Reasons to Live: A Technology-Guided Analysis of Potential Protective Factors in Suicidality

Background and Purpose: Suicide prevention scholarship has focused primarily on why people die by suicide, but improving our understanding of what keeps suicidal people from killing themselves has great potential to improve individual and population-based approaches to suicide prevention. Both theory and practice have focused on why people want to die. For example, the dominant theories in suicide prevention are collectively called, "ideation-to-action" theories (Klonsky et al, 2018). These theories try to explain how people become suicidal and then what compels them to make fatal or near-fatal suicide attempts. Theoretical frameworks have been driven by the need to understand why people attempt suicide or die. Clinical models are informed by research on suicidal thoughts and behaviors (which are informed by theory) as well as risk prevention models. The current study uses a novel approach to understanding how people in the general public decided not to attempt suicide. We report on analysis of a naturally occurring dataset that provides a window into the lived experiences of people who considered dying by suicide but did not. In September 2020, a Reddit user named firegate2233 asked: “Formerly suicidal redditors, what’s something that kept you alive a little while longer and helped you to get through the dark times in your lives?” More than 15,000 people responded in one day.

Methods: This study was approved by the [anonymized university] Institutional Review Board (protocol [anonymized]). Prior to conducting this research we spoke with the original Reddit poster, firegate2233, to verify that he was comfortable with our intended use of the discussion thread he had initiated. All data were anonymized before analysis. In this study, we conducted a computationally assisted content analysis of this large dataset. First, we used a computational analysis called topic modeling to distill a set of proposed thematic categories from the Reddit responses. This method uses techniques from computational linguistics to find commonalities in a large dataset, simliar to factor analysis. Then, two suicide subject matter experts reviewed, refined, and labeled a set of categories and their descriptions based on the topics identified by topic modeling.

Results: The computational content analysis identified 30 topics and human review organized those into four themes related to why someone decided not to attempt suicide: (a) selflessness, (b) hopefulness about the future, (c) sensory pleasures, and (d) fear of being worse off.

Conclusions and Implications: These findings provide an understanding of reasons for living and dying as well as implications for clinical practice. The four themes that emerged from the data complement the categories found in contemporary ideation-to-action frameworks (Klonsky, et al. 2018). For example, our analysis found that people identified hopefulness about the future as their reason for not attempting. This provides a direct contrast with the theoretical construct of hopelessness that is found in most suicide theories and suicide risk assessments. These findings provide a first step at identifying a list of factors that might reduce rather than increase risk. It also provides insight for scholars involved in advancing theoretical frameworks around suicide.