Suicide and Self-Injury
Byoungwook D. Park, B.A.
MA Candidate
Teachers College, Columbia University
Floral Park, New York, United States
Bryan Keller, Ph.D.
Associate Professor of Practice in Applied Statistics
Teachers College, Columbia University
New York, New York, United States
Background: Suicide is a multifaceted problem shaped by interactions between diverse factors (De Berardis et al., 2018). The level of life satisfaction (LS), history of suicide attempts (SA), and severity of suicidal ideation (SI) have been identified as significant risk factors (Chu et al., 2015; Sher & Oquendo, 2023; Zhang et al., 2017). Notably, SI plays a critical role in assessing the overall risk of suicide, as individuals expressing thoughts of killing themselves have a higher likelihood of attempting suicide (Hubers et al., 2018). While the relationship between the history of SA and SI has been examined (e.g., Harmer et al., 2024), their association with LS remains largely unexplored. The current study hypothesizes that (1) LS will be negatively associated with SI for both attempters and non-attempters, and (2) SA history will moderate the relationship between LS and SI. The latter hypothesis offers a novel research direction to enhance understanding of the experiences of individuals with SA history.
Methods: Data were gathered as a part of a cross-sectional survey aimed at validating the Suicidality Scale (SS; Harris et al., 2023), an assessment of SI. Participation in the survey was available to individuals 13 years or older with English proficiency. It was advertised as a suicide study and yielded high participation from suicidal individuals. Individuals between the ages of 13 and 82 (n = 5,115; Mage = 18.64; SD = 7.98) across seven countries completed an online survey that assesses the domain of SI, LS, and history of SA. The SS assessed the severity of SI, the Satisfaction with Life Scale (SWL; Diener et al., 1985) assessed the individual’s perceived level of LS, and a single yes-or-no item assessed SA history (29.1% Yes). Multiple linear regression was employed to evaluate the research hypotheses. First, level of LS was examined as a predictor of SI. Then, the moderating role of SA history in the prediction of SI by LS was examined.
Results: As predicted, the level of LS has a significant negative association with the severity of SI (β = -.98, t = -66.28, p < .001); the Pearson correlation between LS and SI was estimated to be r = -.68. Furthermore, the moderation analysis resulted in a significant regression equation (SIi = 35.68 + 3.47SAi - .89LSi + .17SAi*LSi), with all model coefficients significant at p < .001.
Conclusions: Findings indicate that lower LS is predictive of higher severity of SI. Furthermore, the negative relationship between LS and SI was attenuated by a history of SA, suggesting that individuals with SA history may endorse elevated SI even when experiencing a similar level of LS as those without such history. The regression equation also indicated higher baseline SI for individuals with a history of attempts compared to those without. Moreover, this novel insight illuminates the intricate interaction between the variables under examination. Lower LS serves as a general risk factor, while the history of SA adds a specific layer of risk, potentially amplifying the intensity of suicidal thoughts beyond the expected impact of LS alone. While promising, these results are based on cross-sectional associations. Future research with longitudinal designs aimed at identifying causal relationships would be useful.