OBJECTIVE: To describe the impact of HIV on mortality in men and women aged 15-54 in London. DESIGN: Combination of routine mortality statistics with reports of AIDS deaths adjusted for underreporting and change in address from time of report to time of death. Calculation of standardised mortality ratios (SMRs) for males including and excluding HIV comparing inner London and outer London with the rest of England and Wales. METHODS: Comparison of trends in all cause mortality and SMRs in males over time. Comparison of trends in HIV related deaths with other main causes of deaths in males and females in London. RESULTS: Age standardised rates for the rest of England and Wales showed a continual decline from 1979 to 1996 but rates in inner London males (ages 15-54) stopped declining around 1984-5 leading to a considerable increase in the SMR for inner London from 127 for 1985-7 to 171 for 1994-6. SMRs excluding HIV related deaths for inner London, however, showed no significant change over this time. There was a fall in HIV related mortality in 1996, though HIV was still the leading cause of death in males and second leading cause of death in females in inner London, and the fourth commonest cause of death in males in outer London. CONCLUSION: These data are the first to indicate the impact of HIV on mortality within a significant population in England and Wales. They show that public health priorities in London are different from the rest of the country. Analyses of trends of all cause mortality in people under 65 may mislead unless they take account of HIV.