Ten-year survival analysis of a cohort of heroin addicts in Catalonia: The EMETYST project

Aims. To determine mortality rates and immediate causes of death in a cohort of heroin addicts, and to compare them with other European samples. Design. Longitudinal follow-up study of a cohort for 10.5 years (March/July 1985-December 1995). Setting. Catalonia, Spain. Participants. One hundred and thirty-five heroin addicts. Measurements. (a) Number of total and annual events; (b) annual mortality rate; (c) average annual mortality rate; and (d) standardized mortality ratio (SMR). Kaplan-Meier (log rank test) was used to assess the predictive factors. Findings. During this period, 41 heroin addicts died (30%), the average annual mortality rate was 3.4% and the SMR was 28.5. The most frequent causes of death fell in ICD-9 chapter III (which includes AIDS) (51%) and in chapter XVII (which includes overdose) (30%). Neither the socio-demographic characteristics nor the history of heroin consumption were predictors of survival or cause of death. Conclusions. Compared to other European studies, the cohort in the EMETYST project has the highest SMR and members have a higher chance of dying due to AIDS. The predictors of survival in the long term must be interpreted with caution, with the exceptions of being HIV positive or being diagnosed with AIDS.