Objective: To estimate the effect of HIV-1 infection on subsequent mortality in a complete population.
Design: Prospective cohort study.
Subjects: A total of 7250 haemophilic males were registered in the UK Haemophilia Centre Doctors' Organisation database, 1977-1998. Most were infected with hepatitis C virus. In the early 1980s, 1246 were infected with HIV-1 from contaminated clotting factor concentrate. The main outcome measure was the date of death.
Results: During 1977-1984 annual mortality in severely haemophilic males was 0.9%. For those with HIV, annual mortality increased progressively from 1985 reaching over 10% during 1993-1996 before falling to 5% in 1997-1999, whereas without HIV it remained approximately 0.9% throughout 1985-1999. For moderately/mildly haemophilic males the annual mortality was 0.4% during 1977-1984. Without HIV it remained approximately 0.4% throughout 1985-1999, but with HIV it was similar to that in severe haemophilia with HIV. Survival was strongly related to age at HIV infection. The large temporal changes in mortality with HIV were largely accounted for by HIV-related conditions. Without HIV annual liver disease mortality remained below 0.2% throughout 1985-1999, but with HIV it was 0.2% during 1985-1990, 0.8% during 1991-1996, and 0.8% during 1997-1999.
Conclusion: These data provide a direct estimate of the effect of HIV-1 infection on subsequent mortality in a population with a high prevalence of hepatitis C. From approximately 3 years after HIV infection, large, progressive increases in mortality were seen. From 1997, after the introduction of effective treatment, substantial reductions occurred, although mortality from liver disease remained high.