New Method for Examining Association Between Obesity and Diabetes With Age Among Korean Adults


 This study examined the age-dependence of the association between diabetes and obesity among Korean adults. The data from the Seventh Korean National Health and Nutrition Examination Survey (2017) were analyzed. This paper considered the body mass index (BMI) and waist-to-height ratio (WHT2R) as obesity indices and the fasting blood glucose (FBG) as a diabetes index and used their lump mean values and the regression method. The WEHT2R was a more suitable index for examining the FBG than the BMI. The increase in the WHT2R of females was more sensitive to age than that of males. In contrast, the increase in the FBG of males was more sensitive to age than that of females. The ages that FBG increased to 100.0 mg/DL were 44.7 and 60.5 years for males and females, respectively. Korean adults should reduce their WHT2R to prevent the excessive increase in their FBG level and Korean males should take care of their FBG from a young age. The present results can be used for a preliminary inspection of diseases, such as functional degradation of Langerhans islets and gestational diabetes


Introduction
The incidence of obesity and diabetes mellitus (DM) is increasing worldwide and is a global public health concern. DM is an independent risk factor for coronary artery disease and a group of metabolic disorders characterized by high blood sugar levels [1]. DM is the most prominent disease related to the failure of blood sugar regulation that can cause many complications, such as diabetic ketoacidosis, hyperosmolar hyperglycemic state, cardiovascular disease, stroke, and chronic kidney disease [2,3]. DM is associated with a two to three-fold increase in the risk of cardiovascular disease compared to the non-diabetic population [4]. Patients with an impaired fasting glucose level have more severe coronary artery damage [5] and hyperglycemia is related to diabetic microvascular complications [6]. Cardiovascular disease is a severe complication of diabetes, and many international guidelines recommend that people with diabetes maintain a blood pressure below 140/90 mmHg [7].
Diabetes is classified mainly into two groups: type 1 and type 2. Type 1 diabetes occurs mainly in children, and patients with type 1 diabetes are generally not obese [8]. In contrast, type 2 diabetes is associated with an increased prevalence of obesity among all age groups [9]. Type 1 diabetes is characterized by the loss of insulin-producing beta cells in the pancreatic islets. Type 2 diabetes is characterized by insulin resistance that may be combined with reduced insulin secretion [10]. Several lifestyle factors, such as obesity, lack of physical activity, poor diet, and stress, are important to developing type 2 diabetes [11]. Type 1 diabetes has no known preventive measures, whereas Type 2 diabetes can often be prevented or delayed by maintaining normal body weight, engaging in physical activity, and eating a healthy diet [12].
This study examined the age-dependence of the relationship between obesity and the fasting blood glucose (FBG) level, which is the most fundamental indicator of diabetes [13].
An impaired FBG is an intermediate state of abnormal glucose regulation that increases the risk of developing diabetes and associated vascular disease [14,15]. Obesity is a major worldwide epidemic and a significant risk factor for DM. Moreover, it is a factor in hypertension, dyslipidemia, and cardiovascular damage [16,17]. This study considered the body mass index (BMI) and waist-to-height ratio (WHT2R) as obesity indices. The BMI is a good measure of general adiposity [18], and an above-normal BMI is an established risk factor for ischemic heart disease, stroke, and carcinomas [19].
The WHT2R is a new obesity index presented by the present author that is defined as the waist circumference (WC) divided by the square of the height [20], considering that the WC is associated more with cardio-metabolic mortality than the BMI, and height has an inverse association with mortality [21,22]. This paper used the lump mean values (LMVs) of the FBG and obesity because the FBG varies with obesity in a highly complex manner and thus it is almost impossible to examine the association between the FBG level and obesity individually.
For that purpose, male and female subjects were divided into groups with two successive ages.

Design and participants
The study sample was obtained from the public-use releases of the Seventh Korean National

Statistical analysis
This study investigated the age-dependence of the associations of the BMI and WHT2R with  Figure 1 presents the LMVs of the BMI and FBG of males with respect to age. The BMI of males (blue triangles) increased to a maximum value of 24.9 kg/m 2 with age less than 44.5 years and then decreased, so it was fitted to the blue (Gaussian) curve given by the following equation:

Age-dependence of association between BMI and FBG
where is age, ( ) is the LMV of BMI, and 0 is the minimum value of ( ) when the constant is positive. √ is the standard deviation that determines the width of the distribution.
0 is the value corresponding to the maximum value of ( ). The parameters , , 0 , and 0 are shown in the figure. 2 is the coefficient of determination. The BMI of males appeared 6 to decrease with age after 44.5 years of age because they reduce their weight by controlling their food intake and physical activity for good health.
The FBG of males (red circles) increases continuously with age, so it was fitted to the red straight line expressed as where , , , and ( ) are age, a constant, the slope, and the LMV of the FBG, respectively.
The parameters and are shown in the figure. 2 is the coefficient of determination. The goodness of fit for linear regression was sufficient according to 2 .
The regressed value of the FBG of males increased from 88.6 mg/dL to 109.2 mg/dL with increasing age from 20.5 years old to 64.5 years old. Hence, the increase in the FBG of males with age before 44.5 years of age is associated with an increase in BMI. The FBG of males increased to 100. 0 mg/DL when they were 44.7 years old.  Therefore, the increase in the FBG of females with age is associated with an increase in BMI. 7 The regressed values of the BMI and FBG of females increased from 21.4 kg/m 2 to 24.3 kg/m 2 and from 86.7 mg/dL to 101.3 mg/dL with age from 20.5 to 64.5 years old, respectively. The parameters and are shown in the figure, and the goodness of fit for linear regression was sufficient according to the coefficient of determination ( 2 ). The FBG of females increased to 100. 0 mg/DL when they were 60.5 years of age.

Age-dependence of association between WHT2R and FBG
The WHT2R (blue triangles) and FBG (red circles) of females and males are associated with WHT2R because they increase continuously with age ( Figs. 3 and 4). Thus, the WHT2Rs and   year −1 ) for males. In contrast, the FBG of males increased with age more rapidly than that of females because the slope ( = 0.47 mg/dL • year) of the straight line for males was larger than that ( = 0.33 mg/dL • year) for females.

Discussion
The age dependence of the FBG and its associations with the BMI and WH2TR were investigated using their LMVs. The WHT2R of males, the BMI and WHT2R of females, and the FBG of males and females increased linearly with age, but BMI of males increased with age before 44.5 years of age and then decreased. Hence, the LMVs of the WHT2R of males, the BMI and WHT2R of females, and the FBG of males and females with respect to age were fitted to straight lines given by Eq. (2). In contrast, the BMI of males was fitted to a Gaussian curve given by Eq. (1). This means that the WHT2R is a more suitable index to examine the FBG level than the BMI because the FBG level of females is associated with both the BMI and WHT2R, whereas for males, the FBG level is associated only with the WHT2R.
The increase in the WHT2R of females is more sensitive to age than the increases in the  Furthermore, the present method can be applied to examine the associations of obesity with neutral fat or cholesterol level.