The effect of migration on the incidence and mortality of bloodstream infection: a Danish register-based cohort study.

Abstract Objectives To investigate bloodstream infection (BSI) related to migrant status by comparing the incidence and mortality in migrants with that in non-migrants. Methods In this register-based cohort study we linked a cohort of migrants and non-migrants with a bacteremia database covering two regions in Denmark. We included first-time BSI between January 2000 and December 2015 in individuals ≥18 years. Migrant was divided according to status; refugees or family-reunified migrants. Incidence rate ratio and mortality rate ratio were analyzed using Poisson regression. Results We identified 493 080 non-migrants of which 3 405 had bloodstream infection and 80 740 migrants with 576 cases of these 40 222 family-reunified migrants with 226 cases and 40 518 refugees with 350 cases. Refugees had higher risk of BSI than non-migrants (adjusted IRR:1.19 [95%CI: 1.01-1.40]). Family-reunified migrants and refugees had higher risk of Gram-negative BSI (adjusted IRR:1.23 [95%CI: 1.00-1.51] and 1.57 [95%CI: 1.1.32-1.86]), respectively and lower risk of Gram-positive BSI (adjusted IRR:0.65 [95%CI: 0.51-0.83] and 0.77 [95%CI: 0.63-0.95]), respectively compared with non-migrants. Originating from Southeast Asia and the Pacific resulted in increased risk of BSI compared with non-migrants (adjusted IRR:1.26 [95%CI: 1.07-1.49]). We found no differences in the adjusted 30-day or 90-day mortality according to migrant status. Conclusions Vulnerability towards BSI differs according to migrant status. Refugees had higher risk of BSI overall. Both refugees and family-reunified migrants had higher incidence of Gram-negative BSI than non-migrants. Likewise, migrants from Southeast Asia and the Pacific had higher risk of BSI compared with non-migrants.