CAUSES OF DEATH IN CANADIANS WITH HEMOPHILIA 1980-1995

The life expectancy of individuals with haemophilia was close to that of the general population in the early 1980s. Since then, lift expecta ncy has decreased, due to transfusion-rransmitted virus infections. De aths in individuals with haemophilia were investigated by analysing 24 50 records from the Canadian Hemophilia Registry, for the years 1980-1 995. Deaths were tabulated by age, year and cause, and compared with t hat of the Canadian male population by calculating standardized mortal ity ratios (SMRs). The median life expectancy at I year of age was cal culated for various subpopulations and the impact of various populatio n characteristics was assessed by survival regression modelling. There were 359 deaths and the annual number of deaths increased significant ly after 1986. Risk factors were seropositivity to human immunodeficie ncy virus (relative risk 16.7, 95%CI 11.1-25.1), severe haemophilia (1 .9, 1.3-2.7) and moderate haemophilia (1.8, 1.2-2.6). In HIV antibody negative individuals, the overall death rate was not increased (SMR 0. 3, 95%CI 0.7-1.1) and only haemorrhage was significantly increased. In HIV antibody positive individuals, causes of death which were signifi cantly increased were acquired immunodeficiency syndrome, liver failur e, haemorrhage, lymphoma, liver cancer, nonspecific infections, and tr auma or violence. Deaths due to the acquired immunodeficiency syndrome accounted for only 66% of the excess deaths in individuals who were H IV antibody positive. Life expectancy has markedly decreased since the onset of the HIV epidemic. The impact of HIV is underestimated by con sidering only deaths due to the acquired immunodeficiency syndrome; ot her HIV-linked causes need also to be considered.