Sage Journals: Discover world-class research

Abstract

Diabetes mellitus (DM) is strongly associated with depression, especially in women. This study was designed to investigate the gender-specific association between DM and depressive mood by family history of diabetes. Data from the Korea National Health and Nutrition Examination Survey, a population-based cross-sectional survey in 2020, were used. Of 6,133 participants aged 19 years or older, 4,259 participants were included after excluding participants without data of laboratory or physical examination, medical or family history of diseases, or depression scores of Patient Health Questionnaire–9. We examined associations of glucose and insulin metabolism, and DM with depressed mood by sex and family history of diabetes using logistic regression analyses with three stepwise models. In men, fasting glucose and HbA1c (odds ratio [OR]: 1.25, 95% confidence interval [CI]: [1.10, 1.42]) levels were significantly associated with depressed mood. Men with DM and a family history of diabetes were also significantly associated with depressed mood (OR: 1.84, 95% CI: [1.12, 3.05]), whereas DM without a family history showed no association. In women, glucose and insulin metabolism had no associations with depressed mood, and DM was also not associated with depressed mood regardless of a family history of diabetes. In Korean adults, DM with a family history of diabetes and glucose metabolism showed significant associations with depressed mood in men, but not in women. Our results suggest that men with both DM and a family history of diabetes should be paid more attention to depressed moods, considering ethnic characteristics.

Keywords

diabetes mellitus family history of diabetes sex depression

Introduction

Diabetes mellitus (DM) and depression are the major physical and mental health problems that are closely linked to each other (Danaei et al., 2011; Kessler et al., 2003). A previous study reported that individuals with newly diagnosed diabetes were 30% more likely to have a history of depression compared with those without diabetes (Brown et al., 2005). In other previous studies, the estimated rate of DM due to depression was 6.87% (Campayo et al., 2010) and the risk of general depression gradually increased with worsening glucose metabolism in younger age groups, but not in older age groups (Chen et al., 2016). In addition, another study found that the prevalence of depression in people with DM was almost twice that of people without diabetes (Berge & Riise, 2015).

A major candidate for a common pathway of DM and depressive disorder may be the activation and impairment of the stress system. Chronic stress activates the hypothalamus–pituitary–adrenal (HPA) axis and sympathetic nervous system (SNS), increasing the release of cortisol in the adrenal cortex and adrenaline and noradrenaline in the adrenal medulla (Kyrou & Tsigos, 2009). Chronic hypercortisolemia and persistent SNS activation foster insulin resistance and visceral obesity, and cause metabolic syndrome and DM (Chrousos, 2009). In contrast, chronic stress has behavioral consequences: noradrenaline, cortisol, and other hormones activate the fear system that determines anxiety, anorexia, or bulimia; the same mediators causes tachyphylaxis of the reward system, leading to depression and cravings for food, other substances, and stress (Chrousos, 2009). Excessive cortisol interferes with neurogenesis in the hippocampus, an area related to depression as well as DM (Moulton, Costafreda, et al., 2015).

A previous study on the impact of family history of diabetes on DM control and outcomes reported that people with a family history of type 2 diabetes were diagnosed with DM at a younger age and more obese with a higher incidence of cerebrovascular complication (Alharithy et al., 2018). Another study suggested that DM was associated with a higher risk of major depressive disorder in women than in men (Deischinger et al., 2020).

Previous studies have not considered the influence of the family history of diabetes on the relationship between diabetes and depression. This study was conducted to examine whether the relationship between diabetes and depressed moods is still less relevant to men than women, considering the family history of diabetes.

Materials and Methods

Study Design and Participants

The Korean Centers for Disease Control and Prevention (Korean CDC) conducts the Korea National Health and Nutrition Examination Survey (KNHANES) annually to evaluate the health-related behavior, condition, and nutrition of Koreans. This population-based cross-sectional survey is carried out by a specialized research team under the Korean CDC (Kweon et al., 2014). In this study, we utilized the KNHANES data collected from January to December 2020.

Participants in the KNHANES were selected based on a stratified, multi-stage cluster probability sampling design to achieve nationally representative samples. The study design of the KNHANES was approved by the institutional review board of the Korean CDC (IRB: 2018-01-03-2C-A). All participants in the KNHANES provided written informed consent before participating in the survey.

The research design and method described in this study were derived from the original data usage guidelines provided by the Korean CDC. For this study, we included 4,259 participants aged 19 years or older from the 6,133 participants in KNHANES 2020, after excluding those without data of blood pressure, laboratory examination, physical examination, medical history, family history of diseases, or scores of the Patient Health Questionnaire–9 (PHQ-9) (Han et al., 2008; Kroenke et al., 2001).

Socio-Demographic and Health Status Survey

Trained interviewers asked about participants’ socio-demographic factors, lifestyles, and health status, including medical and family histories. Data on smoking habits and alcohol consumption were obtained through a self-administered questionnaire. Physical examinations were performed at participants’ homes and mobile examination centers. Socioeconomic characteristics such as education level and employment status were classified. Education level was categorized into two groups based on college degree. Absolute family income (in 10,000 Won units) was used in the analyses.

Anthropometric measures were performed using standardized techniques and equipment. Height and weight were measured with a portable stadiometer (Seriter, Bismarck, ND) and a Giant-150 N calibrated balance-beam scale (Hana, Seoul, Korea), respectively. Body mass index (BMI) was calculated as each participant’ weight (kg) divided by height squared (m²). According to a meta-analysis of 239 prospective studies in four continents conducted by The Global BMI Mortality Collaboration, East Asians with BMI of 25 kg/m² or higher show significantly increased risk of death from all causes (Collaboration, 2016). Based on this result, the obesity criterion is set as a BMI of 25 for Koreans by the Korean Society for the Study of Obesity and Korea CDC. Therefore, we classified participants with a BMI of 25 or more into those with obesity.

Regular exercise was defined as high-intensity exercise ≥ 20 min and ≥ 3 times/week, or moderate intensity exercise ≥ 30 min and ≥ 5 times/week, or walking ≥ 30 min and ≥ 5 times/week. For muscle strengthening exercise, exercise days per week were used. Frequencies of alcohol consumption per month and smoking status were also included as covariates possibly associated with diabetes and depressed mood.

Measurements and Definitions of Diabetes and Insulin Resistance

Fasting blood sugar and glycosylated hemoglobin A1c (HbA1c) levels were measured using hexokinase and high-performance liquid chromatography, respectively, after fasting for at least 8 hr. Diabetes was defined as fasting plasma glucose of 126 mg/dl or higher, HbA1c of 6.5% or higher, or medically diagnosed diabetes. Serum insulin level was measured using electrochemiluminescence immunoassay (ECLIA). HOMA-IR (Homeostatic Model Assessment for Insulin Resistance) was calculated with the following formula: HOMA-IR = (fasting insulin [uU/ml] × fasting glucose [mg/dl])/405. Insulin resistance was defined as HOMA-IR of 2.54 or higher (B. Kim et al., 2018).

Measurements and Definitions of Hypertension and Dyslipidemia

For accurate blood pressure assessment, four professional nurses in charge of blood pressure measurement in the Korea CDC’s specialized investigation team measured blood pressure. After giving a break to stabilize blood pressure of study participants, nurses measured their blood pressure 3 times every 5 min. The average value of the second and third blood pressure was calculated and reported as the final blood pressure.

This study referenced the Korean Society of Hypertension guideline to present ethnic differences. Hypertension is defined as blood pressure more than 140/90 mmHg (H. C. Kim et al., 2019). We followed this guideline and divided our participants into two groups: those with hypertension or those without hypertension. Hypertensive status was defined based on a medical history of hypertension and blood pressure measurements. Serum levels of total cholesterol and triglyceride were measured using enzymatic method. Dyslipidemia was defined as having serum levels of total cholesterol of 240 mg or more, or triglycerides of 200 mg/dl or more, or taking lipid-lowering drugs.

Medical and Family History of Diabetes and Other Diseases

Interviewers asked participants about their medically diagnosed hypertension, diabetes, dyslipidemia, cardio- and cerebrovascular diseases, medication for treatment, and family members suffering from these conditions. If participants answered yes, they were defined as having a disease and a family history.

Depressed Mood Measurements

The level of depressed mood was measured using the PHQ-9 (Han et al., 2008) generally used in primary practice for screening depressive disorders and for rating depressive mood. The PHQ-9 contains nine validated depression screener-administered questions about the occurrence of depressive symptoms over the past 2 weeks. Each PHQ-9 question was scored 0 (not at all) to 3 (nearly every day). The total score ranged from 0 to 27, with 27 being the worst. The reliability and validity of the PHQ-9 for grading depression severity and diagnosing depressive disorders have been well established in previous studies (Han et al., 2008; Kroenke et al., 2001). As 82.3% and 95% of participants from the general population in our study had PHQ-9 scores below 5 and 10, respectively, PHQ-9 scores were classified into two groups with a cutoff score of 5, indicating none or minimal depression when the PHQ-9 score was less than 5 (control group) or mild to severe depression when the PHQ-9 score was 5 or higher (depressed mood group).

Statistical Analyses

All statistical analyses were performed using SPSS Statistics version 21.0 (IBM, Armonk, NY, USA). As data were from a nationally representative survey, a complex sampling design, including strata, clusters, and weights, was used for the analyses. Sample weights from the KNHANES were applied in all analyses.

Study participants were divided into three groups according to their diabetic status and a family history of diabetes: control, diabetes without a family history, and diabetes with a family history. Data were analyzed and compared among the three groups. Categorical variables were converted into dummy values for regression analyses. Variables are presented as mean ± standard error or percentage by the general linear model for continuous variables and crosstabs for categorical variables, respectively.

Significant covariates associated with depressed mood (the PHQ-9 ≥ 5) were determined using multiple logistic regression analyses according to gender. To examine associations of glucose and insulin metabolism with diabetic status by a family history of diabetes with depressed mood (PHQ-9 ≥ 5), logistic regression analyses with three models including stepwise covariates were performed according to gender. In Model 1, age and BMI were included as covariates. In Model 2, covariates with a value of p≤ .1 in previous multiple logistic regression analyses were additionally included. In Model 3, all variables were included. The p values < .05 were considered statistically significant.

Results

General Characteristics of Study Participants

Among study participants, the prevalence of diabetes without a family history of diabetes was 6.8 and that of diabetes with a family history was 8.4% (Table 1). Mean age was 45.2 ± 0.5 years for the control group, 61.9 ± 0.9 years for the diabetes group without a family history, and 59.8 ± 0.6 years for the diabetes group with a family history, with the control group showing a significantly (p < .001) younger age. The diabetes group without a family history was older than the group with a family history. However, their difference was not statistically significant (p = .054).

Table 1.

Characteristics of Study Participants According to Diabetes and FHx of Diabetes

Variables	Control	Diabetes without FHx	Diabetes with FHx
n (%)	3,323 (54.2)	418 (6.8)	518 (8.4)
Population size	25,670,246	2,492,626	3,259,226
Variable	M±SE^a	M±SE^a	M±SE^a	p
Age (year)	45.2 ± 0.5	61.9 ± 0.9	59.8 ± 0.6	<.001
Body mass index (kg/m²)^b	24.0 ± 0.1	26.2 ± 0.2	25.7 ± 0.2	<.001
Fasting blood sugar	94.4 ± 0.3	134.3 ± 2.2	137.9 ± 2.2	<.001
Glycated hemoglobin A1c	5.5 ± 0.0	7.1 ± 0.1	7.3 ± 0.1	<.001
Insulin	9.1 ± 0.2	13.3 ± 0.7	11.7 ± 0.5	<.001
HOMA-IR	2.2 ± 0.1	4.6 ± 0.3	4.0 ± 0.2	<.001
Monthly household income (10⁴ won)	525 ± 16	362 ± 22	400 ± 19	<.001
Alcohol consumption (days per month)	3.4 ± 0.1^\	3.3 ± 0.3	3.1 ± 0.2	.589
Muscle strengthening exercise (days per week)	2.4 ± 0.5	2.7 ± 0.2	2.7 ± 0.2	.036
PHQ-9 score^c	2.4 ± 0.1	2.4 ± 0.2	2.5 ± 0.2	.876
	%	%	%	p
Female	50.1	39.2	47.0	<.001
Obesity (BMI ≥ 25)	35.7	56.1	55.5	<.001
Insulin resistance (HOMA-IR ≥ 2.54)	24.6	63.9	59.7	<.001
Depressed mood (PHQ-9 ≥ 5)	17.1	20.4	17.9	.558
Current smokers	18.4	23.7	20.8	.174
Regular exercise^d	45.3	38.5	41.0	.066
Married	69.2	93.1	93.9	<.001
College degree or higher	59.2	29.6	34.6	<.001
The employed	64.6	54.2	57.8	.094
Hypertension	13.7	46.9	48.8	<.001
Dyslipidemia	10.6	37.4	46.2	<.001
Cerebrovascular disease	1.1	5.0	4.4	.001
Cardiovascular disease	1.7	5.8	8.4	<.001
Cancer diseases	3.2	9.0	4.8	.002

Note. FHx = family history; HOMA-IR = Homeostatic Model Assessment for Insulin Resistance; PHQ-9 = Patient Health Questionnaire–9; BMI = body mass index.

Mean ± standard error or % by complex samples: general linear model adjusted for age. ^bWeight in kilograms divided by height squared in meters. ^cPHQ-9 score ranged from 0 to 27, with 27 being the worst. ^dHigh-intensity exercise ≥ 20 min and ≥ 3 times/week, or moderate intensity exercise ≥ 30 min and ≥ 5 times/week or walking ≥ 30 min and ≥ 5 times/week.

The control group showed lower BMI, fasting blood sugar, HbA1c, insulin, HOMA-IR, muscle strengthening exercise days, and lower percentages of participants who were married, who had obesity, insulin resistance, hypertension, dyslipidemia, diabetes, cerebrovascular disease, and cancer compared with the other two groups (all p < .05).

The control group had higher household income, higher percentages of participants who were women, employees, and who had college degrees or higher than the two diabetes groups (all p < .05). The percentage of participants who exercised regularly and smoked and the frequency of alcohol consumption were not significantly different among the three groups (all p > .1).

The diabetes group without a family history had a significantly lower percentage of women, lower prevalence of dyslipidemia, but higher prevalence of cancer than the diabetes group with a family history (p = .022, p = .024, and p = .016, respectively). Otherwise, there were no significant differences between the two groups (all p < .05).

Comparisons of Characteristics of Study Participants According to Sex

The results of comparing age and age-adjusted variables between men and women are shown in Table 2. Significant differences were observed between men and women, except for monthly household income, family history of diabetes, and the prevalence of dyslipidemia and cardiovascular diseases.

Table 2.

Comparisons of Characteristics of Study Participants According to Sex

Variables	Men	Women
n (%)	2,167 (50.8)	2,102 (49.2)
Population size	15,962,426	15,459,672
Variable	M±SE^a	M±SE^a	p
Age (year)	46.4 ± 0.5	48.1 ± 0.5	<.001
BMI (kg/m²)^b	25.2 ± 0.1	23.5 ± 0.1	<.001
Fasting blood sugar	103.4 ± 0.6	97.1 ± 0.5	<.001
Glycated hemoglobin A1c	5.8 ± 0.0	5.7 ± 0.0	<.001
Insulin	10.4 ± 0.3	9.3 ± 0.2	.001
HOMA-IR	2.7 ± 0.1	2.3 ± 0.1	<.001
Monthly household income (10⁴ won)	516 ± 15	500 ± 13	.054
Alcohol consumption (days per month)	4.8 ± 0.1	2.1 ± 0.1	<.001
Muscle strengthening exercise (days per week)	2.3 ± 0.1	1.6 ± 0.0	<.001
PHQ-9 score^c	2.1 ± 0.1	2.8 ± 0.1	<.001
	%	%	p
Family history of diabetes	29.9	32.5	.061
Obesity (BMI ≥ 25)	47.8	30.7	<.001
Insulin resistance (HOMA-IR ≥ 2.54)	35.9	26.1	<.001
Depressed mood (PHQ-9 ≥ 5)	15.1	22.3	<.001
Current smokers	32.9	5.9	<.001
Regular exercise^d	46.6	41.2	<.001
Married	70.1	76.6	<.001
College degree or higher	57.9	50.9	<.001
The employed	73.8	53.5	<.001
Hypertension	21.7	12.7	<.001
Diabetes mellitus	15.6	10.7	<.001
Dyslipidemia	15.1	15.1	.943
Cerebrovascular disease	2.0	1.3	.040
Cardiovascular disease	2.2	1.7	.204
Cancer diseases	3.5	6.0	<.001

Note. HOMA-IR = Homeostatic Model Assessment for Insulin Resistance; PHQ-9 = Patient Health Questionnaire–9; BMI = body mass index.

Mean ± standard error or % by complex samples: general linear model adjusted for age in variables other than age. ^bWeight in kilograms divided by height squared in meters. ^cPHQ-9 score ranged from 0 to 27, with 27 being the worst. ^dHigh-intensity exercise ≥ 20 min and ≥ 3 times/week, or moderate intensity exercise ≥ 30 min and ≥ 5 times/week or walking ≥ 30 min and ≥ 5 times/week.

Women tended to have a higher prevalence of family history of diabetes (29.9% in men vs. 32.5% in women, p = .061), but better indicators of glucose and insulin metabolism were observed in women compared with men (all p≤ .001).

Men were more likely to be obese, consume alcohol, smoke, and regularly engaged in both aerobic and muscle strengthening exercises (all p < .001). Men also had a higher percentage of participants with a college degree or higher, and higher employment rates compared with women (all p < .001). While men had a higher prevalence of hypertension (p < .001), diabetes (p < .001), and cerebrovascular disease (p = .040), women had a higher prevalence of cancer (p < .001).

Multivariate Association of Covariates With Depressed Mood (PHQ-9 ≥ 5)

Multiple logistic regression showed that current smoking was associated with increasing depressed mood in men (odds ratio [OR]: 1.47, 95% confidence interval [CI]: [1.07, 2.46]) and women (OR: 3.49, 95% CI: [2.28, 5.35]) (Table 3). In contrast, in men, older age (OR: 0.97, 95% CI: [0.95, 0.99]), higher household income (OR: 0.90, 95% CI: [0.86, 0.95]), more muscle strengthening exercise (OR: 0.87, 95% CI: [0.79, 0.95]), and employment status (OR: 0.68, 95% CI: [0.48, 0.95]) were associated with decreasing depressed mood. In women, older age (OR: 0.98, 95% CI: [0.97, 0.99]), higher household income (OR: 0.96, 95% CI: [0.93, 0.99]), and employment status (OR: 0.63, 95% CI: [0.50, 0.79]) were associated with decreased depressed mood. However, muscle strengthening exercise was not associated with depressed mood in women.

Table 3.

Multivariate Association of Covariates With PHQ-9^a≥ 5 by Sex

Variables	Men	Women	Total
Variables	OR (95% CI)	OR (95% CI)	OR (95% CI)
Age (year)	0.97 [0.95 – 0.99]***	0.98 [0.97 – 0.99]*	0.98 [0.97 – 0.99]
BMI (kg/m²)	1.05 [0.99 – 1.11]	0.99 [0.94 – 1.04]	1.02 [0.98 – 1.06]
Household income (10⁴ won/month)	0.90 [0.86 – 0.95]***	0.96 [0.93 – 0.99]*	0.93 [0.91 – 0.96]
Alcohol consumption(days/month)	1.01 [0.98 – 1.04]	1.02 [0.99 – 1.06]	1.01 [0.99 – 1.04]
Muscle strengthening exercise (days/week)	0.87 [0.79 – 0.95]**	1.00 [0.92 – 1.10]	0.93 [0.87 – 0.99]
Obesity (BMI ≥ 25) (no = 0)	0.85 [0.55 – 1.31]	1.09 [0.75 – 1.57]	0.95 [0.73 – 1.24]
Current smokers (no = 0)	1.47 [1.07 – 2.46]*	3.49 [2.28 – 5.35]***	1.85 [1.42 – 2.41]
Regular aerobic exercise^b (no = 0)	0.93 [0.66 – 1.30]	1.09 [0.86 – 1.37]	1.05 [0.86 – 1.29]
Married (no = 0)	0.71 [0.42 – 1.19]	0.82 [0.57 – 1.18]	0.75 [0.53 – 1.05]
College degree or higher (no = 0)	0.95 [0.68 – 1.32]	0.84 [0.64 – 1.11]	0.80 [0.65 – 0.99]
The employed (no = 0)	0.68 [0.48 – 0.95]*	0.63 [0.50 – 0.79]***	0.65 [0.54 – 0.79]
Hypertension (no = 0)	1.40 [0.89 – 2.21]	1.31 [0.92 – 1.86]	1.31 [1.02 – 1.69]
FHx of hypertension (no = 0)	1.13 [0.82 – 1.58]	0.98 [0.77 – 1.24]	1.03 [0.85 – 1.25]
Diabetes (no = 0)	1.66 [0.98 – 2.79]	0.79 [0.53 – 1.18]	1.23 [0.98 – 1.53]
FHx of diabetes (no = 0)	1.28 [0.93 – 1.77]	1.20 [0.91 – 1.60]	1.23 [0.98 – 1.53]
Dyslipidemia (no = 0)	1.15 [0.76 – 1.74]	1.25 [0.84 – 1.86]	1.22 [0.92 – 1.61]
Cerebrovascular disease (no = 0)	1.58 [0.53 – 4.60]	0.85 [0.37 – 1.97]	1.19 [0.59 – 2.41]
Cardiovascular disease (no = 0)	1.95 [0.95 – 4.00]	1.74 [0.79 – 3.84]	2.08 [1.21 – 3.56]
Cancer diseases (no = 0)	0.97 [0.37 – 2.57]	1.15 [0.55 – 2.38]	0.91 [0.46 – 1.78]
Sex (men = 0)	1.00	1.94 [1.52 – 2.47]***

Note. By logistic regression analysis of complex model (reference group = 0 in dummy variables). PSQ-9 = Patient Health Questionnaire–9; OR = odds ratio; CI = confidence interval; BMI = body mass index; FHx = family history.

PHQ-9 score ranged from 0 to 27, with 27 being the worst. Two groups of PHQ-9 scores were divided by cut points of 5 and used in the analysis. ^bHigh-intensity exercise ≥ 20 min and ≥ 3 times/week, or moderate intensity exercise ≥ 30 min and ≥ 5 times/week or walking ≥ 30 min and ≥ 5 times/week.

p < .05. **p < .01. ***p < .001.

BMI, alcohol consumption, obesity, regular aerobic exercise, marital status, college degrees or higher, prevalence of hypertension, diabetes, dyslipidemia, cerebrovascular disease, cardiovascular disease, cancer, and family history of hypertension and diabetes were not significantly associated with depressed mood in either men or women. The prevalence of diabetes and cardiovascular disease tended to be associated with increasing depressed mood only in men (p = .059 and p = .067, respectively).

Association of Glucose and Insulin Metabolism With Depressed Mood (PHQ-9 ≥ 5) by Sex

The three models of logistic regression analyses including stepwise covariates showed sex-specific associations of glucose and insulin metabolism with depressed mood (Table 4). In Model 1, age and BMI were included as covariates in the analysis. In Model 2, covariates with a p value ≤ 0.1 in Table 2 were included (variables in Model 1 plus household income, smoking, and employment, and in men, muscle strengthening exercise and prevalence of cardiovascular disease). In Model 3, all variables in this study included.

Table 4.

Multivariate Association of Glucose Metabolism With PHQ-9^a≥ 5 According to Sex

Variables	Men	Women	Total^b
Variables	OR (95% CI)	OR (95% CI)	OR (95% CI)
Fasting blood sugar
Model 1	1.01 [1.00, 1.02]***	1.00 [0.99, 1.00]	1.01 [1.00, 1.01]**
Model 2	1.01 [1.00, 1.01]***	1.00 [0.99, 1.00]	1.00 [1.00, 1.00]
Model 3	1.05 [1.02, 1.07]**	0.97 [0.95, 1.00]	1.00 [1.00, 1.01]
Glycated hemoglobin A1c
Model 1	1.33 [1.17, 1.53]***	0.92 [0.79, 1.07]	1.14 [1.02, 1.27]*
Model 2	1.26 [1.11, 1.43]***	0.90 [0.77, 1.04]	1.10 [0.98, 1.23]
Model 3	1.25 [1.10, 1.42]**	0.88 [0.76, 1.02]	1.01 [0.98, 1.23]
Insulin
Model 1	1.01 [0.99, 1.02]	1.00 [1.00, 1.02]	1.01 [1.00, 1.02]*
Model 2	1.00 [0.99, 1.02]	1.08 [0.99, 1.02]	1.01 [0.99, 1.02]
Model 3	1.00 [0.99, 1.02]	1.01 [1.00, 1.02]	1.01 [0.99, 1.01]
HOMA-IR
Model 1	1.05 [1.01, 1.10]*	1.02 [0.98, 1.05]	1.04 [0.99, 1.07]
Model 2	1.03 [0.99, 1.08]	1.01 [0.98, 1.05]	1.02 [0.98, 1.05]
Model 3	1.02 [0.98, 1.07]	1.01 [0.98, 1.04]	1.02 [0.98, 1.05]
Insulin resistance (HOMA-IR ≥ 2.54)
Model 1	1.38 [1.04, 1.84]*	1.28 [0.99, 1.65]	1.32 [1.08, 1.62]**
Model 2	1.26 [0.94, 1.70]	1.20 [0.92, 1.57]	1.23 [0.99, 1.52]
Model 3	1.22 [0.90, 1.65]	1.21 [0.92, 1.58]	1.22 [0.98, 1.49]

Note. By logistic regression analyses of complex model. PSQ-9 = Patient Health Questionnaire–9; OR = odds ratio; CI = confidence interval; HOMA-IR = Homeostatic Model Assessment for Insulin Resistance.

PHQ-9 score ranged from 0 to 27, with 27 being the worst. Two groups of PHQ-9 scores were divided by cut points of 5 and used in the analysis. Model 1: Adjusted for age and body mass index. Model 2: Adjusted for variables in Model 1 + household income, smoking, and employment, and in men, muscle strengthening exercise and prevalence of cardiovascular disease. Model 3: Adjusted for variables in Model 2 + regular aerobic exercise, college degree or higher, prevalence of cerebrovascular disease and cancer, and in women, muscle strengthening exercise and prevalence of cardiovascular disease. ^bAdjusted for sex and variables in Models 2 and 3 included in men or women.

p < .05. **p < .01. ***p < .001.

In men, fasting glucose and HbA1c levels were significantly associated with increased depressed mood in all three models of analysis. HOMA-IR and insulin resistance (HOMA-IR ≥ 2.54) were significantly associated with increased depressed mood in Model 1 analyses, but not in Model 2 or Model 3 analyses.

In contrast, in women, there were no significant associations between glucose and insulin metabolism and depressed mood. Only borderline associations of fasting glucose and HbA1c levels with lower depressed mood were observed (p = .061 and p = .092 in Model 3 analyses, respectively), whereas insulin resistance tended to be associated with increased depressed mood in Model 1 analysis (p = .061).

Association of Diabetes According to Family History and Depressed Mood (PHQ-9 ≥ 5) by Sex

The same three models in Table 4 were applied in the analyses to evaluate associations of DM by family history of diabetes and gender with depressed mood (Table 5). In men, the association between DM and depressed mood was different according to a family history of diabetes. Diabetes with a family history of diabetes was significantly associated with an increased OR of depressed mood in all three models of analysis (OR: 1.84, 95% CI: [1.12, 3.05] in Model 3), whereas diabetes without a family history was not associated with depressed mood. In women, diabetes was not associated with depressed mood regardless of family history of diabetes (OR: 0.91, 95% CI: [0.54, 1.51] in diabetes without family history and OR: 0.84, 95% CI: [0.52, 1.35] in diabetes with family history, in Model 3).

Table 5.

Association Between Diabetes and Depressed Mood (PHQ-9^a≥ 5) According to FHx of Diabetes and Sex

Sex		Odds ratio (95% confidence interval)	p
Men
	No diabetes (reference)	1.00
Model 1	Diabetes without FHx	1.52 [0.86, 2.68]	.152
Model 1	Diabetes with FHx	2.50 [1.54, 4.06]	<.001
Model 2	Diabetes without FH_X	1.31 [0.76, 2.25]	.337
Model 2	Diabetes with FHx	1.97 [1.17, 3.34]	.012
Model 3	Diabetes without FHx	1.02 [0.60, 1.76]	.936
Model 3	Diabetes with FHx	1.84 [1.12, 3.05]	.017
Women
	No diabetes (reference)	1.00
Model 1	Diabetes without FHx	1.11 [0.70, 1.78]	.650
Model 1	Diabetes with FHx	1.00 [0.64, 1.58]	.987
Model 2	Diabetes without FH	1.00 [0.61, 1.64]	.995
Model 2	Diabetes with FHx	0.85 [0.53, 1.36]	.491
Model 3	Diabetes without FHx	0.91 [0.54, 1.51]	.702
Model 3	Diabetes with FHx	0.84 [0.52, 1.35]	.461
Total^b
	No diabetes (reference)	1.00
Model 1	Diabetes without FHx	1.02 [0.98, 1.05]	.177
Model 1	Diabetes with FHx	1.56 [1.11, 2.22]	.011
Model 2	Diabetes without FH	1.10 [0.77, 1.57]	.603
Model 2	Diabetes with FHx	1.37 [0.96, 1.94]	.080
Model 3	Diabetes without FHx	1.09 [0.77, 1.56]	.627
Model 3	Diabetes with FHx	1.35 [0.95, 1.92]	.088

Note. By logistic regression analysis of complex model. PHQ-9 = Patient Health Questionnaire–9; FHx = family history.

Discussion

Results of this population-based cross-sectional study showed that glucose and insulin metabolism and diabetes were associated with depressed mood according to sex and a family history of diabetes. Glucose and insulin metabolism were significantly associated with depressed mood in men, but not in women. DM with a family history of diabetes was also significantly associated with depressed mood only in men.

Depression and depressive symptoms can increase the risk of developing insulin resistance (Shomaker et al., 2011) and incident diabetes (Hasan et al., 2014). As a result of meta-analysis, the cumulative incidence of DM in people with depression was higher than that in people without depression. The relative risk and hazard ratios for incident diabetes with depressive symptoms have been reported to be 1.67 (95% CI: [1.30, 2.15]) and 1.45 (95% CI: [1.12, 1.87]), respectively (Hasan et al., 2014).

Three possible directions of type 2 diabetes and depression as comorbid conditions have been suggested. First, common etiology plays a role in increasing a person’s risk of both diseases. Second, persons with type 2 diabetes show increased prevalence or risk of developing depression. Third, persons with depression show increased prevalence or risk of developing type 2 diabetes (Berge & Riise, 2015).

Consistent evidence shows that depression and diabetes share common psychological and biological mechanisms. Fetal or maternal stress in the uterus, cumulative exposure to chronic stress due to low socioeconomic status, and poor health behavior in people with genetic predispositions can promote dysregulation of the HPA axis, promote circadian rhythm problems, and function as toxins that can activate innate inflammatory responses (Moulton, Pickup, & Ismail, 2015). Dysregulation of biological pathways can lead to insulin resistance, type 2 diabetes, depression, dementia, and cardiovascular disease (Moulton, Pickup, & Ismail, 2015).

Higher prevalence of depression in women is a universal phenomenon worldwide (Parker & Brotchie, 2010). The reason for such sex-specific difference in depressive disorder remains unclear. However, recent research has suggested that ovarian hormonal changes over menstrual cycles are associated with variations in brain network connections, with a relationship between cycling ovarian hormone levels and symptoms of affective disorder in women being the most important evidence (Altemus et al., 2014; Andreano et al., 2018).

In our study, a significant relationship between diabetes and depressed mood was observed only in men with a family history of diabetes, but not in women, different from a previous study reporting a more significant association of diabetes with depressed mood in women than in men (Deischinger et al., 2020). Other studies suggested that the relationship between psychiatric disorders including major depression and general anxiety disorder and cardiovascular diseases might depend on race and ethnicity (Assari & Lankarani, 2014). In addition, perceived racism in diabetes treatment might be more pronounced in the control of blood sugar in Black men than in Black women (Assari et al., 2017).

The greatest risk of major depression in the Austrian population with diabetes has been observed in overweight women aged 40 to 49 (Deischinger et al., 2020), while a previous study found that depression in middle-aged Korean women was not related to obesity (Jung et al., 2021). In addition, many Korean people with diabetes have normal body weights, unlike Caucasians in the West (Park et al., 1997). In the present study, obesity was not associated with depressed mood in men or women either. These results may indicate ethnic and sex-specific relationship between chronic diseases and mental health.

According to findings of this study, glucose metabolism and diabetes with a family history were significantly associated with depressed mood in men, but not in women, suggesting that women’s depressive mood was more likely to be related to other factors such as hormonal fluctuations during menstruation cycles, lifestyle, socioeconomic status, and cultural background.

There have been a few studies on the impact of a family history of diabetes on patients with diabetes. To the best of our knowledge, studies on associations of diabetes by gender and a family history of diabetes with depressed mood have not been reported yet. Previous studies have reported that people with diabetes with a family history of diabetes smoked more (Alharithy et al., 2018; Chang, 2012) and were diagnosed at a younger age with higher BMI, larger waist circumference, and worse cerebrovascular outcomes (Alharithy et al., 2018; Annis et al., 2005; Jali et al., 2009). In this study, diabetic patients with a family history of diabetes tended to be younger (p = .056) and have more women (p = .022) than those without a family history of diabetes. However, smoking was not significantly related to diabetes regardless of the family history. We conducted additional analyses after adjusting for age and BMI to examine sex-specific association between smoking and diabetes by a family history of diabetes. There were still no associations between them in men (p = .599) or women (p = .508). The prevalence of cerebrocardiovascular disease showed no significant difference by a family history of diabetes either.

Interestingly, although not significant, in our study, patients with both DM and a family history of diabetes were less likely to smoke and drink, and more likely to exercise than those without a family history of diabetes. Depressed mood in men with both DM and a family history of diabetes might be due to being more obsessive and anxious about blood sugar control and preventing complications of diabetes than those without a family history. In Korean society, smoking and alcohol consumption are generous to men. According to the Korean government’s 2021 statistics (https://www.index.go.kr/unify/idx-info.do?idxCd=4238), the smoking rate of men is 31.3% and that of women is 6.9%, which is overwhelmingly high among men. In addition, the excessive drinking rate was 47.0% for men and 24.1% for women, which was also much higher for men. In Korean culture, unlike Western society, people drink alcohol with coworkers, friends, and family rather than drinking alone. Sometimes drinking is even considered essential for better social activities and relationships in men. Therefore, limitations of lifestyle for treating diabetes could cause contractions in social activities or relationships, especially in men.

There are some weaknesses in our research. First, as this was a cross-sectional study, the causal relationship cannot be determined. Second, as data on family history of diabetes were collected based on participants’ answers, there might be an error in missing family members with diabetes. Besides, the data utilized in this study were gathered in 2020 during the ongoing COVID-19 pandemic, thus potentially being influenced by its effects. Consequently, it is conceivable that the study’s findings could differ in the absence of a pandemic. In a previous study, the weighted prevalence of depression in 2016, 2018, and 2020 was 4.0%, 2.4%, and 4.2%, respectively, for men, and 7.2%, 6.0%, and 6.2% respectively, for women in Korea. These percentages indicate a decreasing trend in the prevalence of depression before COVID-19. However, during the COVID-19 pandemic, there has been a shift toward an increasing trend, with a significant rise observed in men compared with women (Jeong et al., 2023). Therefore, it could be assumed that men were more likely to be affected by the COVID-19 pandemic. Further studies will be needed to clarify the trend of depression and the relationship between COVID-19 and depression according to sex. Despite these limitations, this study has a strength in that it is a population-based research using national representative data including various confounding factors.

In conclusion, glucose metabolism and DM showed associations with depressed mood depending on sex and family history of diabetes. Glucose metabolism was significantly associated with depressed mood in men, but not in women. DM with a family history of diabetes was also significantly associated with depressed mood only in men. In general, the prevalence of depression is higher in women than in men, but men with both DM and a family history of diabetes should receive more attention to depressive mood along with consideration of ethnic and cultural differences. Further studies are needed to elucidate the causal relationship between diabetes and mental health by sex and a family history of diabetes along with the mechanisms, ethnic differences, and social environment such as COVID-19 pandemic.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Chan-Hee Song

References

Alharithy

M. K.

Alobaylan

M. M.

Alsugair

Z. O.

Alswat

K. A.

(2018). Impact of family history of diabetes on diabetes control and complications. Endocrine Practice, 24, 773–779.

Altemus

Sarvaiya

Neill Epperson

(2014). Sex differences in anxiety and depression clinical perspectives. Frontiers in Neuroendocrinology, 35, 320–330.

Andreano

J. M.

Touroutoglou

Dickerson

Barrett

L. F.

(2018). Hormonal cycles, brain network connectivity, and windows of vulnerability to affective disorder. Trends in Neuroscience, 41, 660–676.

Annis

A. M.

Caulder

M. S.

Cook

M. L.

Duquette

(2005). Family history, diabetes, and other demographic and risk factors among participants of the National Health and Nutrition Examination Survey 1999–2002. Preventing Chronic Disease, 2, Article A19.

Assari

Lankarani

M. M.

(2014). Race and ethnic differences in the associations between cardiovascular diseases, anxiety, and depression in the United States. International Journal of Travel Medicine and Global Health, 2, 107–113.

Assari

Lee

D. B.

Nicklett

E. J.

Moghani Lankarani

Piette

J. D.

Aikens

J. E.

(2017). Racial discrimination in health care is associated with worse glycemic control among black men but not black women with type 2 diabetes. Frontiers in Public Health, 5, Article 235.

Berge

L. I.

Riise

(2015). Comorbidity between type 2 diabetes and depression in the adult population: Directions of the association and its possible pathophysiological mechanisms. International Journal of Endocrinology, 2015, Article 164760.

Brown

L. C.

Majumdar

S. R.

Newman

S. C.

Johnson

J. A.

(2005). History of depression increases risk of type 2 diabetes in younger adults. Diabetes Care, 28, 1063–1067.

Campayo

de Jonge

Roy

J. F.

Saz

de la Cámara

Quintanilla

M. A.

Marcos

Santabárbara

Lobo

(2010). Depressive disorder and incident diabetes mellitus: The effect of characteristics of depression. American Journal of Psychiatry, 167, 580–588.

10.

Chang

S. A.

(2012). Smoking and type 2 diabetes mellitus. Diabetes & Metabolism Journal, 36, 399–403.

11.

Chen

Zhang

Dai

Zhu

(2016). Association of depression with pre-diabetes, undiagnosed diabetes, and previously diagnosed diabetes: A meta-analysis. Endocrine, 53, 35–46.

12.

Chrousos

G. P.

(2009). Stress and disorders of the stress system. Nature Reviews Endocrinology, 5, 374–381.

13.

Danaei

Finucane

M. M.

Singh

G. M.

Cowan

M. J.

Paciorek

C. J.

Lin

J. K.

Farzadfar

Khang

Y. H.

Stevens

G. A.

Rao

Ali

M. K.

Riley

L. M.

Robinson

C. A.

Ezzati

(2011). National, regional, and global trends in fasting plasma glucose and diabetes prevalence since 1980: Systematic analysis of health examination surveys and epidemiological studies with 370 country-years and 2·7 million participants. Lancet, 378, 31–40.

14.

Deischinger

Dervic

Leutner

Kosi-Trebotic

Klimek

Kautzky

Kautzky-Willer

(2020). Diabetes mellitus is associated with a higher risk for major depressive disorder in women than in men. BMJ Open Diabetes Research & Care, 8, Article e001430.

15.

Global BMI Mortality Collaboration, Di Angelantonio

Bhupathiraju

Wormser

Gao

Kaptoge

de Gonzalez

A. B.

Cairns

Huxley

Jackson

Joshy

Lewington

Manson

Murphy

Patel

Samet

Woodward

Zheng

Zhou

. . . Hu

(2016). Body-mass index and all-cause mortality: Individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet, 20, 776–786.

16.

Han

S. A.

Kwak

J. H.

Pae

C. U.

Steffens

Park

M. H.

(2008). Validation of the Patient Health Questionnaire–9 Korean version in the elderly population: The Ansan Geriatric study. Comprehensive Psychiatry, 49, 218–223.

17.

Hasan

S. S.

Clavarino

A. M.

Mamun

A. A.

Kairuz

(2014). Incidence and risk of diabetes mellitus associated with depressive symptoms in adults: Evidence from longitudinal studies. Diabetes & Metabolic Syndrome, 8, 82–87.

18.

Jali

M. V.

Kambar

Jali

S. M.

Gowda

(2009). Familial early onset of type-2 diabetes mellitus and its complications. North American Journal of Medicine and Science, 1, 377–380.

19.

Jeong

Yim

H. W.

Lee

S. Y.

Jung

D. Y.

(2023). Impact of the COVID-19 pandemic on gender differences in depression based on national representative data. Journal of Korean Medical Science, 38, Article e36.

20.

Jung

K. I.

Ock

S. M.

Song

C. H.

(2021). Weight change and mental health according to obesity in relatively healthy middle-aged Korean women: The Korea National Health and Nutrition Examination Survey 2014, 2016, 2018. Journal of Affective Disorders, 287, 174–181.

21.

Kessler

R. C.

Berglund

Demler

Jin

Koretz

Merikangas

K. R.

Rush

A. J.

Walters

E. E.

Wang

P. S.

(2003). The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R). JAMA, 289, 3095–3105.

22.

Kim

Choi

H. Y.

Kim

Ahn

Lee

Kim

J. G.

Kim

Shin

J. M.

Moon

(2018). The cut-off values of surrogate measures for insulin resistance in the Korean population according to the Korean Genome and Epidemiology Study (KOGES). PLOS ONE, 13, Article e0206994.

23.

Kim

H. C.

Ihm

S.-H.

Kim

G.-H.

Kim

J. H.

Kim

K.-i.

Lee

H.-Y.

Lee

J. H.

Park

J.-M.

Park

Pyun

W. B.

Shin

Chae

S. C.

(2019). 2018 Korean Society of Hypertension guidelines for the management of hypertension: Part I—Epidemiology of hypertension. Clinical Hypertension, 25, Article 16.

24.

Kroenke

Spitzer

R. L.

Williams

J. B. W.

(2001). The PHQ-9. Journal of General Internal Medicine, 16, 606–613.

25.

Kweon

Kim

Jang

M. J.

Kim

Choi

Chun

Khang

Y. H.

(2014). Data resource profile: The Korea National Health and Nutrition Examination Survey (KNHANES). International Journal of Epidemiology, 43, 69–77.

26.

Kyrou

Tsigos

(2009). Stress hormones: Physiological stress and regulation of metabolism. Current Opinion in Pharmacology, 9, 787–793.

27.

Moulton

C. D.

Costafreda

S. G.

Horton

Ismail

C. H.

(2015). Meta-analyses of structural regional cerebral effects in type 1 and type 2 diabetes. Brain Imaging and Behavior, 9, 651–662.

28.

Moulton

C. D.

Pickup

J. C.

Ismail

(2015). The link between depression and diabetes: The search for shared mechanisms. The Lancet Diabetes & Endocrinology, 3, 461–471.

29.

Park

J. Y.

Lee

K. U.

Kim

C. H.

Kim

H. K.

Hong

S. K.

Park

K. S.

Lee

H. K.

Min

H. K.

(1997). Past and current obesity in Koreans with non-insulin-dependent diabetes mellitus. Diabetes Research and Clinical Practice, 35, 49–56.

30.

Parker

Brotchie

(2010). Gender differences in depression. International Review of Psychiatry, 22, 429–436.

31.

Shomaker

L. B.

Tanofsky-Kraff

Stern

E. A.

Miller

Zocca

J. M.

Field

S. E.

Yanovski

S. Z.

Hubbard

V. S.

Yanovski

J. A.

(2011). Longitudinal study of depressive symptoms and progression of insulin resistance in youth at risk for adult obesity. Diabetes Care, 34, 2458–2463.

Men With Both Diabetes and a Family History of Diabetes Were Associated With Depressed Mood in Korean Adults

Abstract

Keywords

Introduction

Materials and Methods

Study Design and Participants

Socio-Demographic and Health Status Survey

Measurements and Definitions of Diabetes and Insulin Resistance

Measurements and Definitions of Hypertension and Dyslipidemia

Medical and Family History of Diabetes and Other Diseases

Depressed Mood Measurements

Statistical Analyses

Results

General Characteristics of Study Participants

Comparisons of Characteristics of Study Participants According to Sex

Multivariate Association of Covariates With Depressed Mood (PHQ-9 ≥ 5)

Association of Glucose and Insulin Metabolism With Depressed Mood (PHQ-9 ≥ 5) by Sex

Association of Diabetes According to Family History and Depressed Mood (PHQ-9 ≥ 5) by Sex

Discussion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

References