(PS9-C71) Idiographic Experiences of Thwarted Belongingness, Perceived Burdensomeness, and Suicide Ideation: An Ecological Momentary Assessment Approach

Background:  Despite decades of research, the prediction and prevention of suicidal thoughts and behaviors (STB) have not significantly improved (Franklin et al., 2017). A major limitation has been the lack of sufficient consideration for the heterogenic nature of suicide risk, even though STBs can be reached through various paths (Kleiman et al., 2022). For example, studies based on the Interpersonal Theory of Suicide (ITS: Van Orden et al., 2010), one of the most prominent theories on STB (Chu et al., 2017), have mostly focused on examining the interaction of thwarted belongingness (TB) and perceived burdensomeness (PB), core constructs of suicide ideation (SI). However, experiences of TB and PB may differ per individual (Hjelmeland & Knizek, 2019); for some the two constructs may co-occur, whereas for others it may be that one precedes the other (e.g., feeling like a burden and therefore feeling like one does not belong). Furthermore, it is uncertain as to whether the interaction of TB and PB is truly an interaction or rather quadratic effects, given that TB and PB are highly correlated (Cero et al., 2015). These possibilities warrant further examination of how idiographic experiences of TB and PB may relate to SI. An ecological momentary assessment approach may be promising as it allows time-intensive examination, which can be used for both group and individual-level analyses.

Method:  A sample of 40 participants with past-year history of STB were recruited from online forums on mental health and suicidality. All participants received 4 EMA surveys per day for 30 days. The EMA surveys contained two items each on passive and active SI, and one item each on TB and PB. To ensure the quality of the data, only those with at least twenty responses (n = 35) were included in the final analyses. The four suicide items were summed to create a suicide score. The gimmeSEM algorithm from the Group Iterative Multiple Model Estimation (GIMME) package (Gates & Molenaar, 2012) in R version 4.2.1 (R Core Team, 2022) was used to examine the relationships among the variables of interest (TB, PB, TB², PB², TB x PB, and SI). Contemporaneous and lagged relationships were examined. Auto-regressive paths were accounted for in lagged relationships and the group cutoff value was set to its default (.75) meaning that a path needed to be significant across 75% of individuals in order to be identified as a group-level path.

Results:  At the group level, only four relationships were found to be significant (i.e., “PB ~ PBlag,” “TB ~ TBlag,” “SI ~ SIlag,” and “PB ~ PB²”). Individual-level findings differed significantly. The interaction between TB and PB was most frequently identified as a significant contemporaneous predictor of SI (n = 7), which was followed by PB (n = 6), TB (n = 4), and PB² (n = 3). In lagged relationships, TBlag (n = 2) and PBlag (n = 1) were identified as significant predictors of SI

Conclusion:  Group-level findings did not support the ITS, especially after accounting for the autoregressive paths. On the other hand, at the individual level, the interaction between TB and PB was identified as a predictor of current SI, which did not hold for lagged relationships. These results suggest the importance of considering heterogeneity in suicide intervention and prevention.