Association of pre-diabetes with increased obesity and depression in older male and female: A secondary analysis of the Korea National Health and Nutrition Examination Survey 2020
Article information
Abstract
Purpose
This study aimed to investigate the association among pre-diabetes, increased obesity, and depression in Korean older male and female.
Methods
Data of 1,005 participants (≥65 years) were extracted from the Korea National Health and Nutrition Examination Survey (KNHANES) VIII-2, 2020. A complex sampling design was adopted to reflect stratification, clustering, and weights. Data were analyzed using t-test, Rao-Scott chi-square test, and simple and multiple logistic regression. In predicting the risk factors for pre-diabetes, after adjusting, multiple logistic regression analysis was performed to confirm the relative contribution of each predictor variable.
Results
Body mass index, waist circumference (WC), and incidence of depression were higher in the pre-diabetes group than in the normal group. According to the simple logistic regression analyses, in male, the odds ratio (OR) of pre-diabetes increased by 1.73, 2.25, and 2.53 for obesity (25.0~29.9 kg/m2), high obesity (≥30.0 kg/m2), and abdominal obesity (AO) (WC≥90 cm), respectively. In female, the OR of pre-diabetes increased by 2.15, 2.30, and 3.12 for high obesity, AO (WC≥85 cm) and depression diagnosis, respectively. In the adjusted analysis, in male, the adjusted odds ratio (AOR) of pre-diabetes increased by 2.73 and 1.97 for high obesity and AO (WC≥90 cm), respectively. In female, the AOR for pre-diabetes increased by 3.74 and 2.19 for depression diagnosis and no treatment for depression, respectively.
Conclusion
This study provides evidence that interventions for obesity and depression should be included differently in pre-diabetes management programs for older male and female.
INTRODUCTION
Diabetes is a chronic disease highly prevalent among older adults. More than a quarter of people aged >65 years suffer from type 2 diabetes mellitus (T2DM), and half of older adults suffer from pre-diabetes [1]. Pre-diabetes is characterized by impaired glucose regulation, with above normal blood glucose levels yet below the diagnostic threshold for T2DM [2]. Pre-diabetes is associated with a chronic disease that can lead to serious vascular complications and premature death [3,4]. The prevalence of pre-diabetes is increasing worldwide [5,6]. If left untreated, 37% of people with pre-diabetes will develop T2DM within 4 years [7]. Both pre-diabetes and T2DM are diagnosed based on glucose criteria, which measure fasting blood glucose (FBS) and hemoglobin A1c (HbA1c) [8]. The American Diabetes Association recommends screening, early detection, and treatment for pre-diabetes to prevent T2DM and its complications [1]. In 2020, the prevalence of T2DM in Korea was higher among those aged ≥65 years (30.1%) than among those aged 30~39 years (16.7%), and it increased with age [9]. Therefore, pre-diabetes is closely related to T2DM [1], in order to prevent T2DM, factors associated with pre-diabetes in older adults must be identified.
Obesity is positively correlated with pre-diabetes [6] and is a major risk factor for exacerbating pre-diabetes [10]. Nevertheless, the prevalence of obesity is highest in older adults owing to decreased basal metabolic rate with aging [11]. In 2020, the obesity rate among older adults in Korea was 35.8% in male and 40.5% in female [12]. The World Health Organization sets the standard for obesity in Westerners as a body mass index (BMI) of ≥30 kg/m2 [10]. The Korean Disease Control and Prevention Agency’s obesity standard is ≥25 kg/m2 [12], and due to differences in body composition, the BMI reference point for obesity in Asians is lower [13]. Obesity is a chronic inflammatory condition associated with a high incidence of impaired fasting glucose [14]. Meanwhile, BMI is commonly used as an indicator of obesity. However, owing to aging, obesity in older adults is characterized by decreased skeletal muscle mass and abdominal obesity (AO) because of the accumulation of abdominal visceral fat [15]. AO is positively associated with pre-diabetes, and central obesity is a major risk factor for exacerbating pre-diabetes [14]. Obesity increases the prevalence of pre-diabetes because fat tissue induces insulin resistance and decreases the function of Langerhans cells that secrete insulin in people with obesity [16]. Insulin resistance and elevated blood sugar levels are more closely related to visceral fat than to fat in other areas. Physiological changes due to aging make controlling blood sugar difficult, making complications more likely to occur in older adults than in younger adults [14]. Therefore, weight loss improves glycemic control in patients with diabetes and prevents pre-diabetes from progressing to T2DM [17].
Pre-diabetes and depression are interrelated [1,18], and it makes individuals vulnerable to T2DM prevention [19]. Depression is a representative mental health problem, and its prevalence in the Korean older population is 13.5%, increasing to 24.0% among people aged >85 years [20]. Patients with depression and pre-diabetes have a higher risk for T2DM than patients with only one of these risk factors [19]. Furthermore, depression makes diabetes treatment difficult because antidepressants significantly increase HbA1c levels, a risk factor for poor blood sugar control [21]. Conversely, depression is a psychological complication that can occur because of diabetes [1]. This has been suggested to be related to abnormalities in neurotransmitter function in patients with diabetes [1,4].
Most previous studies on have included patients with diabetes [2,9,13,17,22], adults of all ages [14,16,23], obesity [2,15,22], depression [5,18,19], conversion from pre-diabetes to T2DM [3,8,21], and diabetes prevalence [4,14]. Obesity increases the incidence of pre-diabetes in male than in female [16]. As depression increases more steeply in female than in male with age, the prevalence of depression has a sex gap [24]. Asians with pre-diabetes are more susceptible to developing T2DM than Westerners [3]. Therefore, there is a need to identify differences in the impact of pre-diabetes in terms of sex. To our knowledge, these studies did not confirm the relationship between obesity and depression as factors affecting pre-diabetes in Asian Korean older male and female, who have a higher conversion rate to T2DM than Westerners [3].
HbA1c is a type of hemoglobin and is an indicator of the average glucose concentration over 2 to 3 months, the World Health Organization and American Diabetes Association suggest an HbA1c value of 5.7%~6.4% as the diagnostic cutoff for pre-diabetes [4,25]. Pre-diabetes is an intermediate stage between healthy adults and T2DM. The probability of developing T2DM within the next 5 years is 25% if the HbA1c value is 5.9% and increases to 50% if the HbA1c value is 6.4%, which is 20 times higher than when the HbA1c value is <5% [25]. This study aimed to investigate the association among pre-diabetes, increased obesity, and depression in Korean older male and female.
METHODS
Ethic statement: This study was approved by the Institutional Review Board (IRB) of Hannam University (IRB- 2023-02-12-0921). Top-coding, bottom-coding, and recategorization methods were applied to the KNHANES Ⅷ-2 (2020) data to ensure the anonymity and confidentiality of personal information. Data were provided after the researcher received approval for data use in accordance with the Korea Disease Control and Prevention Agency’s raw data disclosure and management regulations.
1. Research Design
This descriptive study analyzed the Korea National Health and Nutrition Examination Survey (KNHANES) Ⅷ-2, 2020 data to investigate the association of pre-diabetes with increased obesity and depression in older male and female. This study was described according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines (https://www.strobe-statement.org).
2. Participants and Data Collection
The data of participating older (≥65 years) individuals were extracted from the KNHANES Ⅷ-2. The sample was selected in stages 1 and 2, and households were selected after the sample survey. In this study, data from 1,005 out of 7,359 participants were analyzed, excluding participants aged <65 years, those with a diabetes diagnosis, and non-responders. Overall, 519 and 495 patients comprised the pre-diabetes and normal groups, respectively.
3. Measures
1) Pre-Diabetes Classification Criteria and Blood Measurement
HbA1c levels were classified by American Diabetes Association into normal (<5.7%), pre-diabetes (5.7%~6.4%), and diabetes (≥6.5%) [25], and this classification standard was applied in this study [25]. In the KNHANES Ⅷ-2, a trained expert punctured the vein of a patient who had fasted for >8 hours with a vacutainer needle, collected blood into a vacuum tube, and measured HbA1c levels using high-performance liquid chromatography (Tosoh G8; TOSOH BIOSCIENCE, INC.) [12].
2) Obesity Indicators Measurement
In the KNHANES Ⅷ-2, fasting except water was performed from 7 PM the day before the test. Measurements were taken while wearing a gown and removing belongings, shoes, and socks. Measurement tools, methods, and procedures were applied identically to the subjects. Body weight was measured using GL-6000-20 tool (G-Tech), and the posture was to look straight ahead. The measurement method was performed while exhalation, and the results were recorded when the measurement value was fixed. Height was measured using the SECA 225 tool (SECA), and the subject was in a straight posture with both legs. The measurement method was carried out under inhalation. To measure obesity, BMI was obtained as body weight divided by height squared (kg/m2) [12]. In this study, obesity (BMI=25.0~29.9 kg/m2), and high obesity (BMI≥30.0 kg/m2) were classified according to the obesity criteria of the Korea Disease Control and Prevention Agency [12].
Waist circumference (WC) was measured using a Lufkin w606 tool (Lufkin), with subjects in a forward-facing position. The exact location of the WC is midway between the last rib and the iliac crest [12], and the standard for AO was a WC of ≥90 cm for male and ≥85 cm for female [25].
3) Depression Measurement
In the KNHANES Ⅷ-2, depression-related characteristics include depression diagnosis (Have you ever been diagnosed with depression by a doctor?), depression treatment, current state of depression, and age at depression onset [12].
4. Ethical Considerations
This study was approved by the Institutional Review Board (IRB) of Hannam University (IRB- 2023-02-12-0921). Top-coding, bottom-coding, and recategorization methods were applied to the KNHANES Ⅷ-2 (2020) data to ensure the anonymity and confidentiality of personal information. Data were provided after the researcher received approval for data use in accordance with the Korea Disease Control and Prevention Agency’s raw data disclosure and management regulations.
5. Data Analysis
A complex sample analysis was performed by applying stratification, clustering, and weighting. The participants’ general characteristics included unweighted frequencies, weighted percentages, estimated means, and standard errors. Differences in general characteristics, BMI, WC, and depression between the pre-diabetes and normal groups were analyzed using the t-test and Rao-Scott chi-square test. Simple logistic regression and multiple logistic regression analyses were performed to identify factors affecting pre-diabetes in older male and female.
The variables entered into the logistic regression analysis were those with significant differences between the pre-diabetic and normal groups (age, BMI, WC, and depression diagnosis). Depression-related variables (depression treatment, depression status, and age of depression onset) affect insulin resistance and glycemic outcomes [18], and in patients with pre-diabetes, depression-related variables were more strongly associated with the obesity index [6]. Therefore, these variables were entered in the regression analysis even though there was no significant difference between the two groups.
In this study, simple logistic regression analyses were used to measure the contribution of each variable of obesity, AO, and depression to predicting pre-diabetes. In addition, multiple logistic regression analyses were necessary to confirm the relative contribution of each predictor variable. A multiple logistic regression analysis was performed to identify the risk factors for pre-diabetes after adjusting for BMI, WC, depression diagnosis, depression treatment, current state of depression, age at depression onset, and age. All analyses were conducted using SPSS 26.0 (IBM Corp.).
RESULTS
1. General Characteristics of the Normal and Pre-Diabetes Groups
The mean age of the participants was higher in the pre-diabetes group than in the normal group (t=291.73, p<.001). In the pre-diabetic group, ≥75 years was the most common at 38.6%, and in the normal group, 65~69 years was the most common at 38.2% (χ2=13.25, p=.027). Sex, education level, and economic status were not significantly different between the two groups. The mean HbA1c levels (t=559.67, p<.001) were higher in the pre-diabetes group than in the normal group (Table 1).
General Characteristics of the Normal and Pre-Diabetes Groups (N=1,005)
2. Obesity Indicators and Depression-Related Characteristics of the Normal and Pre-Diabetes Groups
The mean BMI (t=192.59, p<.001) and WC (t=244.30, p<.001) were higher in the pre-diabetes group than in the normal group. In the BMI classification, the pre-diabetes group comprised more individuals with high obesity, obesity, and overweight than the normal group (χ2=35.17, p<.001). AO in WC was higher in the pre-diabetes group (male 57.9% and female 56.3%) than in the normal group (male 33.1% and female 40.0%). A significant difference in AO was observed between the two groups (male: χ2=27.64, p=.001; female: χ2=14.93, p<.001). The incidence of depression diagnosis was higher in the pre-diabetes group (n=35, 7.3%) than in the normal group (n=29, 5.9%) (χ2=9.29, p=.009) (Table 2).
Obesity Indicators and Depression-Related Characteristics of Normal and Pre-Diabetes Groups (N=1,005)
3. Simple and Multiple Logistic Regression Analyses of Factors Affecting the Incidence of Pre-Diabetes in Koreans Older Male and Female
Among the factors affecting the incidence of pre-diabetes in older adults, unadjusted simple logistic regression analyses showed obesity, high obesity, and WC (≥90 cm) in male, and high obesity, WC (≥85 cm), and depression diagnosis in female. The odds ratio for pre-diabetes in older male was 1.73 times (95% confidence interval [95% CI]=1.60~3.80, p<.001) for obesity, 2.25 times (95% CI=1.18~4.27, p=.013) for high obesity compared to normal weight, and AO (WC≥90 cm) was 2.53 times (95% CI=1.49~4.31, p<.001) higher than non-AO (WC<90 cm). In female, high obesity was 2.15 times (95% CI=1.32~3.49, p=.002) compared to normal weight, AO (WC≥85 cm) was 2.30 times (95% CI=1.44~4.08, p<.001) higher than non-AO (WC<85 cm), and depression diagnosis was 3.12 times (95% CI=1.43~8.30, p=.006) than non-depression diagnosis (Table 3).
Simple Logistic Regression Analysis of Factors Affecting the Incidence of Pre-Diabetes in the Koreans Elderly Male and Female (N=1,005)
Among the factors affecting the incidence of pre-diabetes in older adults, the multiple logistic regression analysis showed high obesity and AO (WC≥90 cm) in male, whereas depression diagnosis and non-treatment were the identified factors in female. The adjusted odds ratio for pre-diabetes in male was 2.73 times (95% CI=1.97~4.58, p<.001) higher for high obesity compared to normal weight, and AO (WC≥90 cm) was 1.97 times (95% CI=1.01~3.85, p<.047) higher than non-AO (WC<90 cm). In female, 3.74 times (95% CI=2.69~16.51, p<.001) for depression diagnosis, and 2.19 times (95% CI=1.84~2.69, p<.001) for no treatment for depression than non-depression diagnosis. By contrast, in the adjusted multiple logistic regression analysis, BMI and WC did not influence pre-diabetes in female (Table 4).
Multiple Logistic Regression Analysis of Factors Affecting the Incidence of Pre-Diabetes in the Koreans Older Male and Female (N=1,005)
DISCUSSION
In this study, the prevalence of pre-diabetes was 50.7%. On average, participants meeting the criteria for pre-diabetes were older than those with normal blood glucose levels [26]. The prevalence of pre-diabetes increased by 1.30 times in those aged >65 years than in those aged <50 years [14]. The Professional Practice Committee of the American Diabetes Association reported that the prevalence of pre-diabetes in older adults was over 50% [1], which was similar to that in Korean older adults in this study.
In this study, BMI, WC, and depression diagnosis were higher in the pre-diabetes group than in the normal group. In the unadjusted analysis, BMI and WC were identified as factors affecting pre-diabetes in both sexes, and depression diagnosis was an influential factor only in female patients. Obesity is also associated with pre-diabetes [6]. BMI and fat mass index were higher in the pre-diabetes group than in the normal group [26]. The International Diabetes Federation suggests the use of FBS and HbA1c criteria for diagnosing pre-diabetes [1,4].
In obesity, substances such as hormones, cytokines, and inflammatory markers involved in insulin resistance increase, and excess fat due to obesity induces pre-diabetes by increasing peripheral resistance to glucose uptake [10,16]. Therefore, the authors considered obesity to be associated with pre-diabetes. At similar BMI levels, Asians have higher abdominal and visceral fat levels than white Europeans [1]. AO is also associated with pre-diabetes [6,14]. As an indicator of AO, WC was higher in the pre-diabetes group than in the normal group [26]. Central obesity, including a large WC or waist-to-hip ratio, is a predictor of pre-diabetes and has been demonstrated to increase the risk of pre-diabetes by 1.50 times [23]. Increased visceral fat increases C-reactive protein levels and the risk of pre-diabetes and T2DM [1], and causes the secretion of adipokines (such as tumor necrosis factor, and interleukin-6), which damage cytokines in the pancreatic cells, in turn increasing insulin resistance and exacerbating the progression of pre-diabetes [23]. AO is more associated with pre-diabetes compared with a general obesity index such as BMI, which supports our findings that WC had a greater influence on pre-diabetes than BMI does in both sexes. Higher BMI and WC increase the conversion from pre-diabetes to T2DM [3]; thus, weight management can reduce or delay the risk of T2DM [3,6]. However, individuals with pre-diabetes have a high level of obesity because of failure to control their weight [26]. Depression was more strongly associated with the obesity index in people with pre-diabetes than in the normal glucose group; therefore, depression can hinder weight management in patients with pre-diabetes [6,26].
In the adjusted analysis of this study, the factors affecting pre-diabetes were BMI and WC in male, and depression diagnosis and non-treatment of depression in female. The risk of pre-diabetes in older adults with obesity increased by 1.90 times for male and 1.43 times for female [14]. The incidence of pre-diabetes among older adults with obesity was higher in male than in female, supporting our findings that BMI and WC are factors influencing pre-diabetes in male. The prevalence of depression is higher in the pre-diabetes group than in the normal group [18], and depression can cause pre-diabetes [8]. Depression is associated with the activation of several connected biological systems, including the hypothalamic–pituitary–adrenal axis, and inflammatory response, which can influence insulin resistance and glycemic outcomes [18]. Meanwhile, a study conducted on adults in their 50s using the self-reported patient health questionnaire-9 revealed no difference in depressive symptoms between the pre-diabetes and normal groups [26]. The inconsistent results may be attributed to the differences in the participants and methods of measuring depression.
Depression increases in female with age [24]. Diagnosis of depression is increasing in both T2DM and female [27], and depression is related to diabetes complications and worsens T2DM [28]. This supports the results of this study, which revealed that depression in female patients affects the prevalence of pre-diabetes. In pre-diabetes alone, the incidence of T2DM increased 5.94 times, and in the combination of pre-diabetes and depression, the incidence of T2DM increased 8.29 times [5]. Depression increases cortisol exposure and causes insulin resistance, which leads to the progression of pre-diabetes to T2DM [29]. Further, depression is associated with decreased physical activity [2]. Regarding the relationship between depression and obesity, in the case of depression, male patients increase their weight owing to binge eating, whereas female patients lose weight because of decreased appetite, which may be attributed to differences in hormonal responses to depression [24,26,29].
Weight management is a key intervention for both sexes to prevent the transition from pre-diabetes to T2DM [17]. In patients with T2DM and obesity, moderate weight loss increases glycemic control [16]. Thus, intervention programs should include components that address barriers to weight management [26]. Depression is a mental illness that interferes with weight management [8], and researchers believe that interventions for depression should be a priority, especially for female patients.
A limitation of this study is that the depression-related information collected in the KNHANES Ⅷ-2 (2020) was based on self-reports. However, this was not diagnosed through clinical interviews. Additionally, because pre-diabetes was diagnosed based only on the HbA1c level, the actual prevalence of pre-diabetes may have been underestimated in this study. Future studies are needed to investigate the relationship between depression and obesity in patients with pre-diabetes and to confirm the effectiveness of depression diagnostic methods and treatments to prevent the transition from pre-diabetes to T2DM. In addition, because the relationship between independent and dependent variables was analyzed using data from the KNHANES Ⅷ-2 (2020) in the same year, there were limitations in confirming causal relationships over time. Despite these limitations, this study is meaningful as it empirically confirms that interventions for obesity and depression should be included differently in pre-diabetes management programs for older male and female.
CONCLUSION
As a result of the adjusted analysis of this study, the factors influencing pre-diabetes were BMI and WC for older male, and depression diagnosis and non-treatment of depression for older female. Through the study, differences in influencing factors between sexes were empirically confirmed. In conclusion, pre-diabetes prevention and management programs for Korean older adults should include sex-dependent interventions for obesity and depression. Weight management is necessary for both sexes in pre-diabetes prevention and management programs; however, exercise therapy or lifestyle changes for weight management may not be the optimal strategy for older female with depression. Therefore, older male should adopt exercise methods, physical activity, behavioral therapy, and improving lifestyle habits for weight management, whereas older female should receive depression treatment and psychological therapy.
Notes
Authors' contribution
Contributed to the conception and design of this study, performed the statistical interpretation, drafted the manuscript and critically revised the manuscript, read and approved the final manuscript - HSL
Conflict of interest
No existing or potential conflict of interest relevant to this article was reported.
Funding
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT)(No. 2023R1A2C1006271), and this work was supported by 2023 Hannam University Research Fund.
Data availability
Data is available from the https://knhanes.kdca.go.kr/knhanes/sub03/sub03_02_05.do
Acknowledgements
None.