National Trends in the Prevalence of Suicide Attempts Among Adolescents by Self-Perceived Weight, 2005–2023: A Nationwide Representative Study in South Korea

Study population and data collection

This study utilized data from the Korea Youth Risk Behavior Web-based Survey (KYRBS) for its analysis [13,14]. Since 2005, KYRBS has been conducted annually as an anonymous, self-reported online survey to identify adolescents’ health behaviors, including smoking, drinking, eating habits, and physical activity. The survey is conducted on students from the first year of middle school to the third year of high school. The KYRBS employed a stratified, multistage cluster sampling design to ensure national representativeness. The sampling frame encompassed all middle and high schools nationwide. The sample was evenly distributed between middle and high schools (50:50). These divisions into metropolitan cities, small-to-medium-sized cities, and rural areas were further categorized into 39 regional clusters. Within the selected schools, students were randomly sampled, with one class per grade chosen to participate in the survey. A complete-case analysis was performed, excluding participants with missing values. This analysis covered data from 2005 to 2023, including a total of 1,156,728 participants. The study protocol was approved by the Institutional Review Board of Korea Disease Control and Prevention Agency (2014-06EXP-02-P-A) for primary data collection and Kyung Hee University (KHSIRB-25-125) for secondary analysis. Informed consent was obtained from all participants (or their parents or legal guardians in the case of children under 16) prior to the commencement of the study, with a full explanation provided. The study was conducted in complete accordance with the ethical principles outlined in the Declaration of Helsinki.

Ascertainment of self-perceived weight and suicide attempt

This study analyzes the relationship between self-perceived weight and suicide attempts, incorporating, BMI as a variable. Self-perceived weight was defined using the question, “How do you perceive your body shape?” Those who answered “very thin” or “slightly thin” to this question were classified as underweight, those who answered “average” were classified as normal weight, and those who answered “slightly fat” or “very fat” were classified as overweight [15,16]. To define BMI, the questions “What is your current height?” and “What is your current weight?” were used. Based on the responses, BMI was calculated as weight in kilograms divided by height in meters squared and divided into four groups based on the Korean National Growth Charts: underweight (<5 percentile), normal (5 to 84 percentile), overweight (85 to 94 percentile), and obese (≥95 percentile) [17]. Suicide attempts were defined based on the question, “In the past 12 months, have you attempted suicide?” Respondents who answered “yes” were classified as individuals who had made a suicide attempt [18].

Covariates

This study analyzed the following 12 covariates: sex (male and female), grade (7th to 9th grade and 10th to 12th grade), region of residence (urban and rural), BMI group, self-perceived weight (underweight, normal weight, and overweight), academic achievement (high, middle, and low), economic level (high, middle, and low), suicide attempts (no and yes), sadness (no and yes), stress (low and high), smoking status, and alcohol consumption. All covariates were derived from the survey questionnaire. In particular, the BMI group was classified using the Korean National Growth Charts to ensure age- and sex-specific relevance for adolescents in South Korea [5]. These variables were chosen due to their potential significance to the study, as highlighted in previous research.

In addition, we examined body image distortion, defined as the discrepancy between BMI-based weight status and self-perceived weight. Cases in which the two categories were consistent were classified as “distortion absent,” whereas mismatches were classified as “distortion present.” [15]

Statistical analyses

To examine the influence of self-perceived weight on suicide attempts, we employed a series of analytical models. First, descriptive statistics with frequencies and percentages were computed to summarize the general characteristics of the study population. Given the evolving context of the COVID-19 pandemic, the data were divided into two periods—pre-pandemic and during the pandemic—to facilitate period-specific comparisons [19,20]. To assess temporal trends in self-perceived weight and its association with suicide attempts, linear regression models were employed to compute β coefficients and β_diff with 95% confidence intervals (CIs). To estimate associations, survey-weighted logistic regression models were used to obtain weighted odds ratios (wORs) and their 95% CIs. To further investigate sex-specific differences, we incorporated interaction terms between self-perceived weight (including body image distortion) and sex into logistic regression models to test for potential effect modification. We also assessed multicollinearity among covariates (BMI, self-perceived weight, body image distortion, sex, and age group) using variance inflation factors (VIFs), tolerance values, and condition indices. All VIFs were below 2, tolerance values exceeded 0.65, and the maximum condition index was 17.9, indicating no indication of problematic multicollinearity. All analyses accounted for the complex survey design by incorporating strata, cluster, and weight variables provided by KYRBS, in accordance with the official analytic guidelines, to ensure unbiased variance estimation [21]. All statistical analyses were conducted using SAS software (version 9.4; SAS Institute), and a two-sided p-value of less than 0.05 was considered statistically significant.

Baseline characteristics

This study analyzed a total of 1,156,728 individuals, consisting of 596,030 males (51.53%) and 560,698 females (48.47%), from 2005 to 2023. The BMI analysis of the students who participated in this survey showed that 93,996 (8.13%) students were underweight, 872,361 (75.42%) had a normal weight, 95,975 (8.30%) were overweight, and 94,396 (8.16%) were classified as obese. In addition, when examining self-perceived weight, 318,372 (27.52%) students perceived themselves as underweight, 422,634 (36.54%) as having a normal weight, and 415,722 (35.94%) as overweight. There were significantly more people with a normal BMI than those who perceived themselves as having a normal weight (Table 1 and Supplementary Table 1).

Table 1.

Baseline characteristics of Korean adolescents from the KYRBS, 2005–2023

Sex differences in suicide attempts by body perception

A comparison of suicide attempts based on self-perceived weight suggested that individuals who perceived themselves as overweight (3.97 [95% CI, 3.89 to 4.04]) had the highest rate of suicide attempts, followed by those who perceived themselves as underweight (3.36 [95% CI, 3.28 to 3.44]), while those who perceived themselves as having a normal weight (3.20 [95% CI, 3.14 to 3.27]) had the lowest rate. The same pattern was observed when analyzed separately by sex. Among females, the prevalence of suicide attempts was 4.5% in those perceiving themselves as underweight (4.47 [95% CI, 4.32 to 4.62]), 3.9% in those perceiving themselves as normal weight (3.91 [95% CI, 3.81 to 4.01]), and 5.1% in those perceiving themselves as overweight (5.23 [95% CI, 5.11 to 5.35]). This corresponds to an absolute risk difference of 1.2 percentage points between overweight and normal-weight females, and 0.6 percentage points between underweight and normal-weight females. In males, the prevalence of suicide attempts was 2.7% among those perceiving themselves as underweight (2.73 [95% CI, 2.65 to 2.82]), 2.4% among those perceiving themselves as normal weight (2.43 [95% CI, 2.35 to 2.51]), and 2.6% among those perceiving themselves as overweight (2.60 [95% CI, 2.52 to 2.69]). This indicates to absolute risk differences of 0.3 and 0.2 percentage points, smaller than those observed in females (Figure 1).

Figure 1.

Prevalence of adolescents’ suicide attempts, stratified by sex and self-perceived weight (weighted % [95% confidential interval]). A: Overall. B: Males. C: Females.

In addition, after categorizing individuals into BMI groups and further dividing them based on their self-perceived weight, the analysis of suicide attempts suggested that the highest rates occurred when there was a discrepancy between BMI and self-perceived weight. Underweight males who perceived themselves as overweight had particularly high rates of suicide attempts (7.28 [95% CI, 4.38 to 10.17]), and overweight adolescents who perceived themselves as underweight also showed elevated risks (12.38 [95% CI, 6.57 to 18.18]). Similar mismatch patterns were observed across other BMI categories and in females. Suicide attempts declined before COVID-19 but increased thereafter (Table 2 and Supplementary Table 2).

Table 2.

National trends and weighted prevalence of adolescents’ suicide attempts, stratified by sex and self-perceived weight, with β-coefficients (weighted % [95% CI])

Temporal trends in suicide attempts before and during pandemic

A comparison focusing on the COVID-19 pandemic showed that suicide attempts increased during the pandemic. In the early stage of the pandemic, in 2020, suicide attempts decreased (underweight: 0.73 [95% CI, 0.57 to 0.94]; normal weight: 0.66 [95% CI, 0.52 to 0.85]; overweight: 0.68 [95% CI, 0.56 to 0.83]). However, throughout the pandemic, there was a steady increase. For example, in the case of individuals who perceive themselves as overweight, there was a steady increase during the COVID-19 pandemic, with a 1.02% ([95% CI, 0.79 to 1.32]) increase from 2020 to 2021, 1.30% ([95% CI, 1.01 to 1.68]) from 2021 to 2022, and 1.43% ([95% CI, 1.14 to 1.80]) from 2022 to 2023 (Table 3). A comparison that included both the COVID-19 pandemic and the years prior showed a decreasing trend in suicide attempts before the pandemic, followed by an increase during the COVID-19 pandemic (Supplementary Tables 3 and 4).

Table 3.

Weighted ORs by periods for adolescents’ suicide attempts across time periods, stratified by sex and self-perceived weight (weighted % [95% CI])

Risk factors associated with suicide attempts

Stratified analyses evaluating differences in suicide attempt risk across self-perceived weight categories showed that the association between self-perceived overweight and suicide attempts was more pronounced in specific subgroups (Figure 2). Among females, when the normal weight group was used as the reference, the wOR for suicide attempts was higher in both the underweight group (wOR, 1.16 [95% CI, 1.11 to 1.21]) and the overweight group (1.35 [95% CI, 1.31 to 1.40]), compared to males (underweight: 1.12 [95% CI, 1.07 to 1.17]; overweight: 1.08 [95% CI, 1.05 to 1.13]). Similarly, adolescents residing in urban areas exhibited higher wORs for underweight (1.06 [95% CI, 1.03 to 1.10]) and overweight (1.26 [95% CI, 1.22 to 1.29]) compared to those in rural region. These results indicated that being female, living in urban areas, experiencing elevated depressive symptoms, and consuming alcohol may be risk factors that amplify the association between self-perceived weight and suicide attempts (Supplementary Tables 5 and 6).

Figure 2.

Forest plot of weighted ORs (95% CI) for suicide attempts by self-perceived weight categories, with normal weight as the reference. BMI, body mass index; CI, confidence interval; OR, odds ratio.

Association between body image distortion and suicide attempts

Individuals with body image distortion exhibited significantly higher odds of suicide attempts compared to male without distortion across both sexes. Compared with males without body image distortion, those with body image distortion had higher odds of attempting suicide (wOR, 1.13 [95% CI, 1.09 to 1.18]). The association was substantially stronger among females, with those reporting body image distortion showing more than a twofold increase in the odds of suicide attempts (wOR, 2.12 [95% CI, 2.05 to 2.20]). This trend was more pronounced during the COVID-19 pandemic than in the pre-pandemic period (Supplementary Table 7).

Key finding

This study is the first to analyze the relationship between self-perceived weight and suicide attempts, including risk factors, in a total of 1,156,728 adolescents from 2005 to 2023. While suicide attempts showed a declining trend before the COVID-19 pandemic, there was a continuous increase during the pandemic. The results indicated that the number of individuals with a normal BMI was much higher than those who perceived themselves as having a normal weight. This suggests a potential mismatch between actual body weight and self-perceived weight, indicating that many individuals with a normal BMI may not recognize their weight status accurately. The analysis of suicide attempts based on self-perceived weight suggested that individuals who perceived themselves as overweight had the highest rate of suicide attempts, followed by those who perceived themselves as underweight, with the lowest rates found among those who perceived themselves as normal weight. This suggests that self-perceived weight may contribute to identifying those at risk. A more detailed analysis indicated that suicide attempts were highest when there was a significant discrepancy between BMI based weight and self-perceived weight. Additionally, in all self-perceived weight groups, females exhibited a higher rate of suicide attempts compared to males. Furthermore, female sex, urban residence, high levels of depressive symptoms, and alcohol consumption may serve as potential risk factors that amplify the association between self-perceived weight and suicide attempts.

Plausible underlying mechanisms

Suicide attempts were higher in individuals with self-perceived weight of being overweight, followed by those who perceived themselves as underweight, and lowest among those who viewed themselves as having a normal weight. First, overweight perception is often accompanied by stigma, which can foster internalized shame and fear of negative social evaluation, leading to psychological distress and suicidality [22-24]. Second, adolescents who see themselves as overweight are more likely to face bullying in appearance-focused peer cultures, contributing to low self-esteem, social isolation, and depressive symptoms [25]. Third, persistent overweight perception may increase the risk of eating disorders, further elevating the likelihood of suicide attempts [26]. Finally, overweight status based on BMI itself is associated with psychological problems such as depression and anxiety, which can heighten vulnerability to suicide attempts [27]. In contrast, individuals who perceive themselves as having a normal weight tend to exhibit psychological health, resulting in lower rates of suicide attempts [28].

In addition to overweight perception, underweight perception was also associated with elevated risk of suicide attempts. Underweight perception could contribute to body dissatisfaction, particularly in sociocultural contexts where physical strength or a fuller body shape is idealized. Adolescents who perceive themselves as underweight might also face stigma, being viewed as weak or unattractive, which could increase vulnerability to bullying and social isolation [29]. Furthermore, underweight perception has been linked to higher levels of anxiety, depression, and body image distortion, all of which could act as risk factors for suicidality [30].

A high rate of suicide attempts was observed when there was a significant discrepancy between actual body shape and self-perceived weight. This can be explained by several factors. Perceiving oneself as overweight despite not being objectively overweight may be a symptom of depression, which could lead to an increased likelihood of suicide attempts [31]. In addition, adolescents with distorted body image may experience mental health issues such as aggressive behavior, delinquency, and suicidal thoughts, which can ultimately result in suicide attempts [7,32].

Females generally exhibit higher rates of suicide attempts compared to males. This pattern may be partly explained by the higher prevalence of mental health conditions such as major depressive disorder and post-traumatic stress disorder among females [33], as well as stronger social pressures and body dissatisfaction related to body image [34,35]. Previous studies have similarly reported that while females are more likely to attempt suicide [36], males have a higher mortality rate, suggesting that attempts by males are more often fatal [37].

Comparison of previous studies

Previous studies have suggested that self-perceived weight influences suicidal consideration [7,12]. However, many of these studies were conducted on relatively small samples, and research on Korean adolescents was limited to data up to 2018, failing to account for the potential influence of COVID-19 pandemic [38]. Other studies, also based on small sample sizes, focused exclusively on adults, making it difficult to establish any correlations with adolescents [39]. Adolescents are more susceptible to external influences than adults, and problems that arise during adolescence can continue into adulthood, so they need to be addressed in greater depth. Additionally, these studies used data only up to 2016, leaving out any analysis of the post-COVID-19 pandemic [39].

Another body of research analyzed the influences of self-perceived weight and BMI on suicidal consideration, but these studies were also limited by using data only up to 2012 and focusing solely on adults [12]. Furthermore, one study indicated that perceiving oneself as underweight was not associated with suicidal consideration, but this research was conducted over a very short period, highlighting the need for longer-term studies [7]. Most previous studies have only examined the correlation with suicidal consideration. However, the focus of this study is on suicide attempts, which involves taking actual actions toward suicide. Therefore, this issue requires more careful consideration and should be evaluated with greater importance.

Clinical and policy implications

This study suggests that self-perceived weight may be associated with an increased risk of suicide attempts. Considering that the greater the difference between actual BMI and self-perceived weight, the greater the risk of suicide attempts, it can be concluded that distorted self-perceived weight can increase the risk of suicide attempts. Accordingly, the introduction of programs to correct distorted body perception and prevent suicide is required [40]. Also, policy makers need to promote campaigns to raise social awareness of distorted self-perceived weight, regulate mass media content that emphasizes idealized and unrealistic body images, and encourage body positive campaigns to create a culture that embraces diverse body types [41].

It is also necessary to introduce government-level media guidelines that restrict advertisements or programs that have a negative influence on self-perceived weight. Indeed, the UK is implementing the “Be Real Campaign” to help adolescents break away from unrealistic beauty standards and focus on health and self-esteem. In addition, incorporating structured mental health literacy programs tailored for students into regular school curricula provides an effective strategy for early intervention, equipping adolescents with the knowledge and skills to recognize, manage, and seek help for mental health challenges at an early stage [42]. School-based BodyThink programs adapted for adolescents have also been shown to enhance body image and reduce appearance-related stress and depression in South Korea [43]. Thus, This study indicates that distorted body perception among adolescents may be associated with an increased risk of suicide, and suggests that multidimensional policy efforts are required to address this issue, encompassing both individual-level mental health education and improvements in the sociocultural environment.

Strengths and limitations

This study has several notable strengths. It utilizes nationally representative data from a large sample collected over an extended period, including the COVID-19 pandemic, providing more reliable results and identifying pandemic-related trends. Additionally, the focus is on suicide attempts, which involve actual self-harm behaviors and pose a more immediate and severe risk compared to suicidal consideration. Furthermore, the study specifically analyzes adolescents, a group at higher risk of suicide attempts due to their impulsivity, with issues during adolescence often extending into adulthood, making this focus particularly important.

However, this study has several limitations. First, the study follows a cross-sectional design, which makes it difficult to establish causal relationships. To address this limitation, the study references various previous studies to infer and present potential causality. Nevertheless, it is important to acknowledge that the underlying mechanisms behind the observed associations cannot be fully analyzed. Second, this study has the limitation of analyzing data from a single country. To address this, the study references previous research conducted in other countries, such as China and the United States, to provide a broader contextual understanding [33,44]. Furthermore, since suicide attempts are recognized by the WHO as a global issue, it is important to note that this cannot be viewed as a study confined solely to South Korea. Third, this study relies on self-reported data, which carries inherent limitations such as recall bias and social desirability bias. However, since this study focuses on self-perceived weight and the emotions resulting from it, self-reported data provided the advantage of capturing these variables more accurately. Fourth, various risk factors not addressed in this study may also have influenced the relationship between self-perceived weight and suicide attempts. While this study included variables such as academic grade level and educational attainment, other factors, such as social media and internet use, as well as cyberbullying, may also play a role. Moreover, protective factors such as family support or school commitment, which could mitigate the risk of suicide attempts, were not included due to limitations in the dataset. However, due to the limitations of the available data, these variables could not be analyzed in this study. Therefore, further research is needed to explore these factors in greater depth.

In conclusion, this study is the first to analyze the relationship between self-perceived weight and suicide attempts including risk factors among a total of 1,156,728 adolescents from 2005 to 2023. The results indicated that a substantial proportion of adolescents with a normal BMI misperceived their weight status, with 36.54% perceiving themselves as normal weight but 63.46% perceiving themselves as underweight or overweight despite having a normal BMI. The analysis of suicide attempts based on self-perceived weight indicated that adolescents who perceived themselves as overweight had the highest suicide attempts, followed by those who viewed themselves as underweight. This suggests that self-perceived weight may serve as a predictor of suicide attempts. Furthermore, suicide attempts were higher among females than males. These findings should be interpreted in the context of social and economic conditions, as socioeconomic inequalities may shape both body image and suicidality. Therefore, this study recommends the introduction of targeted programs and public campaigns to correct distorted self-perceived weight.