NFS Journal 4 (2016) 914

Contents lists available at ScienceDirect

NFS Journal

journal homepage: http://www.journals.elsevier.com/nfs-journal/

Original article

The effect of coffee consumption on food group intake, nutrient intake, and metabolic
syndrome of Korean adults2010 KNHANES (V-1)
Fangfang Song a, JiEun Oh a, KyungWon Lee b, Mi Sook Cho a,
a
Department of Nutritional Science and Food Management, Ewha Womans University, Seoul 120-750, Korea
b
Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI, USA

a r t i c l e i n f o a b s t r a c t

Article history: Background: Coffee is a popular beverage in Korea recent years. The purpose of this study was to investigate the
Received 9 October 2015 relationship between coffee consumption and the risk of metabolic syndrome in Korean adults based on the 2010
Accepted 27 April 2016 Korean National Health and Nutrition Examination Survey (KNHANES V-1).
Available online 3 May 2016 Methods: Dietary intake status and the factors of metabolic syndrome were assessed. Three groups (no coffee
consumption, moderate intake, and high intake) were divided into tertile according to black coffee cream
Keywords:
(include brewed coffee) consumption per day.
Coffee consumption
Nutrient intake
Results: Our results showed that the Tertile 3 group consumed more calories from fat, and niacin was higher than
INQ in the Tertile 1 and Tertile 2 group. INQ for protein and vitamin B1 was signicantly higher in no coffee consump-
Metabolic syndrome tion group than the other groups and in Tertile 3 exhibited signicantly higher niacin intake. The subjects in
KNHANES Tertile 3 showed signicantly higher consumption in grain and oil intake, and Tertile 1 group showed higher con-
sumption in milk and dairy products. In the logistic regression analysis, adjusting for sex, age, energy intake,
smoking, and drinking, being in the high coffee consumption group (Tertile 3) was signicantly and inversely as-
sociated with abdominal obesity (OR = 0.76, CI = 0.710.82), hypertension (OR = 0.70, CI = 0.540.87), high
glucose(OR = 0.71, CI = 0.610.86). However, no signicant association was found between coffee consump-
tion and metabolic syndrome.
Conclusion: Coffee consumption has not a considerably relationship with nutrient intake. Appropriate consump-
tion of coffee may have potentially helpful effects on certain metabolic risk factors, such as abdominal obesity,
hypertension, and high glucose.
2016 The Authors. Published by Elsevier GmbH on behalf of Society of Nutrition and Food Science e.V. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
In Korea, the culture of coffee consumption has become more com-
mon through the adoption of the Westernized diet, and it has continu-
ously and rapidly increased in recent years [1]. In research based on
the 20072009 (KNHANES V-1), 8056 subjects (52.3%) of the 15,389
subjects included in the analysis reported consuming coffee more than
once a day, and 50.7% of the coffee consumers reported that their aver-
age coffee consumption was more than twice a day [2]. Recently, coffee
was reported to be the most frequently consumed foods among Korean
adults after rice (17 times per week), cabbage (14 times per week), and
alcoholic beverages (10 times per week) [3]. According to the 2007
KNHANES of the Korea Centers for Disease Control and Prevention, the
average daily energy intake among Koreans from coffee/instant coffee
mix was 15.3 kcal per day, which ranked 24th for its contribution to
Corresponding author at: 52, Ewhayeodae-gil, Seodaemun-gu, Seoul 120-750, Korea.
Tel.: +82 2 3277 2826; fax: +82 2 3277 2862.
E-mail address: misocho@ewha.ac.kr (M.S. Cho).
energy intake among Koreans. It was shown that coffee was the only
item categorized as a beverage within the top 30 foods ranked in
order of their contribution to energy intake [4].
Coffee contains diverse ingredients, such as Caf Royal, caffeol,
chlorogenic acid, potassium, niacin, and magnesium. Such ingredients
as caffeine are known to affect diverse physiological functions in our
bodies [5]. Previous research conducted in many countries on the rela-
tionship between coffee and health has shown a potential effect on
health by frequency of coffee consumption. There was one nding
which showed a signicantly lower level of metabolic syndrome pres-
ent in Japanese people who drink coffee frequently [68]. However, no
signicant association between coffee and metabolic syndrome was
found in cohort studies in the U.S. and Europe [9,10]. Additionally, the
majority of the precedent studies suggested that higher or more fre-
quent consumption of coffee reduces the morbidity rate of diabetes
[1113]. The relationship between coffee consumption and diabetes
has been relatively consistent in studies. In comparison to overseas
studies, domestic studies on coffee and health are insufcient. Speci-
cally, a general consensus on the risk of metabolic syndrome has not
been established to date.

http://dx.doi.org/10.1016/j.nfs.2016.04.002
2352-3646/ 2016 The Authors. Published by Elsevier GmbH on behalf of Society of Nutrition and Food Science e.V. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
10 F. Song et al. / NFS Journal 4 (2016) 914
As the consumption of coffee beverages has rapidly grown and the
frequency of coffee consumption by adults has become especially
high, a majority of studies have been conducted on the physicochemical
characteristics of coffee or the product quality and organoleptic proper-
ties [14]. However, there have been an inadequate number of studies on
food and nutrient intake regarding coffee consumption [15]. There is
normally a quantitative correlation between nutrient intake and energy
intake. Therefore, an evaluation on the level of nutrient intake should be
accompanied by considerations of energy intake [16]. A study by Lutsey
et al. [17] on coffee consumption and food intake did not nd any signif-
icant relationships between the risk of metabolic syndrome and the in-
take of food groups, including grains, fruits, vegetables, nuts, sh,
poultry, and dairy products. While previous studies have focused
more on eating patterns followed by coffee consumption or the con-
struction of nutrients and calorie intake among the food groups, there
is a gradual trend currently toward the study of the food itself ingested
during a meal [18]. However, there is not yet an understanding of the re-
lationship between coffee consumption and chronic diseases. Therefore,
nothing much is known about the inuence of coffee consumption on
the diet of Korean adults, and there is no sufcient nutrient education
underway in this eld.
The 2010 KNHANES was a nationally funded, cross-sectional, nation-
wide survey that used a multistage sampling design to collect data re-
garding the health behaviors, nutrition, and socio-demographics of the
general Korean population [19,20]. Currently, the nutrition survey has
adopted 3 methods, such as health behavior investigation and health
examination investigation, food frequency questionnaire and dietary in-
takes (24 h recall method), and dietary habits research (research on
items related to dietary life and eating habits). Additionally, the health
behavior survey and checkup survey are underway [21].
Therefore, this study aimed to understand the level of coffee con-
sumption in Korean adults and to compare and analyze the true state
of food group and nutrient intake, followed by coffee consumption
(Tertile), INQ, and index related to metabolic disease, by utilizing the
2010 KNHANES. We hypothesized that higher coffee consumption af-
fects dietary intake and increase risk of metabolic syndrome.
2. Materials and methods
2.1. Study population
The analysis in this study was conducted using 2010 KNHANES, V-1
data. Overall, 8958 subjects participated in this research. The following
subjects were excluded from the analysis: (1) subjects who were youn-
ger than 19 years of age (2171 subjects), (2) subjects whose energy
intakes were less than 500 kcal/day or exceeded 5000 kcal/day
(1065 subjects), (3) pregnant women (771 subjects), and (4) subjects
who had missing values (145 subjects). In total, 4806 individuals were
included in the analysis after excluding the subjects who met those
criteria.
2.2. Data collection and denitions
This study utilized data on sex, age, alcohol consumption, smoking
status, body mass index(BMI), and coffee consumption, which were col-
lected through the health inquiry survey of the KNHANE Survey.
BMI was divide the subjects into a low weight group (b18.5 kg/m2),
normal weight group (18.5 kg/m2 BMI b 23 kg/m2), overweight
group (23 kg/m2 BMI b 25 kg/m2), and obese group (25 kg/m2 )
[22]. To classify the groups according to coffee consumption, this
study used variables from the second round of food code to examine
the data on food intake of the subjects by using 24-h recall method-
used-study in KNHANES. The second round of food code is a classica-
tion value to nd foods that can be considered the same because their
commercial names are identical and their water contents are similar.
The coffee intake was analyzed by the black coffee include brewed
coffee, calculated by adding up the weight (g) variables. The subjects
were classied into 3 groups (Tertile 1, Tertile 2, and Tertile 3) based
on their daily coffee consumption by utilizing tertile and by calculating
the average coffee intake of each group.
2.3. Dietary assessment
This study analyzed and displayed intake per 1000 kcal of each nutri-
ent based on the food intake survey conducted through the 24-h recall
method in the nutrient survey part of the data from the 2010
KNHANES (V-1). Nutrient intake was calculated from intake per nutri-
ent of the subjects for analysis using data from the 24-h recall method
[2,18]. Nutrients utilized in the analysis were total energy, protein, car-
bohydrate, fat, ber, calcium, phosphorus, iron, sodium, vitamin A, vita-
min B1, vitamin B2, niacin, and vitamin C. The energy ratio acquired from
carbohydrates, fats, and proteins was calculated using the intake of
those 3 nutrients.
This study calculated daily intake of different food groups by classi-
fying them into the following 6 food groups: grains; meat, sh, egg,
and beans; vegetables; fruits; milk and dairy products; and fat, oils,
and sugars. These were grouped based on recommendations in the re-
vised edition (2010) of the Nutrition Intake Criteria of Koreans from
the Korea Nutrition Society, after examining the food code data which
focused on the nutrition code data provided by KNHANES [23].
This study used the Index of Nutritional Quality (INQ) to evaluate
the quality of a meal. This index can evaluate the degree of satisfaction
in energy intake from each nutrient based on nutrient density. The
INQ was used to compare nutrient intake per 1000 kcal of each nutrient
and the recommended dietary intake per 1000 kcal. In this study, eval-
uation of the INQ was conducted on the following 12 nutrients: protein,
ber, calcium, phosphorus, iron, sodium, potassium, vitamin A, vitamin
B1, vitamin B2, niacin, and vitamin C. In cases where the INQ exceeds 1, a
specic nutrient was sufciently consumed. In cases where INQ is less
than 1, more of that nutrient should be consumed to satisfy the recom-
mended dietary intake [23,24].

INQ nutrient intake per 1; 000 kcal of calorie intake
nutrient intake per 1; 000 kcal of recommended calorie intake
2.4. Risk factors of metabolic syndrome
This study closely examined the following 8 factors of metabolic
risks: BMI, waist circumference, diastolic blood pressure, systolic blood
pressure, concentration of blood triglycerides, blood cholesterol concen-
tration, HDL cholesterol concentration, and fasting blood glucose. These
factors were used to study the association between coffee consumption
and metabolic syndrome. Metabolic syndrome was dened as any
Asian, according to the WHO, meeting any 3 out of the following 5
criteria recommended by the 2001 National Cholesterol Education
Program-Adult Treatment Panel III (NCEP-ATP III): (1) abdominal obesi-
ty: waist circumference for males 90 cm and females 85 cm; (2) high
blood pressure: systolic blood pressure 133 mmHg or diastolic blood
pressure 85 mmHg, or medication status in the case of a patient who
has past medical history of high blood pressure; (3) hypertriglyc-
eridemia: blood triglycerides 150 mg/dL or taking medication;
(4) low HDL cholesterol: in males b 40 mg/dL, in females b50 mg/dL, or
taking medication; and (5) fasting blood glucose 100 mg/dL or taking
medication. Obesity is dened as BMI 25.0 kg/m2 according to the
criteria of the Asia-Pacic region of the WHO [24,25].
2.5. Statistical analyses
Data analysis was conducted in SAS (Statistical Analysis System, ver-
sion 9.3, SAS Institute, Cary, NC, USA). Composite sampling based on the
outcome data of KNHANES was applied to the data treatment and the
 F. Song et al. / NFS Journal 4 (2016) 914 11

weighted-value application system, as recommended by the Korean Table 2

Centers for Disease Control and Prevention. The general aspects and Food groups intake status of all subjects by coffee consumption group.

data of the subjects were shown in the form of frequency and percentile, Coffee consumption group
which were analyzed using a chi-square test. Analysis of variance Tertile 1 Tertile 2 Tertile 3 p-value
(ANOVA) was performed to compare averages among the 3 groups
Grains (g) 359.29 5.21a 349.03 4.95a 326.54 4.71b b;.0001
(Tertile 1, Tertile 2, and Tertile 3) regarding their nutrients intake,
Meat, sh, eggs and 216.01 4.91 215.61 5.72 205.85 5.21 0.2476
food group intake, and evaluation on meal quality. The post hoc com- beans (g)
parison was performed using Turkey's multiple range test on variables Vegetables (g) 357.10 7.22 354.74 7.77 358.34 7.16 0.9333
showing a signicant difference as a result of the analysis of variance. Fruits (g) 219.18 12.17 223.46 17.97 204.25 13.66 0.3574
a a
To analyze the relationship between metabolic syndrome and relevant Milk and dairy 99.05 6.58 79.29 5.04 66.31 5.43b 0.0017
products (g)
risk factors taking into account the composite sampling, the following Oil and sugars (g) 13.47 0.57c 16.31 0.59b 26.19 0.93a b.0001
models were used: a logistic regression analysis model, which did not
Values are expressed as mean SD adjusted for sex, age, and energy intake.
adjust for disturbance factors related to metabolic syndrome (Model
Mean values in the row on the coffee intake quartile are signicantly different (*p b 0.05,
1: unadjusted); an analysis model, which adjusted for the factors of **p b 0.01, ***p b 0.001) by ANOVA.
age, sex, and energy intake (Model 2: age, sex, energy intake- a, b, Mean values with different superscripts are signicantly different by Tukey's multiple
adjusted); and an analysis model, which adjusted for the factors of rang test.
age, sex, energy intake, smoking, and drinking (Model 3: age, sex, ener-
gy intake, smoking, drinking adjusted). Every statistical analysis used in The results of analysis of INQ, which evaluates the quality of a meal
this study tested signicance at the level of = 0.05. by using the proportion of nutrient content in diet by the recommended
intake of each nutrient per 1000 kcal, are shown in Table 4. The
3. Results INQ index of protein (p b 0.05), vitamin B1 (p b 0.01), and niacin
(p b 0.001) showed statistically signicant results. The Tertile 1 group
Analysis on a total of 4806 individuals (male: 1960 and female: showed a signicantly higher results in protein (p b 0.05), and vitamin
2846) was conducted, and their general characteristics are presented B1 (p b 0.01) compared to the Tertile 3 group; it also showed a signi-
in Table 1 with grouping according to coffee intake. The age group be- cantly lower level of niacin (p b 0.001).
tween 30 years old and 50 years old was the largest group accounting To estimate the relationship between coffee consumption and met-
for 44.07% of participants; 77.85% were drinkers and 22.15% were abolic syndrome, this study performed a logistic repercussion analysis
smokers. In the category of BMI, 42.05% were of standard weight, the and obtained odds ratios (OR) and condence intervals (95% CI),
highest of all groups; the proportion of subjects who were obese was which are shown in Table 5. Among the index related to abdominal obe-
31.25% and overweight was 22.36%. sity, high blood pressure, and high fasting blood glucose were associated
Results of analysis on food group intake by coffee consumption are with coffee consumption. Analysis of the risk factors for metabolic syn-
shown in Table 2. Signicant differences were found for the following drome was performed by comparing the no coffee consumption group.
groups: grains; milk and dairy products; and fat, oils, and sugars. Tertile Model 2, in which confounding factors (sex, age, energy intake) have
1 group had the highest average daily intake of the grains food group, as been adjusted, showed that the risk of abdominal obesity was lower in
well as milk and dairy product food group (p b 0.001). The oil and sugar the Tertile 3 group (0.77; 95% CI = 0.720.82) (p b 0.05). The risk of hy-
food group was consumed the most by Tertile 3 (p b 0.01). pertension was lower in the Tertile 3 group (0.69; 95% CI = 0.640.76)
The average daily nutrient intake of subjects by coffee consumption (p b 0.05), and the high fasting blood glucose in Tertile 3 group was also
is shown in Table 3. A statistically signicant difference was shown in lower (0.71; 95% CI = 0.600.86) (p b 0.05). In Model 1 and Model 3,
the intake of fat from energy (p b 0.01), total energy (p b 0.001), protein where diverse confounding factors (sex, age, energy intake, smoking
(p b 0.05), vitamin B1 (p b 0.05), and niacin (p b 0.01) by coffee con- status, and drinking status) were adjusted, the risk of abdominal obesity
sumption. Protein accounted for the highest proportion of calories in was found to be lower in Tertile 3 (0.76; 95% CI = 0.710.81) (p b 0.05).
the no coffee consumption group (p b 0.05), which also had the lowest Compared to Tertile 1, the risk of blood pressure increasing was shown
intake of fat (p b 0.01). The Tertile 3 group, which had the highest coffee to be lower in the Tertile 3 group, and the risk of high blood pressure de-
consumption, also had the highest level of average daily energy intake creased signicantly (0.70; 95% CI = 0.540.87) (p b 0.001) as coffee
(p b 0.001). The Tertile 1 group had the highest average daily protein consumption increased. The risk of high glucose was lower (0.71; 95%
and vitamin B1 intake (p b 0.05), whereas had the lowest average CI 0.610.86) (p b 0.05) in the Tertile 3 group. According to these results,
daily niacin intake (p b 0.01). risks for abdominal obesity, high blood pressure, and high fasting blood

Table 1
General characteristics of all subjects (N, %).

Characteristic Tertile 1 Tertile 2 Tertile 3 Total

(N = 4806 )

Gender Male 620(37.17) 495(33.56) 845(50.81) 1960(49.84)

Female 1048(62.83) 980(66.44) 818(49.19) 2846(50.16)
Age 19 age b 30 295(17.69) 86(5.83) 162(9.74) 543(19.79)
30 age b 50 496(29.74) 543(36.81) 812(44.83) 1851(44.07)
50 age b 65 454(27.22) 488(33.08) 463(27.84) 1395(24.32)
65 423(25.36) 358(24.27) 236(14.19) 1017(11.83)
Drinking No 579(34.71) 429(29.08) 339(20.38) 1344(22.15)
Yes 1092(65.29) 1046(71.92) 1324(79.62) 3462(77.85)
Smoking Never 1109(66.49) 985(66.78) 828(49.79) 2922(53.75)
Former 335(20.08) 294(19.93) 353(21.23) 982(20.57)
Current 224(13.43) 196(13.29) 482(28.98) 902(25.68)
BMI (kg/m2) b18.5 79(4.74) 49(3.32) 68(4.09) 196(4.34)
18.5 BMI b 23 703(42.15) 620(42.03) 697(41.91) 2020(42.05)
23 BMI b 25 380(22.78) 340(23.05) 378(22.73) 1098(22.36)
25 506(30.34) 466(28.02) 520(31.27) 1492(31.25)
12 F. Song et al. / NFS Journal 4 (2016) 914

Table 3
Nutrition status of all subjects by coffee consumption group.

Coffee consumption group

Tertile 1 Tertile 2 Tertile 3 p-value

% Energy
Carbohydrate1 68.82 0.31 68.74 0.39 67.89 0.35 0.0632
Protein 14.69 0.12 14.51 0.16 14.29 0.15 0.1038
Fat 16.39 0.26b 16.75 0.29b 17.82 0.25a 0.0002
Energy (kcal)2 1924.11 23.35b 1913.80 25.45b 2105.51 22.41a b.0001
Carbohydrate (g) 324.02 2.35 325.19 3.03 327.65 2.67 0.5933
Protein (g) 72.50 0.65a 71.40 0.78a 69.49 0.75b 0.0108
Fat (g) 38.24 0.62 37.99 0.70 39.04 0.65 0.5133
Fiber (g) 8.02 0.17 7.92 0.18 7.86 0.22 0.7446
Calcium (mg) 537.53 11.13 525.61 9.73 522.70 10.74 0.5634
Phosphorus (mg) 1202.71 10.94 1194.23 10.37 1184.62 10.12 0.4488
Iron (mg) 15.62 0.31 15.35 0.35 15.63 0.39 0.7886
Sodium (mg) 5075.66 80.6 5069.86 94.28 5096.09 84.21 0.9762
Potassium (mg) 3094.09 39.26 3112.43 39.65 3119.69 42.10 0.0621
Vitamin A (g RE) 839.79 28.24 864.95 36.80 852.04 34.78 0.8360
Vitamin B1 (mg) 1.36 0.04a 1.32 0.04a 1.26 0.04b 0.0052
Vitamin B2 (mg) 1.25 0.04 1.22 0.04 1.20 0.05 0.3089
Niacin (mg) 16.41 0.20b 16.47 0.19b 17.45 0.21a 0.0010
Vitamin C (mg) 112.44 3.02 113.89 2.92 108.33 2.74 0.2864

a, b, mean values with different superscripts are signicantly different by Tukey's multiple rang test.
1
Values are expressed as mean SD adjusted for sex, age, and energy intake.
2
Mean values in the row on the coffee intake group are signicantly different (*pb 0.05, **p b 0.01, ***p b 0.001) by ANOVA.

glucose decreased signicantly as coffee intake increased. However, no Kim [27], which also found high intake of protein and vitamin B1 and
signicant relationship was shown between the coffee intake and risk low intake of niacin in the low coffee consumption group. Such a result
for metabolic syndrome. can be observed because the high coffee intake group consumes more
foods from the meat group and vegetable group that have high protein
4. Discussion and vitamin B1 contents. For food group intake according to coffee

In the present study, we evaluated the relationship between coffee

Table 5
consumption and dietary intake status and metabolic syndrome using
Crude and adjusted odds ratios (95% condence interval) of the risk of metabolic syn-
the KNHANES data (2010). We found that coffee consumption has not drome with coffee consumption group of all subjects.
very clear relationship with food and nutrition intake, whereas has po-
Tertile 1 Tertile 2 Tertile 3 p for trend
tentially helpful effects on some of metabolic risk factors.
Lutsey et al.'s research [17] determined no signicant relationship Obesity (BMI 25 kg/m2)
between risk of metabolic syndrome and intake of food groups, such Model 1a 1.0 0.90 (0.741.09) 0.96 (0.791.17) 0.4150
Model 2 1.0 0.92 (0.761.12) 0.91 (0.751.12) 0.8841
as grains, fruits, vegetables, nuts, poultry, dairy products, and coffee in- Model 3 1.0 0.92 (0.761.12) 0.92 (0.751.13) 0.9466
take. Among the high coffee consumption group, when controlling for
sex, age, and energy intake, the daily energy intake was shown to be Abdominal obesity (waist circumstances 90 cm in men, 85 cm in women)
Model 1 1.0 0.84 (0.701.02) 0.68 (0.570.81) 0.0025
at the highest level. In research by Christina et al. [26], the highest coffee Model 2 1.0 1.04 (0.861.26) 0.77 (0.720.82) 0.0139
intake group also had the highest energy intake. In the lowest coffee Model 3 1.0 1.03 (0.791.27) 0.76 (0.710.81) 0.0133
consumption group, the average daily intake of protein and vitamin B1
Hypertension (SBP 130 mmHg or DBP 85 mmHg)
was the highest; however, the average daily intake of niacin was signif- Model 1 1.0 1.01 (0.841.19) 0.81 (0.680.96) 0.0144
icantly lower. This result coincided with a study by Bae, Kim [18], and Model 2 1.0 0.72 (0.660.88) 0.69 (0.640.76) 0.0006
Model 3 1.0 0.72 (0.660.88) 0.70 (0.540.87) 0.0011

Table 4 Hypertriglyceridemia (serum TG 150 mg/dL)

Index of Nutrition Quality (INQ) of all subjects by coffee consumption group. Model 1 1.0 1.02 (0.821.28) 1.10 (0.891.36) 0.4348
Model 2 1.0 1.09 (0.861.38) 1.05 (0.841.31) 0.7180
Coffee Consumption Group Model 3 1.0 1.11 (0.881.41) 1.02 (0.811.28) 0.3182
Tertile 1 Tertile 2 Tertile 3 p-value Low HDL-cholesterol (HDL b 40 mg/dL in men, b50 mg/dL in women)
Protein (g) 1.49 0.03a 1.49 0.03a 1.45 0.03b 0.0161 Model 1 1.0 0.95 (0.781.16) 0.81 (0.660.99) 0.1093
Fiber (g) 0.38 0.02 0.38 0.02 0.37 0.02 0.4008 Model 2 1.0 1.11 (0.911.35) 1.06 (0.851.32) 0.6323
Calcium (mg) 0.80 0.03 0.79 0.03 0.78 0.03 0.4867 Model 3 1.0 1.12 (0.921.35) 1.04 (0.831.29) 0.4956
Phosphorus (mg) 1.77 0.03 1.75 0.03 1.73 0.03 0.1491
High glucose (plasma glucose 100 mg/dL)
Iron (mg) 1.66 0.08 1.59 0.06 1.59 0.08 0.2159
Model 1 1.0 1.15 (0.841.58) 0.67 (0.470.87) 0.0094
Sodium (mg) 3.73 0.11 3.68 0.11 3.67 0.13 0.7224
Model 2 1.0 0.78 (0.670.91) 0.71 (0.600.85) 0.0232
Potassium (mg) 0.91 0.02 0.92 0.02 0.94 0.02 0.1502
Model 3 1.0 0.78 (0.670.91) 0.71 (0.610.86) 0.0266
Vitamin A (g RE) 1.29 0.10 1.29 0.07 1.28 0.08 0.9940
Vitamin B1(mg) 1.21 0.03a 1.17 0.03ab 1.12 0.03b 0.0011 Metabolic syndrome
Vitamin B2(mg) 0.95 0.03 0.94 0.04 0.93 0.04 0.5472 Model 1 1.0 1.02(0.821.27) 1.01(0.811.26) 0.9471
Niacin (mg) 1.11 0.02b 1.12 0.03b 1.19 0.03a 0.0002 Model 2 1.0 1.17 (0.941.46) 1.14 (0.911.42) 0.7764
Vitamin C (mg) 1.17 0.06 1.17 0.06 1.12 0.06 0.3399 Model 3 1.0 1.18 (0.951.47) 1.12 (0.891.40) 0.5917
Values are expressed as mean SD adjusted for sex, age, and energy intake. Mean OR (95% CI), ORs from the Tertile 2, Tertile 3 relative to the Tertile 1. p from multiple logis-
values in the row on the coffee intake quartile are signicantly different (*p b 0.05, tic regression analysis p b 0.05.
**p b 0.01, ***p b 0.001) by ANOVA. a, b, mean values with different superscripts are a
Model 1: crude; Model 2: adjusted for age, sex, and energy intake; Model 3: adjusted
signicantly different by Tukey's multiple rang test. for age, sex, energy intake, smoking, and drinking.
 F. Song et al. / NFS Journal 4 (2016) 914
consumption, the results were identical to YY et al.'s research [15],
which reported a higher intake of meat, sugar, fat, and oils by the high
coffee intake group. However, there is a need for more diet surveys on
choice and preference of food consumed.
Additionally, INQ evaluates the quality of a meal using a ratio divid-
ing nutrient content in the body by a recommended intake per
1000 kcal of that nutrient. The results were similar to the results of YJ
& MH's research [18], where a high INQ of vitamin B1 and vitamin B2
was found in a low coffee consumption group. Additionally, a low INQ
of protein was found in a low coffee consumption group in YY et al.'s
study [15]. This result can be observed because the no coffee intake
group consumed high of grains, milk, and dairy products; therefore, in-
take of protein and vitamin B1, which are contained in those food
groups, also increased.
This study analyzed the inuence of coffee consumption on meta-
bolic syndrome and risk factors of metabolic syndrome through cross-
sectional study. Although prior research has studied the relevance of
coffee intake and index related to metabolic disease, a denite associa-
tion between coffee consumption and BMI could not be found in the
existing literature [10,28] related to coffee consumption and metabolic
syndrome. Lisanne and et al.'s research [29] showed a lower level of
waist circumference was associated with higher coffee consumption, al-
though results were not signicant. In this study, a signicant relation-
ship between coffee intake and BMI was not found as it was in the
previous study, but the relationship between coffee consumption and
abdominal obesity was found.
Several studies have reported that there is no clear relationship be-
tween coffee consumption and blood pressure [30,31], and there is
also research suggesting the opposite. The analysis of results from this
study showed that there is lower risk of high blood pressure if the
level of coffee consumption is higher. Coffee drinking increased BP in
non-habitual drinkers but not in habitual coffee drinkers [32]. In
meta-analyses conducted by Arthur et al. [33] and Steffen [34], long-
term coffee consumption was not associated with a clear increase in
BP or risk of hypertension. In this study, results also showed that the
risk of high fasting blood glucose was lower in the high coffee consump-
tion group, and the risk of high fasting blood glucose decreased signi-
cantly as coffee intake increased. Recently published research found
that consuming 5 cups of coffee per day decreased insulin resistance
[35], and that coffee consumption may reduce the risk of type 2 diabetes
mellitus and hypertension [36]. It was suggested in previous studies
that there is a visible preventive effect on diabetes, which is closely re-
lated to metabolic syndrome [7,13,37]. The research showed that the
risk of high fasting blood decreased if coffee intake was higher, which
was also consistent with this study.
In a meta-analysis [38,39], the intake of coffee, especially unltered
coffee, contributed signicantly to the increase in TC, LDL-C, and TG.
However, in this study, we cannot nd a signicant association between
coffee consumption and LDL-C or TG. A signicant negative (benecial)
association between coffee consumption and prevalence of metabolic
syndrome was observed both in men and women [39], while in
studies conducted by Lisanne et al. [29] and Lutsey et al. [17], no signif-
icant relationship between coffee intake and incidence of metabolic
syndrome was determined. Similarly, no signicant relationship be-
tween coffee consumption and risk of metabolic syndrome was shown
in this study.
In a meta-analysis study [40], coffee consumption may have an effect
to be favorably related with health such like to reduce risk of diabetes
and to protect by metabolic impairment. Our study found that, more
grains and milk, dairy products were consumed in the no coffee con-
sumption group, and coffee intake have not a very clear association
with food group and nutrition intake. No clear relationship was found
between that and the risk of metabolic syndrome or intake of coffee
can potentially help in preventing the occurrence of chronic diseases,
such as abdominal obesity, high blood pressure, and high fasting blood
glucose.
13
In conclusion, coffee consumption has not a clear association with
food group intake, nutrient intake, INQ. Appropriate consumption of
coffee may have potentially helpful effects on some of the metabolic
risk factors, such as abdominal obesity, hypertension and high glucose.
However, in this study, we did not nd a signicant relationship be-
tween coffee consumption and metabolic syndrome. As a widely con-
sumed beverage in South Korea, coffee can be included as part of a
healthy diet for the general public and also for those with increased
metabolic syndrome risk factors. Due to methodological limitations,
we cannot exclude that the observed associations on coffee consump-
tion and metabolic syndrome are due to other healthy lifestyle behav-
iors, and further prospective studies are needed that better adjust for
potential confounding factors. In the future, we need to collect data on
dietary intake and conduct research using the cohort method.
Conict of interest
The authors declare that they have no conict of interest.
References
[1] Y.G. Oh, K.J. Kim, A study of the effect of food-related lifestyle on coffee consumption
behavior, Asia-Pac. J. Bus. Ventur. Entrep. 7 (2012) 6575.
[2] J.W. Shin, Effect of Sustainability Management on Consumers Behavioral Intention
at Coffee Houses and Korean Status of Coffee Intake(MS Thesis) Seoul National Uni-
versity, Seoul, Korea, 2012 2526.
[3] Ministry of Health and Welfare, Korea Centers for Disease Control and Prevention,
Korea Health Statistics 2010: Korea National Health and Nutrition Examination Sur-
vey (KNHANES V-1), 2011.
[4] Korea Centers for Disease Control and Prevention &; Korea Health Industry
Development Institute, The Fourth Korea Health and Nutrition Examination Survey
Manual (20072009). Seoul, Korea, 2010.
[5] T. Ranheim, B. Halvorsen, Coffee consumption and human healthbenecial or det-
rimental? Mechanisms for effects of coffee consumption on different risk factors for
cardiovascular disease and type 2 diabetes mellitus, Mol. Nutr. Food Res. 49 (2005)
274284.
[6] A. Hino, H. Adachi, M. Enomoto, K. Furuki, Y. Shifetoh, M. Ohtsuka, S.I. Kumagae, Y.
Hirai, A. Jalaldin, A. Satoh, T. Imaizumi, Habitual coffee but not green tea consump-
tion is inversely associated with metabolic syndrome: an epidemiological study in a
general Japanese population, Diabetes Res. Clin. Pract. 76 (2007) 383389.
[7] E.K. Kim, J.S. Choe, E.K. Kim, Correlation of nutrient intake, obesity-related anthropo-
metrics, and blood lipid status with instant coffee-mix intakes in Gangneung and
Samcheok residents, Korean J. Commun. Nutr. 18 (2013) 134141.
[8] H. Takami, M. Nakamoto, H. Uemura, S. Katsuura, M. Yamaguchi, M. Hiyoshi, F.
Sawachika, T. Juta, K. Arisawa, Inverse correlation between coffee consumption
and prevalence of metabolic syndrome: baseline survey of the Japan multi-institu-
tional collaborative cohort (J-MICC) study in Tokushima, Japan, J. Epidemiol. 23
(2013) 1220.
[9] H. Matsuura, K. Mure, N. Nishio, N. Kitano, N. Nagai, T. Takeshita, Relationship be-
tween coffee consumption and prevalence of metabolic syndrome among
Japanese civil servant, J. Epidemiol. 22 (2012) 160166.
[10] M.T. Driessen, L.L. Koppes, L. Veldhuis, D. Samoocha, J.W.R. Twisk, Coffee consump-
tion is not related to the metabolic syndrome at the age of 36 years: the Amsterdam
growth and health longitudinal study, Eur. J. Clin. Nutr. 63 (2009) 536542.
[11] R.M. VanDam, Coffee consumption and risk of type 2 diabetes, cardiovascular dis-
eases, and cancer, Appl. Physiol. Nutr. Metab. 33 (2008) 12691283.
[12] S. van Dieren, C.S. Uiterwaal, Y.T. Schouw, A. D.L., J.M.A. Boer, A. Spijkerman, D.E.
Grobbee, J.W.J. Beulens, Coffee and tea consumption and risk of type 2 diabetes,
Diabetologia 52 (2009) 25612569.
[13] R. Huxley, C.M.Y. Lee, F. Barzi, L. Timmermeister, S. Czernichow, V. Perkovic, D.E.
Grobbee, D. Batty, M. Woodward, Coffee, decaffeinated coffee, and tea consumption
in relation to incident type 2 diabetes mellitus: a systematic review with meta-
analysis, Arch. Intern. Med. 169 (2009) 20532063.
[14] S.Y. Lee, I.K. Hwang, M.H. Park, H.S. Seo, Sensory characteristics of diluted espresso
(Americano) in relation to dilution rates, Korean J. Food Cookery Sci. 23 (2007)
839847.
[15] Y.Y. Jee, Y.Y. Bae, M.H. Kim, H.K. Jo, E.Y. Kim, Evaluation of nutrient intake and bone
status of female college students according to the calorie consumption from coffee
containing beverage, KFN 22 (2009) 430442.
[16] K. Vynvke, E.C. Fernandez, F.P. Marta, C.G. Magdalena, W.D. Keyzer, G.G. Marcela,
et al., Validation of the diet quality index for adolescents by comparison with bio-
markers, nutrition and food intakes: the HELENA study, BJN 109 (2013) 20672078.
[17] P.L. Lutsey, L.M. Steffen, J. Stevents, Dietary intake and the development of the met-
abolic syndrome: the atherosclerosis risk in communities study, Circulation 117
(2008) 754761.
[18] Y.J. Bae, M.H. Kim, A study evaluating nutrient intake and diet quality in female col-
lege students according to coffee consumption, J Korean Diet. Assoc. 15 (2009)
128138.
14 F. Song et al. / NFS Journal 4 (2016) 914
[19] K.J. Yun, J.D. Han, M.K. Kin, Y.M. Park, K.H. Baek, K.H. Song, K. Kil, H.S. Kwon, Effect of
maternal age at children on insulin resistance: the 2010 Korean National Health and
nutrition examination survey, Clin. Endocrinol. 82 (2015) 824830.
[20] C.S. Sim, Y. Kim, H. Lee, C.Y. Park, J.O. Ham, B.K. Lee, Iron deciency increase blood
lead levels in boys and per-menarche girls surveyed in KNHANES 20102011, Envi-
ron. Res. 130 (2014) 16.
[21] Korea Centers for Disease Control and Prevention, The Fourth Korea National Health
and Nutrition Examination Survey 20072009 (KNHANES IV), Seoul, Korea, 2010.
[22] S.M. Grundy, J.I. Cleeman, S.R. Daniels, K.A. Donato, R.H. Eckel, B.A. Franklin, D.J.
Gordon, R.M. Krauss, P.J. Savage, S.C. Smith, J.A. Spertus, C. Fernando, Diagnosis and
management of the metabolic syndrome: an American heart association/national
heart, lung, and blood institute scientic statement: executive summary, Circulation
112 (2005) 27352752.
[23] K.W. Lee, Understanding the Cognitive Structures of the Koreans toward Korean
Food and Korean-Style Dietary Patterns(PhD Dissertation) Ewha Womans Universi-
ty, Seoul, Korea, 2013 2526.
[24] K.S. Lim, Nutrient density and nutrient-cost evaluation for the representative menus
of the school lunch program in Korea, Korean J. Nutr. 30 (1997) 12441257.
[25] R.H. Eckel, S.M. Grundy, P.Z. Zimmet, The metabolic syndrome, Lancet 365 (2005)
14151428.
[26] C. Bamia, P. Lagiou, M. Jenab, A. Trichopoulou, V. Fedirko, K. Aleksandrova, et al.,
Coffee, tea and decaffeinated coffee in relation to hepatocellular carcinoma in a
European population: multicentre, prospective cohort study, IJC 136 (2015)
18991905.
[27] H.J. Kim, Green Tea and Coffee Consumption and Metabolic Conditions in Korean
Adults(MS Thesis) Department of Food and Nutrition Graduate School, Yeungnam
University, Gyeongsan-si, Korea, 2013 5560.
[28] D.R. Bouchard, R. Ross, L. Janssen, Coffee, tea and their additives: association with
BMI and waist circumference, Eur. J. Obes. 3 (2010) 345352.
[29] L. Balk, T. Hoekstra, J. Twisk, Relationship between long-term coffee consumption
and consumption of the metabolic syndrome: the Amsterdam growth and health
longitudinal study, Eur. J. Epidemiol. 24 (2009) 203209.
[30] M.K. Choi, Y.S. Lee, The relationships among coffee consumption, blood pressure,
and serum lipids in Korean adult men and women, Korean J. Food Nutr. 20 (2007)
460466.
[31] J.M. Geleijnse, Habitual coffee consumption and blood pressure: an epidemiological
perspective, Vasc. Health Risk Manag. 4 (2008) 963970.
[32] Y. Je, S. Jeong, T. Park, Coffee consumption patterns in Korean adults: the Korean
National Health Nutrition Examination Survey (20012011), Asia Pac. J. Clin. Nutr.
23 (2014) 691702.
[33] A.E. Mesas, L.M. Luz, R.A. Fernando, L.G. Esther, The effect of coffee on blood pressure
and cardiovascular disease in hypertensive individuals: a systematic review and
meta-analysis, Am. J. Clin. Nutr. 94 (2011) 11131126.
[34] S. Mark, K. Carol, H. Donald, E. Patricia, M. Mohammad, The effect of coffee con-
sumption on blood pressure and the development of hypertension: a systematic re-
view and meta-analysis, J. Hypertens. 30 (2012) 22452254.
[35] N.W. Wedick, A.M. Brennan, Q. Sun, F.B. Hu, C.S. Mantzoros, D.M. Rob, Effects of caf-
feinated and decaffeinated coffee on biological risk factors for type 2 diabetes: a ran-
domized controlled trial, Nutr. J. 10 (2011) 93101.
[36] J.H. O'Keefe, S.K. Bhatti, H.R. Patil, J.J. DiNicolantonio, S.C. Lucan, C.J. Lavie, Effect of
habitual coffee consumption on cardiometabolic disease, cardiovascular health,
and all-cause mortality, J. Am. Coll. Cardiol. 62 (2013) 10431051.
[37] J. Tuomilehto, G. Hu, S. Bidel, J. Lindstrom, P. Jousilahti, Coffee consumption and risk
of type 2 diabetes mellitus among middle-aged nish men and women, JAMA 291
(2004) 12131219.
[38] L. Cai, D. Ma, Y. Zhang, Z. Liu, P. Wang, The effect of coffee consumption on serum
lipids: a meta-analysis of randomized controlled trials, Eur. J. Clin. Nutr. 66 (2012)
827.
[39] G. Grosso, U. Stepaniak, A. Micek, R. Topor-Madry, H. Pikhart, K. Szafraniec, A. Pajak,
Association of daily coffee and tea consumption and metabolic syndrome: results
from the polish arm of HAPIEE study, Eur. J. Nutr. 54 (2014) 11291137.
[40] S. Buscemi, S. Marventano, M. Antoci, A. Cagnetti, G. Castorina, F.,.M. Marranzano, A.
Mistretta, Coffee and metabolic impairment: an undated review of epidemiological
studies, NFS J. 3 (2016) 17.