STUDY OBJECTIVE--The aim was to quantify all cause mortality among injecting drug users. DESIGN--This was a retrospective analysis of 1989 data on injecting drug users and mortality obtained from three independent agencies: the Procurator Fiscal's Office, the General Register Office, and the Scottish HIV-test register. SETTING--Greater Glasgow, Scotland. SUBJECTS--Drug injectors, estimated population 9424. MAIN RESULTS--81 names were found using the three sources to identify deaths. After removing duplicates, 51 deaths were found. This represented a mortality rate of 0.54% in the estimated population. Among female injectors the mortality rate was 0.85%, significantly higher than the rate of 0.42% among male injectors (95% CI for the true difference in mortality rates between female and male injectors was 0.31%-0.55%). Over 90% of deaths were attributed to overdose or suicide. Although AIDS caused only one death, 19% of cases (5/27) whose HIV antibody status could be ascertained were positive. The mortality rate among HIV positive injectors (3.8%) was significantly higher than among HIV negative injectors (0.49%). CONCLUSIONS--Comprehensive coverage using three data sources revealed a far greater annual number of all cause deaths among injectors than would have been expected from previous research. The observed mortality rate was lower than in previous studies where the denominators used to calculate rates had an element of underenumeration. For the foreseeable future it is unlikely that AIDS will have much impact on mortality among injectors in Glasgow, because of the low prevalence of HIV infection among injectors in the city, and because HIV positive injectors are dying for reasons other than AIDS; rather, overdose and suicide will continue to be the main causes of death.