Negative affectivity and disinhibition as moderators of an interpersonal pathway to suicidal behavior in borderline personality disorder

Abstract

This longitudinal study examined whether personality traits moderate the link between interpersonal dysfunction and suicidal behavior in a high-risk sample of 458 individuals diagnosed with borderline personality disorder (BPD). Participants were assessed annually for up to 30 years (mean number of follow-ups = 7.82). Using multilevel structural equation modeling, we examined i) longitudinal, within-person relationships among interpersonal dysfunction, suicidal ideation, and suicide attempts; and ii) moderation of these relationships by negative affectivity and disinhibition. Negative affectivity predicted a stronger within-person coupling between interpersonal dysfunction and suicidal ideation. Disinhibition predicted a stronger coupling between ideation and suicide attempts. Assessing negative affectivity and disinhibition in a treatment setting may guide clinician vigilance toward those at highest risk for interpersonally triggered suicidal behaviors.

Keywords: borderline personality disorder; disinhibition; interpersonal; negative affectivity; suicide.

Figure 1.

Confirmatory factor analysis of baseline measures of personality. Model fit: χ² (19) = 54.70, p < .001; CFI = .94; RMSEA = .07; SRMR = .06. All paths are significant at p ≤ .001.

Figure 2.

Multilevel structural equation model examining within- and between-person associations between interpersonal dysfunction, suicidal ideation, and suicide attempts. Interpersonal dysfunction, ideation, and attempts are decomposed into within- and between-person components using latent decomposition. Covariates are excluded for parsimony. All coefficients are standardized except for the indirect effects. Standard deviations are in parentheses. *p < .05.

Figure 3.

A) Multilevel structural equation model examining the moderating influence of negative affectivity and disinhibition on within-person associations between interpersonal dysfunction, suicidal ideation, and suicide attempts. Only paths relevant to primary hypotheses are shown (see Table S5 for all path estimates). Covariates and latent decomposition of variables are excluded for parsimony. All coefficients are standardized except for the indirect effect. Standard deviations are in parentheses. B) Moderation of an interpersonal pathway to suicide attempts by personality. High negative affectivity is associated with a stronger within-person coupling between interpersonal dysfunction and suicidal ideation. C) High disinhibition is associated with a stronger within-person coupling between ideation and attempts. *p < .05.