Table 2. Class I or II Evidence in Meta-analyses of the Association Between Antidepressant Use and Risk of Adverse Health Outcomes.
| Source | Adverse Health Outcome | Exposed/ Unexposed |
Prevalence Based on Cohort Studies, % | No. of Included Studies per Association | Random-Effects Measure, ES (95% CI) | Result | Criteria for Level-of-Evidence Classification | AMSTAR 2 Quality | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of Cases/ Total Population |
P Value for Random Effects | Heterogeneity, I2 (P Value) |
PI, 95% CI |
SSE/ESB | LS | CE | ||||||||
| Morales et al,51 2018 | Autism spectrum disorders (prepregnancy maternal exposure) | Any AD users/ no AD users |
0.8 | 7 | RR: 1.48 (1.29 to 1.71) | Increased risk for AD | 22 877/ 2 400 720 |
6.8 × 10−8 | 24 (.24) | 1.09 to 2.02 | No/NP | Yes | I | Moderate |
| Andalib et al,52 2017 | Autism spectrum disorders (pregnancy maternal exposure; unadjusted estimates only) | SSRI/ non-SSRI users |
0.9 | 7 | OR: 1.84 (1.60 to 2.11) | Increased risk for SSRI | 58 178/ 5 868 692 |
1.2 × 10−17 | 0 (.73) | 1.53 to 2.20 | No/NP | Yes | I | Moderate |
| Barbui et al,80 2009 | Suicide attempt and completion in children and adolescents | SSRI/ non-SSRI users |
9.4 | 5 | OR: 1.92 (1.51 to 2.44) | Increased risk for SSRI | 6531/ 61 522 |
1.0 × 10−7 | 0 (.47) | 1.30 to 2.84 | No/NP | Yes | I | High |
| Khanassov et al,42 2018 | Osteoporotic fractures | SSRI/ non-SSRI users |
6.5 | 24 | RR:1.67 (1.56 to 1.80) | Increased risk for SSRI | 136 449/ 1 546 913 |
1.3 × 10−44 | 88 (<.001) | 1.23 to 2.27 | No/NP | Yes | II | Moderate |
| Man et al,48 2018 | ADHD in children | Prenatal exposure to AD/ no AD users |
2.4 | 7 | RR: 1.39 (1.21 to 1.61) | Increased risk for AD | 57 552/ 2 886 904 |
5.1 × 10−6 | 79 (<.001) | 0.90 to 2.15 | No/NP | Yes | II | Moderate |
| Fu et al,49 2018 | Cataract development | TCA/nonusers or no users of any other AD |
NA | 3 | OR: 1.19 (1.11 to 1.28) | Increased risk for TCA | 215 298/ 431 171 |
2.0 × 10−6 | 58 (.09) | 0.50 to 2.52 | No/NP | Yes | II | Moderate |
| Laporte et al,55 2017 | Severe bleeding at any site | SSRI + SNRI/ non-users or no users of any other AD |
2.4 | 44 | OR: 1.41 (1.27 to 1.57) | Increased risk for SSRI + SNRI | 75 215/ 1 443 029 |
2.2 × 10−10 | 90 (<.001) | 0.77 to 2.59 | No/no | Yes | II | Low |
| Jiang et al,58 2016 | Postpartum hemorrhage | Any AD users/non-AD users | 6.8 | 17 | RR: 1.32 (1.17 to 1.48) | Increased risk for AD | 49 155/ 651 715 |
3.3 × 10−6 | 85 (<.001) | 0.84 to 2.07 | No/NP | Yes | II | Low |
| Jiang et al,64 2015 | Upper GI bleeding | SSRI + other non-AD/ no SSRI use only + other non-AD |
0.7 | 22 | OR: 1.55 (1.35 to 1.78) | Increased risk for SSRI | 56 182/ 592 508 |
9.2 × 10−12 | 89 (<.001) | 0.83 to 2.91 | No/no | Yes | II | Moderate |
| Huang et al,66 2014 | Preterm birth | Any AD users/ no AD users |
0.8 | 28 | RR: 1.68 (1.52 to 1.86) | Increased risk for AD | 24 669/ 3 063 709 |
3.6 × 10−23 | 44 (.008) | 1.23 to 2.30 | Yes/NP | Yes | II | Moderate |
| Wu et al,70 2013 | Osteoporotic fractures | TCA users/ non-TCA users |
1.4 | 12 | RR: 1.45 (1.31 to 1.60) | Increased risk for TCA | 178 237/ 831 912 |
6.2 × 10−13 | 76 (<.001) | 1.04 to 2.01 | Yes/no | Yes | II | Moderate |
| Ross et al,71 2013 | Apgar score at 5 min | Any AD users/ no AD users |
2.1 | 15 | SMD: −0.33 (−0.47 to −0.20) | Increased risk for AD | 1473/ 71 828 |
7.5 × 10−7 | 58 (.003) | −1.02 to 0.36 | No/no | Yes | II | Moderate |
| Oderda et al,76 2012 | Hip fracture | TCA and/or SSRI users/no AD users |
7.4 | 18 | OR: 1.78 (1.53 to 2.07) | Increased risk for TCA or SSRI | 49 276/ 210 577 |
5.2 × 10−14 | 89 (<.001) | 1.00 to 3.19 | Yes/yes | Yes | II | Critically low |
| Barbui et al,80 2009 | Suicide attempt and completion in adults | SSRI/ non-SSRI users |
4.5 | 7 | OR: 0.59 (0.48 to 0.72) | Decreased risk for SSRI | 7164/ 147 383 |
5.2 × 10−7 | 59 (.02) | 0.33 to 1.05 | No/NP | Yes | II | High |
Abbreviations: AD, antidepressant; ADHD, attention-deficit/hyperactivity disorder; Apgar score, appearance (skin color), pulse (heart rate), grimace (reflex irritability), activity (muscle tone), and respiration; AMSTAR, A Measurement Tool to Assess Systematic Reviews; CE, class of evidence; ES, effect size; ESB, excess significance bias; GI, gastrointestinal; LS, largest study with significant effect; NA, not applicable; NP, not pertinent because of fewer-than-expected number of observed studies; OR, odds ratio; PI, prediction interval; RR, relative risk; SMD, standardized mean difference; SNRI, serotonin-norepinephrine reuptake inhibitor; SSE, small-study effect; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant.