Table 1Antidepressants vs placebo

Domains pertaining to strength of evidenceMagnitude of effectStrength of evidence
Number of studies;
# of subjects
Risk of bias (Design/ Risk of bias)ConsistencyDirectnessPrecisionMagnitude of effectHigh, Moderate, Low, Insufficient
Escitalopram versus placebo (Emslie 2009, Wagner 2006)20, 22
Suicide deaths (Wagner 2006)22
 1; 268Medium (RCT/Unclear)N/AIndirectImpreciseNo eventsInsufficient
Adverse events suggestive of self-harm, with a suicidal tendency (Emslie 2009)20
 1; 316Medium (RCT/Unclear)N/AIndirectImprecise0 vs 0.6% (1/157)Low
Fluoxetine versus placebo (TADS)19, 21, 124, 125
Suicide deaths at 36 weeks
 1; 221Low (RCT/Low)N/AIndirectImpreciseNo eventsInsufficient
Suicide attempts at 36 weeks
 1; 221Low (RCT/Low)N/AIndirectImprecise6.4% (7/109) vs 5.4% (6/112), P not reportedLow
Paroxetine versus placebo (DeRubeis 2005, Emslie 2006)18, 23
Suicide deaths (DeRubeis 2005)23
 1; 180Low (RCT/Low)N/AIndirectImprecise0.8% (1/120) vs 0Low
Suicide behavior (Emslie 2006)18
 1; 206Low (RCT/Low)N/AIndirectImprecise2% (2/104) vs 0Low

From: APPENDIX L, STRENGTH OF EVIDENCE RATINGS FOR PRIMARY STUDIES RELATED TO PHARMACOTHERAPY

Cover of Suicide Prevention Interventions and Referral/Follow-Up Services: A Systematic Review
Suicide Prevention Interventions and Referral/Follow-Up Services: A Systematic Review [Internet].
O'Neil ME, Peterson K, Low A, et al.
Washington (DC): Department of Veterans Affairs; 2012 Mar.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.