Association of Antidepressant Use With Adverse Health Outcomes: A Systematic Umbrella Review

Elena Dragioti; Marco Solmi; Angela Favaro; Paolo Fusar-Poli; Paola Dazzan; Trevor Thompson; Brendon Stubbs; Joseph Firth; Michele Fornaro; Dimitrios Tsartsalis; Andre F. Carvalho; Eduard Vieta; Philip McGuire; Allan H. Young; Jae Il Shin; Christoph U. Correll; Evangelos Evangelou

doi:10.1001/jamapsychiatry.2019.2859

Association of Antidepressant Use With Adverse Health Outcomes: A Systematic Umbrella Review

Elena Dragioti^*, Marco Solmi, Angela Favaro, Paolo Fusar-Poli, Paola Dazzan, Trevor Thompson, Brendon Stubbs, Joseph Firth, Michele Fornaro, Dimitrios Tsartsalis, Andre F. Carvalho, Eduard Vieta, Philip McGuire, Allan H. Young, Jae Il Shin, Christoph U. Correll, Evangelos Evangelou

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

144 Citations (Scopus)

619 Downloads (Pure)

Abstract

Importance: Antidepressant use is increasing worldwide. Contrasting evidence from meta-analyses is available on their safety. Objective: To grade the evidence across meta-analyses of observational studies assessing the association between antidepressants and adverse health outcomes, using an umbrella review approach. Data Sources: We searched PubMed, Scopus, and PsycINFO, up to April 5, 2019. Evidence Review: Only meta-analyses of observational studies with a cohort or case-control study design were eligible. Two independent reviewers recorded the data and assessed the methodological quality of the selected meta-analyses. Evidence was ranked as convincing, highly suggestive, suggestive, weak, and not significant according to established criteria. Results: Forty-five eligible meta-analyses describing 120 associations including data from 1012 individual effect size estimates were included, after scrutinizing 252 potential articles. Seventy-four of the 120 associations (61.7%) had a nominally statistically significant effect (p≤0.05) using random-effects models. Large heterogeneity (I2>50%) was present (43.3%), while small-study effects (14.2%) and excess significance bias (7.5%) were less common. While convincing meta-analytic evidence emerged from both main and sensitivity analysis for the association between antidepressants and risk of suicide attempt/completion in children/adolescents (9.4%, OR=1.92, 95%CI=1.51-2.44 in adjusted studies, SSRIs only, and in high-quality studies, 3.6% OR=1.88, 95%CI=1.47-2.40), autism spectrum disorders (ASD) with antidepressants before (0.8%, RR=1.48, 95%CI=1.29-1.71 in any antidepressant, adjusted studies, and high-quality studies) and during pregnancy (0.9%, OR=1.84, 95%CI=1.60-2.11 in both high-quality studies and SSRIs only, and 0.1% OR=1.80, 95%CI=1.54-2.10 in European studies), pre-term birth in prospective cohort studies (0.4%, RR=1.87, 95%CI=1.52-2.30), and in studies of mixed antidepressants (0.4%, RR=1.59, 95%CI=1.31-1.93), and low APGAR scores in SSRIs only studies (5.7%, SMD=-0.27, 95%CI -0.37 to -0.16), none of these associations remained at convincing evidence after sensitivity analysis for confounding by indication. Conclusions and Relevance: Most putative adverse health outcomes associated with antidepressants are not supported by convincing evidence, and the few of those supported by convincing evidence are affected by confounding by indication. Antidepressants appear to be overall safe for the treatment of psychiatric disorders. More studies matching for underlying disease are needed to clarify the degree of confounding by indication and other biases. No absolute contraindication to antidepressants emerges from this umbrella review. Registration: PROSPERO registration 2018: CRD42018103462.

Original language	English
Pages (from-to)	1241-1255
Number of pages	15
Journal	JAMA Psychiatry
Volume	76
Issue number	12
Early online date	2 Oct 2019
DOIs	https://doi.org/10.1001/jamapsychiatry.2019.2859 https://doi.org/10.1001/jamapsychiatry.2019.2859
Publication status	Published - 2 Oct 2019

Keywords

antidepressants; adverse outcomes; umbrella review; meta-analysis; safety

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

AD_UR_manuscript_R2_clean version Accepted author manuscript, 376 KB
JAMA_AdverseHealthOutcomesAntiDepres_2019Accepted author manuscript, 1.2 MB
Association of Antidepressant Use With_DRAGIOTI_Accepted21June2019_GREEN AAMAccepted author manuscript, 538 KB

Cite this

Dragioti, E., Solmi, M., Favaro, A., Fusar-Poli, P., Dazzan, P., Thompson, T., Stubbs, B., Firth, J., Fornaro, M., Tsartsalis, D., Carvalho, A. F., Vieta, E., McGuire, P., Young, A. H., Shin, J. I., Correll, C. U., & Evangelou, E. (2019). Association of Antidepressant Use With Adverse Health Outcomes: A Systematic Umbrella Review. JAMA Psychiatry, 76(12), 1241-1255. https://doi.org/10.1001/jamapsychiatry.2019.2859, https://doi.org/10.1001/jamapsychiatry.2019.2859

@article{1d9ac7e5185a40919e797bedfd3e2fb5,

title = "Association of Antidepressant Use With Adverse Health Outcomes: A Systematic Umbrella Review",

abstract = "Importance: Antidepressant use is increasing worldwide. Contrasting evidence from meta-analyses is available on their safety. Objective: To grade the evidence across meta-analyses of observational studies assessing the association between antidepressants and adverse health outcomes, using an umbrella review approach. Data Sources: We searched PubMed, Scopus, and PsycINFO, up to April 5, 2019. Evidence Review: Only meta-analyses of observational studies with a cohort or case-control study design were eligible. Two independent reviewers recorded the data and assessed the methodological quality of the selected meta-analyses. Evidence was ranked as convincing, highly suggestive, suggestive, weak, and not significant according to established criteria. Results: Forty-five eligible meta-analyses describing 120 associations including data from 1012 individual effect size estimates were included, after scrutinizing 252 potential articles. Seventy-four of the 120 associations (61.7%) had a nominally statistically significant effect (p≤0.05) using random-effects models. Large heterogeneity (I2>50%) was present (43.3%), while small-study effects (14.2%) and excess significance bias (7.5%) were less common. While convincing meta-analytic evidence emerged from both main and sensitivity analysis for the association between antidepressants and risk of suicide attempt/completion in children/adolescents (9.4%, OR=1.92, 95%CI=1.51-2.44 in adjusted studies, SSRIs only, and in high-quality studies, 3.6% OR=1.88, 95%CI=1.47-2.40), autism spectrum disorders (ASD) with antidepressants before (0.8%, RR=1.48, 95%CI=1.29-1.71 in any antidepressant, adjusted studies, and high-quality studies) and during pregnancy (0.9%, OR=1.84, 95%CI=1.60-2.11 in both high-quality studies and SSRIs only, and 0.1% OR=1.80, 95%CI=1.54-2.10 in European studies), pre-term birth in prospective cohort studies (0.4%, RR=1.87, 95%CI=1.52-2.30), and in studies of mixed antidepressants (0.4%, RR=1.59, 95%CI=1.31-1.93), and low APGAR scores in SSRIs only studies (5.7%, SMD=-0.27, 95%CI -0.37 to -0.16), none of these associations remained at convincing evidence after sensitivity analysis for confounding by indication. Conclusions and Relevance: Most putative adverse health outcomes associated with antidepressants are not supported by convincing evidence, and the few of those supported by convincing evidence are affected by confounding by indication. Antidepressants appear to be overall safe for the treatment of psychiatric disorders. More studies matching for underlying disease are needed to clarify the degree of confounding by indication and other biases. No absolute contraindication to antidepressants emerges from this umbrella review. Registration: PROSPERO registration 2018: CRD42018103462. ",

keywords = "antidepressants; adverse outcomes; umbrella review; meta-analysis; safety",

author = "Elena Dragioti and Marco Solmi and Angela Favaro and Paolo Fusar-Poli and Paola Dazzan and Trevor Thompson and Brendon Stubbs and Joseph Firth and Michele Fornaro and Dimitrios Tsartsalis and Carvalho, {Andre F.} and Eduard Vieta and Philip McGuire and Young, {Allan H.} and Shin, {Jae Il} and Correll, {Christoph U.} and Evangelos Evangelou",

year = "2019",

month = oct,

day = "2",

doi = "10.1001/jamapsychiatry.2019.2859",

language = "English",

volume = "76",

pages = "1241--1255",

journal = "JAMA Psychiatry",

issn = "2168-622X",

publisher = "American Medical Association",

number = "12",

}

Dragioti, E, Solmi, M, Favaro, A, Fusar-Poli, P , Dazzan, P, Thompson, T, Stubbs, B, Firth, J, Fornaro, M, Tsartsalis, D, Carvalho, AF, Vieta, E, McGuire, P , Young, AH, Shin, JI, Correll, CU & Evangelou, E 2019, 'Association of Antidepressant Use With Adverse Health Outcomes: A Systematic Umbrella Review', JAMA Psychiatry, vol. 76, no. 12, pp. 1241-1255. https://doi.org/10.1001/jamapsychiatry.2019.2859, https://doi.org/10.1001/jamapsychiatry.2019.2859

TY - JOUR

T1 - Association of Antidepressant Use With Adverse Health Outcomes

T2 - A Systematic Umbrella Review

AU - Dragioti, Elena

AU - Solmi, Marco

AU - Favaro, Angela

AU - Fusar-Poli, Paolo

AU - Dazzan, Paola

AU - Thompson, Trevor

AU - Stubbs, Brendon

AU - Firth, Joseph

AU - Fornaro, Michele

AU - Tsartsalis, Dimitrios

AU - Carvalho, Andre F.

AU - Vieta, Eduard

AU - McGuire, Philip

AU - Young, Allan H.

AU - Shin, Jae Il

AU - Correll, Christoph U.

AU - Evangelou, Evangelos

PY - 2019/10/2

Y1 - 2019/10/2

N2 - Importance: Antidepressant use is increasing worldwide. Contrasting evidence from meta-analyses is available on their safety. Objective: To grade the evidence across meta-analyses of observational studies assessing the association between antidepressants and adverse health outcomes, using an umbrella review approach. Data Sources: We searched PubMed, Scopus, and PsycINFO, up to April 5, 2019. Evidence Review: Only meta-analyses of observational studies with a cohort or case-control study design were eligible. Two independent reviewers recorded the data and assessed the methodological quality of the selected meta-analyses. Evidence was ranked as convincing, highly suggestive, suggestive, weak, and not significant according to established criteria. Results: Forty-five eligible meta-analyses describing 120 associations including data from 1012 individual effect size estimates were included, after scrutinizing 252 potential articles. Seventy-four of the 120 associations (61.7%) had a nominally statistically significant effect (p≤0.05) using random-effects models. Large heterogeneity (I2>50%) was present (43.3%), while small-study effects (14.2%) and excess significance bias (7.5%) were less common. While convincing meta-analytic evidence emerged from both main and sensitivity analysis for the association between antidepressants and risk of suicide attempt/completion in children/adolescents (9.4%, OR=1.92, 95%CI=1.51-2.44 in adjusted studies, SSRIs only, and in high-quality studies, 3.6% OR=1.88, 95%CI=1.47-2.40), autism spectrum disorders (ASD) with antidepressants before (0.8%, RR=1.48, 95%CI=1.29-1.71 in any antidepressant, adjusted studies, and high-quality studies) and during pregnancy (0.9%, OR=1.84, 95%CI=1.60-2.11 in both high-quality studies and SSRIs only, and 0.1% OR=1.80, 95%CI=1.54-2.10 in European studies), pre-term birth in prospective cohort studies (0.4%, RR=1.87, 95%CI=1.52-2.30), and in studies of mixed antidepressants (0.4%, RR=1.59, 95%CI=1.31-1.93), and low APGAR scores in SSRIs only studies (5.7%, SMD=-0.27, 95%CI -0.37 to -0.16), none of these associations remained at convincing evidence after sensitivity analysis for confounding by indication. Conclusions and Relevance: Most putative adverse health outcomes associated with antidepressants are not supported by convincing evidence, and the few of those supported by convincing evidence are affected by confounding by indication. Antidepressants appear to be overall safe for the treatment of psychiatric disorders. More studies matching for underlying disease are needed to clarify the degree of confounding by indication and other biases. No absolute contraindication to antidepressants emerges from this umbrella review. Registration: PROSPERO registration 2018: CRD42018103462.

AB - Importance: Antidepressant use is increasing worldwide. Contrasting evidence from meta-analyses is available on their safety. Objective: To grade the evidence across meta-analyses of observational studies assessing the association between antidepressants and adverse health outcomes, using an umbrella review approach. Data Sources: We searched PubMed, Scopus, and PsycINFO, up to April 5, 2019. Evidence Review: Only meta-analyses of observational studies with a cohort or case-control study design were eligible. Two independent reviewers recorded the data and assessed the methodological quality of the selected meta-analyses. Evidence was ranked as convincing, highly suggestive, suggestive, weak, and not significant according to established criteria. Results: Forty-five eligible meta-analyses describing 120 associations including data from 1012 individual effect size estimates were included, after scrutinizing 252 potential articles. Seventy-four of the 120 associations (61.7%) had a nominally statistically significant effect (p≤0.05) using random-effects models. Large heterogeneity (I2>50%) was present (43.3%), while small-study effects (14.2%) and excess significance bias (7.5%) were less common. While convincing meta-analytic evidence emerged from both main and sensitivity analysis for the association between antidepressants and risk of suicide attempt/completion in children/adolescents (9.4%, OR=1.92, 95%CI=1.51-2.44 in adjusted studies, SSRIs only, and in high-quality studies, 3.6% OR=1.88, 95%CI=1.47-2.40), autism spectrum disorders (ASD) with antidepressants before (0.8%, RR=1.48, 95%CI=1.29-1.71 in any antidepressant, adjusted studies, and high-quality studies) and during pregnancy (0.9%, OR=1.84, 95%CI=1.60-2.11 in both high-quality studies and SSRIs only, and 0.1% OR=1.80, 95%CI=1.54-2.10 in European studies), pre-term birth in prospective cohort studies (0.4%, RR=1.87, 95%CI=1.52-2.30), and in studies of mixed antidepressants (0.4%, RR=1.59, 95%CI=1.31-1.93), and low APGAR scores in SSRIs only studies (5.7%, SMD=-0.27, 95%CI -0.37 to -0.16), none of these associations remained at convincing evidence after sensitivity analysis for confounding by indication. Conclusions and Relevance: Most putative adverse health outcomes associated with antidepressants are not supported by convincing evidence, and the few of those supported by convincing evidence are affected by confounding by indication. Antidepressants appear to be overall safe for the treatment of psychiatric disorders. More studies matching for underlying disease are needed to clarify the degree of confounding by indication and other biases. No absolute contraindication to antidepressants emerges from this umbrella review. Registration: PROSPERO registration 2018: CRD42018103462.

KW - antidepressants; adverse outcomes; umbrella review; meta-analysis; safety

UR - http://www.scopus.com/inward/record.url?scp=85073050563&partnerID=8YFLogxK

U2 - 10.1001/jamapsychiatry.2019.2859

DO - 10.1001/jamapsychiatry.2019.2859

M3 - Article

C2 - 31577342

AN - SCOPUS:85073050563

SN - 2168-622X

VL - 76

SP - 1241

EP - 1255

JO - JAMA Psychiatry

JF - JAMA Psychiatry

IS - 12

ER -

Association of Antidepressant Use With Adverse Health Outcomes: A Systematic Umbrella Review

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this