Biological risk factors for suicidal behaviors: a meta-analysis.

B P Chang,K H Bentley,K R Fox,M K Nock,E M Kleiman,J D Ribeiro,J C Franklin

doi:10.1038/tp.2016.165

B P Chang, K H Bentley + Show 5 more

Open Access

https://doi.org/10.1038/tp.2016.165

Copy DOI

Abstract

Prior studies have proposed a wide range of potential biological risk factors for future suicidal behaviors. Although strong evidence exists for biological correlates of suicidal behaviors, it remains unclear if these correlates are also risk factors for suicidal behaviors. We performed a meta-analysis to integrate the existing literature on biological risk factors for suicidal behaviors and to determine their statistical significance. We conducted a systematic search of PubMed, PsycInfo and Google Scholar for studies that used a biological factor to predict either suicide attempt or death by suicide. Inclusion criteria included studies with at least one longitudinal analysis using a biological factor to predict either of these outcomes in any population through 2015. From an initial screen of 2541 studies we identified 94 cases. Random effects models were used for both meta-analyses and meta-regression. The combined effect of biological factors produced statistically significant but relatively weak prediction of suicide attempts (weighted mean odds ratio (wOR)=1.41; CI: 1.09–1.81) and suicide death (wOR=1.28; CI: 1.13–1.45). After accounting for publication bias, prediction was nonsignificant for both suicide attempts and suicide death. Only two factors remained significant after accounting for publication bias—cytokines (wOR=2.87; CI: 1.40–5.93) and low levels of fish oil nutrients (wOR=1.09; CI: 1.01–1.19). Our meta-analysis revealed that currently known biological factors are weak predictors of future suicidal behaviors. This conclusion should be interpreted within the context of the limitations of the existing literature, including long follow-up intervals and a lack of tests of interactions with other risk factors. Future studies addressing these limitations may more effectively test for potential biological risk factors.

Highlights

Suicide is a pressing public health problem
Some studies have shown that nonsuppression during the dexamethasone suppression test predicts future suicidal behaviors (for example, odds ratios (ORs)410),[10,11] whereas others report that dexamethasone suppression test nonsuppression predicts significantly fewer suicidal behaviors.[12]
We investigated the ability of protective biological factors to reduce the likelihood of suicide attempt and suicide death

Summary

Introduction

Suicide is a pressing public health problem. Each year there are an estimated one million suicide deaths worldwide in addition to ~ 25 million non-lethal suicide attempts.[1,2,3] Despite a large increase in suicide research over the last five decades, there has been no appreciable decline in the rates of suicidal behaviors.[4]This lack of progress highlights the need for a new suicide research strategy.[5]The study of potential biomarkers may aid researchers and clinicians in the prediction of future suicidal behaviors as well as help shed light on the etiology and underlying mechanisms associated with suicidal behavior. Each year there are an estimated one million suicide deaths worldwide in addition to ~ 25 million non-lethal suicide attempts.[1,2,3] Despite a large increase in suicide research over the last five decades, there has been no appreciable decline in the rates of suicidal behaviors.[4]. This lack of progress highlights the need for a new suicide research strategy.[5]. Some studies have shown that nonsuppression during the dexamethasone suppression test predicts future suicidal behaviors (for example, odds ratios (ORs)410),[10,11] whereas others report that dexamethasone suppression test nonsuppression predicts significantly fewer suicidal behaviors.[12]

Methods

Results

Conclusion