 (mg), vitamin B2 (mg), niacin (mg NE), and vitamin C (mg). The ratio of energy intake from each macronutrient to the total energy was calculated as the percentage of energy intake from carbohydrates, fats, and protein. The intake of other nutrients was also calculated as a ratio per 1000 kcal of total energy. 2.4. Pulmonary function assessment Trained technicians measured pulmonary function, including FVC and FEV1, using a rolling-dry seal spirometer (model 2130, SensorMedics, Yorba Linda, CA, USA), according to the guidelines of the American Thoracic Society and European Respiratory Society (ATS/ERS Task Force, 2005). Pulmonary function measured by spirometry is a good predictor of mortality in the general population and in patients with respiratory disease, including asthma, dyspnea, COPD, and bronchial disorder (Petty, 2005). Spirometry was performed at least three times in adults older than 40, those without severe lung disease (tuberculosis, pneumothorax, and pulmonary emphysema), surgery history within the 3 months, and experience of stroke or myocardial infarction within the 3 months. All spirometry results were examined as to whether the results met the ATS/ERS criteria for acceptability and repeatability (Korean Academy of Tuberculosis and Respiratory Diseases, 2005). To obtain the acceptable data, subjects should not be interfered by cough or obstruction of mouthpiece during the first second of expiration. Also, the spirometry test must properly initiate and terminate without hesitation during the expiration. Finally, two largest spirograms were obtained, satisfying the following conditions: two acceptable measurements must be within 0.15 L, with initial expiration volume of less than 5% of the FVC or 0.15 L. 2.5. Statistical analyses Statistical analyses were performed using SPSS version 21.0 (SPSS, Inc., Chicago, IL, USA). Dietary patterns were analyzed using the

Y. Cho et al./Food and Chemical Toxicology 74 (2014) 177–183

179

Food groups

Food item

Processed grains

White rice

Whole grains

Unpolished rice, Barley, Glutinous rice

Noodles

Instant noodle (Ramyon), Chinese noodle, wheat noodle

Rice cakes

Korean traditional steamed rice food

Snacks

Crackers, Cookies

Bread

Loaf bread, Cakes, Sandwich

Potatoes

Potatoes, Sweet potatoes

Legumes

Soybeans, Soybean curd, Soybean paste, Soybean milk

Vegetables

Cabbage, White radish, Soybean sprout, Spinach, Cucumber, Hot pepper, Carrot, Pumpkin, Tomatoes

Mushrooms

Mushrooms

Fruits

Mandarin orange, Orange, Persimmon, Apple, Pear, Watermelon, Strawberry, Oriental melon, Grapes, Peach, Banana, Pineapple

Meat

Beef, Chicken, Pork, ham, Sausage, Bacon

Eggs

Egg

Fish

Mackerel, Tuna, Yellow corbina, Pollock, Anchovy, Fish cakes, Squid, Shellfish

Seaweeds

Seaweed, Laver

Milk and dairy products

Milk, Yogurt, Ice cream, Cheese

Beverages

Carbonated beverages, Coffee, Green tea,

Alcohol

Beer, Soju, Makgeolli (Traditional rice wine)

Fast foods

Hamburger, Pizza, Fried foods

Kimchi

Traditional salted vegetables Fig. 1. Food groups from the food frequency questionnaire used in dietary pattern analysis.

exploratory principal component factor analysis based on the Korean nutrition database. The Kaiser criterion (eigenvalues > 1.0) was used to identify the number of factors to be retained. The retained scree plots were rotated by an orthogonal transformation (Varimax) to obtain a more condensed structure with advanced interpretability. After the Varimax rotation, factor scores for each dietary pattern and individual were calculated by summing the intake of each food group using factor loading. Continuous and categorical variables were described as mean ± standard error values or number and percentages of subjects, respectively. Associations between dietary pattern scores and nutrient intake were examined by correlation analysis with and without adjustment for covariates. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for levels of pulmonary function across the tertiles of dietary patterns, with the lowest tertile group as the reference. The first model was unadjusted, and the second was adjusted for age, BMI, smoking status, number of cigarette packs smoked per day, secondhand smoking, exercise, education, income, occupation, and residence area. 3. Results 3.1. Dietary patterns and characteristics of the population Two different types of dietary patterns were derived from the factor analysis: a balanced diet or a refined diet (Table 1). A balanced diet was characterized as the consumption of various foods including vegetables, fish, meat, seaweeds, and mushrooms. A refined diet was characterized by high intakes of snacks, breads, dairy

Table 1 Factor loading matrix for the dietary patterns of study population. Food groups

Processed grains Whole grains Noodles Bread Rice cakes Snacks Legumes Potatoes Vegetables Mushrooms Fruit Meat Eggs Fish Seaweeds Dairy products Beverages Alcohol Fast foods Kimchi Total % of variance explained

Componenta Factor 1 Balanced diet

Factor 2 Refined diet

−0.09 0.05 0.03 0.12 0.21 0.02 0.17 0.53 0.73 0.64 0.66 0.37 0.42 0.67 0.64 0.43 0.13 0.03 0.13 0.15 15.8

−0.12 0.01 0.20 0.68 0.49 0.68 −0.01 0.17 −0.05 0.02 0.19 0.44 0.27 0.09 0.04 0.30 0.13 −0.29 0.45 −0.07 9.8

Extraction method: principal component analysis. Rotation method: varimax with Kaiser normalization. a Factor loadings are only displayed for values ≤ −0.30 or ≥0.30.

180

Table 2 Basic characteristics by quartile of dietary pattern score in the population. Pattern 1: Balanced diet T1 (n = 2538)

T2 (n = 2539)

T3 (n = 2538)

P-value

T1 (n = 2538)

T2 (n = 2539)

T3 (n = 2538)

P-value

56.9 ± 0.1 24.2 ± 0.04

60.9 ± 0.2a 24.3 ± 0.1a

56.0 ± 0.2b 24.3 ± 0.1a

53.9 ± 0.2c 24.0 ± 0.1b