Since the outbreak of COVID-19 in Hubei, China, in December 2019, restrictions have been imposed on people’s daily lives to address the threat of the virus; however, these restrictions have had prolonged negative consequences. The severing of social relationships owing to measures such as social distancing has resulted in mental health problems in the general population, and fear of quarantine has led to depression, anxiety, stress, and despair, thereby impairing people’s quality of life. In particular, 59.8% of adolescents in South Korea have reported dominant feelings of “anxiety and worry” as a result of the COVID-19 pandemic [37]. Thus, this study analyzed the factors associated with mental health issues and suicide among adolescents by categorizing a representative sample of South Korean adolescents into four groups: a healthy group, suicide-ideation group, suicide-planning group, and suicide-attempt group. We sought to specifically identify the predictors of related mental health issues. The following results were obtained.

First, among 57,925 South Korean adolescents, 6.9% showed suicidal ideation, 2.2% reported planning suicide, and 1.9% reported attempting suicide over the 12 months preceding the data collection. These rates are lower than those reported in a 2019 study conducted in the United States among children and adolescents, wherein 18.8% reported seriously considering attempting suicide, 15.7% reported planning suicide, and 8.9% reported attempting suicide [41]. The above data show that adolescents in South Korea and the US are more likely to engage in suicidal ideation than to actually plan or attempt suicide [42]. However, in Ghana’s Global School-based Student Health Survey conducted in 2012 for high school students, the prevalence of suicidal thoughts, suicide plans, and suicide attempts was 18.2%, 22.5%, and 22.2%, respectively. Therefore, further research is needed on these differences in suicide-related variables between countries [43]. Moreover, although there is a large discrepancy between suicidal ideation and suicidal behavior among adolescents, this contrasts the situation of middle-aged and older adults [44]. Still, as adolescents have relatively less life experience and tolerance of various forms of shocks and stressors than do their older counterparts, they can be more impulsive regarding suicide. This indicates the need for national policies in South Korea that exclusively target suicide prevention among adolescents.

Second, all demographic characteristics examined in this study, except for fathers’ highest education level, significantly predicted adolescents’ mental health status. In the healthy group, male high schoolers with higher academic achievement were mentally healthier. In the suicide-ideation and suicide-attempt groups, female high schoolers with poor grades showed a higher risk of mental health problems. When compared to the healthy group, the risk of mental health problems was higher among students who were not living with family members, those who had a lower socioeconomic status, and those who reported planning or attempting suicide. The risk of stress, sadness and despair, loneliness, or GAD was higher among those who had experienced greater financial difficulty as a result of the COVID-19 pandemic. The risk was also higher among those with inadequate sleep, smokers, alcohol users, those with experience using drugs, and those with smartphone overdependence as compared to their counterparts. The level of financial assistance received had little effect.

The risk to attempt suicide was higher among female high schoolers with “poor” academic achievement, those living with non-family members, and those with lower economic status as compared to their counterparts. The risk of suicidal ideation, suicide planning, and suicide attempt was higher among those with inadequate sleep, smokers, alcohol users, those with experience using drugs, and those with smartphone overdependence. The findings for sex in this study are similar to those of previous South Korea-based studies, in which the risks of suicidal ideation, suicide planning, and suicide attempt were higher among female than male students [45,46,47]. A possible reason for this is that male students tend to interpret an event cognitively, while female students are more heavily influenced by and sensitive to the emotional aspects of the event [47, 48]. Between 2010 and 2017, the mean suicide rate among adolescents aged 10–19 years in South Korea was 5.45% for boys and 4.30% for girls, differing from the US rates of 7.57% for boys and 2.57% for girls [49]. Both the South Korean and US data show differences between sexes, which indicates that a sex-specific approach should be taken to address this issue.

As aforementioned, the results showed that sleep, smoking, alcohol, drugs, and smartphone overdependence were associated with the risk of suicide. This shows the necessity of helping students recover psychologically and improve their quality of life; this can be done by providing various forms of support for their mental and psychological conditions within the educational system, as past studies also show that drinking, smoking, drug use, or excessive experience of smartphones in adolescence harms physical health [22, 23, 50]. According to the SEYLE Cluster Randomized Controlled Trial (German Clinical Trials Register DRKS00000214), which encompassed 11,110 high school students in 10 EU countries, specific school-based suicide prevention interventions were effective in mitigating the dangerous effects of self-harm behavior; these interventions could also be further developed by increasing parental participation [51]. Parental support and encouragement may be one of the key factors for adolescents to be able to avoid suicidal thoughts. This suggests that the role of parents in recognizing adolescents as part of the family is essential for adolescent suicide prevention, as parental affirmation may help adolescents perceive themselves as valuable. Thus, to promote adolescent suicide prevention in South Korea, education systems and parents should facilitate or provide various forms of psychological support to adolescents, as these efforts may help strengthen adolescent mental health and, in turn, quality of life.

Furthermore, humanity has now already experienced how much our daily lives can change very quickly upon the onset of a pandemic, such as that provoked by COVID-19. Thus, we should conduct in-depth discussions on how our daily lives would be like in the upcoming post-COVID-19 period. To systematically respond to these forthcoming periods of human history, we could develop and apply digital platform education programs aimed at improving adolescent health behaviors and mental health, especially among those who had their lives impacted by the COVID-19 pandemic. Once more, education systems can play a significant role on the topic, as they can provide adolescents with various forms of psychological support so as to strengthen their mental health and quality of life.

Third, we comparatively analyzed mental health and suicide-related factors. The results showed that negative emotions such as depression among adolescents increase the odds of their progressing from suicidal ideation to attempting suicide. This supports previous findings that adolescents with emotional problems exhibit high levels of depressive symptoms, stress, and impulse to drop out of school [52], and that loneliness is a powerful predictor of suicidal ideation [53, 54]. Contrastingly, Naragon-Gainey and Watson [55] reported that while depression is a significant factor in suicidal ideation, GAD is not. Another study shows that emotional intelligence associates with health indices, and that decreasing emotional intelligence relates to increasing suicidal behavior [56]. In an analysis of the components and behavioral risk reduction of adolescent mental health programs for those aged 10–19 across 18 studies, the interventions were shown to universally improve adolescent mental health and reduce risk behaviors [57]. Thus, programs that cultivate one’s ability to control and positively manage emotions should be provided for adolescents who are psychologically unstable and who are repeatedly exposed to negative emotions; this could help to promote psychological stability and functionality.

Fourth, we identified factors that predict adolescents’ mental health status. Specifically, sex, academic achievement, economic status, impact of COVID-19 on economic status, sleep, alcohol use, drug use, smartphone overdependence, and suicide-related factors were identified as significant predictors. Many studies have similarly reported that sex, academic achievement, economic status, health behaviors, and emotional characteristics influence adolescents’ mental health and suicidal ideation [58, 59]. Further, studies exploring the influence that COVID-19-induced changes in subjective economic status have on adolescents’ stress have found that deterioration in economic status has a negative impact on adolescents’ mental health [60,61,62].

Past studies on suicide have generally examined predictors of suicidal ideation. Contrastingly, this study classified adolescents into four groups, namely the healthy, suicide-ideation, suicide-planning, and suicide-attempt groups. Thus, the significance of this study lies in considering suicide not as a simple, isolated event, but rather as a complex and dynamic process that forms a continuum from suicidal ideation through planning suicide to attempting suicide. Thus, in examining the demographic characteristics and the transition to suicidal attempt in individuals in the healthy group, we shed light on the specific directions that can be taken for interventions targeting each stage of mental health crisis among adolescents, from suicidal ideation to suicidal planning and suicide attempt. Further, while past studies have generally focused on depression when examining factors associated with suicide in adolescents, we confirmed that sadness and despair and GAD elevate the risk of suicidal ideation, suicide planning, and attempting suicide. These findings collectively showcase the need for programs that, through accurate mental health evaluations, focus on these factors and enable the early screening and implementation of interventions for adolescents at risk of suicide. Another strength of this study is that it provides an opportunity to identify new means of maintaining adolescents’ mental health amid the prolonged COVID-19 pandemic.

Comparisons of the key variables are displayed in Fig. 1. The suicide-attempt group showed the highest level of stress (4.136); this group was followed by the suicide-ideation group (3.938), suicide-planning group (3.923), and healthy group (3.065). The difference was significant (p < 0.001).

Analysis of covariance and post-hoc test for stress, sadness and despair, loneliness, and generalized anxiety disorder in the health, suicide-ideation, suicide-planning, and suicide-attempt groups. The basic control variables included sex, education level, academic achievement, father’s highest education level, mother’s highest education level, living arrangement, economic status, impact of COVID-19 on economic status, whether the family received financial assistance, sleep, smoking, drinking, drug use, and smartphone overdependence

The suicide-attempt group showed the highest level of sadness and despair (1.807); this group was followed by the suicide-ideation group (1.684), suicide-planning group (1.680), and healthy group (1.192). The difference was significant (p < 0.001).

The suicide-attempt group showed the highest level of loneliness (3.645); this group was followed by the suicide-ideation group (3.331), suicide-planning group (3.305), and healthy group (2.281). The difference was significant (p < 0.001).

The suicide-attempt group showed the highest level of GAD (10.252); this group was followed by the suicide-planning group (8.685), suicide-ideation group (8.131), and healthy group (3.289). The difference was significant (p < 0.001).

Table 3 shows the results of the multiple logistic regression analysis that was performed to identify the predictors of mental health status among South Korean adolescents.

The ORs for stress, sadness and despair, loneliness, and GAD were higher among boys than girls (stress OR = 1.779; sadness and despair OR = 1.632; loneliness OR = 1.768; GAD OR = 1.851).

Regarding education level, the ORs for stress (OR = 1.113) and GAD (OR = 1.071) were higher among high schoolers than middle schoolers. Regarding academic achievement, the ORs were lower among students with good grades than those with poor grades (stress in high schoolers OR = 0.901; sadness and despair in middle schoolers OR = 0.843, in high schoolers OR = 0.770; loneliness in middle schoolers OR = 0.893, in high schoolers OR = 0.940; GAD in middle schoolers OR = 0.908, in high schoolers OR = 0.937).

The ORs for stress, sadness and despair, and loneliness were higher among students whose fathers were college graduates or higher than among those whose fathers had high school or lower education (stress in college or higher OR = 1.134, in unknown OR = 1.094; sadness and despair in college or higher OR = 1.086; loneliness in college or higher OR = 1.073). The OR for GAD was higher among students whose mothers were college graduates than among those whose mothers had below high school education (OR = 1.083).

Regarding living arrangement, the OR for loneliness (OR = 1.182) was higher among students who lived with a non-family member than those who lived with a family member.

Regarding economic status, the ORs were lower among students with middle and high status than those with low status (stress in middle OR = 0.781, in high OR = 0.704; sadness and despair in middle OR = 0.857; loneliness in middle OR = 0.754, in high OR = 0.715; GAD in middle OR = 0.767, in high OR = 0.743).

The ORs were higher among those who had experienced low economic status as a result of the COVID-19 pandemic than among those who did not experience this (stress in really OR = 1.310, in changed OR = 1.507, in critically OR = 1.540; sadness and despair in really OR = 1.068, in changed OR = 1.294, in critically OR = 1.645; loneliness in really OR = 1.336, in changed OR = 1.585, in critically OR = 1.536; GAD in really OR = 1.196, in changed OR = 1.382, in critically OR = 1.500).

Regarding financial assistance, the ORs for sadness and despair, loneliness, and GAD were higher among students who lived with a non-family member than those who lived with a family member (sadness and despair OR = 1.147; loneliness OR = 1.133; GAD OR = 1.115).

The ORs were higher among those who had adequate and average sleep than among those who had inadequate sleep (stress in average OR = 2.242, in inadequate OR = 3.197; sadness and despair in average OR = 1.399, in inadequate OR = 2.132; loneliness in average OR = 1.666, in inadequate OR = 2.206; GAD in average OR = 1.639, in inadequate OR = 2.759).

The ORs for sadness and despair (OR = 1.405) and loneliness (OR = 1.256) were higher among smokers than non-smokers. The ORs for all four mental health conditions were higher among those with inadequate sleep, alcohol users, those with drug user experience, and individuals with smartphone overdependence.