Suicide and Self-Harm in Intellectual Disability: A Systematic Review and Meta-Analysis.

Only 10 studies were included in the meta-analysis, and most planned subgroup and moderator analyses could not be done. Both outcomes showed very high between-study differences, so the combined results depend strongly on which studies were included. The final suicide death and self-harm results came after removing an outlier study from each analysis, and the authors state that these outliers may reflect real differences in the intellectual disability population rather than statistical artefacts. Several post-preregistration adjustments were needed, which may have changed study weighting and the final pooled estimate. All included studies were from high-income countries, and the authors note that 75% of suicides occur in low- and middle-income settings. The review included self-harm as a suicide-related outcome, but self-harm may not always be intended to cause death, so the self-harm estimate may mix different behaviours. Age differences could not be studied because age was often not reported and adjustment for age was inconsistent.

The authors declare no conflicts of interest.

Do not read the lower suicide death risk as reassurance that self-harm is less serious. The self-harm risk was higher after removing an outlier study, and the initial analyses for both outcomes were not significant.