Employment Status and Alcohol-Attributable Mortality Risk-A Systematic Review and Meta-Analysis

Abstract

Being unemployed has been linked to various health burdens. In particular, there appears to be an association between unemployment and alcohol-attributable deaths. However, risk estimates presented in a previous review were based on only two studies. Thus, we estimated updated sex-stratified alcohol-attributable mortality risks for unemployed compared with employed individuals. A systematic literature search was conducted in August 2020 using the following databases: Embase, MEDLINE, PsycINFO, and Web of Science. The relative risk (RR) of dying from an alcohol-attributable cause of death for unemployed compared with employed individuals was summarized using sex-stratified random-effects DerSimonian-Laird meta-analyses. A total of 10 studies were identified, comprising about 14.4 million women and 19.0 million men, among whom there were about 3147 and 17,815 alcohol-attributable deaths, respectively. The pooled RRs were 3.64 (95% confidence interval (CI): 2.04-6.66) and 4.93 (95% CI 3.45-7.05) for women and men, respectively. The findings of our quantitative synthesis provide evidence that being unemployed is associated with an over three-fold higher risk of alcohol-attributable mortality compared with being employed. Consequently, a global public health strategy connecting brief interventions and specialized care with social services assisting those currently unemployed is needed.

Keywords: alcohol use; alcohol-attributable mortality; employment status; inequality; mortality; public health; socioeconomic status; systematic literature review and meta-analysis.

Figure 1

The PRISMA flow chart of the study selection for the search conducted in 2013 (including studies published up to February 2013) and 2020 (including studies published between January 2013 and August 2020). SES, socioeconomic status.

Figure 2

Random-effects meta-regression for employment status. RR, relative risk; CI, confidence interval; UK, United Kingdom. * This estimate is referring to long-term unemployment, whereas the other estimate by Herttua et al. 2008 [20] is referring to short-term unemployment. For women refer to [19,20,32,35,36,37,38,39], for men refer to [19,20,32,33,34,36,37,38,39].