Metabolic syndrome and risk of suicide death in South Korea: a nationwide cohort study

Article information

Korean J Intern Med. 2026;41(6):1114-1125
Publication date (electronic) : 2026 November 1
doi : https://doi.org/10.3904/kjim.2025.371
1Department of Endocrinology and Metabolism, Chosun University Hospital, Chosun University School of Medicine, Gwangju, Korea
2Division of Endocrinology and Metabolism, Department of Internal Medicine, Yeouido St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
3Department of Statistics and Actuarial Science, Soongsil University, Seoul, Korea
Correspondence to: Hyuk-Sang Kwon, M.D., Ph.D., Division of Endocrinology and Metabolism, Department of Internal Medicine, Yeouido St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 10, 63-ro, Yeongdeungpo-gu, Seoul 07345, Korea, Tel: +82-2-3799-1039, Fax: +82-2-786-1479, E-mail: drkwon@catholic.ac.kr, https://orcid.org/0000-0003-4026-4572
Kyungdo Han, Ph.D., Department of Statistics and Actuarial Science, Soongsil University, 369 Sangdo-ro, Dongjak-gu, Seoul 06978, Korea, Tel: +82-2-820-7025, Fax: +82-2-823-1746, E-mail: hkd917@naver.com, https://orcid.org/0000-0002-6096-1263
Received 2025 October 28; Revised 2026 April 22; Accepted 2026 May 28.

Abstract

Background/Aims

South Korea has one of the highest suicide rates worldwide. This study examined the association between metabolic syndrome (MetS) and suicide death using nationwide data.

Methods

The study used the National Health Insurance Service database in South Korea. A total of 3,951,398 individuals were followed for up to 11 years. The risk of suicide death was analyzed in those with and without MetS and its components. Kaplan–Meier analysis was conducted to assess the incidence of suicide death, and Cox proportional hazards regression was used to calculate adjusted hazard ratios (HRs) and 95% confidence intervals (CIs).

Results

During 10,657,552.48 person-years, the incidence of suicide death was higher in individuals with MetS than in those without MetS (0.35 vs. 0.26 per 1,000 person-years). In adjusted analyses, MetS was associated with suicide death (HR: 1.135; 95% CI: 1.088–1.185). Compared with individuals without any metabolic components, those with one to five components showed HRs of 1.09 (1.032–1.151), 1.223 (1.155–1.295), 1.232 (1.154–1.314), 1.358 (1.258–1.466), and 1.434 (1.277–1.609), respectively.

Conclusions

MetS showed a possible association with suicide death. Further studies are needed to confirm these findings.

Graphical abstract

INTRODUCTION

Suicide is a major public health issue and a leading cause of death worldwide. Globally, 703,000 people die by suicide each year, accounting for 1.3% of all deaths, and reducing suicide mortality has been identified as a global priority by the World Health Organization [1]. Among the member countries of the Organization for Economic Co-operation and Development, South Korea has had one of the highest suicide rates since 2003 [2]. The burden of suicide continues to increase, and identifying clinical factors associated with suicide risk may help in the early detection of high-risk individuals.

Metabolic syndrome (MetS) is a cluster of metabolic abnormalities, including impaired glucose metabolism, abdominal obesity, atherogenic dyslipidemia, and elevated blood pressure (BP) [3]. Its prevalence has increased worldwide [4,5], and in Korea, the age-adjusted prevalence rose from 27.1% in 2001 to 33.2% in 2020 [6].

MetS and its components are associated with psychological distress [7], depressive symptoms [8–10], psychological characteristics (e.g., hostility and anger) [11], and adverse personality traits [12]. Given that psychiatric disorders are major risk factors for suicide [13], it is clinically relevant to examine whether MetS is associated with suicide risk. However, few studies have evaluated the relationship between MetS and suicide mortality [14]. A previous study reported an association between elevated BP and suicide death but not with MetS overall [14]. In addition, most prior studies have focused on suicidal ideation rather than suicide mortality [15–17].

Based on this gap, we conducted a nationwide population-based study to examine the association between MetS and suicide death.

METHODS

Data

We used health insurance claims data and health examination results from the National Health Insurance Service (NHIS) of South Korea. Nearly all Koreans (97.2% of the population) are covered by the NHIS, which is a non-profit, single-payer organization by the Korean government [18]. The NHIS supports annual or biennial health checkups for all insured Koreans aged > 40 years and employees aged > 20 years. Therefore, the NHIS database contains a complete set of health information pertaining to almost the entire South Korean population, including sociodemographic information, anthropometric measurements, blood tests, completed questionnaires on lifestyle behaviors, claims for diseases (all diagnostic codes in the International Classification of Diseases, ICD-10), and treatment information.

Study cohort

The population for this study consisted of 40% randomly sampled individuals who completed health examinations from January 1 to December 31, 2009, among all Korean NHIS beneficiaries. According to the regulations of the Korean National Health Insurance Sharing Services, which limits total sample numbers and data sizes, we could only obtain a 40% simple random sampling database of the subjects who underwent a health screening examination in 2009. The index year was set at 2009, and 4,234,415 adults (aged ≥ 20 years) were initially identified. Individuals with missing data for baseline characteristics and covariates were excluded from the analysis (n = 272,763). We also excluded deaths occurring within one year from baseline (lag period) (n = 10,254). The final analytic cohort included 3,951,398 adults. A flowchart of the study population sampling is shown in Supplementary Figure 1.

This study was approved by the Chosun University Hospital Institutional Review Board (No. CHOSUN 202308030).

Exposure

The health screening examinations that the subjects underwent as NHIS beneficiaries included the measurement of body weight and height using an electronic scale, and waist circumference (WC) at the middle point between the rib cage and iliac crest by trained examiners. All blood samples were collected after fasting, and BP was measured using a sphygmomanometer after 5 minutes of rest. Additional details are provided in the Methods section in the Supplementary Appendix. Using the NHIS data, we defined the presence of the following medical conditions using claims data under their respective ICD-10 codes and medication prescriptions, where applicable: type 2 diabetes, ICD-10 codes E11–14 and prescription for antidiabetic medication; hypertension, ICD-10 codes I10, I11, I12, I13, and I15 and prescription for antihypertensive medication; dyslipidemia, ICD-10 code E78 and prescription of dyslipidemia agents; depression, ICD-10 codes F32 and F33. As for chronic kidney disease (CKD), we referred to the estimated glomerular filtration rate in the data, with a threshold of < 60 mL/ min/1.73 m2 using the Modification of Diet in Renal Disease Study equation. We defined MetS using the harmonized International Diabetes Federation criteria and adopted the Korean-specific WC cutoff for abdominal obesity [19,20]. Individuals with at least three of the following five components were diagnosed with MetS: WC of ≥ 85 cm in women or ≥ 90 cm in men, fasting plasma glucose (FPG) of ≥ 100 mg/dL or drug treatment for type 2 diabetes, BP of ≥ 130/85 mm Hg or drug treatment for hypertension, serum high-density lipoprotein cholesterol (HDL) of < 50 mg/dL in women or < 40 mg/dL in men or drug treatment for dyslipidemia, and serum triglyceride (TG) of ≥ 150 mg/dL or drug treatment for dyslipidemia.

Outcomes and follow-up

The primary outcome was death by suicide, defined as ICD-10 codes X60–X84 as recorded by the Korea National Statistical Office. The study population was followed from baseline to the date of death, or until December 31, 2021, whichever came first.

Statistical analyses

We obtained mean ± standard deviation for the continuous variables and numbers and percentages for the categorical variables. We performed group comparisons using student’s t-test or one-way analysis of variance for continuous variables and chi-squared tests for categorical variables. We calculated incidence rates by dividing the number of events by 1,000 person-years. Kaplan–Meier curves were used to estimate the cumulative incidence of suicide death, and differences between groups were assessed using the log-rank test. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Model 1 was unadjusted. Model 2 was adjusted for age, sex, income, smoking status, alcohol consumption, and regular exercise. Model 3 further included depression. Model 4 additionally included body mass index (BMI) to account for general adiposity. Subgroup analyses were performed according to age, sex, and depression status, and interactions were tested. Additional analyses were conducted according to the number and combinations of MetS components. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA), and a two-sided p value < 0.05 was considered statistically significant.

RESULTS

Characteristics of the study sample

Table 1 presents the basic characteristics of the subjects according to the presence of MetS. Individuals with MetS, compared with those without MetS, were more likely to be older, male, and have a smoking status, while also less likely to have an alcohol drinking status and exercise more. Individuals with MetS had also higher values for BMI, WC, total cholesterol, low-density lipoprotein (LDL), TG, BP, and FPG, as well as lower HDL, compared with individuals without MetS. Obesity, diabetes, hypertension, dyslipidemia, CKD, and depression were more common in individuals with MetS.

Characteristics of the subjects

The baseline characteristics according to number of MetS components are shown in Supplementary Table 1. Subjects with an increasing number of MetS components were significantly more likely to be male and older, less likely to be smokers, less likely to be drinkers, less likely to exercise more, and had a higher prevalence of obesity, diabetes, hypertension, dyslipidemia, CKD, and depression.

Suicide death risk according to the presence of MetS

As shown in Table 2, during 10,657,552.48 person-years of follow-up, the incidence rate of suicide death was higher in participants with MetS at baseline than in those without MetS (0.35 vs. 0.26 per 1,000 person-years). In Cox proportional hazards models, MetS at baseline was associated with suicide death in the unadjusted model (Model 1: HR, 1.347; 95% CI, 1.296–1.400). After adjustment for age, sex, income, smoking status, alcohol consumption, and regular exercise (Model 2), the HR was 0.978 (95% CI, 0.940–1.017). Additional adjustment for depression (Model 3) yielded an HR of 0.959 (95% CI, 0.922–0.998). In the fully adjusted model (Model 4, including BMI), MetS was significantly associated with suicide death (HR, 1.135; 95% CI, 1.088–1.185).

Hazard ratios (95% confidence interval) for suicide death according to presence of metabolic syndrome (MetS) and its components

Figure 1A shows the cumulative incidence of suicide death according to MetS status, with a higher incidence observed in individuals with MetS.

Figure 1

Kaplan–Meier curve for cumulative incidence probability of suicide death according to presence of metabolic syndrome (A) and number of metabolic syndrome components (B). Figure 1A shows the significantly increased cumulative risk for suicide death in the presence of metabolic syndrome compared with the case without metabolic syndrome. Figure 1B shows the graded cumulative increased risk of suicide death according to a number of metabolic syndrome components.

Suicide death risk according to MetS components

Table 2 shows the risk of suicide death according to each MetS component. The column for model 4 gives the multivariable-adjusted HRs of suicide death for elevated FPG, elevated TG, elevated BP, and greater WC at baseline compared with those without MetS. Meanwhile, we found no significant association between low HDL levels and suicide death. Supplementary Figure 2 shows that each MetS component revealed a significant cumulative increased suicide death risk compared with those without each MetS component.

The sum of MetS components showed a positive association with a higher risk of suicide death. Compared with individuals without any metabolic component, those with one, two, three, four, or five metabolic components at baseline had a significantly higher risk of suicide death (model 4 in Table 2), with confounding factors fully adjusted. Figure 1B shows the graded cumulative increased risk of suicide death according to a number of MetS components.

Subgroup analysis

Subgroup analyses were performed according to age, sex, and depression status, and p values for interaction were assessed (Fig. 2, Supplementary Table 2). The association between MetS and suicide death was similar across age, sex, and depression status.

Figure 2

Hazard ratios (HRs) (95% confidence interval) for suicide death according to metabolic syndrome (MetS) in different subgroups. IR, incidence rate.

In analyses of individual MetS components, no significant interactions were observed for WC, HDL, or TG. For FPG, a significant interaction with age was identified, while the pattern was similar across subgroups. For BP, interactions with sex and depression were observed, while the pattern was similar across subgroups (Supplementary Table 2, Supplementary Fig. 3).

In analyses according to the number of MetS components, a significant interaction with sex was observed, while the pattern was similar across subgroups. No significant interactions were found for age or depression (Fig. 3A, Supplementary Table 3).

Figure 3

Stratified analysis according to possible combinations of metabolic syndrome components within the same number of metabolic syndrome components (A) and hazard ratios (HRs) (95% confidence interval) for suicide death according to number of metabolic syndrome components in different subgroups (B). WC, waist circumference; GLU, fasting plasma glucose; BP, blood pressure; HDL, high-density lipoprotein cholesterol; TG, triglyceride; IR, incidence rate.

Additional analyses were conducted according to combinations of MetS components within the same number of components (Fig. 3B, Table 3). Overall, the analysis results were consistent with the main results.

Stratified analysis according to possible combinations of MetS components within the same number of MetS components

DISCUSSION

To our knowledge, this investigation is the first large-scale study to show the association of MetS with death by suicide as an outcome. In this nationwide population-based study of more than 3.9 million adults covering a period of 11 years, we found an association between MetS and suicide death compared with those without MetS at baseline. The risk of suicide death appeared to increase in a graded manner with the number of MetS components.

We could not determine the mechanisms that connect MetS and increasing risk of suicide mortality. Rather than having a single cause, suicide is a result of the complex interaction of many factors that occur throughout the individual’s life course [21]. Although many risk factors have been identified, they do not fully explain why individuals progress to suicide [22]. The same risk factor might have a different effect according to when it occurs in someone’s life [21].

Our findings suggest that metabolic factors may be related to suicide risk over time. Kim et al. [15] reported that MetS is associated with suicidal ideation. Ko et al. [17] revealed that elevated TG and elevated FPG are associated with suicidal ideation, but not with the presence of MetS. These studies evaluated limited suicidal ideation as an outcome. Suicidal ideation represents an important stage, but not all individuals who experience such thoughts proceed to suicide death, suggesting that additional processes influence this transition [21]. To date, only one large population-based report has examined MetS in relation to suicide mortality. Chang et al. [14] reported that suicide mortality is not associated with MetS but associated with elevated BP in a Taiwanese population. However, their study lacked consideration of confounders, such as health behaviors or characteristics. In our study, we adjusted for a range of possible confounders and observed an association between MetS and suicide death.

MetS is a clustering of interrelated metabolic risk factors and associated with an increased risk of cardiovascular diseases, type 2 diabetes, and all-cause mortality [23,24]. Individuals with MetS may experience a cumulative burden of chronic illness and ongoing medical care [20], and the accumulation of physical illness has been associated with suicidal behavior [25–27]. The graded increase in suicide death observed with a greater number of MetS components may reflect the overall burden of metabolic dysregulation. Awareness of serious complications related to metabolic disorders may be associated with perceptions of illness burden or self-rated health, which has been suggested as a potential mediator between metabolic factors and suicidal ideation [28,29]. In our study, the association between MetS and suicide death was generally similar across age groups, including both younger and older individuals, with no significant interaction by age. These findings suggest that the relationship may be present across the adult lifespan. Further studies are needed.

Although the role of MetS in the pathobiology of suicide remains unclear, studies have reported reduced central serotonergic activity in individuals with MetS [30], and serotonergic dysfunction has been implicated in people with suicidal ideation [31]. Increased blood levels of fatty acids might raise levels of circulating tryptophan, which may, in turn, affect brain serotonin function [32].

Obesity stigma may also be relevant. Individuals with obesity commonly experience persistent social stigma and may be less likely to receive adequate care [33]. In this context, BMI was included to account for general adiposity. Given its close relationship with other metabolic components, including WC, the possibility of overadjustment or collinearity cannot be excluded. These findings warrant cautious interpretation.

This study has several strengths. First, it is a large-scale, nationally representative cohort study. Second, the use of nationwide registers with individual-level data allowed the investigation of a relatively rare outcome, such as suicide death. We examined suicide death as the primary outcome, rather than suicidality, including suicidal ideation, as in prior studies. In addition, the classification of suicide by the Korea National Statistical Office is considered reliable. These features support the interpretation of the observed association between MetS and suicide death based on real-world data. This study has several limitations. First, suicide death is a multifactorial outcome influenced by psychiatric, psychosocial, and socioeconomic factors [34]. Although depression was included as a covariate, other established risk factors—such as prior suicide attempts, personality traits, substance use severity, chronic pain, marital status, and major life stressors—were not available in this dataset. In addition, depression was defined using claims data, and undiagnosed or subclinical cases could not be assessed, which may have resulted in residual confounding. Second, reverse causation cannot be excluded. Subclinical psychiatric vulnerability may precede behavioral changes contributing to MetS and represent a shared antecedent of both MetS and suicide death. Although the longitudinal design establishes temporal sequencing, this possibility remains. Third, the study population was derived from individuals who participated in the national health screening program. Participation may be higher among economically active individuals, and socially marginalized populations may be underrepresented, which may result in selection bias. The program, however, covers a large proportion of the Korean population. Fourth, information on the progression of MetS-related conditions, non-lethal suicidal behavior, and family history of suicide was not available. In addition, detailed information on other chronic conditions, such as cancer or heart failure, was not fully accounted for in the analysis. Fifth, generalizability may be limited, as the study was conducted in a South Korean population. Replication in other populations is needed. Lastly, suicide is influenced by cultural, lifestyle, and substance use factors, which were not captured in the NHIS database.

In this nationwide population-based cohort study, MetS showed a possible association with suicide mortality. Further studies are needed to confirm this association.

KEY MESSAGE

1. Metabolic syndrome (MetS) is associated with psychological distress and mood disorders, which may be related to suicide risk.

2. In South Korea, MetS showed a possible association with suicide mortality over 11 years of follow-up, with a graded pattern according to the number of MetS components.

3. These findings suggest a potential link between metabolic health and suicide risk.

Supplementary Information

Notes

CRedit authorship contributions

Jin Hwa Kim: conceptualization, methodology, validation, writing - original draft, writing - review & editing, visualization; Young Sang Lyu: conceptualization, writing - review & editing; Mee Kyoung Kim: conceptualization, writing - review & editing; Sang Yong Kim: conceptualization, writing - review & editing; Ki-Hyun Baek: conceptualization, writing - review & editing; Ki-Ho Song: conceptualization, writing - review & editing; Kyungdo Han: conceptualization, methodology, resources, investigation, data curation, formal analysis, validation, software, writing - review & editing; Hyuk-Sang Kwon: conceptualization, methodology, resources, validation, writing - review & editing, visualization, supervision, project administration, funding acquisition

Conflicts of interest

The authors disclose no conflicts.

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2021R1A2C2013890).

References

1. World Health Organization. Suicide worldwide in 2019 global health estimates Geneva: World Health Organization; 2021.
2. OECD. Health at a Glance 2021. Health at a Glance 2021: OECD Indicators Paris: OECD Publishing; 2021.
3. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/ National Heart, Lung, and Blood Institute Scientific Statement. Circulation 2005;112:2735–2752.
4. Hirode G, Wong RJ. Trends in the prevalence of metabolic syndrome in the United States, 2011–2016. JAMA 2020;323:2526–2528.
5. Ranasinghe P, Mathangasinghe Y, Jayawardena R, Hills AP, Misra A. Prevalence and trends of metabolic syndrome among adults in the asia-pacific region: a systematic review. BMC Public Health 2017;17:101.
6. Park D, Shin MJ, Després JP, Eckel RH, Tuomilehto J, Lim S. 20-year trends in metabolic syndrome among Korean adults from 2001 to 2020. JACC Asia 2023;3:491–502.
7. Roohafza H, Sadeghi M, Naghnaeian M, Shokouh P, Ahmadi A, Sarrafzadegan N. Relationship between metabolic syndrome and its components with psychological distress. Int J Endocrinol 2014;2014:203463.
8. Skilton MR, Moulin P, Terra JL, Bonnet F. Associations between anxiety, depression, and the metabolic syndrome. Biol Psychiatry 2007;62:1251–1257.
9. Park SJ, Roh S, Hwang J, et al. Association between depression and metabolic syndrome in Korean women: results from the Korean National Health and Nutrition Examination Survey (2007–2013). J Affect Disord 2016;205:393–399.
10. Koponen H, Jokelainen J, Keinänen-Kiukaanniemi S, Kumpusalo E, Vanhala M. Metabolic syndrome predisposes to depressive symptoms: a population-based 7-year follow-up study. J Clin Psychiatry 2008;69:178–182.
11. Goldbacher EM, Matthews KA. Are psychological characteristics related to risk of the metabolic syndrome? A review of the literature. Ann Behav Med 2007;34:240–252.
12. Sutin AR, Costa PT Jr, Uda M, Ferrucci L, Schlessinger D, Terracciano A. Personality and metabolic syndrome. Age (Dordr) 2010;32:513–519.
13. Fazel S, Runeson B. Suicide. N Engl J Med 2020;382:266–274.
14. Chang JC, Yen AM, Lee CS, et al. Metabolic syndrome and the risk of suicide: a community-based integrated screening samples cohort study. Psychosom Med 2013;75:807–814.
15. Kim SM, Lee CS, Park CS, et al. Metabolic syndrome and suicidal ideation in Korean based on the 2010 Korean National Health and Nutrition Examination Survey. Psychiatry Investig 2014;11:325–329.
16. Cho KI, Kim BH, Je HG, Jang JS, Park YH. Gender-specific associations between socioeconomic status and psychological factors and metabolic syndrome in the Korean population: findings from the 2013 Korean National Health and Nutrition Examination Survey. Biomed Res Int 2016;2016:3973197.
17. Ko JK, Han KM, Shin C, et al. Association of metabolic syndrome and its components with suicidal ideation and depression in adults: a nationally representative sample of the Korean population. J Affect Disord 2019;249:319–326.
18. Lee J, Lee JS, Park SH, Shin SA, Kim K. Cohort profile: the National Health Insurance Service-National Sample Cohort (NHIS-NSC), South Korea. Int J Epidemiol 2017;46:e15.
19. Lee SY, Park HS, Kim DJ, et al. Appropriate waist circumference cutoff points for central obesity in Korean adults. Diabetes Res Clin Pract 2007;75:72–80.
20. Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009;120:1640–1645.
21. Knipe D, Padmanathan P, Newton-Howes G, Chan LF, Kapur N. Suicide and self-harm. Lancet 2022;399:1903–1916.
22. O’Connor RC, Nock MK. The psychology of suicidal behaviour. Lancet Psychiatry 2014;1:73–85.
23. Wilson PW, D’Agostino RB, Parise H, Sullivan L, Meigs JB. Metabolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation 2005;112:3066–3072.
24. Wang K, Yang QF, Chen XL, et al. Metabolic syndrome and its components predict the risk of type 2 diabetes mellitus in the mainland Chinese: a 3-year cohort study. Int J Endocrinol 2018;2018:9376179.
25. Scott KM, Hwang I, Chiu WT, et al. Chronic physical conditions and their association with first onset of suicidal behavior in the world mental health surveys. Psychosom Med 2010;72:712–719.
26. Webb RT, Kontopantelis E, Doran T, Qin P, Creed F, Kapur N. Suicide risk in primary care patients with major physical diseases: a case-control study. Arch Gen Psychiatry 2012;69:256–264.
27. Waern M, Rubenowitz E, Runeson B, Skoog I, Wilhelmson K, Allebeck P. Burden of illness and suicide in elderly people: case-control study. BMJ 2002;324:1355.
28. Henson KE, Brock R, Charnock J, Wickramasinghe B, Will O, Pitman A. Risk of suicide after cancer diagnosis in England. JAMA Psychiatry 2019;76:51–60.
29. Ford ES. Risks for all-cause mortality, cardiovascular disease, and diabetes associated with the metabolic syndrome: a summary of the evidence. Diabetes Care 2005;28:1769–1778.
30. Muldoon MF, Mackey RH, Korytkowski MT, Flory JD, Pollock BG, Manuck SB. The metabolic syndrome is associated with reduced central serotonergic responsivity in healthy community volunteers. J Clin Endocrinol Metab 2006;91:718–721.
31. Muldoon MF, Mackey RH, Williams KV, Korytkowski MT, Flory JD, Manuck SB. Low central nervous system serotonergic responsivity is associated with the metabolic syndrome and physical inactivity. J Clin Endocrinol Metab 2004;89:266–271.
32. Virkkunen M, Goldman D, Nielsen DA, Linnoila M. Low brain serotonin turnover rate (low CSF 5-HIAA) and impulsive violence. J Psychiatry Neurosci 1995;20:271–275.
33. Rubino F, Puhl RM, Cummings DE, et al. Joint international consensus statement for ending stigma of obesity. Nat Med 2020;26:485–497.
34. Gramaglia C, Gambaro E, Bartolomei G, et al. Increased risk of metabolic syndrome in antidepressants users: a mini review. Front Psychiatry 2018;9:621.

Article information Continued

Funded by : Korea government (MSIT)
Award ID : 2021R1A2C2013890
Funding : This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2021R1A2C2013890).

Figure 1

Kaplan–Meier curve for cumulative incidence probability of suicide death according to presence of metabolic syndrome (A) and number of metabolic syndrome components (B). Figure 1A shows the significantly increased cumulative risk for suicide death in the presence of metabolic syndrome compared with the case without metabolic syndrome. Figure 1B shows the graded cumulative increased risk of suicide death according to a number of metabolic syndrome components.

Figure 2

Hazard ratios (HRs) (95% confidence interval) for suicide death according to metabolic syndrome (MetS) in different subgroups. IR, incidence rate.

Figure 3

Stratified analysis according to possible combinations of metabolic syndrome components within the same number of metabolic syndrome components (A) and hazard ratios (HRs) (95% confidence interval) for suicide death according to number of metabolic syndrome components in different subgroups (B). WC, waist circumference; GLU, fasting plasma glucose; BP, blood pressure; HDL, high-density lipoprotein cholesterol; TG, triglyceride; IR, incidence rate.

Table 1

Characteristics of the subjects

Characteristics Total Metabolic syndrome
No Yes p value
N 3,951,398 2,969,433 981,965
Sex, male 2,154,120 (54.52) 1,573,790 (53) 580,330 (59.1) < 0.0001
Age group (yr) < 0.0001
 20–39 1,219,104 (30.85) 1,076,058 (36.24) 143,046 (14.57)
 40–64 2,210,557 (55.94) 1,614,226 (54.36) 596,331 (60.73)
 ≥ 65 521,737 (13.2) 279,149 (9.4) 242,588 (24.7)
Low income 769,305 (19.47) 584,459 (19.68) 184,846 (18.82) < 0.0001
Smoke (current) 1,027,080 (25.99) 767,435 (25.84) 259,645 (26.44) < 0.0001
Alcohol drink 1,906,690 (48.25) 1,462,497 (49.25) 444,193 (45.24) < 0.0001
Regular exercise 709,597 (17.96) 523,467 (17.63) 186,130 (18.95) < 0.0001
Obesity 1,289,193 (32.63) 689,378 (23.22) 599,815 (61.08) < 0.0001
Diabetes 344,208 (8.71) 99,708 (3.36) 244,500 (24.9) < 0.0001
Hypertension 1,007,961 (25.51) 439,002 (14.78) 568,959 (57.94) < 0.0001
Dyslipidemia 686,652 (17.38) 299,069 (10.07) 387,583 (39.47) < 0.0001
CKD 275,432 (6.97) 166,466 (5.61) 108,966 (11.1) < 0.0001
Depression 120,511 (3.05) 74,383 (2.5) 46,128 (4.7) < 0.0001
Age, yr 47.28 ± 14.04 44.92 ± 13.51 54.41 ± 13.20 < 0.0001
Height (cm) 163.82 ± 9.23 164.03 ± 9.01 163.17 ± 9.84 < 0.0001
Body weight (kg) 63.86 ± 11.62 62.02 ± 10.68 69.41 ± 12.56 < 0.0001
BMI (kg/m2) 23.70 ± 3.23 22.96 ± 2.88 25.94 ± 3.20 < 0.0001
WC (cm) 80.24 ± 9.11 77.89 ± 8.14 87.34 ± 8.16 < 0.0001
FPG (mg/dL) 97.33 ± 23.85 92.66 ± 17.33 111.45 ± 33.4 < 0.0001
SBP (mmHg) 122.47 ± 15.07 119.43 ± 13.92 131.66 ± 14.67 < 0.0001
DBP (mmHg) 76.32 ± 10.07 74.66 ± 9.51 81.34 ± 10.06 < 0.0001
TC (mg/dL) 195.03 ± 36.81 192.65 ± 35.12 202.23 ± 40.69 < 0.0001
HDL (mg/dL) 56.07 ± 27.84 58.37 ± 27.95 49.15 ± 26.35 < 0.0001
LDL (mg/dL) 113.58 ± 38.63 113.39 ± 37.40 114.16 ± 42.12 < 0.0001
TG (mg/dL)a) 112.56 (112.5–112.63) 97.09 (97.03–97.15) 176.03 (175.85–176.22) < 0.0001
Death 249,830 (6.32) 142,286 (4.79) 107,544 (10.95) < 0.0001
Suicide death 12,290 (0.31) 8,563 (0.29) 3,727 (0.38) < 0.0001
Follow up duration
 Mean ± SD 11.05 ± 1.47 11.12 ± 1.30 10.85 ± 1.86 < 0.0001
 Median (Q1–Q3) 11.29 (11.1–11.56) 11.29 (11.1–11.55) 11.29 (11.07–11.58) 0.9374

Data are mean ± standard deviations or number (%).

BMI, body mass index; CKD, chronic kidney disease; DBP, diastolic blood pressure; FPG, fasting plasma glucose; HDL, high-density lipoprotein cholesterol; LDL, low-density lipoprotein cholesterol; SBP, systolic blood pressure; TG, triglyceride; WC, waist circumference.

a)

Geometric mean (95% confidence interval).

Table 2

Hazard ratios (95% confidence interval) for suicide death according to presence of metabolic syndrome (MetS) and its components

N Suicide death (n) Person-years IRa) Model 1 Model 2 Model 3 Model 4
MetS
No 2,969,433 8,563 33,005,702.82 0.26 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
Yes 981,965 3,727 10,657,552.48 0.35 1.347 (1.296, 1.4) 0.978 (0.94, 1.017) 0.959 (0.922, 0.998) 1.135 (1.088, 1.185)
N of MetS components
0 1,042,722 2,310 11,704,145.2 0.20 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
1 1,084,034 3,203 12,038,543.86 0.27 1.347 (1.277, 1.421) 1.008 (0.955, 1.064) 1.006 (0.953, 1.062) 1.09 (1.032, 1.151)
2 842,677 3,050 9,263,013.76 0.33 1.667 (1.579, 1.759) 1.047 (0.99, 1.106) 1.04 (0.984, 1.099) 1.223 (1.155, 1.295)
3 566,266 2,079 6,178,305.38 0.34 1.703 (1.605, 1.807) 0.984 (0.926, 1.046) 0.969 (0.912, 1.03) 1.232 (1.154, 1.314)
4 312,712 1,251 3,379,768.01 0.37 1.873 (1.748, 2.006) 1.028 (0.958, 1.103) 1 (0.932, 1.073) 1.358 (1.258, 1.466)
5 102,987 397 1,099,479.09 0.36 1.826 (1.642, 2.032) 0.98 (0.879, 1.092) 0.938 (0.842, 1.045) 1.434 (1.277, 1.609)
MetS components
 WC No 3,171,769 9,746 35,131,295.56 0.28 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
Yes 779,629 2,544 8,531,959.73 0.30 1.074 (1.029, 1.122) 0.871 (0.834, 0.91) 0.864 (0.827, 0.903) 1.155 (1.095, 1.219)
 FPG No 2,703,792 7,454 30,094,387.3 0.25 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
Yes 1,247,606 4,836 13,568,868 0.36 1.438 (1.387, 1.491) 1.064 (1.025, 1.104) 1.059 (1.02, 1.099) 1.116 (1.075, 1.159)
 BP No 2,228,475 5,617 24,915,464.93 0.23 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
Yes 1,722,923 6,673 18,747,790.37 0.36 1.578 (1.523, 1.635) 1.078 (1.039, 1.119) 1.072 (1.032, 1.112) 1.172 (1.128, 1.218)
 HDL No 2,811,736 8,880 31,164,621.91 0.28 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
Yes 1,139,662 3,410 12,498,633.38 0.27 0.957 (0.92, 0.996) 0.933 (0.896, 0.972) 0.912 (0.876, 0.95) 0.981 (0.941, 1.022)
 TG No 2,607,249 7,224 28,894,551.79 0.25 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
Yes 1,344,149 5,066 14,768,703.51 0.34 1.372 (1.323, 1.422) 1.025 (0.988, 1.063) 1.007 (0.971, 1.044) 1.119 (1.077, 1.162)

BMI, body mass index; BP, blood pressure; FPG, fasting plasma glucose; HDL, high-density lipoprotein; IR, incidence rate; TG, triglyceride; WC, waist circumference.

Model 1: unadjusted. Model 2: adjusted for age, sex, income, smoking status, alcohol drinking, and regular exercise. Model 3: Model 1 plus depression. Model 4: Model 2 plus BMI.

a)

Incidence per 1,000 person-years.

Table 3

Stratified analysis according to possible combinations of MetS components within the same number of MetS components

MetS components N Suicide death (n) Person-years IRa) HR (95% CI)
Sum WC FPG BP HDL TG Model 1 Model 2 Model 3 Model 4
0 No No No No No 1,042,722 2,310 11,704,145.2 0.20 1 (Ref.) 1 (Ref.) 1 (Ref.) 1 (Ref.)
1 No No No No Yes 168,073 554 1,884,116.05 0.29 1.49 (1.358, 1.634) 1.033 (0.941, 1.134) 1.026 (0.935, 1.126) 1.134 (1.032, 1.245)
1 No No No Yes No 209,943 375 2,354,843.21 0.16 0.807 (0.723, 0.9) 0.933 (0.836, 1.041) 0.935 (0.838, 1.043) 0.982 (0.88, 1.096)
1 No No Yes No No 397,219 1,410 4,370,238.82 0.32 1.634 (1.529, 1.746) 1.068 (0.999, 1.143) 1.067 (0.998, 1.141) 1.149 (1.073, 1.229)
1 No Yes No No No 234,403 714 2,598,704.52 0.27 1.392 (1.28, 1.514) 1.024 (0.941, 1.114) 1.022 (0.94, 1.112) 1.065 (0.979, 1.159)
1 Yes No No No No 74,396 150 830,641.26 0.18 0.915 (0.776, 1.079) 0.718 (0.609, 0.847) 0.719 (0.61, 0.849) 1.029 (0.869, 1.219)
2 No No No Yes Yes 118,576 344 1,326,892.98 0.26 1.313 (1.172, 1.471) 0.998 (0.891, 1.118) 0.974 (0.87, 1.092) 1.11 (0.99, 1.245)
2 No No Yes No Yes 136,249 629 1,504,901.93 0.42 2.117 (1.938, 2.312) 1.205 (1.103, 1.318) 1.2 (1.098, 1.312) 1.389 (1.269, 1.52)
2 No Yes No No Yes 75,238 328 834,201.94 0.39 1.992 (1.774, 2.236) 1.17 (1.042, 1.314) 1.164 (1.037, 1.308) 1.321 (1.175, 1.485)
2 Yes No No No Yes 38,310 100 429,057.68 0.23 1.181 (0.967, 1.443) 0.789 (0.645, 0.964) 0.779 (0.638, 0.952) 1.155 (0.942, 1.417)
2 No No Yes Yes No 82,840 228 903,862.98 0.25 1.277 (1.115, 1.463) 0.971 (0.847, 1.114) 0.965 (0.841, 1.107) 1.095 (0.954, 1.257)
2 No Yes No Yes No 48,306 106 533,633.28 0.20 1.006 (0.828, 1.222) 0.886 (0.729, 1.077) 0.879 (0.724, 1.069) 0.963 (0.792, 1.17)
2 Yes No No Yes No 26,010 44 290,519.59 0.15 0.767 (0.569, 1.034) 0.684 (0.508, 0.922) 0.679 (0.504, 0.916) 0.993 (0.735, 1.342)
2 No Yes Yes No No 197,549 907 2,126,110.78 0.43 2.16 (2, 2.332) 1.153 (1.065, 1.248) 1.152 (1.064, 1.246) 1.27 (1.173, 1.375)
2 Yes No Yes No No 86,774 259 952,052.55 0.27 1.378 (1.212, 1.566) 0.845 (0.742, 0.961) 0.843 (0.741, 0.959) 1.258 (1.099, 1.44)
2 Yes Yes No No No 32,825 105 361,780.04 0.29 1.47 (1.209, 1.788) 0.944 (0.776, 1.149) 0.94 (0.773, 1.143) 1.364 (1.118, 1.665)
3 No No Yes Yes Yes 125,974 421 1,378,444.37 0.31 1.546 (1.394, 1.715) 0.946 (0.852, 1.051) 0.912 (0.821, 1.013) 1.083 (0.973, 1.205)
3 No Yes No Yes Yes 62,130 217 685,142.88 0.32 1.604 (1.396, 1.843) 1.026 (0.892, 1.18) 0.999 (0.869, 1.149) 1.158 (1.006, 1.332)
3 Yes No No Yes Yes 34,176 77 381,779.2 0.20 1.021 (0.814, 1.282) 0.715 (0.57, 0.897) 0.695 (0.553, 0.872) 1.039 (0.825, 1.308)
3 No Yes Yes No Yes 107,150 546 1,162,176.01 0.47 2.379 (2.167, 2.612) 1.171 (1.065, 1.288) 1.167 (1.061, 1.283) 1.368 (1.242, 1.506)
3 Yes No Yes No Yes 60,837 207 672,113.2 0.31 1.56 (1.353, 1.798) 0.896 (0.777, 1.033) 0.89 (0.771, 1.026) 1.369 (1.179, 1.589)
3 Yes Yes No No Yes 25,105 69 278,698.04 0.25 1.254 (0.987, 1.594) 0.721 (0.567, 0.916) 0.719 (0.566, 0.914) 1.07 (0.839, 1.364)
3 No Yes Yes Yes No 42,197 140 446,117.44 0.31 1.588 (1.339, 1.883) 0.988 (0.832, 1.173) 0.98 (0.825, 1.163) 1.132 (0.953, 1.346)
3 Yes No Yes Yes No 29,124 81 317,269.6 0.26 1.292 (1.035, 1.613) 0.898 (0.719, 1.121) 0.89 (0.713, 1.112) 1.356 (1.081, 1.7)
3 Yes Yes No Yes No 11,204 34 123,166.17 0.28 1.398 (0.996, 1.961) 1.051 (0.749, 1.475) 1.044 (0.744, 1.465) 1.549 (1.101, 2.178)
3 Yes Yes Yes No No 68,369 287 733,398.48 0.39 1.981 (1.752, 2.239) 1.036 (0.915, 1.174) 1.028 (0.908, 1.165) 1.549 (1.359, 1.765)
4 No Yes Yes Yes Yes 123,200 517 1,317,763.17 0.39 1.986 (1.805, 2.184) 1.06 (0.962, 1.169) 1.022 (0.927, 1.127) 1.226 (1.11, 1.354)
4 Yes No Yes Yes Yes 72,569 265 793,508.68 0.33 1.691 (1.489, 1.92) 1.011 (0.889, 1.149) 0.974 (0.857, 1.108) 1.512 (1.321, 1.731)
4 Yes Yes No Yes Yes 27,058 90 298,142.88 0.30 1.529 (1.238, 1.887) 0.943 (0.763, 1.164) 0.908 (0.735, 1.121) 1.364 (1.101, 1.691)
4 Yes Yes Yes No Yes 66,517 305 722,285.41 0.42 2.138 (1.898, 2.41) 1.076 (0.953, 1.213) 1.069 (0.948, 1.206) 1.649 (1.451, 1.875)
4 Yes Yes Yes Yes No 23,368 74 248,067.87 0.30 1.509 (1.197, 1.902) 0.914 (0.725, 1.154) 0.907 (0.719, 1.145) 1.393 (1.1, 1.765)
5 Yes Yes Yes Yes Yes 102,987 397 1,099,479.09 0.36 1.827 (1.643, 2.033) 0.985 (0.884, 1.098) 0.943 (0.846, 1.051) 1.476 (1.313, 1.659)

BMI, body mass index; BP, blood pressure; CI, confidence interval; FPG, fasting plasma glucose; HDL, high-density lipoprotein; HR, hazard ratio; IR, incidence rate; MetS, metabolic syndrome; TG, triglyceride; WC, waist circumference.

Model 1: unadjusted. Model 2: adjusted for age, sex, income, smoking status, alcohol drinking, and regular exercise. Model 3: Model 1 plus depression. Model 4: Model 2 plus BMI.

a)

Incidence per 1,000 person-years.