(PS11-32) Examination of the Moderating Role of Internalizing Symptoms in the Direct and Indirect Relationships Between Alcohol Use, Rumination, and Suicidal Ideation

Abstract Text: Alcohol use has been widely associated with suicidal ideation (SI) Conner et al., 2003; Stack, 2000). Moreover, past literature has uncovered rumination, a known suicide risk factor (Rogers & Joiner, 2017), to be elevated following alcohol use, suggesting its potency in the relationship between alcohol use and SI (Steele and Josephs, 1990). The nature of this relationship is understudied, and more work is necessary to understand the dynamic interaction between alcohol use and SI. The current study aimed to address the indirect relationship between alcohol use and suicidal ideation through rumination, as well as whether internalizing symptom severity (depression and anxiety) moderated these paths. Participants were 278 community-based adults with current suicidal ideation and/or a lifetime suicide attempt, aged 18 to 62 years (Mage = 28.05, SD = 6.26, 65.5% cisgender man, 56.5% White), who completed a series of self-report measures. A moderated mediation model was conducted in PROCESS (Model 59) to examine mediating and moderating relationships. First, there were significant interactions between alcohol use and both anxiety (B = .02, SE = .01 , p = .045) and depression (B = .04, SE = .01, p < .001) in predicting rumination (Path A). Specifically, alcohol use was positively and significantly associated with rumination at high, but not low and mean, levels of anxiety (B = .19, SE = .07, p =.008) and depression (B = .28, SE = .07, p < .001). Second, there were significant interactions between rumination and both anxiety (B = -.03, SE = .01, p = .001) and depression (B = -.03, SE = .01, p = .002) in predicting SI (Path B). Contrary to expectations, however, rumination was positively and significantly associated with SI at low (ps < .001) and mean (ps < .001), but not high (ps = .086 and .533), levels of anxiety and depression. Third, there were significant interactions between alcohol use and both anxiety (B = .05, SE = .01, p < .001) and depression (B = .03, SE = .01, p = .001) in predicting SI (Path C). These relations were positive and significant at high (Bs = .29 and .32, SE = .07 and .08, ps < .001), but not low and mean, levels of anxiety and depression. Finally, contrary to our expectations, rumination did not account for the relation between alcohol use and SI at any levels of anxiety or depression. The indirect relationship between alcohol use and SI through rumination was non-significant. However, internalizing symptoms of depression and anxiety significantly moderated the relationships between alcohol use, rumination, and SI, though varying levels of these symptoms impacted these associations differently. These results reveal how internalizing symptoms may impact risk factors for suicide differentially, suggesting notable clinical and research implications. Relatedly, future work may benefit from evaluating temporal precedence of alcohol use, and related internalizing factors, on SI by addressing this relationship longitudinally. Further implications and limitations will be addressed.