Objective: To analyse overall and cause-specific hazards of death in a cohort of heroin users, separately by gender, survival and other risk factors.
Design and setting: Longitudinal study of intravenous heroin users; subjects were enrolled between 1975 and 1999 in public health services of the Provinces of Modena and Ferrara and were included in a treatment program.
Main outcome measures: For each gender, age-standardized mortality rates and standardized mortality ratios (SMR) for all causes and for specific causes. Kaplan-Meier method was used for estimating survival probability and Cox regression model to estimate hazard ratios (HR) of death.
Results: in the cohort of 4.644 intravenous drug users, 801 deaths were observed. In both sexes, mortality due to AIDS was lower in subjects enrolled in 1990-99 than among those enrolled in 1980-89. Mortality caused by overdose was particularly high in males enrolled between 1995-99. (SMRs in males and females were respectively 12.12 (95% CI 11.22-13.08) and 20.26 (17.23-23.83). Survival probability at 20 years of observation was 62% (60% for males and 68% for females). Risk of death was highest in males, in subjects enrolled after age 25, in subjects with a low educational level and in unemployed persons.
Conclusion: Gender and socioeconomic conditions are important determinants of mortality among heroin users. The increase in deaths from heroin overdose in subjects enrolled in the recent years requires particular attention.