Suicide and interpersonal violence are important contributors to mortality and morbidity among young people. The authors set out to clarify psychological and behavioural pathways linking victimisation to suicidal outcomes among young adults in a North Indian city. Specifically, the study tested a theory-driven structural equation model (SEM) that treated suicidal ideation and attempts as downstream outcomes of victimisation, with intermediary roles for physical violence and depressive symptoms.
Data came from a cross-sectional survey of young adults aged 18–22 years recruited by random sampling from six colleges. The analytic sample comprised 752 respondents.
Victimisation was represented using three observable indicators: dating violence, campus bullying and cyberbullying. Physical violence was assessed via involvement in physical fights. Depressive symptoms and suicidal behaviour were measured by survey items; the authors combined suicidal ideation and suicide attempts into a single ordinal outcome variable with five ordered levels ranging from "did not consider suicide" to "had a suicide attempt requiring medical treatment."
For modelling, the investigators specified a theory-based SEM in which victimisation predicted suicidal ideation and attempts directly and indirectly via physical violence and depression. Because the outcome was categorical, they used a probit link function with theta parameterisation for categorical endogenous variables. Model outputs included standardised path coefficients, p-values, and global fit statistics.
The median age of participants was 19 years and 64.4% were women. Reported prevalences in the past 12 months included: involvement in a physical fight in 16.6% of respondents, depressive symptoms in 24.2%, and serious consideration of suicide in 4.5%.
The SEM treated victimisation as an upstream latent construct indicated by dating violence, campus bullying and cyberbullying. Physical violence and depressive symptoms were positioned as mediators between victimisation and the ordinal suicidal outcome. The analytic approach accounted for the categorical nature of the outcome using a probit link and theta parameterisation. The investigators reported standardised path coefficients and statistical significance for direct and indirect effects, together with global measures of model fit.
Primary model findings were that the association between victimisation and the combined suicidal ideation/attempts outcome was predominantly indirect. Victimisation was associated with higher levels of physical aggression (physical fights) and with greater depressive symptoms, and these mediators, in turn, were associated with increased suicidal ideation and attempts. Direct effects of victimisation on suicidal ideation and attempts were described as negligible. The global fit of the model was reported as acceptable, supporting the plausibility of the specified pathways in this sample.
The authors interpret the SEM results to indicate that experiences of bullying, dating violence and cyberbullying may elevate risk for suicidal behaviour chiefly by promoting physical aggression and depressive symptomatology. In other words, victimisation appears to set in motion a chain of adverse behavioural and psychological responses that increase suicide risk, rather than exerting a strong direct effect on suicidal behaviour in this sample.
Based on the observed pathways, the authors argue for multi-level suicide prevention strategies among youth. Recommended components include measures to prevent and respond to bullying and interpersonal violence on campus and online, life-skills education to reduce escalation into physical aggression, and routine screening and treatment for depression. The implication is that addressing upstream victimisation and its mediators could reduce suicidal behaviour among young adults.
Ethical approval was obtained from the Institute Ethics Committee of the Postgraduate Institute of Medical Education and Research, Chandigarh (letter number INT/IEC/2018/000791, dated May 24, 2018). Students provided written informed consent and investigators maintained privacy and confidentiality of data. The authors declared no competing interests. The data supporting the study findings are publicly available at the cited DOI, and the codebook and statistical analysis code are provided as supplementary material with the article.