Citation

Köhler CA, Evangelou E, Stubbs B, Solmi M, Veronese N, Belbasis L, Bortolato B, Melo MCA, Coelho CA, Fernandes BS, Olfson M, Ioannidis JPA, Carvalho AF. J. Psychiatr. Res. 2018; 103: 189-207.

Affiliation

Institute for Clinical Research and Education in Medicine (IREM), Padova, Italy; Department of Psychiatry, University of Toronto, Toronto, ON, Canada; Center for Addiction and Mental Health (CAMH), Toronto, ON, Canada. Electronic address: andre.carvalho@camh.ca.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.jpsychires.2018.05.020

PMID

29886003

Abstract

The development of depression may involve a complex interplay of environmental and genetic risk factors. PubMed and PsycInfo databases were searched from inception through August 3, 2017, to identify meta-analyses and Mendelian randomization (MR) studies of environmental risk factors associated with depression. For each eligible meta-analysis, we estimated the summary effect size and its 95% confidence interval (CI) by random-effects modeling, the 95% prediction interval, heterogeneity with I², and evidence of small-study effects and excess significance bias. Seventy meta-analytic reviews met the eligibility criteria and provided 134 meta-analyses for associations from 1283 primary studies. While 109 associations were nominally significant (P < 0.05), only 8 met the criteria for convincing evidence and, when limited to prospective studies, convincing evidence was found in 6 (widowhood, physical abuse during childhood, obesity, having 4-5 metabolic risk factors, sexual dysfunction, job strain). In studies in which depression was assessed through a structured diagnostic interview, only associations with widowhood, job strain, and being a Gulf War veteran were supported by convincing evidence. Additionally, 8 MR studies were included and provided no consistent evidence for the causal effects of obesity, smoking, and alcohol consumption. The proportion of variance explained by genetic risk factors was extremely small (0.1-0.4%), which limited the evidence provided by the MR studies. Our findings suggest that despite the large number of putative risk factors investigated in the literature, few associations were supported by robust evidence. The current findings may have clinical and research implications for the early identification of individuals at risk for depression.

Copyright © 2018 Elsevier Ltd. All rights reserved.

Language: en

Keywords

Depression; Meta-analyses; Prevention; Psychiatry; Risk factors; Umbrella review