Employment history indicators and mortality in a nested case-control study from the Spanish WORKing life social security (WORKss) cohort.

Employment has transitioned from stable to more flexible schemes. Little is known about the effects of dynamic working lives and mortality. We examined the association of employment, unemployment and inactivity on mortality among workers born in 1926-1988, in a nested case-control study of workers from the Spanish WORKss cohort. Cases were all deaths that occurred during 2004-2013 and controls were living persons, matched for sex and age at the time the case occurred. We had information on employment from 1981 to 2013. Logistic regression was used to measure the associations between the 3 employment history indicators separately by sex. There were 53,989 cases and an equal number of controls (n = 107,978). More than 16 years employed showed a protective effect against mortality in women (OR = 0.88, 95%CI: 0.81, 0.94) and men (OR = 0.76, 95%CI: 0.70, 0.79). The number of spells and time in unemployment and inactivity were significantly related to mortality in men, but not women. Sensitivity analyses by labor relationship showed stronger associations of unemployment (OR = 1.42, 95%CI: 1.13, 1.78) and inactivity (OR = 1.34; 95%CI: 1.09, 1.65) for temporary workers. Employment gaps are detrimental to health and have worse effects if the gaps occur without unemployment benefits or after temporary contracts. These results may drive improvement of labor and social policies that protect workers against the potential negative effects of dynamic work lives