|
Abstract
|
BACKGROUND: Rumination is a cognitive process that has been implicated in the onset and maintenance of a variety of psychological difficulties. The purpose of this review and meta-analysis was to examine the nature and strength of the relationship between rumination and non-suicidal self-injury (NSSI).
METHODS: The protocol for this review was pre-registered (CRD42019148186). A literature search of electronic databases PsycINFO, CINAHL, MEDLINE, and Web of Science was performed from the earliest date available to March 2020. Thirty-nine eligible papers were identified. An additional seven papers were identified from a search conducted in September 2021, resulting in a total of 46 papers. Separate meta-analyses were undertaken for NSSI frequency and NSSI history, with studies grouped by rumination type (depressive, transdiagnostic, anger, brooding, reflection, catastrophising, overall). Moderator analyses were also conducted along with a narrative synthesis of adjusted associations and longitudinal studies.
RESULTS: Rumination had a positive small association with NSSI frequency and a positive moderate association with NSSI history. The adjusted associations yielded mixed findings and most longitudinal research found rumination to be associated with prospective NSSI. LIMITATIONS: Most included studies had a moderate risk of bias and used a student sample. A limitation of this review was that only English language papers were included.
CONCLUSIONS: Findings indicate that rumination is associated with NSSI, but more so the likelihood of engaging in NSSI overall than the frequency. Rumination-focused techniques for NSSI may therefore be of benefit. Further research is needed to understand this association, particularly with longitudinal studies that focus on state rumination rather than stable trait rumination. PRACTITIONER POINTS: Rumination was found to have a small to moderate association with NSSI. Some individuals who engage in NSSI may benefit from psychological techniques that target rumination. Most of the studies included had a moderate risk of bias.
Language: en |