Happiness and Its Association With Suicide Ideation and Attempt in Korea: The Roles of Socio-Environmental, Psychological, and Health-Related Factors Authors: Minjae Choi, Joshua Kirabo Sempungu, Mi-Hyui Kim, Joonhee Han, Yo Han Lee Published: 2024 ABSTRACT BackgroundThe association between happiness and suicide behaviour across different gender and age groups remains unclear, with few studies identifying potential confounding or mediating factors that explain this association. We aimed 1) to examine the association of happiness with suicide ideation/attempt and 2) to assess the relative contribution of potential factors in explaining the associations in South Korea.MethodsWe used data from the Korea Community Health Survey 2021, with 214,070 respondents aged over 20 years. Happiness was measured by using Cantril’s ladder of life satisfaction. Suicide ideation and suicide attempt were each measured by a single question. A hierarchical logistic regression model was used to identify the association between happiness and suicide ideation/attempt and estimate the relative importance of each socio-environmental, psychological, and health-related factor.ResultsUnhappiness was associated with suicide ideation/attempt across gender and age groups. The associations appeared stronger in younger rather than older adults, particularly for women aged 20–39 and 40–59 years. Psychological factors contributed the most to explaining the association across all genders and age groups. Socio-environmental factors explained the associations more among individuals aged 40–59 years and health-related factors were more pronounced in explaining the association among those aged 60 years and above, compared to other age groups.ConclusionUnhappiness is associated with suicide ideation/attempt. Psychological factors could largely explain this association, suggesting that improving population mental health could play an important role in increasing happiness and preventing suicide behaviour. INTRODUCTION There has been a growing focus on happiness, or more broadly subjective well-being, in academia, governments, and international organisations.1 This shift occurs as societies transition into the post-industrial era, where public policy evolves from a primary focus on economic development to a greater emphasis on enhancing well-being.23 This indicates that some non-economic factors become increasingly important for shaping happiness.4 The United Nations General Assembly has measured happiness among its member countries and used it to help guide public policy for increasing well-being, resulting in the annual publication of the World Happiness Report since 2011.5 Suicide behaviour has also been considered an alternative indicator of well-being as it may reflect an extremely low level of well-being.267 This suggests that studying both suicide behaviour and happiness may contribute to a better understanding of well-being. For instance, if happiness is protective of suicide behaviour, then public policies or interventions aimed at enhancing happiness may also have a positive impact on suicide prevention. However, there are a few previous studies that investigated happiness and suicide behaviour together in the general population.289101112131415 Happiness and suicide behaviour seem to be related to each other, but the relationship between them is not clear. Some studies reported happiness is associated with reduced suicide,2816 whereas other studies found weak or no statistical evidence for a correlation between happiness and suicide.91017 Additionally, another previous study showed a positive correlation between happiness and suicide at the population level.18 Most previous studies used country or area-aggregated happiness scores and suicide rates to investigate the associations,16 indicating that using smaller area-level units provides stronger evidence of the association between happiness and suicide.219 However, there was limited research on the association between happiness and suicide behaviour using individual-level data.111213141520 One study found that happiness is protective against suicide at the individual level.20 Other studies identified an association between happiness and suicide ideation and behaviour.1112131415 Furthermore, few studies investigated the gender- and age-specific effects on the strength of the associations between happiness and suicide behaviour.2820 Previous studies showed that the strength of the association between happiness and suicide rates slightly increased with age at the population level.810 However, no statistical evidence of age differences in the association was observed at the individual level. Further investigation is needed to explore other suicide-related outcomes, including suicide ideation and attempt across gender and age groups. A wide array of factors, including socioeconomic, psychological, and health-related factors, appears to contribute to both happiness and suicide and many of the factors are related to both happiness21222324 and suicide behaviour.252627 It might be important to explain why happiness was associated with suicide behaviour by identifying the explanatory factors to better understand the association. Few studies investigated the extent to which the association between happiness and suicide behaviour can be explained by factors. One previous study investigated the association between happiness and suicide at the population level in Hong Kong.2 It found a strong correlation between happiness and suicide at a small area level and suggested that area socioeconomic factors explain this association.2 Another study identified that depression explained the association between happiness and suicide at the individual level.20 However, further investigations are needed to consider the possible confounding or mediating effects of a range of factors, such as psychological distress and health behaviour, related to both happiness and suicide behaviour when investigating the association between an individual level of happiness and suicide behaviour. In South Korea (hereafter Korea), with a population of 51 million population in the East Asian region, suicide rates have been the highest among Organisation for Economic Co-operation and Development (OECD) countries with an age-standardised suicide rate of 21.16 in 2019.28 Moreover, Korea showed the lowest happiness scores in 2019 among OECD countries with available data, according to the World Happiness Report (Fig. 1).29 Although there was no statistical evidence of the association between happiness and suicide at the national level, Korea may provide a unique context for investigating the relationship between happiness and suicide behaviour at an individual level (Fig. 1). Scatterplot of age-standardised suicide rates and happiness scores among 31 Organisation for Economic Co-operation and Development member countries in 2019, with Spearman correlation coefficient (ρ) and 95% CIs. CI = confidence interval. In this study, we aimed 1) to examine the association of happiness with suicide ideation/attempt and 2) to assess the relative contribution of a range of factors in explaining the association between happiness and suicide ideation/attempt in Korea. METHODS Data source and study population We used data from the publicly available Korea Community Health Survey (KCHS), obtained from the Korea Disease Control and Prevention Agency website (https://chs.kdca.go.kr/chs). The KCHS is an annual cross-sectional survey of Korean adults conducted by the Korea Disease Control and Prevention Agency from 2008, aimed at providing information on community-level health status to develop and evaluate community health plans or initiatives.30 In the KCHS, the samples were drawn using a stratified multistage sampling method to represent the Korean population and report on their health status (e.g., physical health, mental health, and health behaviour) and social and economic characteristics.30 Annually, over 200,000 respondents participate in the KCHS with trained interviewers.30 More detailed information on the KCHS has been reported elsewhere.30 In this study, we extracted data from KCHS collected in 2021 because happiness, suicide ideation, and suicide attempts were measured only in that year. Among those aged over 20 years, our study samples were participants who responded to all questions used in this study. The final sample was 214,070 adults, including 97,344 (45.5%) men and 116,726 (54.5%) women. Measures Suicide ideation was measured using a question: ‘Have you ever considered killing yourself over the past year?’ Similarly, suicide attempt was assessed using a single question: ‘Have you ever attempted killing yourself over the past year?’ Each question is answered with either yes or no. Happiness was measured using Cantril’s ladder of life satisfaction, which has been previously employed in studies on happiness51018: ‘All things considered, how satisfied are you with your life as a whole these days?’ Responses were reported using a Likert scale that ranged from 1 (less satisfied) to 10 (more satisfied), where a higher score corresponded to a greater level of happiness. We dichotomised happiness scores as happiness and unhappiness based on the median value (median = 7). To investigate the contribution of socio-environmental factors to the association between happiness and suicide ideation/attempt, education attainment, employment status, household income, marital status, and region were included in the analysis. Educational attainment was classified as college or higher, high school, or middle school or less. Employment status was categorised into employed or self-employed and unemployed or economically inactive. For household income, we categorised into high-, middle-, and low-income groups. Marital status was categorised into three groups: married, unmarried, and divorced or bereaved. The region was determined by district-level residence and categorised into three groups: metropolitan, urban, and rural area. For psychological factors, we included variables of neighbourhood trust, perceived stress, and depression. Trust was measured by a single question: ‘Can people in your neighbourhoods trust and rely on each other?’ with a ‘yes’ or ‘no’ answer. Perceived stress was assessed using a single question: ‘How much stress do you usually feel?’ Participants who responded with ‘very much’ or ‘much’ were defined as having perceived stress. Depression was assessed using the Patient Health Questionnaire-9 (PHQ-9), a widely used scale for depression screening. The PHQ-9 comprises nine questions and scores each of the nine criteria of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, based on the frequency of depressive symptoms.31 Each question is answered on a 4-point Likert scale (0 = ‘not at all’ to 3 = ’nearly everyday’), and the nine items are summed, resulting in scores ranging from 0 to 27, where a higher score corresponded to a greater level of depressive symptoms. We categorised respondents into three groups: normal (0–9 scores), moderate depression (10–19 scores), and severe depression (20–27 scores).32 In the context of health-related factors, we included a set of variables including self-rated health, unmet healthcare, smoking, and drinking status. Self-rated health was assessed by a single question: ‘In general, would you say your health is very good, good, fair, poor, or very poor?’ and categorised into two groups: very good, good, and fair as a single group; and poor and very poor as the other group. Unmet healthcare need was measured by asking a question: ‘In the past year, have you needed hospital care or examination but did not receive it?’ We categorised participants into two groups: those without unmet healthcare experiences and those with unmet healthcare experiences. Smoking status was assessed by asking a single question ‘Do you currently smoke cigarettes?’ and smokers were defined as individuals who reported they were ‘currently smoking cigarettes’ at the time of the survey. Drinking status was measured by asking a single question, ‘In the past year, how often did you drink?’ We defined participants who reported ‘no drink within the last year’ or ‘no drink within their lifetime’ as non-drinkers, and those who reported otherwise as drinkers. For demographic factors, we included gender and age. Age was stratified into three groups: 20–39 years (young-aged adults), 40–59 years (middle-aged adults), and over 60 years (older-aged adults). Statistical analysis A descriptive analysis was conducted to estimate the prevalence of suicide ideation/attempt by happiness, gener and age group. To assess the association between happiness and suicide ideation/attempt, we used logistic regression to estimate the odds ratio (OR) and 95% confidence intervals (CIs) for suicide ideation/attempt. A Hierarchical logistic regression model was used to investigate the relative contribution of each individual or grouped factor to the association between happiness and suicide ideation/attempt. As the baseline model, we estimated the age-adjusted OR for all age groups and crude OR for age-specific groups. Furthermore, each individual explanatory variable and grouped variables (i.e., socio-environmental, psychological, and health-related factors) were included in separate models (Models 2–17).3334 Subsequently, we calculated the percentage change in the OR for both suicide ideation and suicide attempt using the provided formula35: [OR(Baseline Model) − OR(Explanatory Model)]/[OR(Baseline Model) − 1] × 100. To test the heterogeneity of the association between happiness and suicide ideation/attempt by gender and age group, we tested interaction by adding interaction terms in the final model (full adjustment). In all analyses, except for the descriptive analysis of the general characteristics of the sample, we applied sample weights assigned to all participants based on the data provided by KCHS to represent the general population in Korea. All statistical analyses were conducted using SAS version 9.4 and statistical significance was set as P value < 0.05 (SAS Institute, Cary, NC, USA). Ethics statement This study was exempted from Institutional Review Board (IRB) review at Korea University because it used secondary data without personal identifiers (IRB approval number: KUIRB-2023-0226-01). Informed consent was waived because of the retrospective nature of the study. RESULTS Table 1 summarises the characteristics of the study sample and presents the distribution of all the variables. Table 2 shows the weighted prevalence of suicide ideation and suicide attempt for men and women, respectively. The prevalence of suicide ideation and suicide attempt was higher in women than men and increased with lower socioeconomic status, psychological distress and health status. For happiness, respondents in the unhappy group showed about five times greater prevalence of suicide ideation/attempt among both men and women (e.g., prevalence of suicide ideation in the happy group among women: 2.7% vs. the unhappy group: 12.4%). Table 1General characteristics of the study populationCharacteristicsTotalMenWomenP valueaAge group, yr< 0.00120–3944,557 (20.8)21,541 (22.1)23,016 (19.7)40–5973,466 (34.3)34,364 (35.3)39,102 (33.5)60+96,047 (44.9)41,439 (42.6)54,608 (46.8)Education level< 0.001College or more71,171 (33.2)36,524 (37.5)34,647 (29.7)High school70,353 (32.9)35,744 (36.7)34,609 (29.7)Less than middle school72,546 (33.9)25,076 (25.8)47,470 (40.7)Employment status< 0.001Employed or self-employed134,469 (62.8)70,947 (72.9)63,522 (54.4)Unemployed or inactive79,601 (37.2)26,397 (27.1)53,204 (45.6)Household income< 0.001High72,010 (33.6)34,441 (35.4)37,569 (32.2)Middle74,371 (34.7)35,802 (36.8)38,569 (33.0)Low67,689 (31.6)27,101 (27.8)40,588 (34.8)Marital status< 0.001Married136,466 (63.7)66,270 (68.1)70,196 (60.1)Unmarried34,464 (16.1)19,678 (20.2)14,786 (12.7)Divorced or bereaved43,140 (20.2)11,396 (11.7)31,744 (27.2)Region< 0.001Metropolitan59,918 (28.0)26,978 (27.7)32,940 (28.2)Urban85,613 (40.0)39,393 (40.5)46,220 (39.6)Rural68,539 (32.0)30,973 (31.8)37,566 (32.2)Neighbourhood trust< 0.001Yes152,023 (71.0)68,005 (69.9)84,018 (72.0)No62,047 (29.0)29,339 (30.1)32,708 (28.0)Perceived stress< 0.001No166,809 (77.9)77,261 (79.4)89,548 (76.7)Yes47,261 (22.1)20,083 (20.6)27,178 (23.3)Depression< 0.001No206,820 (96.6)95,065 (97.7)111,755 (95.7)Moderate6,616 (3.1)2,068 (2.1)4,548 (3.9)Severe634 (0.3)211 (0.2)423 (0.4)Self-rated health< 0.001Good177,295 (82.8)84,682 (87.0)92,613 (79.3)Poor36,775 (17.2)12,662 (13.0)24,113 (20.7)Unmet healthcare need< 0.001No203,651 (95.1)93,605 (96.2)110,046 (94.3)Yes10,419 (4.9)3,739 (3.8)6,680 (5.7)Smoking< 0.001Non-smoker179,794 (84.0)66,145 (68.0)113,649 (97.4)Smoker34,276 (16.0)31,199 (32.0)3,077 (2.6)Drinking< 0.001Non-drinker94,833 (44.3)30,152 (31.0)64,681 (55.4)Drinker119,237 (55.7)67,192 (69.0)52,045 (44.6)Happiness0.074Happy91,679 (42.8)41,893 (43.0)49,786 (42.7)Unhappy122,391 (57.2)55,451 (57.0)66,940 (57.4)Suicide ideation< 0.001Yes14,966 (7.0)4,945 (5.1)10,021 (8.6)No199,104 (93.0)92,399 (94.9)106,705 (91.4)Suicide attempt< 0.001Yes728 (0.3)273 (0.3)455 (0.4)No213,342 (99.7)97,071 (99.7)116,271 (99.6)Total214,070 (100.0)97,344 (100.0)116,726 (100.0)Values are presented as number (%).aP value for gender differences. General characteristics of the study population Values are presented as number (%). aP value for gender differences. Table 2Weighted prevalence of suicide ideation and suicide attempt with 95% confidence intervalsCharacteristicsWeighted prevalence of suicide ideationWeighted prevalence of suicide attemptTotalMenWomenTotalMenWomenAge group, yr20–395.5 (5.3–5.8)3.4 (3.1–3.7)7.9 (7.5–8.3)0.4 (0.4–0.5)0.3 (0.2–0.4)0.5 (0.4–0.7)40–595.7 (5.5–5.9)4.7 (4.4–5.0)6.8 (6.4–7.1)0.3 (0.3–0.4)0.3 (0.2–0.4)0.3 (0.3–0.4)60+8.6 (8.3–8.8)6.6 (6.2–6.9)10.2 (9.8–10.6)0.3 (0.2–0.3)0.3 (0.2–0.4)0.3 (0.2–0.4)Education levelCollege or more4.5 (4.3–4.6)3.1 (2.9–3.4)5.9 (5.6–6.3)0.2 (0.1–0.2)0.1 (0.1–0.2)0.2 (0.2–0.3)High school6.6 (6.4–6.9)5.1 (4.8–5.4)8.3 (7.9–8.7)0.5 (0.4–0.5)0.4 (0.3–0.5)0.5 (0.4–0.6)Less than middle school10.9 (10.6–11.2)9.3 (8.8–9.9)11.8 (11.4–12.2)0.5 (0.4–0.6)0.5 (0.4–0.6)0.5 (0.4–0.6)Employment statusEmployed or self-employed5.0 (4.8–5.1)3.7 (3.5–3.9)6.7 (6.4–7.0)0.2 (0.2–0.3)0.2 (0.2–0.2)0.3 (0.2–0.4)Unemployed or inactive9.3 (9.0–9.5)8.1 (7.7–8.5)9.9 (9.6–10.2)0.5 (0.5–0.6)0.6 (0.5–0.7)0.5 (0.4–0.6)Household incomeHigh4.1 (3.9–4.2)2.7 (2.5–2.9)5.5 (5.2–5.8)0.2 (0.1–0.2)0.2 (0.1–0.2)0.2 (0.1–0.3)Middle6.1 (5.8–6.3)4.5 (4.2–4.8)7.7 (7.3–8.0)0.3 (0.2–0.3)0.3 (0.2–0.3)0.3 (0.3–0.4)Low12.5 (12.1–12.9)10.6 (10.0–11.2)14.0 (13.5–14.5)0.8 (0.7–0.9)0.7 (0.5–0.8)0.9 (0.7–1.0)Marital statusMarried5.2 (5.1–5.4)3.9 (3.7–4.1)6.6 (6.3–6.8)0.2 (0.2–0.2)0.2 (0.1–0.2)0.2 (0.2–0.3)Unmarried6.4 (6.1–6.7)4.6 (4.3–5.0)8.9 (8.4–9.5)0.5 (0.4–0.6)0.4 (0.3–0.5)0.7 (0.5–0.8)Divorced or bereaved12.1 (11.7–12.5)11.8 (11.0–12.7)12.2 (11.7–12.7)0.7 (0.6–0.8)0.9 (0.6–1.1)0.6 (0.5–0.7)RegionMetropolitan6.3 (6.1–6.5)4.6 (4.3–4.9)8.0 (7.6–8.3)0.4 (0.3–0.5)0.4 (0.3–0.5)0.4 (0.3–0.5)Urban6.7 (6.4–6.9)5.0 (4.7–5.2)8.3 (8.0–8.7)0.3 (0.3–0.4)0.3 (0.2–0.3)0.4 (0.3–0.4)Rural6.7 (6.5–6.9)4.8 (4.5–5.1)8.7 (8.3–9.0)0.3 (0.3–0.4)0.2 (0.2–0.3)0.4 (0.3–0.5)Neighbourhood trustYes5.0 (4.9–5.2)3.7 (3.5–3.9)6.3 (6.1–6.5)0.2 (0.2–0.3)0.2 (0.2–0.3)0.2 (0.2–0.3)No9.2 (8.9–9.5)6.7 (6.4–7.1)11.8 (11.4–12.3)0.6 (0.5–0.6)0.4 (0.3–0.5)0.7 (0.6–0.8)Perceived stressNo3.5 (3.4–3.6)2.6 (2.5–2.8)4.4 (4.2–4.5)0.1 (0.1–0.1)0.1 (0.1–0.2)0.1 (0.1–0.2)Yes15.7 (15.3–16.2)11.8 (11.3–12.4)19.2 (18.6–19.9)1.0 (0.9–1.1)0.9 (0.7–1.0)1.2 (1.0–1.3)DepressionNo4.8 (4.7–5.0)3.6 (3.5–3.8)6.1 (5.9–6.3)0.2 (0.2–0.2)0.2 (0.1–0.2)0.2 (0.2–0.2)Moderate51.2 (49.6–52.8)49.0 (46.2–51.9)52.4 (50.4–54.3)4.1 (3.4–4.7)4.1 (3.0–5.2)4.1 (3.3–4.8)Severe80.9 (76.6–85.1)80.4 (73.3–87.6)81.2 (75.9–86.4)14.3 (10.6–18.0)13.7 (7.5–19.8)14.7 (10.0–19.5)Self-rated healthGood4.7 (4.6–4.8)3.5 (3.3–3.7)5.9 (5.7–6.1)0.2 (0.2–0.2)0.2 (0.1–0.2)0.2 (0.2–0.3)Poor19.5 (18.9–20.1)17.2 (16.2–18.1)20.9 (20.1–21.6)1.3 (1.1–1.5)1.4 (1.1–1.7)1.2 (1.0–1.4)Unmet healthcare needNo6.0 (5.8–6.1)4.4 (4.2–4.5)7.5 (7.3–7.7)0.3 (0.3–0.3)0.3 (0.2–0.3)0.3 (0.3–0.4)Yes18.0 (17.0–19.1)15.2 (13.7–16.8)20.0 (18.6–21.3)1.3 (1.0–1.6)1.4 (0.9–1.9)1.3 (0.9–1.6)SmokingNon-smoker6.3 (6.2–6.5)4.1 (3.9–4.3)7.8 (7.6–8.0)0.3 (0.2–0.3)0.2 (0.1–0.2)0.3 (0.3–0.3)Smoker7.5 (7.1–7.9)6.2 (5.8–6.5)21.6 (19.7–23.5)0.8 (0.6–0.9)0.5 (0.4–0.7)3.3 (2.5–4.1)DrinkingNon-drinker8.0 (7.7–8.2)6.4 (6.0–6.8)8.8 (8.5–9.1)0.4 (0.3–0.4)0.4 (0.3–0.5)0.3 (0.3–0.4)Drinker5.7 (5.5–5.9)4.2 (4.1–4.4)7.7 (7.4–8.0)0.3 (0.3–0.4)0.3 (0.2–0.3)0.4 (0.4–0.5)HappinessHappy2.0 (1.8–2.1)1.3 (1.1–1.4)2.7 (2.5–2.9)0.1 (0.0–0.1)0.1 (0.0–0.1)0.1 (0.0–0.1)Unhappy10.0 (9.8–10.3)7.5 (7.2–7.8)12.4 (12.1–12.8)0.6 (0.5–0.6)0.5 (0.4–0.5)0.6 (0.6–0.7)Total6.5 (6.4–6.7)4.8 (4.6–5.0)8.2 (8.0–8.4)0.3 (0.3–0.4)0.3 (0.3–0.3)0.4 (0.3–0.4)Values are presented as % (95% confidence interval). Weighted prevalence of suicide ideation and suicide attempt with 95% confidence intervals Values are presented as % (95% confidence interval). Fig. 2 shows the gender- and age-specific prevalence of suicide ideation and suicide attempt by happiness. The prevalence of suicide ideation is the highest among individuals aged 60 years and older while the prevalence of suicide attempt is the highest among those aged 20–39 years. The prevalence of both suicide ideation and attempt was higher among those in the unhappy group than in the happy group across all age groups. For example, among women aged 20–39 years, the prevalence of suicide ideation was 2.1% in the happy group and 12.5% in the unhappy group. Furthermore, the gap in the prevalence of suicide attempt between the happy and unhappy groups widened more among individuals aged 20–39 years than among those aged 60 years and older (e.g., 0.06% for happy women aged 20–39 years vs. 0.92% for unhappy women aged 20–39 years). Gender and age-specific prevalence of (A) suicide ideation and (B) suicide attempt by level of happiness. Table 3 shows the association between happiness and suicide ideation in both genders in the base model and models including a range of explanatory factors. In Model 1, there was statistical evidence of an association between unhappiness and suicide ideation across all age groups in both men and women. For example, compared to individuals of all ages in the happy group, those who were unhappy had 6.20- and 5.11-times higher odds of suicide ideation in men and women, respectively (95% CI, 5.53–6.95 and 4.70–5.55 for men and women, respectively). The strength of the association between unhappiness and suicide ideation was greater in individuals aged 20–39 and 40–59 years than in those aged 60 years and older. For instance, in Model 1, the OR of suicide ideation among males aged 20–39 years was 7.27 (95% CI, 5.64–9.36), followed by those aged 40–59 years (OR, 7.02; 95% CI, 5.79–8.52), and those over 60 years (OR, 4.92; 95% CI, 4.18–5.78). Among explanatory factors, psychological factors accounted for the greatest reduction in the magnitude of association between unhappiness and suicide ideation among all age groups in both men and women in Model 15 (41.3–55.8%). Particularly, among men and women aged 20–39 years, the ORs of suicide ideation decreased the most, by 53.4% and 55.8%, respectively. For socio-environmental factors, the ORs of suicide ideation decreased by 7.7–29.3% after controlling all socio-environmental factors (Model 14). The highest reduction in the ORs was observed among men and women aged 40–59 years (29.3% for men and 15.6% for women) after accounting for all socio-environmental factors. For health-related factors, the odds of suicide ideation decreased by 15.4–35.4% after controlling all health-related factors (Model 16). Particularly, the ORs of suicide ideation substantially decreased by 35.4% and 34.6% among men and women aged 60 years and above. After including all variables in the model (Model 17), the ORs decreased by 59.9% to 67.8%, and the association between unhappiness and suicide ideation remained significant in all age groups for both genders. Table 3Gender and age-specific ORs (95% confidence intervals) for associations between happiness and suicide ideation adjusted by explanatory variables and the explained fractionModelsMenWomenAll aged20–39 yr40–59 yr60+ yr-All aged20–39 yr40–59 yr60+ yr-Model 1Age-adjusted or crude6.20 (5.53–6.95)7.27 (5.64–9.36)7.02 (5.79–8.52)4.92 (4.18–5.78)5.11 (4.70–5.55)6.85 (5.78–8.12)5.40 (4.64–6.28)3.98 (3.55–4.46)Model 2M1 + Education5.81 (5.18–6.51)7.07 (5.49–9.12)6.40 (5.26–7.77)4.60 (3.9–5.41)4.91 (4.52–5.34)6.71 (5.65–7.95)5.14 (4.42–5.98)3.82 (3.41–4.28)Change, %7.53.110.48.24.82.45.85.2Model 3M1 + Employment status5.90 (5.27–6.62)7.20 (5.6–9.27)6.12 (5.03–7.43)4.76 (4.05–5.60)5.09 (4.69–5.53)6.84 (5.77–8.11)5.36 (4.61–6.24)3.95 (3.53–4.43)Change, %5.71.015.14.00.40.00.80.8Model 4M1 + Income5.42 (4.84–6.08)6.78 (5.26–8.74)5.79 (4.76–7.04)4.35 (3.69–5.11)4.77 (4.39–5.18)6.53 (5.5–7.74)4.89 (4.2–5.68)3.73 (3.33–4.18)Change, %14.97.820.514.68.35.511.68.3Model 5M1 + Marital status5.63 (5.02–6.32)7.15 (5.56–9.21)6.13 (5.04–7.45)4.43 (3.76–5.22)4.94 (4.55–5.36)6.69 (5.63–7.94)5.13 (4.41–5.96)3.87 (3.45–4.34)Change, %10.91.814.912.54.12.76.13.5Model 6M1 + Region6.21 (5.54–6.96)7.28 (5.65–9.38)7.04 (5.80–8.54)4.92 (4.19–5.79)5.12 (4.71–5.56)6.86 (5.78–8.13)5.40 (4.64–6.28)3.99 (3.56–4.47)Change, %−0.2−0.2−0.2−0.1−0.2−0.2−0.1−0.4Model 7M1 + Neighbourhood trust5.76 (5.14–6.46)6.71 (5.20–8.66)6.52 (5.37–7.92)4.63 (3.93–5.45)4.78 (4.4–5.19)6.41 (5.41–7.6)4.92 (4.22–5.73)3.81 (3.4–4.27)Change, %8.48.98.37.48.07.410.95.5Model 8M1 + Perceived stress4.72 (4.20–5.29)5.19 (4.01–6.72)5.56 (4.58–6.76)3.70 (3.14–4.35)3.64 (3.35–3.96)4.62 (3.88–5.5)3.80 (3.26–4.43)2.99 (2.66–3.36)Change, %28.633.124.331.235.838.136.333.3Model 9M1 + Depression4.88 (4.34–5.49)5.12 (3.93–6.68)5.69 (4.67–6.93)4.00 (3.39–4.72)3.93 (3.61–4.28)4.86 (4.08–5.8)4.34 (3.72–5.06)3.11 (2.77–3.5)Change, %25.434.222.223.628.633.924.029.0Model 10M1 + Self-rated health5.24 (4.67–5.88)6.53 (5.06–8.43)6.13 (5.04–7.45)3.78 (3.21–4.46)4.40 (4.05–4.78)6.13 (5.17–7.28)4.81 (4.13–5.6)3.18 (2.83–3.58)Change, %18.511.714.828.917.212.213.426.6Model 11M1 + Unmet healthcare need5.95 (5.31–6.67)7.05 (5.47–9.08)6.73 (5.55–8.17)4.69 (3.99–5.52)4.92 (4.53–5.34)6.63 (5.59–7.87)5.19 (4.46–6.04)3.81 (3.4–4.27)Change, %4.83.54.95.84.63.64.75.5Model 12M1 + Smoking5.98 (5.33–6.71)7.19 (5.56–9.29)6.63 (5.46–8.06)4.76 (4.05–5.61)4.96 (4.57–5.39)6.53 (5.5–7.74)5.24 (4.51–6.1)3.93 (3.5–4.4)Change, %4.21.26.53.93.65.53.51.6Model 13M1 + Drinking6.17 (5.51–6.92)7.22 (5.61–9.30)7.00 (5.77–8.50)4.90 (4.17–5.76)5.11 (4.71–5.55)6.81 (5.74–8.07)5.40 (4.65–6.28)3.96 (3.54–4.44)Change, %0.50.70.30.60.00.7−0.10.4Model 14Socio-environmental factors5.03 (4.48–5.65)6.71 (5.21–8.65)5.26 (4.31–6.41)3.91 (3.31–4.61)4.62 (4.26–5.02)6.40 (5.39–7.6)4.71 (4.05–5.48)3.62 (3.23–4.06)(M2–M6) change, %22.58.829.325.811.87.715.611.8Model 15Psychological factors3.83 (3.04–4.31)3.92 (2.99–5.14)4.54 (3.72–5.54)3.14 (2.65–3.71)2.97 (2.73–3.24)3.58 (3–4.29)3.15 (2.69–3.7)2.49 (2.22–2.81)(M7–M9) change, %45.653.441.345.552.055.851.149.8Model 16Health-related factors4.90 (4.36–5.50)6.30 (4.86–8.16)5.61 (4.61–6.82)3.53 (2.99–4.17)4.18 (3.84–4.54)5.49 (4.62–6.52)4.25 (3.65–4.96)2.95 (2.62–3.31)(M10–M13) change, %25.115.423.535.422.723.226.034.6Model 17All adjusted2.96 (2.62–3.34)3.51 (2.67–4.62)3.35 (2.73–4.11)2.26 (1.90–2.70)2.59 (2.37–2.83)3.27 (2.73–3.92)2.73 (2.33–3.21)2.11 (1.87–2.38)(M1–M13) change, %62.359.961.067.861.361.260.662.8P value for difference< 0.001Change was calculated by [OR(Model 1) − OR(Explanatory Model)]/[OR(Model 1) − 1] × 100.OR = odds ratio. Gender and age-specific ORs (95% confidence intervals) for associations between happiness and suicide ideation adjusted by explanatory variables and the explained fraction Change was calculated by [OR(Model 1) − OR(Explanatory Model)]/[OR(Model 1) − 1] × 100. OR = odds ratio. Table 4 similarly shows statistical evidence of associations with suicide attempt, with much stronger associations observed among individuals aged 20–39 and 40–59 years than among those aged 60 years and older, particularly for women. For example, the least association with suicide attempt was observed among men aged 60 years and above (OR, 4.08; 95% CI, 1.95–8.51), while the strongest association was found among women aged 40–59 years (OR, 16.76; 95% CI, 8.49–33.08) in Model 1. After accounting for explanatory factors in each model, findings were similar to the association of these factors with suicide ideation. For example, psychological factors accounted for the largest reduction in the association between unhappiness and suicide ideation across all gender and age groups (55.5–71.3%; Model 15). For socio-environmental factors, the ORs of suicide attempt in males and females aged 40–59 years decreased by 48.1% and 32.1%, respectively, after controlling for all socio-environmental factors (Model 14). For health-related factors, among men and women aged 60 years and above, the ORs of suicide attempt decreased by 54.4% and 42.2%, respectively, after controlling for all health-related factors (Model 16). After including all explanatory variables in the same model (Model 17), the ORs decreased by 74.3% to 88.9%, and the association between happiness and suicide attempt remained significant only in women aged 20–39 and 40–59 years. Table 4Gender and age-specific ORs (95% confidence intervals) for associations between happiness and suicide attempt adjusted by explanatory variables and the explained fractionModelsMenWomenAll aged20–39 yr40–59 yr60+ yr-All aged20–39 yr40–59 yr60+ yr-Model 1Age-adjusted or crude OR5.78 (3.55–9.39)5.52 (2.39–12.77)7.71 (3.13–19.01)4.08 (1.95–8.51)11.64 (7.87–17.21)15.34 (7.53–31.27)16.76 (8.49–33.08)5.61 (3.01–10.45)Model 2M1 + Education5.07 (3.13–8.23)4.97 (2.15–11.48)6.43 (2.67–15.5)3.92 (1.83–8.38)10.45 (7.07–15.45)14.11 (6.93–28.73)14.45 (7.27–28.73)5.29 (2.83–9.86)Change, %14.712.219.15.011.28.614.66.9Model 3M1 + Employment status5.20 (3.21–8.42)5.25 (2.28–12.09)6.11 (2.48–15.06)3.83 (1.84–7.97)11.53 (7.79–17.05)15.32 (7.52–31.22)16.31 (8.26–32.22)5.55 (2.98–10.34)Change, %12.05.923.97.91.00.22.81.2Model 4M1 + Income4.84 (2.97–7.91)4.95 (2.13–11.49)5.63 (2.23–14.24)3.72 (1.77–7.83)9.78 (6.6–14.48)13.31 (6.51–27.22)13.41 (6.69–26.86)4.81 (2.6–8.88)Change, %19.512.531.011.417.514.221.317.4Model 5M1 + Marital status4.76 (2.87–7.87)4.88 (2.05–11.62)5.73 (2.21–14.82)3.44 (1.64–7.23)10.75 (7.27–15.91)14.04 (6.86–28.73)14.77 (7.44–29.32)5.42 (2.91–10.07)Change, %21.414.029.620.68.39.112.74.1Model 6M1 + Region5.74 (3.53–9.34)5.44 (2.35–12.59)7.69 (3.12–18.99)4.06 (1.94–8.5)11.62 (7.86–17.18)15.33 (7.53–31.22)16.74 (8.49–33.03)5.59 (3–10.42)Change, %0.71.80.30.40.20.10.10.3Model 7M1 + Neighbourhood trust5.37 (3.34–8.64)5.17 (2.3–11.59)7.13 (2.93–17.33)3.82 (1.84–7.94)10.37 (7.03–15.3)13.30 (6.54–27.07)14.77 (7.49–29.13)5.28 (2.86–9.76)Change, %8.57.78.78.311.914.212.67.0Model 8M1 + Perceived stress4.00 (2.45–6.52)3.77 (1.61–8.8)5.77 (2.35–14.19)2.44 (1.17–5.11)7.00 (4.71–10.4)9.74 (4.68–20.27)9.88 (5.02–19.45)3.23 (1.72–6.07)Change, %37.238.728.953.143.639.043.751.6Model 9M1 + Depression3.63 (2.19–6.03)3.26 (1.32–8.03)4.86 (1.95–12.15)2.70 (1.24–5.92)6.23 (4.16–9.32)7.08 (3.35–14.95)9.44 (4.72–18.89)3.42 (1.81–6.47)Change, %44.950.042.544.650.857.646.547.5Model 10M1 + Self-rated health4.34 (2.66–7.09)4.28 (1.84–9.99)5.91 (2.4–14.57)2.80 (1.31–5.96)9.15 (6.15–13.6)12.33 (6.04–25.18)13.60 (6.87–26.9)3.96 (2.08–7.52)Change, %30.027.426.941.523.421.020.135.8Model 11M1 + Unmet healthcare need5.36 (3.29–8.72)5.13 (2.21–11.95)7.28 (2.96–17.93)3.67 (1.77–7.61)10.98 (7.42–16.25)14.56 (7.13–29.74)15.88 (8.05–31.33)5.20 (2.79–9.7)Change, %8.88.56.413.36.15.55.68.7Model 12M1 + Smoking5.18 (3.18–8.45)4.85 (2.07–11.33)7.02 (2.86–17.24)3.77 (1.81–7.85)10.12 (6.84–14.98)12.48 (6.12–25.46)14.65 (7.4–29.01)5.40 (2.9–10.04)Change, %12.414.910.310.014.220.013.44.6Model 13M1 + Drinking5.73 (3.52–9.32)5.46 (2.36–12.64)7.65 (3.1–18.9)4.05 (1.94–8.46)11.62 (7.86–17.18)15.24 (7.49–31.04)16.75 (8.49–33.05)5.62 (3.02–10.47)Change, %1.01.30.90.80.20.70.0−0.3Model 14Socio-environmental factors4.02 (2.43–6.65)4.36 (1.84–10.36)4.48 (1.74–11.58)3.17 (1.46–6.91)8.96 (6.05–13.27)12.20 (5.96–24.95)11.71 (5.81–23.61)4.60 (2.48–8.53)(M2–M6) change, %36.725.548.129.325.222.032.121.8Model 15Psychological factors2.83 (1.72–4.68)2.55 (1.04–6.21)3.99 (1.62–9.81)1.88 (0.87–4.09)4.48 (2.98–6.74)5.43 (2.49–11.84)6.63 (3.32–13.22)2.35 (1.24–4.46)(M7–M9) change, %61.665.855.571.367.369.164.370.7Model 16Health-related factors3.71 (2.27–6.08)3.58 (1.52–8.48)5.17 (2.09–12.77)2.40 (1.14–5.06)7.84 (5.26–11.67)10.03 (4.9–20.56)11.79 (5.94–23.39)3.66 (1.93–6.96)(M10–M13) change, %43.242.837.954.435.737.031.642.2Model 17All adjusted1.86 (1.09–3.18)1.75 (0.69–4.44)2.60 (0.99–6.86)1.34 (0.57–3.18)3.49 (2.30–5.28)4.29 (1.95–9.42)5.06 (2.46–10.38)1.81 (0.94–3.49)(M1–M13) change, %82.083.576.188.976.677.174.382.4P value for difference0.001Change was calculated by [OR(Model 1) − OR(Explanatory Model)]/[OR(Model 1) − 1] × 100.OR = odds ratio. Gender and age-specific ORs (95% confidence intervals) for associations between happiness and suicide attempt adjusted by explanatory variables and the explained fraction Change was calculated by [OR(Model 1) − OR(Explanatory Model)]/[OR(Model 1) − 1] × 100. OR = odds ratio. Associations between other covariables (e.g., low education, low income, low trust, stress, depression, poor self-rated health, smoking, and drinking) and suicide ideation/attempt were observed in both men and women after accounting for all variables (Supplementary Table 1). DISCUSSION This study explored and identified negative associations of happiness, in the expected direction, with the risk of suicide ideation/attempt in the general population in Korea. The associations between unhappiness and suicide ideation/attempt appeared to be stronger in younger rather than older adults. Among potential explanatory factors for the association between unhappiness and suicide ideation/attempt, psychological factors contributed most to the explanation of the association across all age groups and both genders. Furthermore, the contribution of socio-environmental factors for the association was found to be higher among individuals aged 40–59 years compared to those aged 20–39 and 60 years and older. Health-related factors had a higher influence among individuals aged 60 years and older compared to those aged 20–39 years. Altogether, after adjusting for socio-environmental, psychological, and health-related factors, the associations were reduced by about 60% for suicide ideation and 80% for suicide attempt, respectively, in both genders. This study showed that unhappiness was associated with suicide ideation/attempt, which aligns with previous studies.2820363738 According to an economic theory of suicide based on the intertemporal choice theory, people may consider or attempt suicide when they perceive that the utility of continuing to live does not exceed the utility of killing themselves in the future.9 Conversely, individuals with high utility may have a lower risk of suicide as they have more to lose.9 Given that happiness may be defined as a feeling of satisfaction with life,20 people may perceive being happy as the important utility of continuing to live. Recently, as society undergoes more socioeconomic development, there has been a greater focus on improving well-being or happiness rather than only focusing on economic growth.39 This shift might be associated with psychological changes that, in turn, should impact the utility of continuing to live. Particularly, our age-stratified analyses showed that associations between unhappiness and suicide ideation/attempt appeared to be stronger in younger rather than older adults. Although average scores and prevalence of happiness were higher in individuals aged 20–39 and 40–59 years compared to those aged 60 years and older (Supplementary Table 2), the gap in prevalence of suicide ideation/attempt between those aged 20–39 and 40–59 years in the happy and unhappy groups was greater compared to those aged 60 years and above. This suggests that individuals aged 20–39 and 40–59 years may have a stronger aspiration of being happy. Younger generations in Korea often describe Korea as a country where experiences of constant social pressure have led to anxiety and frustration, making them feel more stressed and in despair.4041 Thus, well-being and work-life balance have gradually become more important under social pressure, particularly among the younger generations.42 Our study found that psychological factors play an important role in explaining the associations between unhappiness and suicide ideation/attempt across all age groups, which is consistent with a previous study.20 When individuals experience heightened unhappiness, it may worsen stress levels, leading to poor mental health conditions. Alternatively, poor mental health conditions may exacerbate unhappiness, suggesting both may contribute to suicide ideation and attempt as a way to resolve conflict.43 Given that the pathways could be associated with psychological distress, such as stress and depression, unhappiness might be an early sign of future poor mental health conditions or the beginning of suicide ideation or behaviour,20 indicating that psychological support needs to be provided appropriately. It might be also important to consider monitoring and improving population mental health, resulting in increased happiness and less suicide behaviour. A self-reported measure of happiness could provide information on population mental health and suicide risk that may otherwise have remained unnoticed, particularly among younger adults. Other different factors explained associations between happiness and suicide ideation/attempt across different age groups. The contribution of socio-environmental factors for the association was found to be greater among individuals aged 40–59 years compared to those aged 20–39 and 60 years and older, suggesting that individuals aged 40–59 years, the main active economic population, may take on greater economic pressures, particularly income, mostly because they may support their family and parents. For health-related factors, particularly, the contribution of subjective health status to the association between unhappiness and suicide ideation/attempt was higher among individuals aged 60 years and older compared to those aged 20–39 years. Previous studies identified subjective health status was associated with happiness among older adults,444546 suggesting that they may perceive health as an important consideration of being happy. Socio-environmental and health-related factors may also be important to explain the association between happiness and suicide ideation/attempt among certain groups. Although this study included several associated factors of happiness and suicide ideation/attempt, the associations between unhappiness and suicide ideation/attempt statistically remained after full adjustments. It suggests that unhappiness is independently associated with suicide ideation/attempts, or other factors not included in this study but potentially important should be considered in further studies. For example, the purpose in life or meaning of life was associated with life satisfaction and suicide ideation/attempt, particularly among youths previously in other studies.4748 To our knowledge, little research was conducted on the association between unhappiness and suicide ideation/attempt and no study has combined three groups of explanatory factors to explain the association between unhappiness and suicide ideation/attempt among the general population20 using individual-level data. The strengths of the current study include the use of nationally representative data for the general population in Korea and the availability of several measures of explanatory factors associated with unhappiness and suicide ideation/attempt. Furthermore, the large sample size may allow gender and age-stratified analyses. However, there are some methodological limitations in this study. As this study is a cross-sectional study design, it may constrain the inference of causal relationships between happiness and suicide ideation/attempt. Although a self-reported happiness measurement was widely used among several countries and international organizations, its contexts and dimensions may vary across different individuals,20 suggesting that understanding happiness might be challenging when using a single self-reported question because reports of happiness could be biased by social desirability or selective recall. Other social and psychological factors (i.e., social network, social support, subjective economic condition, purpose in life, and hopelessness) were not included in our analysis due to limited data. Further investigation should be necessary to better understand happiness and suicide behaviour, considering these methodological limitations. The current findings found that individuals who perceive themselves as unhappy were more likely to have suicide ideation/attempt across all age groups. Improving population happiness and decreasing suicide behaviour may not be determined by one single risk factor,4950 as they are considered to be the result of an interrelationship among several factors, including socio-environmental, psychological, and health-related factors. Improving the population mental health could play an important role in increasing happiness and preventing suicide behaviour. Socio-environmental and health-related factors also should be considered as important contributors to the association between unhappiness and suicide ideation/attempt. Additionally, further investigation should be necessary to better understand the interrelationship among potential explanatory factors.