Lee et al.

BMC Endocrine Disorders (2018) 18:33

https://doi.org/10.1186/s12902-018-0259-x

RESEARCH ARTICLE Open Access

The product of fasting plasma glucose and

triglycerides improves risk prediction of
type 2 diabetes in middle-aged Koreans
Joung-Won Lee1,2, Nam-Kyoo Lim1 and Hyun-Young Park1*

Abstract
Background: Screening for risk of type 2 diabetes mellitus (T2DM) is an important public health issue. Previous studies
report that fasting plasma glucose (FPG) and triglyceride (TG)-related indices, such as lipid accumulation product (LAP)
and the product of fasting glucose and triglyceride (TyG index), are associated with incident T2DM. We
aimed to evaluate whether FPG or TG-related indices can improve the predictive ability of a diabetes risk
model for middle-aged Koreans.
Methods: 7708 Koreans aged 40–69 years without diabetes at baseline were eligible from the Korean Genome and
Epidemiology Study. The overall cumulative incidence of T2DM was 21.1% (766 cases) in men and 19.6% (797 cases) in
women. Therefore, the overall cumulative incidence of T2DM was 20.3% (1563 cases). Multiple logistic regression
analysis was conducted to compare the odds ratios (ORs) for incident T2DM for each index. The area under the
receiver operating characteristic curve (AROC), continuous net reclassification improvement (cNRI), and integrated
discrimination improvement (IDI) were calculated when each measure was added to the basic risk model for diabetes.
Results: All the TG-related indices and FPG were more strongly associated with incident T2DM than WC in our study
population. The adjusted ORs for the highest quartiles of WC, TG, FPG, LAP, and TyG index compared to the lowest,
were 1.64 (95% CI, 1.13–2.38), 2.03 (1.59–2.61), 3.85 (2.99–4.97), 2.47 (1.82–3.34), and 2.79 (2.16–3.60) in men, and
1.17 (0.83–1.65), 2.42 (1.90–3.08), 2.15 (1.71–2.71), 2.44 (1.82–3.26), and 2.85 (2.22–3.66) in women, respectively.
The addition of TG-related parameters or FPG, but not WC, to the basic risk model for T2DM (including age,
body mass index, family history of diabetes, hypertension, current smoking, current drinking, and regular
exercise) significantly increased cNRI, IDI, and AROC in both sexes.
Conclusions: Adding either TyG index or FPG into the basic risk model for T2DM increases its prediction and
reclassification ability. Compared to FPG, TyG index was a more robust T2DM predictor in the stratified sex
and fasting glucose level. Therefore, TyG index should be considered as a screening tool for identification
of people at high risk for T2DM in practice.
Keywords: TyG index, Type 2 diabetes mellitus, Risk model

* Correspondence: [email protected]
1
Division of Cardiovascular Diseases, Center for Biomedical Sciences, Korea
National Institute of Health, 187 Osongsaengmyeng 2-ro, Osong-eup,
Heungdeok-gu, Cheongju-si, Chungcheongbuk-do 361-951, South Korea
Full list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 2 of 10

Background the prediction of risk of future T2DM. To date, few

Type 2 diabetes mellitus (T2DM) is one of the most studies have been undertaken in Korea that compare
prevalent non-communicable diseases in the middle-aged the predictive ability for incident T2DM of the sim-
population worldwide, largely because of recent ple model and composite models, which include
changes in diet and lifestyle [1, 2]. In the Korean blood test results [11, 22]. Recently, a risk model for
National Health and Nutrition Examination Surveys T2DM that included blood test results was proposed
(KNHANES), the prevalence of diabetes among adults based on cohort data, and its reclassification ability
aged 30 and over was found to have slightly in- was significantly improved when glycated
creased, from 12.4% in 2011 to 13.0% in 2014 [3, 4]. hemoglobin (HbA1c) was included [11]. However,
According to an estimate by the International Dia- HbA1c is not assessed in the routine health examin-
betes Federation, 46.3% of cases of diabetes in Ko- ation and this study did not consider reclassification
reans aged 20–79 were undiagnosed in 2015 [5]. ability when other blood test results apart from
However, lifestyle intervention can reduce the risk of HbA1c were included in the T2DM risk model [23].
incident T2DM and mortality in individuals at high Therefore, in the present study, we aimed to iden-
risk of diabetes [6, 7]. Therefore, it is important to tify which of the TG-related indices that can be de-
screen high-risk individuals for T2DM regularly to rived from general check-up data would improve the
ensure early diagnosis. For this reason, risk models prediction ability of the simple T2DM risk model in
for diabetes have been proposed in previous studies, middle-aged Koreans.
and, recently, T2DM risk prediction models have been
reported in Korea [8–14]. Methods
Obesity is the most significant risk factor for inci- Study population
dent T2DM [15, 16]. Body mass index (BMI) has The Korean Genome and Epidemiology Study
been used as a surrogate marker for obesity and in- (KoGES) consists of a gene-environment model and
cluded as one of the variables in most risk models for population-based studies [24]. KoGES: Ansan and
T2DM [9, 11–14]. However, BMI does not reflect Ansung study is an ongoing prospective cohort study
central obesity. Lee et al. selected waist circumference conducted in urban (Ansan) and rural (Ansung)
(WC) instead of BMI in their diabetes risk model areas in Korea with biennial follow-ups, which
considering its association with diabetes [10]. In the started in 2001. 10038 people underwent an initial
systematic review, more than 30% of the diabetes risk examination, and 9001 subjects were included after
models stating its components included both BMI exclusion of 1037 who refused to participate or died.
and WC [8]. To improve the prediction ability of a Thirty-five participants were not suitable for the
risk model for incident T2DM, blood parameters are present study because of their age, and subjects with
also frequently included [8]. a history of diabetes at baseline or incomplete data
Also, the increase of fasting plasma glucose (FPG) were also excluded. Finally, 7708 people aged 40 to
in the normal range is associated with increased in- 69 years remained eligible for the current study.
cident T2DM [17]. Of these, serum triglyceride (TG) Written informed consent was obtained from all
has been used to identify people at high risk for subjects. The Institutional Review Board of the Ko-
T2DM, alongside obesity [18]. In addition, lipid ac- rean Centers for Disease Control and Prevention ap-
cumulation product (LAP) and the product of fasting proved the study protocol.
plasma glucose and triglyceride (TyG index), com-
posite indices including TG, have been proposed as Measurements and surveys
predictors of T2DM [19, 20]. In particular, TyG Height and weight were measured to the nearest
index has been used as a marker of insulin resist- 0.1 cm and 0.1 kg using a digital stadiometer and a
ance [19]. Although the simple diabetes risk model scale, respectively. Resting blood pressure while sit-
is convenient for self-assessment, more accurate pre- ting was measured by trained technicians using a
diction models that include blood parameters are standard mercury sphygmomanometer. Blood sam-
also required to facilitate more accurate clinical con- ples were collected after fasting for at least 8 h. The
sultations [8]. In Korea, most people are registered Friedewald formula was used to indirectly estimate
with the National Health Insurance (NHI), which low-density lipoprotein cholesterol levels in subjects
provides biannual medical check-ups for middle-aged with plasma TG < 400 mg/dl [25]. Diabetes was de-
people, including the measurement of key blood pa- fined by FPG > 126 mg/dl, 2 h post-challenge plasma
rameters [21]. Therefore, risk models for incident glucose > 200 mg/dl, HbA1c > 6.5%, or prescription
T2DM that are based on the data obtained from for anti-diabetic medication [26]. Subjects were
these medical check-ups would be of great use for questioned by trained interviewers regarding their
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 3 of 10

socio-demographics, family history of diabetes, and Macros were used to calculate cNRI and IDI, and data
lifestyle factors including smoking and alcohol con- analysis was performed using SAS 9.4 and MedCalc [32].
sumption. The subjects’ smoking and alcohol con-
sumption status was subdivided according to their Results
past and present habits. Regular exercise was defined Baseline characteristics
as subjects’ exercise was over 90 min as the sum of Table 1 indicates the baseline characteristics of the study
moderate and vigorous physical activity a day [27]. subjects. The ages of the subjects were 51.4 ± 8.6 years
LAP and TyG index were calculated as follows: for men and 52.0 ± 8.9 years for women. The prevalence
of hypertension was higher in men than in women
(29.5% vs. 27.0%, P = 0.0149). By contrast, the percentage
LAP for men ¼ ½WCðcmÞ‐65  TGðmmol=LÞ
of subjects with a family history of diabetes was higher
LAP for women ¼ ½WCðcmÞ‐58  TGðmmol=LÞ
in women than in men (11.0% vs. 9.2%, P = 0.0067).
TyG index ¼ ln ½TGðmg=dlÞ  FPGðmg=dlÞ=2
There were also significant differences in anthropomet-
ric indices between sexes. Mean BMI and WC were 24.1
± 2.9 kg/m2 and 83.3 ± 7.6 cm in men, and 24.7 ± 3.2 kg/
Statistical analysis m2 and 81.2 ± 9.5 cm in women, respectively. Mean TG,
Data are expressed as numbers and proportions for FPG, LAP, and TyG index were 170.8 ± 111.6 mg/dl,
discrete variables and mean ± SD for continuous vari- 84.5 ± 9.0 mg/dl, 38.1 ± 34.0, and 8.7 ± 0.5 in men, and
ables, and were analyzed using Chi-square and Student’s 140.9 ± 75.6 mg/dl, 81.1 ± 7.7 mg/dl, 39.0 ± 30.6, and 8.5
t-tests, respectively. ± 0.5 in women, respectively. TG-related indices also
Based on the maximized Youden index, we calcu- showed significant differences between the sexes, with
lated sex-specific cut-off points of each index for the exception of LAP.
T2DM. Multiple logistic regression analysis was con-
ducted and adjusted for age, BMI, hypertensive sta- Table 1 Characteristics of the study population at baseline
tus, family history of diabetes, current smoking and Variables Men Women P-
value
alcohol consumption status, and regular exercise. (n = 3636) (n = 4072)
The basic model was derived from the published Age, years 51.4 ± 8.6 52.0 ± 8.9 0.0011
diabetes risk model for middle-aged Koreans and in- Current Smoker, n(%) 1753 (48.2) 142 (3.5) <.0001
cluded the variables listed above [10, 11]. We tested Current Drinker, n(%) 2580 (71.0) 1085 (26.7) <.0001
multicollinearity for all covariates based on the vari-
Hypertension, n(%) 1074 (29.5) 1101 (27.0) 0.0149
able inflation factor (VIF). The area under the re-
Family history of diabetes, n(%) 333 (9.2) 449 (11.0) 0.0067
ceiver operating characteristic curve (AROC) was
calculated for each risk model of incident T2DM, in- Regular exercise, n(%) 1712 (47.1) 1654 (40.6) <.0001
dicating the diagnostic power of each model for inci- BMI, kg/m2 24.1 ± 2.9 24.7 ± 3.2 <.0001
dent T2DM during the follow-up period [28]. WC, cm 83.3 ± 7.6 81.2 ± 9.5 <.0001
Differences in AROC between the basic model and TG, mg/dl 170.8 ± 111.6 140.9 ± 75.6 <.0001
each composite model were analyzed using the
TyG index 8.7 ± 0.5 8.5 ± 0.5 <.0001
method of DeLong et al. [29]. Pencina et al. have
LAP 38.1 ± 34.0 39.0 ± 30.6 0.1925
suggested category-based net reclassification im-
provement (NRI) and integrated discrimination im- SBP, mmHg 121.2 ± 16.6 119.9 ± 19.1 0.0029
provement (IDI) for calculating the usefulness of a DBP, mmHg 81.6 ± 10.8 78.5 ± 11.6 <.0001
new marker in prediction models [30]. The FPG, mg/dl 84.5 ± 9.0 81.1 ± 7.7 <.0001
category-based NRI measures the accuracy of reclas- 2hPG, mg/dl 110.5 ± 32.0 117.2 ± 28.3 <.0001
sification based on how well the subjects are reclas-
Total cholesterol, mg/dl 190.3 ± 34.6 188.9 ± 33.9 0.0670
sified as upwards for events and downwards for
HDL-C, mg/dl 43.8 ± 10.0 46.0 ± 10.0 <.0001
non-events. However, the category-based NRI can be
affected by the number and choice of categories LDL-C, mg/dl 112.4 ± 34.2 114.7 ± 30.7 0.0017
[31]. The continuous (category-free) NRI (cNRI) is an ex- HbA1c, mg/dl 5.6 ± 0.3 5.5 ± 0.4 0.1636
panded method to solve limitation of the categories. They P values are from t-tests or chi-square tests for analysis of variance for
also proposed the IDI that calculates the extent of average continuous variables and categorical variables
Abbreviations: BMI body mass index, WC waist circumference, TG triglycerides;
sensitivity and ‘1-specificity’ when a new marker is added TyG index, the product of fasting glucose and triglycerides; LAP lipid
to the basic model [30]. We calculated the cNRI and IDI accumulation product, SBP systolic blood pressure, DBP diastolic blood
pressure, FPG fasting plasma glucose, HDL-C high-density lipoprotein
to compare the prediction and reclassification abilities of cholesterol, LDL-C low density lipoprotein cholesterol, HbA1c
each measure when added to the basic model of diabetes. glycated hemoglobin
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 4 of 10

Incidence of T2DM according to each index category 0.599, 0.623, and 0.644 in women respectively. The
During the 10 year follow-up, the overall cumula- cut-off points for predicting T2DM were 84.00 cm,
tive incidence of T2DM was 21.1% (766 cases) in 172.00 mg/dl, 87.00 mg/dl, 30.50, and 8.86 in men
men and 19.6% (797 cases) in women. Table 2 and 78.17 cm, 122.00 mg/dl, 84.00 mg/dl, 35.84, and
shows the overall cumulative incidence of T2DM, 8.52 in women for WC, TG, FPG, LAP, and TyG
categorized by quartiles for each index. In men, the index, respectively.
cumulative incidences of T2DM across the quar-
tiles of TyG index were (lowest-highest) 13.3 and
31.5% and those of FPG were 12.9 and 35.9% and Odds ratios for incident T2DM for each composite
those of WC were 16.4 and 27.7%. In women, the predictive model
values for TyG index were 11.1 and 30.9% and Table 4 shows the odds ratio (OR) for incident
those FPG were 13.9 and 28.4% and those WC T2DM in higher quartiles compared to the first
were 13.4 and 24.5%. The increase in the cumula- quartile of each index. The unadjusted ORs for all
tive incidence of T2DM with higher category of TG-related indices for incident T2DM were higher
WC was less marked than that with increasing TyG than that of WC. These trends were similar after ad-
index or FPG. justment for age, BMI, hypertensive status, family
history of diabetes, smoking, alcohol consumption
status and regular exercise. When the highest quar-
Cut-off points of each index for predicting T2DM tile for each index was compared to the lowest, the
Table 3 shows the AROC values and cut-off points of adjusted ORs of WC, TG, FPG, LAP, and TyG index
the indices for predicting T2DM. The AROCs for were 1.64 (95% confidence interval (CI), 1.13–2.38),
WC, TG, FPG, LAP, and TyG index were 0.579, 2.03 (1.59–2.61), 3.85 (2.99–4.97), 2.47 (1.82–3.34),
0.592, 0.660, 0.602, and 0.623 in men 0.576, 0.627, and 2.79 (2.16–3.60) in men, and 1.17 (0.83–1.65),

Table 2 Number of incident type 2 diabetes cases according to the quartiles for each measure
Categories Men (n = 3636) Women (n = 4072)
Quartile Diabetes (%) Quartile Diabetes (%)
WC (cm) Q1 (< 78.0) 143 (16.4) Q1 (< 74.0) 128 (13.4)
Q2 (78.0–83.2) 162 (17.6) Q2 (74.0–80.6) 198 (18.1)
Q3 (83.3–88.2) 207 (22.3) Q3 (80.7–87.8) 222 (22.2)
Q4 (≥88.3) 254 (27.7) Q4 (≥87.9) 249 (24.5)
TG (mg/dl) Q1 (< 106.0) 137 (15.2) Q1 (< 92.5) 126 (12.4)
Q2 (106.0–142.9) 167 (18.4) Q2 (92.5–121.9) 136 (13.6)
Q3 (143.0–200.9) 196 (21.5) Q3 (122.0–165.9) 225 (21.9)
Q4 (≥201.0) 266 (29.0) Q4 (≥166.0) 310 (30.3)
FPG (mg/dl) Q1 (< 78.0) 106 (12.9) Q1 (< 76.0) 133 (13.9)
Q2 (78.0–82.9) 132 (14.7) Q2 (76.0–79.9) 143 (16.3)
Q3 (83.0–89.9) 190 (19.5) Q3 (80.0–84.9) 206 (18.3)
Q4 (≥90.0) 338 (35.9) Q4 (≥85.0) 315 (28.4)
LAP Q1 (< 16.7) 137 (15.1) Q1 (< 18.8) 126 (12.4)
Q2 (16.7–29.4) 140 (15.4) Q2 (18.8–30.7) 167 (16.4)
Q3 (29.5–49.1) 212 (23.3) Q3 (30.8–50.1) 190 (18.7)
Q4 (≥49.2) 277 (30.5) Q4 (≥50.2) 314 (30.8)
TyG index Q1 (< 8.4) 121 (13.3) Q1 (< 8.2) 113 (11.1)
Q2 (8.4–8.6) 153 (16.8) Q2 (8.2–8.4) 140 (13.8)
Q3 (8.7–9.0) 206 (22.7) Q3 (8.5–8.7) 229 (22.5)
Q4 (≥9.1) 286 (31.5) Q4 (≥8.8) 315 (30.9)
Abbreviations: WC waist circumference, TG triglycerides, FPG fasting plasma glucose, LAP lipid accumulation product, TyG index the product of fasting glucose
and triglycerides
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 5 of 10

Table 3 The area under the ROC curve (AROC) and cut-off points for indices to predict type 2 diabetes
Index AROC (95% CI) Cut-off point Sensitivity (%) Specificity (%) Youden index
Men
WC (cm) 0.579 (0.563–0.595) 84.00 56.01 57.11 0.13
TG (mg/dl) 0.592 (0.576–0.608) 172.00 46.74 67.84 0.15
FPG (mg/dl) 0.660 (0.645–0.676) 87.00 51.31 72.89 0.24
LAP 0.602 (0.586–0.618) 30.50 62.79 55.54 0.18
TyG index 0.623 (0.607–0.638) 8.86 52.09 66.59 0.19
Women
WC (cm) 0.576 (0.561–0.592) 78.17 69.26 43.54 0.13
TG (mg/dl) 0.627 (0.612–0.642) 122.00 66.37 54.41 0.21
FPG (mg/dl) 0.599 (0.584–0.614) 84.00 39.52 75.69 0.15
LAP 0.623 (0.607–0.637) 35.84 57.59 61.59 0.19
TyG index 0.644 (0.629–0.659) 8.52 67.25 55.85 0.23
Abbreviations: AROC area under the receiver operating characteristic curve, WC waist circumference, TG triglycerides, FPG fasting plasma glucose, LAP lipid
accumulation product, TyG index, the product of fasting glucose and triglycerides

2.42 (1.90–3.08), 2.15 (1.71–2.71), 2.44 (1.82–3.26), practice because of measurement inaccuracy [34].
and 2.85 (2.22–3.66) in women, respectively. The Among the TG-related indices analyzed in the
calculated VIF for all covariates in the multivariate present study, the AROC and reclassification ability
model, were below 5.0, indicating no severe multi- of TyG index were higher than for TG or LAP.
collinearity among covariates [33]. Although the inclusion of LAP also improved the
predictive ability of the risk model for incident dia-
Effects of the addition of each index to the basic model betes, this required the additional inclusion of WC,
of T2DM on cNRI, IDI, and AROC while this was not required to demonstrate im-
Table 5 shows the reclassification and discrimination provements in the risk model by the inclusion of
abilities when each measure was added to the basic TG or TyG index [20]. Most studies that used TyG
model of diabetes. The AROCs for the addition of index to predict incident T2DM used it as a surro-
TyG index or FPG to the basic model were larger gate for insulin resistance, and reported that its
than those for the other indices in both sexes. The predictive ability is better than that of the homeo-
cNRI for TG, FPG, LAP, and TyG index were 22.7% stasis model of assessment (HOMA-IR) [19, 35, 36].
(P < 0.0001), 48.0% (P < 0.0001), 23.6% (P < 0.0001), However, Abbasi and Reaven showed that the cor-
and 38.7% (P < 0.0001) in men, and 28.6% (P < relation between insulin-mediated glucose uptake
0.0001), 21.3% (P < 0.0001), 21.3% (P < 0.0001), and (IMGU) and TyG index is not better than that be-
36.0% (P < 0.0001) in women, respectively. By con- tween IMGU and TG or HOMA-IR [37, 38]. Never-
trast, the addition of WC made cNRI only 5.6% (P = theless, it was noted that TyG index is a practical
0.1689), in men and − 2.3% (P = 0.5666), in women. measure for use in T2DM prediction because of its
The IDI was also higher when TG-related indices or cost-efficiency [37]. Moreover, the use of HbA 1c
FPG was added than when WC was added to the has been recommended for diagnosis and screening
basic model of T2DM in both sexes. of patients. Lim et al. added HbA 1c to a diabetes
risk model and demonstrated an increase in pre-
Discussion dictive ability for T2DM in a middle-aged Korean
We showed that the predictive ability of the simple cohort, while Ahn et al. also demonstrated through
T2DM risk model was increased when TG-related longitudinal validation analysis that adding FPG or
indices or FPG was added. Recently, several dia- HbA 1c to the simple diabetes risk model also im-
betes risk models have been proposed to screen proves its predictive ability [11, 22]. Thus, in gen-
high-risk groups, some of which included blood pa- eral, the predictive ability of a diabetes risk model
rameters. WC was included as one of the compo- that includes laboratory parameters such as TG,
nents in the model because increased WC increases FPG, or HbA 1c is better than that of the simple
T2DM risk [8]. However, WC is not easily used in diabetes risk model [11, 14, 22].
Table 4 Multiple logistic regression analyses for each measure in predicting type 2 diabetes (odds ratios and 95% confidence intervals)
Lee et al. BMC Endocrine Disorders (2018) 18:33

Categories Unadjusted Adjusteda)

WC TG FPG LAP TyG index WC TG FPG LAP TyG index
ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI)
Men
Q1 Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref.
Q2 1.09 (0.85–1.39) 1.26 (0.99–1.62) 1.16 (0.88–1.53) 1.03 (0.79–1.33) 1.32 (1.02–1.70) 1.07 (0.81–1.40) 1.21 (0.94–1.55) 1.18 (0.89–1.56) 1.04 (0.79–1.36) 1.26 (0.97–1.64)
Q3 1.46 (1.15–1.85) 1.54 (1.21–1.95) 1.64 (1.27–2.12) 1.71 (1.35–2.18) 1.91 (1.49–2.45) 1.35 (1.00–1.83) 1.39 (1.08–1.79) 1.68 (1.29–2.19) 1.70 (1.28–2.25) 1.82 (1.41–2.36)
Q4 1.95 (1.55–2.46) 2.29 (1.82–2.88) 3.78 (2.96–4.82) 2.47 (1.96–3.11) 2.99 (2.36–3.79) 1.64 (1.13–2.38) 2.03 (1.59–2.61) 3.85 (2.99–4.97) 2.47 (1.82–3.34) 2.79 (2.16–3.60)
Women
Q1 Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref. Ref.
Q2 1.43 (1.12–1.82) 1.11 (0.86–1.44) 1.20 (0.93–1.55) 1.39 (1.08–1.78) 1.28 (0.98–1.66) 1.14 (0.88–1.48) 1.02 (0.78–1.33) 1.18 (0.91–1.53) 1.26 (0.97–1.64) 1.19 (0.91–1.55)
Q3 1.86 (1.46–2.36) 1.98 (1.56–2.51) 1.38 (1.09–1.75) 1.63 (1.27–2.07) 2.32 (1.82–2.97) 1.27 (0.95–1.69) 1.69 (1.32–2.16) 1.32 (1.04–1.68) 1.35 (1.03–1.78) 1.97 (1.53–2.53)
Q4 2.10 (1.66–2.66) 3.07 (2.44–3.87) 2.45 (1.95–3.06) 3.15 (2.51–3.97) 3.59 (2.83–4.55) 1.17 (0.83–1.65) 2.42 (1.90–3.08) 2.15 (1.71–2.71) 2.44 (1.82–3.26) 2.85 (2.22–3.66)
a)
Adjusted for age, body mass index, status of hypertension, family history of diabetes, smoking status, alcohol consumption, and regular exercise
Abbreviations: ORs odds ratios, WC waist circumference, TG triglycerides, FPG fasting plasma glucose, LAP lipid accumulation product, TyG index, the product of fasting glucose and triglycerides
Page 6 of 10
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 7 of 10

Table 5 Reclassification and discrimination results associated with the risk prediction of incident type 2 diabetes according to each
measure
Parameter Men Women
AROC p-value* cNRI p-value IDI p-value AROC p-value* cNRI p-value IDI p-value
Basic modela) 0.615 0.621
Basic model + WC 0.619 0.1176 0.056 0.1689 0.002 0.0140 0.621 0.3233 −0.023 0.5666 0.000 1.0000
Basic model + TG 0.632 <.0001 0.227 <.0001 0.007 <.0001 0.652 <.0001 0.286 <.0001 0.019 <.0001
Basic model + FPG 0.692 <.0001 0.480 <.0001 0.064 <.0001 0.652 <.0001 0.214 <.0001 0.018 <.0001
Basic model + LAP 0.634 <.0001 0.236 <.0001 0.007 <.0001 0.644 <.0001 0.213 <.0001 0.016 <.0001
Basic model + TyG index 0.656 <.0001 0.387 <.0001 0.023 <.0001 0.666 <.0001 0.360 <.0001 0.029 <.0001
a)
Basic model: age, body mass index, status of hypertension, family history of diabetes, smoking status, alcohol consumption, and regular exercise
*P-value for AROC means vs. Basic model
Abbreviations: AROC area under the receiver operating characteristic curve, WC waist circumference, TG triglycerides, LAP lipid accumulation product, TyG index
the product of fasting glucose and triglycerides, cNRI continuous net reclassification improvement, IDI Integrated Discrimination Improvement

The advantages of the simple diabetes risk model universe coverage, the health promotion that offered
are that it is inexpensive and patients can use it for the general health check-ups for 17.6 million Koreans
self-assessment [8]. Weighed against the improved in 2016, and its following examination rate (77.7%),
predictive accuracy obtained by the addition of HbA1c there should be less burden caused by blood tests
and the Oral Glucose Tolerance Test (OGTT) to the than other countries [43]. Therefore, TyG index is
T2DM risk model, there are substantial practical con- recommended as a screening tool for the prediction
straints on screening the whole population using such of T2DM. Although TyG index predicts incident
measurements [22]. As an effective alternative, a T2DM well, its usefulness is inconsistent, when com-
two-pronged approach to screening has been pro- pared to fasting glucose [38, 44]. For predicting
posed [22, 39]. This approach consists of using the T2DM risk, TyG index was not better than FPG or
non-laboratory score for the general population and OGTT in Isfahan Diabetes Prevention Study [38].
using the laboratory score for patients in a higher-risk Wang et al. reported that TyG index, LAP, and vis-
group for T2DM. For screening as part of the Korean ceral adiposity were not superior to FPG or WC
NHI program, TG-related indices or FPG are more alone as diabetes predictors among Chinese [44]. In
suitable as one of the components of the T2DM risk the Vascular-Metabolic CUN cohort, Navarro-Gonzalez et
model than OGTT or HbA1c, given that these mea- al. compared the prediction ability of TyG index and FPG
surements are not currently included in primary for onset T2DM [35]. Association between indexes and
screening measurements in Korea [23]. To the best of their discrimination for onset T2DM were different de-
our knowledge, there is no TyG index criterion yet. pending on the fasting glucose subgroup. When the
On the other hand, several studies conducted in highest quartile for each index was compared to the
Europe and Asian have shown that the risk of inci- lowest, the hazard ratios (HRs) of TyG index and
dent T2DM was increased with increasing TyG index FPG were 3.0 and 7.3 in the impaired fasting glucose
[40–42]. Some researchers have proposed a cut-off group. On the other hand, TyG index showed stron-
value of TyG index of 8.8 for incident T2DM and in- ger association with onset DM than FPG in the nor-
sulin resistance [41, 42]. The subjects in the present mal fasting glucose group (HRs: 6.8 vs. 4.6). The
study were classified by the cut-off value of TyG discrimination of TyG index for onset T2DM was
index of 8.8. The adjusted odds ratio (95% CI) of in- also better than that of FPG in the normal fasting
cident T2DM in the subjects with TyG index ≥8.8 glucose group (AROC: 0.75 vs. 0.66). In the present
was 1.95 (1.73–2.20) compared to the counterparts. study, we found the association between some meta-
We also have proposed sex-specific cut-off points of bolic syndrome (MetS) components and incident
TyG index (≥ 8.86 in men, ≥ 8.52 in women) as pre- T2DM (Appendix 2). Therefore, we compared the
dictors of T2DM based on the maximized Youden discrimination of each index for incident T2DM in
index. Compared to the counterparts, the adjusted the stratified MetS components (Appendix 3). In the
odds ratio (95% CI) of incident T2DM in the men elevated FPG group, the discrimination of FPG
with TyG index ≥8.86 and in the women with TyG (AROC: 0.506) for incident T2DM was inferior to
index ≥8.52 were 2.01 (1.69–2.39) and 2.16 (1.82– that of other indices. On the other hand, TyG index
2.56), respectively (Appendix 1). Considering NIH’s was a more robust discrimination index than other
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 8 of 10

indices in the group stratified by sex and MetS com- Appendix 2

ponents. Our findings indicate that TyG index is not Table 7 Multiple logistic regression analyses for each MetS
only a better predictor for incident diabetes than WC, component in predicting type 2 diabetes (odds ratios and
LAP, and TG, but it also has a better reclassification 95% confidence intervals)
ability. On the other hand, association between FPG Components Men Women
of MetSa)
and its(their) reclassification ability for incident Unadjusted Adjusted h)
Unadjusted Adjusted h)
T2DM were different depending on sex. The present ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI)
study used community-based long-term prospective Central 1.71 (1.42–2.06) 1.35 (1.10–1.65) 1.56 (1.33–1.83) 1.16 (0.97–1.38)
cohort data. Therefore, the temporal relationship be- obesityb)

tween each measure and incident diabetes is clear. Elevated TGc) 1.70 (1.45–2.00) 1.57 (1.31–1.87) 2.26 (1.93–2.64) 1.90 (1.60–2.26)
Our findings indicate that TyG index is not only a Low HDL-Cd) 1.14 (0.96–1.34) 1.01 (0.84–1.22) 1.53 (1.28–1.83) 1.21 (1.00–1.47)
better predictor for incident diabetes than WC, LAP, Elevated 5.43 (4.09–7.21) 5.16 (3.85–6.92) 3.65 (2.39–5.57) 3.55 (2.29–5.50)
and TG, but it also has a better reclassification abil- FPGe)

ity. It was also noted that association between FPG Elevated BPf) 1.79 (1.52–2.10) 1.47 (1.24–1.75) 1.65 (1.41–1.93) 1.29 (1.09–1.54)

and its reclassification ability for incident T2DM were MetSg) 2.22 (1.86–2.66) 2.24 (1.87–2.69)i) 2.22 (1.88–2.61) 2.03 (1.70–2.41)i)
different depending on sex. There were, however, a a)
Modified NCEP-ATP III criteria [47] with the Korean cut-off for WC, b)WC ≥
90 cm for men ≥85 for women, Lee et al. [48], c)TG ≥ 150 mg/dl, d)HDL-C <
few limitations to this study. Firstly, even though 40 mg/dl for men < 50 mg/dl for women, e)FPG ≥ 100 mg/dl, f)Systolic blood
KoGES is designed to include subjects who live both pressure ≥ 130 mmHg or diastolic blood pressure ≥ 85 mmHg, g)Number of
MetS components ≥3, h)Adjusted for other components of MetS, age,
in rural and urban areas, the data are not representa- family history of diabetes, smoking status, alcohol consumption, and
tive of the entire Korean population [45]. Secondly, regular exercise, i)Adjusted for age, family history of diabetes, smoking
status, alcohol consumption, and regular exercise
dietary intake was not included in the analysis. More- Abbreviations: MetS Metabolic syndrome, ORs odds ratios, TG triglyceride,
over, Baik et al. had reported that the usefulness of HDL-C high-density lipoprotein cholesterol, FPG fasting plasma glucose,
dietary information in cardiovascular disease risk pre- BP blood pressure, WC, waist circumference

diction models [46]. However, dietary information has

been excluded from the questionnaire in medical
check since 2009. Finally, our results were not vali-
dated using a separate Korean dataset. Therefore,
additional studies are required to validate our data.

Conclusions
In conclusion, TG-related indices and FPG were more
accurate than WC in the prediction of incident
T2DM. In the subgroup categorized by sex and fast-
ing glucose level, TyG index was a more robust pre-
dictor for onset T2DM than other indexes. Therefore,
TyG index can be a useful screening tool for incident
T2DM in middle-aged Koreans.

Appendix 1
Table 6 Multiple logistic regression analyses for cutoff-points of
TyG index in predicting type 2 diabetes (odds ratios and 95%
confidence intervals)
Cutoff- Men Women
points
Unadjusted Adjusteda) Unadjusted Adjusteda)
ORs (95% CI) ORs (95% CI) ORs (95% CI) ORs (95% CI)
TyG index 2.05 (1.75–2.41) 1.93 (1.62–2.29) 2.41 (2.05–2.83) 2.03 (1.71–2.41)
≥8.8b)
TyG index 2.13 (1.82–2.51) 2.01 (1.69–2.39) 2.57 (2.18–3.02) 2.16 (1.82–2.56)
≥8.86/8.52c)
a)
Adjusted for age, body mass index, status of hypertension, family history of
diabetes, smoking status, alcohol consumption, and regular exercise, b)TyG
index ≥8.8 cm, Lee et al. [42], c)TyG index ≥8.86 cm for men ≥8.52 for
women in the present study
Abbreviations: ORs odds ratios; TyG index, the product of fasting glucose
and triglycerides
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 9 of 10

Appendix 3
Table 8 The area under the ROC curve (AROC) for indices to predict type 2 diabetes stratified by MetS components
MetS componentsa) Central obesityb) Elevated TGc) Low HDL-Cd) Elevated FPGe) Elevated BPf)
Subgroup No Yes No Yes No Yes No Yes No Yes
Index AROC AROC AROC AROC AROC AROC AROC AROC AROC AROC
Men (n = 2929) (n = 707) (n = 1970) (n = 1666) (n = 2326) (n = 1310) (n = 3423) (n = 213) (n = 2092) (n = 1544)
WC 0.553 0.532 0.535 0.569 0.570 0.584 0.571 0.573 0.565 0.559
TG 0.589 0.534 0.546 0.559 0.598 0.574 0.588 0.591 0.572 0.588
FPG 0.648 0.680 0.637 0.676 0.661 0.666 0.621 0.506 0.649 0.661
LAP 0.584 0.537 0.546 0.582 0.598 0.599 0.595 0.604 0.582 0.591
TyG index 0.618 0.569 0.605 0.607 0.631 0.605 0.607 0.586 0.602 0.620
Women (n = 2680) (n = 1392) (n = 2754) (n = 1318) (n = 1276) (n = 2796) (n = 3983) (n = 89) (n = 2639) (n = 1433)
WC 0.556 0.519 0.536 0.564 0.520 0.583 0.576 0.506 0.570 0.545
TG 0.610 0.626 0.567 0.559 0.611 0.622 0.629 0.596 0.614 0.621
FPG 0.590 0.598 0.581 0.617 0.550 0.616 0.586 0.566 0.598 0.587
LAP 0.601 0.625 0.553 0.590 0.565 0.630 0.623 0.570 0.608 0.609
TyG index 0.628 0.643 0.595 0.596 0.621 0.643 0.641 0.597 0.633 0.634
Modified NCEP-ATP III criteria [47] with the Korean cut-off for WC, WC ≥ 90 cm for men ≥85 for women, Lee et al. [48], TG ≥ 150 mg/dl, HDL-C < 40 mg/dl for
a) b) c) d)

men < 50 mg/dl for women, e)FPG ≥ 100 mg/dl, f) Systolic blood pressure ≥ 130 mmHg or diastolic blood pressure ≥ 85 mmHg
Abbreviations: MetS Metabolic syndrome, TG triglyceride, HDL-C high-density lipoprotein cholesterol, FPG fasting plasma glucose, BP blood pressure, AROC area
under the receiver operating characteristic curve, WC waist circumference, LAP lipid accumulation product, TyG index, the product of fasting glucose
and triglycerides

Abbreviations Ethics approval and consent to participate

AROC: Area under the receiver operating characteristic curve; BMI: Body mass The Institutional Review Board of the Korean Centers for Disease Control and
index; CI: Confidence interval; cNRI: Continuous net reclassification Prevention approved the KoGES protocol. Written informed consent was
improvement; FPG: Fasting plasma glucose; HbA1c: Glycated hemoglobin; obtained from all subjects.
HRs: Hazard ratios; IDI: Integrated discrimination improvement; IMGU: Insulin-
mediated glucose uptake; KNHANES: Korea National Health and Nutrition Competing interests
Examination; KNIH: Korea National Institute of Health; KoGES: Korean Genome The authors declare that they have no competing interest.
and Epidemiology Study; LAP: Lipid accumulation product; NHI: National Health
Insurance; OGTT: Oral Glucose Tolerance Test; ORs: Odds ratios; T2DM: Type 2
diabetes mellitus; TG: Triglyceride; TyG index: Product of fasting glucose and Publisher’s Note
triglyceride; VIF: Variable inflation factor; WC: Waist circumference Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Acknowledgements
Epidemiologic data used in this study were from the Korean Genome and Author details
Epidemiology Study (KoGES) of the Korea Centers for Disease Control & 1
Division of Cardiovascular Diseases, Center for Biomedical Sciences, Korea
Prevention, Republic of Korea. National Institute of Health, 187 Osongsaengmyeng 2-ro, Osong-eup,
Heungdeok-gu, Cheongju-si, Chungcheongbuk-do 361-951, South Korea.
2
Funding Department of Public Health Sciences, Graduate School, Korea University,
This study was supported by an intramural grant of the Korea National Seoul, South Korea.
Institute of Health, Korea 4800–4845-302 (2017-NI63001–00). The funders had
no role in the design of the study and collection, analysis, and interpretation Received: 18 July 2017 Accepted: 16 May 2018
of data and in writing the manuscript.

Availability of data and materials References

The information of KoGES can be obtained at [http://www.nih.go.kr/NIH/ 1. Esteghamati A, Gouya MM, Abbasi M, Delavari A, Alikhani S, Alaedini F,
eng/main.jsp > Research infrastructure > KoGES]. The detailed cohort profile Safaie A, Forouzanfar M, Gregg EW. Prevalence of diabetes and impaired
and how to access the data are described in the following source: KIM YJ, fasting glucose in the adult population of Iran: National Survey of risk
Han BG, KoGES group. Cohort profile: the Korean genome and epidemiology factors for non-communicable diseases of Iran. Diabetes Care. 2008;
study (KoGES) Consortium. International journal of epidemiology, 2016, 46.2: 31(1):96–8.
e20-e20. The data that support the findings of this study are available from the 2. Kim DJ. The epidemiology of diabetes in Korea. Diabetes Metab J. 2011;
Korea National Institute of Health (KNIH) but restrictions apply to the availability 35(4):303–8.
of these data. Data can be accessible upon reasonable request and with ap- 3. Ministry of Health and Welfare of Korea KCfDCaP. Korea health statistics
proval of a designated research proposal review committee of the KNIH. 2011: Korea National Health and Nutrition Examination Survey (KNHANES V-
2). Seoul: Ministry of Health and Welfare of Korea; 2012.
Authors’ contributions 4. Ministry of Health and Welfare of Korea KCfDCaP. Korea health
JWL carried out the data analysis and wrote the manuscript. NKL contributed statistics 2014: Korea National Health and Nutrition Examination
to study design and advised statistical analyses. HYP contributed to study Survey (KNHANES VI-2). Sejong: Ministry of Health and Welfare of
design and critically reviewed the paper. All authors read and approved the Korea; 2015.
final manuscript. 5. Federation ID. IDF Diabetes Atlas. 7th ed; 2015.
Lee et al. BMC Endocrine Disorders (2018) 18:33 Page 10 of 10

6. Li G, Zhang P, Wang J, An Y, Gong Q, Gregg EW, Yang W, Zhang B, Shuai Y, 30. Pencina MJ, D'Agostino RB, D'Agostino RB, Vasan RS. Evaluating the added
Hong J. Cardiovascular mortality, all-cause mortality, and diabetes incidence predictive ability of a new marker: from area under the ROC curve to
after lifestyle intervention for people with impaired glucose tolerance in the reclassification and beyond. Stat Med. 2008;27(2):157.
Da Qing diabetes prevention study: a 23-year follow-up study. Lancet 31. Pencina MJ, D'Agostino RB, Steyerberg EW. Extensions of net reclassification
Diabetes Endocrinol. 2014;2(6):474–80. improvement calculations to measure usefulness of new biomarkers. Stat
7. Tuomilehto J, Schwarz P, Lindström J. Long-term benefits from lifestyle Med. 2011;30(1):11–21.
interventions for type 2 diabetes prevention time to expand the efforts. 32. Kennedy K, Pencina M. A SAS macro to compute added predictive ability of
Diabetes Care. 2011;34(Supplement 2):S210–4. new markers predicting a dichotomous outcome. In: SouthEeast SAS Users
8. Noble D, Mathur R, Dent T, Meads C, Greenhalgh T. Risk models and scores Group Annual Meeting Proceedings: 2010; 2010.
for type 2 diabetes: systematic review. Bmj. 2011;343:d7163. 33. Vatcheva KP, Lee M, McCormick JB, Rahbar MH. Multicollinearity in
9. Chien K, Cai T, Hsu H, Su T, Chang W, Chen M, Lee Y, Hu F. A prediction regression analyses conducted in epidemiologic studies. Epidemiol. 2016;
model for type 2 diabetes risk among Chinese people. Diabetologia. 2009; 6(227). https://doi.org/10.4172/2161-1165.1000227. https://www.omicsonline.
52(3):443–50. org/open-access/multicollinearity-in-regression-analyses-conducted-in-
10. Lee Y-H, Bang H, Kim HC, Kim HM, Park SW, Kim DJ. A simple screening inepidemiologic-studies-2161-1165-1000227.php?aid=69442.
score for diabetes for the Korean population development, validation, and 34. Sebo P, Beer-Borst S, Haller DM, Bovier PA. Reliability of doctors'
comparison with other scores. Diabetes Care. 2012;35(8):1723–30. anthropometric measurements to detect obesity. Prev Med. 2008;47(4):
11. Lim N-K, Park S-H, Choi S-J, Lee K-S, Park H-Y. A risk score for predicting the 389–93.
incidence of type 2 diabetes in a middle-aged Korean cohort. Circ J. 2012; 35. Navarro-González D, Sánchez-Íñigo L, Pastrana-Delgado J, Fernández-
76(8):1904–10. Montero A, Martinez JA. Triglyceride–glucose index (TyG index) in
12. Lindström J, Tuomilehto J. The diabetes risk score. Diabetes Care. 2003;26(3): comparison with fasting plasma glucose improved diabetes prediction in
725–31. patients with normal fasting glucose: the vascular-metabolic CUN cohort.
13. Mann DM, Bertoni AG, Shimbo D, Carnethon MR, Chen H, Jenny NS, Prev Med. 2016;86:99–105.
Muntner P. Comparative validity of 3 diabetes mellitus risk prediction 36. Vasques ACJ, Novaes FS, MdS d O, JRM S, Yamanaka A, Pareja JC, Tambascia
scoring models in a multiethnic US cohort the multi-ethnic study of MA, MJA S, Geloneze B. TyG index performs better than HOMA in a Brazilian
atherosclerosis. Am J Epidemiol. 2010;171(9):980–8. population: a hyperglycemic clamp validated study. Diabetes Res Clin Pract.
14. Wilson PW, Meigs JB, Sullivan L, Fox CS, Nathan DM, D’Agostino RB. 2011;93(3):e98–e100.
Prediction of incident diabetes mellitus in middle-aged adults: the 37. Abbasi F, Reaven G. Statin-induced diabetes: how important is insulin
Framingham offspring study. Arch Intern Med. 2007;167(10):1068–74. resistance? J Intern Med. 2015;277(4):498–500.
15. Consultation WE: Waist circumference and waist-hip Ratio 2011. 38. Janghorbani M, Almasi SZ, Amini M. The product of triglycerides and
16. Ko G, Chan J, Woo J, Lau E, Yeung V, Chow C, Wai H, Li J, So W, Cockram C: glucose in comparison with fasting plasma glucose did not improve
Simple anthropometric indexes and cardiovascular risk factors in Chinese. diabetes prediction. Acta Diabetol. 2015:1–8.
Int J Obes Relat Metab Disord 1997, 21(11):995–1001. 39. Wannamethee S, Papacosta O, Whincup P, Thomas M, Carson C, Lawlor D,
17. Janghorbani M, Amini M. Normal fasting plasma glucose and risk of Ebrahim S, Sattar N. The potential for a two-stage diabetes risk algorithm
prediabetes and type 2 diabetes: the Isfahan diabetes prevention study. combining non-laboratory-based scores with subsequent routine non-
Rev Diabet Stud. 2011;8(4):490. fasting blood tests: results from prospective studies in older men and
18. Zhang M, Gao Y, Chang H, Wang X, Liu D, Zhu Z, Huang G. women. Diabet Med. 2011;28(1):23–30.
Hypertriglyceridemic-waist phenotype predicts diabetes: a cohort study in 40. Zheng R, Mao Y. Triglyceride and glucose (TyG) index as a predictor of
Chinese urban adults. BMC Public Health. 2012;12(1):1081. incident hypertension: a 9-year longitudinal population-based study. Lipids
Health Dis. 2017;16(1):175.
19. Lee S-H, Kwon H-S, Park Y-M, Ha H-S, Jeong SH, Yang HK, Lee J-H, Yim H-W,
41. Navarro-González D, Sánchez-Íñigo L, Fernández-Montero A, Pastrana-
Kang M-I, Lee W-C. Predicting the development of diabetes using the
Delgado J, Martinez JA. TyG index change is more determinant for
product of triglycerides and glucose: the Chungju metabolic disease cohort
forecasting type 2 diabetes onset than weight gain. Medicine. 2016;95(19)
(CMC) study. PLoS One. 2014;9(2):e90430.
42. Lee DY, Lee ES, Kim JH, Park SE, Park C-Y, Oh K-W, Park S-W, Rhee E-J, Lee
20. Kahn HS. The lipid accumulation product is better than BMI for identifying
W-Y. Predictive value of triglyceride glucose index for the risk of incident
diabetes a population-based comparison. Diabetes Care. 2006;29(1):151–3.
diabetes: a 4-year retrospective longitudinal study. PLoS One. 2016;11(9):
21. Song YJ. The south Korean health care system. JMAJ. 2009;52(3):206–9.
e0163465.
22. Ahn CH, Yoon JW, Hahn S, Moon MK, Park KS, Cho YM. Evaluation of non-
43. National Health Insurance Service. 2016 National health screening statistical
laboratory and laboratory prediction models for current and future diabetes
yearbook http://www.nhis.or.kr/menu/boardRetriveMenuSet.xx?menuId=F3328.
mellitus: a cross-sectional and retrospective cohort study. PLoS One. 2016;
44. Wang B, Zhang M, Liu Y, Sun X, Zhang L, Wang C, Linlin L, Ren Y, Han C,
11(5):e0156155.
Zhao Y. Utility of three novel insulin resistance-related lipid indexes for
23. Korean National Health Insurance Service: National Health Insurance
predicting type 2 diabetes mellitus among people with normal fasting
[http://www.nhis.or.kr/static/html/wbd/g/a/wbdga0606.html].
glucose in rural China. J Diab. 2018. https://www.ncbi.nlm.nih.gov/pubmed/
24. Kim Y, Han B-G. Cohort profile: the Korean genome and epidemiology 29322661.
study (KoGES) consortium. Int J Epidemiol. 2016;46(2):e20–e20. 45. Shin C, Abbott R, Lee H, Kim J, Kimm K. Prevalence and correlates of
25. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of orthostatic hypotension in middle-aged men and women in Korea: the
low-density lipoprotein cholesterol in plasma, without use of the Korean health and genome study. J Hum Hypertens. 2004;18(10):717–23.
preparative ultracentrifuge. Clin Chem. 1972;18(6):499–502. 46. Baik I, Cho N, Kim S, Shin C. Dietary information improves cardiovascular
26. Association AD. Diagnosis and classification of diabetes mellitus. Diabetes disease risk prediction models. Eur J Clin Nutr. 2013;67(1):25.
Care. 2010;33(Supplement 1):S62–9. 47. Third Report of the National Cholesterol Education Program. (NCEP) expert
27. Park SK, Ryoo J-H, Oh C-M, Choi J-M, Choi Y-J, Lee KO, Jung JY. The risk of panel on detection, evaluation, and treatment of high blood cholesterol in
type 2 diabetes mellitus according to 2-hour plasma glucose level: the Korean adults (adult treatment panel III) final report. Circulation. 2002;106(25):
genome and epidemiology study (KoGES). Diabetes Res Clin Pract. 2017. 3143–421.
https://www.ncbi.nlm.nih.gov/pubmed/?term=The+risk+of+type+2+diabetes 48. Lee SY, Park HS, Kim DJ, Han JH, Kim SM, Cho GJ, Kim DY, Kwon HS, Kim SR,
+mellitus+according+to+2-hour+plasma+glucose+level%3A+the+Korean Lee CB. Appropriate waist circumference cutoff points for central obesity in
+genome+and+epidemiology+study. Korean adults. Diabetes Res Clin Pract. 2007;75(1):72–80.
28. Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a
fundamental evaluation tool in clinical medicine. Clin Chem. 1993;39(4):561–77.
29. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under
two or more correlated receiver operating characteristic curves: a
nonparametric approach. Biometrics. 1988;44(3):837–45.