- Claims of adverse health outcomes associated with antidepressant use may not be supported by strong evidence and may be exaggerated by confounding by indication, according to an umbrella review of meta-analyses of observational studies.
Why this matters
- Although prior studies have suggested a risk for adverse health outcomes associated with antidepressant use, the credibility of those studies had not been quantified.
- This systematic umbrella review of 45 meta-analyses of observational studies evaluated the association between antidepressant use and adverse health outcomes.
- Funding: NIHR Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London.
- 45 meta-analyses had 120 associations and included 1012 individual study estimates of adverse health outcomes.
- Of the 120 associations studied, 74 (61.7%) associations were nominally statistically significant (P≤.05); only 22 (18.3%) reached a P≤1×10−6.
- There was large heterogeneity in 43.3% of the associations (I2>50%), small-study effects for 14.2% of the associations, and excess significant bias for 7.5% of the associations.
- Evidence from meta-analyses of randomised clinical trials was not graded.
Coauthored with Chitra Ravi, MPharm