Mortality of Cohort of Very Young Injecting Drug Users in Prague, 1996-2010

Abstract

To determine the mortality in a cohort of very young injecting drug users (IDUs), and the factors associated with it. A database linkage prospective (follow-up) cohort study. A convenience sample of clients of 2 low-threshold facilities, 1 drug treatment clinic, and one special facility for children with severe behavioural disorders, who were all younger than 19 and older than 15, was interviewed one or more times in 1996-8 and asked to agree with their being interviewed again after 10 or more years. 151 (65 male, 86 female) IDUs recruited in October 1996 - December 1998. Database linkage study compared unique identifiers (IDs) of the recruited subjects with the general register of deaths to determine the life status, and the causes of death of those deceased. Where necessary, we examined the death protocols directly. Altogether, 8 deaths were registered between recruitment and 31st December 2008 (1,660 person-years). All the deceased were male, and all their deaths were "unnatural"--that is, caused by drug overdose or accident. This translates into the crude mortality rates for the whole cohort being 4.8 deaths per 1,000 person-years (PY), and into a specific mortality ratio in the males SMR=14.4 with the peak at the age of 15-20 (SMR=60.1), declining to SMR=8.2 at the age of 25-30. Except gender, we found no "predictors of death" in this high-risk cohort. The overall mortality in the cohort was substantially higher than in the general population; in the male part of the cohort of young injecting drug users it was excessively high in the first three years after recruitment, and caused by external causes exclusively; the mortality in the female sub-cohort was zero, i.e. lower than in the general population of the same age range. Our findings suggest a need to develop targeted prevention of overdoses and other unnatural deaths in young male drug injectors.