Mortality risk in migrant and non-migrant individuals with chronic hepatitis B virus infection: a French hospital-based cohort study (ANRS CO22 HEPATHER).

Background: Migrants in Europe are disproportionately affected by hepatitis B virus (HBV) infection, especially those coming from endemic countries. We aimed to determine whether migrant status was associated with all-cause mortality risk in people living with chronic HBV infection integrated into a hospital-based care pathway in France.

Methods: We analysed clinical and socio-behavioural data collected over 8 years of follow-up among patients with chronic HBV infection enrolled in the French prospective multicentre cohort ANRS CO22 HEPATHER. Migrant status was tested as a binary variable (non-migrants versus migrants) and according to three categories (low, moderate, and high) of HBV endemicity in the migrants' region of birth. The association between migrant status and all-cause mortality risk was assessed using a multivariable Cox proportional hazards model. A competing risks analysis was conducted for liver-related and non-liver-related mortality.

Results: Of the 5597 study participants, accounting for 33,222.8 person-years (PY), 68.1% were migrants, mainly from Sub-Saharan Africa and Asia. During follow-up, 247 patients died and the all-cause mortality rate [95% confidence interval (CI)] was 7.4 [6.6-8.4]/1000 PY. Migrants had a lower mortality rate than non-migrants: 4.5 [3.7-5.5]/1000 PY versus 13.5 [11.4-15.8]/1000 PY (p < 0.001), irrespective of migrants' region of birth and time since arrival in France. After adjustment for sex, age, living in poverty, alcohol use, tobacco smoking, diabetes, and HBV disease phase, the all-cause mortality risk was still lower in migrants than in non-migrants (adjusted hazard ratio [95% CI] 0.58 [0.43-0.78], p < 0.001). All three migrant HBV endemicity categories had a lower risk of all-cause and non-liver-related mortality than non-migrants. By contrast, these differences were not significant for liver-related mortality.

Conclusions: A lower all-cause, liver-related and non-liver-related mortality risk was found among migrants with chronic HBV infection in France compared to non-migrants. However, after multivariable adjustment, the liver-related mortality risk was similar between migrants and non-migrants, indicating that mortality advantage for migrants is explained by the protective adjustment factors, such as younger age, less advanced liver disease and fewer unhealthy behaviours. In contrast, these factors did not fully explain the observed mortality advantage for both non-liver-related and all-cause mortality.

Trial registration: ClinicalTrials.gov registry number: NCT01953458.