Long-term mortality and hospitalisation among Italian veterans deployed to the Balkans: a retrospective longitudinal study based on individual linkage with national health data

Abstract

Introduction: The potential health risks among veterans deployed to the Balkans have long been debated internationally. In Italy concern about cancer risk in veterans deployed to the Balkans in 1995-2004 (SEBAL cohort) originated from an initial study reporting an excess risk of Hodgkin's lymphoma. Subsequent studies have not fully dispelled these concerns due to methodological shortcomings and the lack of a national cancer registry. Our aim is to investigate long-term mortality and hospitalisation risk in SEBAL veterans for cancer and other major pathologies in comparison to the general population and to non-deployed military peers.

Methods: By linking with national archives of causes of death (1999-2018) and hospital discharges (2005-2018), long-term mortality (20 years) and hospitalisation (14 years) of the SEBAL cohort (71,144 soldiers) were evaluated, for the first time, for major disease groups and 27 cancer types. Standardized Mortality Ratios (SMR) and Standardized Hospitalisation Ratios (SHR) were used to compare health outcomes versus the Italian population (IP) and a military cohort never deployed abroad (114,260 Carabinieri).

Results: The SEBAL cohort showed a lower risk of death and hospitalisation than both control populations for all natural causes (SMR = 0.40 and SHR = 0.80 vs. IP; SMR = 0.72 and SHR = 0.80 vs. Carabinieri), infectious, endocrine, respiratory, circulatory, digestive and genitourinary diseases, with the only exception of a significantly higher hospitalisation risk for musculoskeletal diseases vs. IP (SHR = 1.27). Reduced risks of hospitalisation and death were found for all cancers (SHR = 0.82 and SMR = 0.56 vs. IP; SHR = 0.91 and SMR = 0.78 vs. Carabinieri), head and neck, respiratory and digestive cancers. Marginally increased hospitalisation risks vs. IP, with no effect on mortality, were observed for thyroid cancer (+39%) and skin melanoma (+32%). For all other solid cancers, hospitalisation and death risks were lower or not significantly different. We found no increased risk for hematological neoplasms, including Hodgkin's lymphoma.

Conclusion: Overall, the results of this observational study of unprecedented follow-up and completeness do not support specific long-term risks of mortality or hospitalisation linked to peacekeeping missions to the Balkans.

Keywords: Balkans; depleted uranium; environmental health; hospitalisation; mortality; neoplasms; occupational medicine; veterans.