The prevalence of self-injurious behaviour in autism: a meta-analytic study

Catherine Steenfeldt-Kristensen, Chris A. Jones, Caroline Richards*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)
214 Downloads (Pure)


Self-injurious behaviour is purportedly common in autism, but prevalence rates have not yet been synthesised meta-analytically. In the present study, data from 14,379 participants in thirty-seven papers were analysed to generate a pooled prevalence estimate of self-injury in autism of 42% (confidence intervals 0.38–0.47). Hand-hitting topography was the most common form of self-injury (23%), self-cutting topography the least common (3%). Sub-group analyses revealed no association between study quality, participant intellectual disability or age and overall prevalence rate of self-injury. However, females obtained higher prevalence rates than males (p =.013) and hair pulling and self-scratching were associated with intellectual disability (p =.008 and p =.002, respectively). The results confirm very high rates of self-injury in autism and highlight within group risk-markers.

Original languageEnglish
Pages (from-to)3857-3873
JournalJournal of Autism and Developmental Disorders
Issue number11
Early online date15 Apr 2020
Publication statusE-pub ahead of print - 15 Apr 2020


  • Autism
  • Intellectual disability
  • Prevalence
  • Self-harm
  • Self-injurious behaviour

ASJC Scopus subject areas

  • Developmental and Educational Psychology


Dive into the research topics of 'The prevalence of self-injurious behaviour in autism: a meta-analytic study'. Together they form a unique fingerprint.

Cite this