Impulsivity‐Related Traits in Eating Disorders With Self‐Harm or Suicidal Behaviour: A Systematic Review and Meta‐Analysis
Maria Gemescu, Cezar Giosan, Ana Maria Olguța Barizi, Carmen‐Andreea Petre, Ana‐Alecsandra Gușoaie, Elena‐Luiza Costache, Ana‐Patricia Darabont, Teodora‐Maria Neagoe, Rareș‐Mihnea IosifescuABSTRACT
Background
Although impulsivity is commonly associated with eating disorders (EDs) comorbid with self‐harm, no prior reviews have investigated whether ED groups with co‐occurring self‐harm exhibit increased impulsivity‐related traits.
Objective
This systematic review and meta‐analysis synthesized the differences between ED groups with vs. without self‐harm/suicidal behaviour, with respect to impulsivity‐related domains.
Method
Following the PRISMA 2020 guidelines, we searched for comparative observational studies using electronic databases, reference lists, personal reference collection and requests sent to authors in the field for unpublished data. We included studies conducted from 1994 onwards that compared clinical ED groups with vs. without self‐harm/suicidal behaviour on impulsivity‐related domains, as assessed using questionnaires. We used a two‐stage screening process conducted by independent reviewers and the Joanna Briggs Institute's tool for risk of bias assessment. Results were synthesized narratively and quantitatively, using separate random‐effects meta‐analyses for each impulsivity‐related domain.
Results
According to meta‐analytic results, clinical ED groups with self‐harm/suicidal behaviour exhibited significantly increased scores on Negative Urgency ( k = 10, g = 0.55), Lack of Premeditation ( k = 8, g = 0.34), Lack of Perseverance ( k = 8, g = 0.20) and Sensation Seeking ( k = 8, g = 0.10). Only two studies assessed Positive Urgency, both reporting non‐significant differences.
Conclusion
Negative Urgency emerged as the prime impulsivity‐related trait associated with self‐harm in ED groups, therefore strengthening the evidence for emotion‐driven impulsivity as a transdiagnostic factor of psychopathology. The small number of studies included, considerable heterogeneity and predominance of female samples should be taken into account when interpreting current findings.