A six-year study found mostly stable low levels, but also groups whose suicidal thoughts increased or declined over time.
Researchers analyzed data collected annually from 2018 to 2023 through the Korea Children and Youth Panel Survey. The nationally representative cohort included 2,607 students who were in fourth grade at the start of the study and ninth grade at the final assessment.
The analysis identified three groups: one with consistently low levels of suicidal thoughts, one with elevated levels that gradually increased, and one that began with very high levels before declining and later showing a modest rebound. The researchers also examined personal, family and school factors measured at the start of the study.
Three paths over six years
The low-stable group included 73.5% of participants, or 1,915 children. Their reported suicidal thoughts remained consistently minimal. The moderate-increasing group included 19.2%, or 501 children, and showed persistently elevated levels that gradually increased. The initially-high declining group included 7.3%, or 191 children; it began with extremely high levels, then declined substantially before a modest rebound.
Self-esteem was the strongest baseline correlate of group membership. Compared with the moderate-increasing group, children with higher self-esteem were more likely to be in the low-stable group and less likely to be in the initially-high declining group. Boys were more likely than girls to belong to the low-stable group. Parental warmth and rejection were also associated with trajectory membership, while school factors did not independently predict membership after individual and family factors were considered.
Why different paths matter
The findings indicate that suicidal thoughts did not follow one pattern for all children during the transition from elementary to middle school. That variation suggests that a universal prevention approach may not be sufficient on its own, and that prevention and early identification could be adapted to children’s different developmental trajectories.
The group whose initially high levels declined also shows that substantial reductions in suicidal thoughts occurred for some children, although this study does not establish what caused that change. The results point to self-esteem and family relationships as factors that may be important to consider when identifying children who could need additional support.
Evidence and caveats
This was a prospective, six-year analysis of a nationally representative South Korean panel survey, rather than an experiment. Researchers used a statistical method to group children according to reported patterns over time and examined baseline factors associated with group membership. These associations do not show that self-esteem, family relationships or other factors caused a particular trajectory.
Suicidal thoughts were measured each year with one four-point question. That measure could not capture differences in frequency, intensity, persistence, or whether thoughts involved a plan or intent. The researchers also treated the ordinal responses as continuous and used a model that assumes no true variation within each identified group. All predictors were measured at baseline, so the analysis could not assess how changing circumstances influenced later suicidal thoughts. The trajectory groups should therefore be understood as broad patterns of self-reported thoughts, not complete or clinical categories.
// Source
Humanities and Social Sciences Communications · 2026 · DOI: 10.1057/s41599-026-09241-1
Authors: Changmin Yoo
Institutions: Inha University
