Mortality risk among new onset injection drug users


David Vlahov New York Academy of Medicine 1216 Fifth Avenue New York, NY 10029 USA E-mail:


Aims  To characterize mortality experience among those who only recently started injection.

Design  Prospective study.

Setting  Independent study clinic within high drug use neighborhoods.

Participants  In 1988–1989, we enrolled 256 adult injection drug users (IDUs) recruited through street outreach who had initiated injection within the prior 2 years.

Measurements  Consenting participants underwent venipuncture for HIV antibody testing and interviews. We prospectively ascertained date and cause of death through follow-up contact and registry linkages. Analyses included standardized mortality ratios (SMRs) with local, state and national mortality data, adjusted for age, gender and race.

Findings  Baseline median age was 30 years, 70% were male, 95% were African-American and 90% injected within the prior 6 months. We identified 69 deaths through October 2000; mortality rate was 3.3/100 person-years. The adjusted SMR with the USA (and Baltimore) as the reference for IDUs was 4.40 (2.43) for 1991–1992, which increased to 8.12 (4.13) by 1993–1994, decreased to 4.43 (2.13) by 1997–1998 and increased slightly to 5.35 (2.79) during 1999–2000. Excluding HIV-related mortality, SMRs remained elevated. Decline in SMRs was not linked to drug abuse treatment.

Conclusions  These data demonstrate excess mortality among new-onset IDUs compared with demographically similar peers in the general population, indicating the need for interventions to prevent premature death among young IDUs.