Causally-informative analyses of the effect of job displacement on all-cause and specific-cause mortality from the 1990s Finnish recession until 2020: A population registry study

Abstract

AbstractBackgroundJob loss is related to morbidity and mortality, but generation of causal evidence is challenging due to confounding factors. Finland suffered a severe economic recession in the early 1990s with unemployment reaching almost 25%, with many job losses due to mass layoff or company closure. Such job displacements are assumed to be exogenous to the individual and create a natural experiment for causal inference.MethodsWe evaluate the causal relationship between job displacement and mortality using register data from Finland between 1988-2020 (N=590,823 individuals [43·3 % female] aged 25-55 and securely employed by the private sector at baseline, N=93,199 total deaths by 2020).FindingsJob displacement is associated with increased risk of all-cause mortality even after accounting for sex, age, marital status, and education (HR=1·09 [1·07, 1·10]). Risks of death by suicide, violence, alcohol, accidents, and disease are higher for displaced individuals at all follow-up periods examined. Risks of death from cancer and ischaemic heart disease are higher for displaced individuals only in later follow up periods.InterpretationOur analyses support the causal influence of job displacement on all-cause and specific-cause mortality, even up to 30 years after the recession; this risk varies by cause of death and by length of follow-up. Future work should evaluate stress and substance use as potential pathways from job displacement to mortality.FundingThis work was supported by the Biology of Trauma Initiative at the Broad Institute of MIT and Harvard and the Academy of Finland Centre of Excellence in Complex Disease Genetics.Research in ContextEvidence before this studyJob loss and unemployment are stressful life events. We searched PubMed from database inception until August 24, 2024 with the search terms “(mortality) AND (job loss) OR (job displacement)” to identify research utilizing natural experiment job displacement designs to estimate the effect of job loss on all-cause and/or cause-specific mortality. No restrictions were applied to the language of publication or article type. We also reviewed references to identify additional relevant studies. Meta-analytic estimates indicate that job loss negatively influences mental and physical health and confers risk for mortality. That said, unemployment and health have a complex relationship, where poor health can increase likelihood of job loss, which then in turn worsens health; therefore, natural experiments are required for causal inference. Causally-informative studies of job loss suggest that job loss causally increases risk of all-cause mortality up to 20 years after job loss and some specific causes of death within 5-10 years after job loss.Added value of this studyThis study expands the length of follow-up after job loss (up to 30 years) and evaluates a broader range of specific causes of death, such as violence. The long follow-up permits investigation into deaths resulting from diseases that develop over long periods of time, like heart disease and cancer.Implications of all available evidenceJob loss causally increases risk of all-cause and specific-cause mortality and the risk of mortality from specific causes varies by duration after job loss, with external causes of death showing increased risk shortly after job loss and long-term illnesses like cancer or heart disease showing increased risk long after the job loss event. Substance use and stress are implicated as possible mechanisms for future research.