Relatively few of the adverse health outcomes attributed to antidepressants are supported by convincing evidence, reported the authors of a systematic review of 45 meta-analyses.
The authors did find convincing evidence linking the use of antidepressants and suicide attempt or completion among people under age 19 years and use of the medication and autism risk among offspring. “However, the few [studies] with convincing evidence associations did not reflect causality, and none of them remained at the convincing evidence level after accounting for confounding by indication,” wrote Elena Dragioti, PhD, of the Pain and Rehabilitation Centre at Linköping (Sweden) University and coauthors. The study was published in JAMA Psychiatry.
Dr. Dragioti and coauthors undertook a systematic “umbrella review” grading the evidence from the 45 meta-analyses of 695 observational studies into the association between antidepressant use and the risk of adverse health outcomes. All the meta-analyses included a control group not exposed to antidepressants, with the exception of one that compared the risk of gastrointestinal bleeding between two classes of antidepressants.
They found 120 possible adverse health associations described in the meta-analyses, 61.7% of which related to maternal and pregnancy-related adverse health outcomes. Two-thirds of the adverse health outcome associations involved selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs).
However, among the 120 adverse health associations, only three (2.5%) were supported by “convincing” evidence. One was the association between SSRIs and increased risk of suicide attempts and completion in children and adolescents. Convincing evidence also was found between any antidepressant use before pregnancy and autism spectrum disorder and between SSRI use during pregnancy and autism spectrum disorder. The evidence for the association with suicide risk was deemed high quality, but the two associations with autism spectrum disorder were only of moderate quality.
The authors commented that these findings needed to be considered when prescribing antidepressants in adolescents and children, particularly as another networked meta-analysis had found fluoxetine was the only antidepressant that worked better than placebo in children and adolescents. “In addition, they wrote.
The review found that 11 adverse health outcomes (9.2%) had “highly suggestive” evidence linking them to antidepressant use. These were ADHD in children, cataract development, severe bleeding at any site, upper gastrointestinal tract bleeding, postpartum hemorrhage, preterm birth, lower Apgar score at 5 minutes, osteoporotic fracture, and hip fracture.
Seven of those – ADHD in children, lower Apgar score, severe bleeding at any site, cataract development, osteoporotic features, preterm birth, and upper GI bleeding – had moderate-quality evidence. However, the authors noted that the effect sizes were small and had low prevalence.
The study also found highly suggestive evidence linking antidepressant use to a decreased risk of suicide attempts or completion in adults.
The authors said several of those adverse events in adults, such as GI bleeding and osteoporotic fractures, could be prevented with medication, so the advantages of antidepressant use in adults could outweigh the disadvantage of those preventable safety issues.
Twenty-one adverse health outcomes showed either suggestive, weak, or no evidence for their association with antidepressant use.
They also conducted a sensitivity analysis that limited the analysis to cohort studies, prospective cohort studies, studies that controlled for confounding by the treatment indication, and studies from North America. This showed that none of the associations for which there was originally deemed to be convincing evidence retained that same rank.
“Overall, the results showed that the association between antidepressant use and adverse health outcomes was not supported by robust evidence and that the underlying disease likely inflated the findings in a relevant way,” the authors wrote.
However, when they looked solely at prospective cohort studies, the association between preterm birth and use of any antidepressant was upgraded to having convincing evidence.
When the analysis focused on SSRIs only, the association with lower Apgar scores at 5 minutes also was upgraded to having convincing evidence. Similarly, the evidence for an association with preterm birth also was found to be convincing when the analysis was limited to other or mixed antidepressants.
Dr. Dragioti and coauthors cited several limitations, including the inability of some randomized, controlled trials to address adverse outcomes.
“Antidepressant use appears to be safe for the treatment of psychiatric disorders, but more studies matching for underlying disease are needed to clarify the degree of confounding by indication and other biases,” the authors wrote.
The study was funded by several entities, including the National Institute for Health Research’s Biomedical Research Centre at South London and Maudsley NHS Foundation Trust. Dr. Dragioti reported no disclosures. Four authors declared funding, consultancies, personal fees, royalties, or shares in the pharmaceutical sector. No other conflicts of interest were declared.
SOURCE: Dragioti E et al. JAMA Psychiatry. 2019 Oct 2. doi: 10.1001/jamapsychiatry.2019.2859.