(PS2-A1) Love Is (not) All You Need: Understanding the Role of Relationship Status and Dysfunction in Predicting Self-directed Violence

Suicide remains a major public health concern with widespread impacts. Thus, it is critical to enhance understanding of modifiable risk factors for behaviors strongly linked to suicide (suicide attempts and nonsuicidal self-injury [NSSI], collectively known as self-directed violence) to mitigate suicide risk through interventions. One protective factor identified in research is being married relative to being divorced, likely because of the practical and emotional support that partners can provide. Simultaneously, difficulties within relationships can confer deleterious effects on health outcomes, including suicide risk. However, research focused on marital status rarely examines relationship functioning, obfuscating the joint contribution of relationship status and relationship dysfunction for suicide risk. In line with the convention theme of using science to better understand and intervene in the context of health emergencies, this investigation aims to clarify the role of modifiable interpersonal risk factors for self-directed violence.

The current study examined relationship status, relationship dysfunction, and the combination of these as predictors of self-directed violence. United States veterans (N = 1049) completed a survey about multiple indicators of health and well-being, including history of NSSI and suicide attempts. Veterans in a relationship also rated the degree of trouble in their relationship (hereafter referred to as “dysfunction”) on a 7-point scale. Logistic regression models were used to capture how (a) relationship status (married or unmarried but living together vs divorced or separated), (b) continuous ratings of relationship dysfunction, and (c) being divorced compared to being in a low or high dysfunction relationship (categories created based on dysfunction scores) predicted likelihood of self-directed violence. Analyses controlled for intrapersonal factors associated with suicide risk (sex, age, race, mental health diagnoses).

Findings indicated that veterans in a relationship did not have reduced risk for self-directed violence compared to veterans who were divorced/separated. However, among veterans in a relationship, increasingly more dysfunction was associated with greater risk for self-directed violence (AOR = 1.22, p < .001). Finally, risk for self-directed violence was greater in veterans endorsing high dysfunction relationships compared with low dysfunction relationships (AOR = 2.38, p < .001) and compared to divorced veterans (AOR = 1.83, p = .009). These results suggest that it is insufficient to only consider relationship status when evaluating interpersonal risk factors for suicide. Given that veterans in high dysfunction relationships had the greatest risk for self-directed violence, it may be important to improve access to evidence-based couple services to mitigate this risk. Importantly, this study found that a single item assessing relationship dysfunction predicted risk over and above intrapersonal risk factors. Integrating such single-item measures into clinical practice could improve identification and subsequent tailored intervention for veterans at greater risk for suicide.