Longitudinal associations between non-suicidal self-injury and borderline personality disorder in adolescents: a literature review

BACKGROUND: Borderline personality disorder (BPD) in adolescent samples is similar to BPD in adults concerning clinical characteristics. A notable difference is that adolescents with BPD – and adolescents in general – are more likely than adults to present with acute symptoms such as non-suicidal self-injury (NSSI) and suicidal behaviours. BPD is the only disorder in the Diagnostic and Statistical Manual- 5th Edition that includes a criterion of NSSI. Additionally, NSSI is purported to be a developmental precursor of BPD under the biosocial developmental model. Though much cross-sectional data have illustrated the robust association of adolescent NSSI and BPD, no review to date has summarized the longitudinal associations between these phenomena. The aim of this literature review was to summarize what is known about the longitudinal associations between adolescent NSSI and BPD symptoms. Information on the developmental course of NSSI in relation to BPD would be helpful to clinicians, as the rate of NSSI is high in adolescent populations, and research indicates early, possibly BPD-specific interventions are imperative. METHODS: A literature search was conducted using Embase, MEDLINE, and PsycINFO databases and cited reference searches. Criteria included studies of adolescents (age ≤ 18 at baseline) from either epidemiological or clinical samples, incorporating a longitudinal design, with predictors and outcomes of interest, including both NSSI and BPD diagnosis/symptoms/traits. RESULTS: Six independent samples were identified that matched our search criteria. The articles were grouped and reported on separately by population type (epidemiological vs. clinical), and directionality of relations. We identified two epidemiological and four clinical samples. Five samples examined the longitudinal associations of NSSI preceding BPD, three samples measured BPD in adolescence (baseline age ≤ 18), and two of those samples measured BPD at baseline. Both epidemiological studies revealed significant longitudinal associations between NSSI and later BPD/BPD symptoms; however, they differed notably in their methodologies hindering data synthesis across studies. In the clinical studies, findings of the association or predictive relations were not consistent. This is potentially due to differing methodologies, or differences in treatment effectiveness and responsiveness across the samples. CONCLUSIONS: This review highlights the paucity of data that are available examining the longitudinal association between NSSI and BPD within adolescent samples. Thus, it is not possible to reliably comment on how NSSI and BPD are related over time. Future studies will benefit from the measurement of BPD symptoms in very early adolescence, and concurrent measurement of NSSI as well as other forms of suicidal behaviours across adolescence.