Effect of Alcohol Use Disorder and Alcohol Intake on Risk of Subsequent Depression: A Dose-Response Meta-Analysis of Cohort Studies

Abstract

Background: Alcohol misuse is a highly prevalent and attributed to a great burden of public health. Cross-sectional studies reported an association between alcohol misuse and depression. However, whether preceded alcohol misuse is a risk factor for subsequent depression remains undetermined. Moreover, different assessments based on the existence of alcohol use disorder (AUD) and various intake levels of alcohol may lead to different results. We did a dose-response meta-analysis to estimate the longitudinal association between AUD, alcohol intake level and depression. Methods: We searched Pubmed, Embase and PsycINFO for cohort studies investigating alcohol use on risk of subsequent depression up to April 2019. Effect of Alcohol use disorder (AUD) and alcohol intake on depression onset were analyzed separately. In analysis of alcohol intake, we first compared the risk between heavy drinking and non-heavy drinking, following by a qualitative analysis and a dose-response analysis in comparison of different levels of intake versus abstinence. A random-effects model was used to pool the summary RR and 95% confidence intervals (95%CI). Subgroup analysis and meta-regression was performed to trace heterogeneity. Publication bias was evaluated using multiple methods. Findings: Forty-two studies comprising 338426 participants were included in the current analysis. The summary RR of incident depression for AUD was 1.57 (1.41-1.76) compared to non-AUD. Summary RR of depression was 1.14 (1.06, 1.23) on heavy drinking versus non-heavy drinking. In qualitative analysis for various level of alcohol consumption compared to abstinence, light-moderate drinking was inversely associated with depression risk while heavy drinking was found to possibly elevate the risk of depression. Similarity was demonstrated in dose-response analysis, there was a non-linear association between alcohol intake and depression, with a decreased risk observed when consumption approximately between 0 and 29 g/day and an increased but non-significant risk when the intake was >42 g/day. Interpretation: Alcohol use disorder (AUD) is significantly associated with subsequent depression and heavy drinking probably enhanced the risk. Alarm and attention should be sent to alcohol problematic users especially to people with AUD, regardless of the absence of depression yet. Funding: This work was supported by the funding of National Key R&D Program of China (2018YFC1314400 and 2018YFC1314401 to YP). Declaration of Interest: The authors declare that they have no conflict of interest.