Trends in injection drug use behaviors over 10 years among street youth

Abstract

Objectives

To determine the temporal trends of initiation into injection drug use, current injection, and recent receptive sharing of injection paraphernalia among street youth.

Design

Data from two cohort studies conducted between 1995 and 2005 were combined.

Methods

Recruitment was done on an ongoing basis. Interviews were performed semi-annually. Overall and annual drug injection incidence rates were calculated with the person-time method. Poisson regression was used to assess the predictive power of calendar year on incidence rate. Generalized estimating equations (GEE) were used to assess linear trends in current injection among street youth and in recent receptive sharing of syringe and other injection paraphernalia among current injection drug users.

Results

By 31 March 2005, 1633 subjects had completed 8875 questionnaires. Most subjects were born in Canada (94%), their mean age at entry was 20 years, 68% were boys and almost half (44%) had injected drugs before recruitment. Among 778 never injectors at entry, 130 subjects initiated injection in 1898 person-years of follow-up (incidence rate: 6.8 per 100 person-years). When controlling for age, calendar year was not a significant predictor of incidence rate. Prevalence of current injection was stable (around 30%). Odds of both sharing behaviors decreased by approximately 10% per year.

Conclusions

In addition to injection paraphernalia sharing reduction efforts, interventions aimed at preventing initiation into injection drug use among high-risk youth are needed.

Introduction

All around the world, numerous uncontrolled epidemics of HIV and hepatitis C virus (HCV) infections are related to injection drug use (Roy et al., 2002, NIH CDCS, 2002, UNAIDS/WHO, 2005). Since the vast majority of initiations into drug injection occur during late adolescence and early adulthood (WHO ICG, 1994), and sharing of drug injection equipment is common among young new injection drug users (IDUs) (Kipke et al., 1996, Fennema et al., 1997, Kral et al., 2000, Mullen and Barry, 2001, Smyth et al., 2001, Thorpe et al., 2001, Roy et al., 2001, Montgomery et al., 2002, Novelli et al., 2005), these epidemics affect young people in particular.

Despite this situation, very little information is available on temporal trends of drug injection and sharing of injection paraphernalia in young vulnerable populations. In Amsterdam, recent injection was found to be more common among drug users aged 30 years old or less recruited into a cohort study between 1985 and 1989 than among those recruited into a cross-sectional study in 1998 (77% versus 56%) (Welp et al., 2002). The authors concluded that this trend could reflect a reduction in the rate of initiation into injection among young drug users in Amsterdam. However, they also acknowledged that the two samples were not fully comparable due to different recruitment strategies. Other analyses based on the Amsterdam cohort study showed no significant changes between 1986 and 1995 in the borrowing of used needles among IDU under 26 years of age entering the cohort (Fennema et al., 1996).

In North America, street youth form a population in which injection drug use is highly prevalent. According to various studies, between 30% (Clatts et al., 1998) and 40% (Kral et al., 1997, Pfeifer and Oliver, 1997, Gleghorn et al., 1998) of street youth in the United States have injected drugs at least once. In Canada, these proportions range from 17% (DeMatteo et al., 1999) to approximately 40% (Ochnio et al., 2001, Roy et al., 2006a). Canadian public health authorities are very concerned about these figures, particularly in major cities where injection drug use is a significant cause of HIV and HCV transmission (Patrick et al., 2001, Hankins et al., 2002). In Montréal, where the present studies were conducted, HIV and HCV infection incidence rates are estimated to be 4 per 100 person-years and 27 per 100 person-years respectively among IDUs attending needle exchange programs (Parent et al., 2006, Roy et al., 2006b).

Using data from two prospective cohort studies of street youth, we conducted analyses to determine the temporal trends of injection drug use in terms of initiation and current injection. We also examined temporal trends in recent receptive sharing of syringe and other injection paraphernalia among youth injecting drugs. The present analyses focus on the predictive effect of calendar year on each behavior.