The influence of needle exchange programs on injection risk behaviors and infection with hepatitis C virus among young injection drug users in select cities in the United States, 1994–2004

Abstract

Objective

Our purpose was to assess whether participation in needle exchange programs (NEPs) influenced incident hepatitis C virus (HCV) infection through effects on injection risk behaviors among young injection drug users (IDUs) in the United States.

Methods

Data were drawn from three multi-site studies carried out in four major cities that enrolled IDUs over the period 1994–2004. Bivariate and multivariate analyses were conducted to assess relationships among sociodemographic characteristics, NEP use, injection risk behaviors, and prevalent or incident HCV infection.

Results

Of the total participants (n = 4663), HCV seroprevalence was 37%; among those who initially tested negative and completed follow-up at three, six, or 12 months (n = 1288), 12% seroconverted. Nearly half of participants reported NEP (46%) use at baseline. Multivariate results showed no significant relationship between NEP use and HCV seroconversion. Controlling for sociodemographic characteristics, IDUs reporting NEP use were significantly less likely to share needles (aOR = 0.77, 95% CI = 0.67–0.88). Additionally, controlling for sociodemographic characteristics and program use, sharing needles, sharing other injection paraphernalia, longer injection duration, and injecting daily were all positively related to prevalent infection.

Conclusions

Our results suggest an indirect protective effect of NEP use on HCV infection by reducing risk behavior.

Introduction

Hepatitis C virus (HCV) is one of five recognized hepatitis viruses worldwide and one of three (including hepatitis A and B viruses) most common in the United States. Transmission occurs through contact with infected blood and may cause acute as well as chronic illness, which in turn can result in more serious health problems (NIH, 2002). Currently in the United States, infection with HCV is the leading cause of liver disease and the leading cause of all liver transplants (Rustgi, 2007).

The incidence of acute HCV infection in the United States has shown a steady decline since its all-time high in the late 1980s (Wasley et al., 2008). While this decline is not fully understood, researchers speculate that it began with targeted risk reductions efforts for the prevention of HIV/AIDS and accelerated after an HCV-antibody test, developed in 1990, virtually eliminated transmission through receipt of blood or blood products. Despite the decline since 1992, a small increase in the incidence of acute infection was observed from 2005 to 2006 (Wasley et al., 2008). Moreover, current estimates (among the civilian, non-institutionalized adult population) indicate that about 3.2 million persons in the United States are chronically infected with HCV (Armstrong et al., 2006).

Because the virus is most efficiently transmitted parenterally by contaminated needle or syringe, persons who inject drugs have been the focus of considerable study. In both the United States and other countries, sharing needles (Thomas et al., 1995, Hahn et al., 2001, Hahn et al., 2002, Todd et al., 2007), not using clean needles (Garfein et al., 1998, Kapadia et al., 2002), sharing injection equipment (other than needles/syringes) (Hagan et al., 2001, Diaz et al., 2001, Thorpe et al., 2002), daily injection (Thomas et al., 1995, Hahn et al., 2001, Van den Hoek et al., 1990), cocaine injection (Thomas et al., 1995, Garfein et al., 1998, Diaz et al., 2001), crack-cocaine injection (Santibanez et al., 2005), injecting with others rather than alone (Hagan et al., 2007), and greater number of years injecting (Hahn et al., 2001, Todd et al., 2007, Garfein et al., 1998, Diaz et al., 2001) have all been shown to be associated with prevalent or incident HCV infection.

In addition to injection risk behaviors, investigations also have examined relationships between participation in programs such as drug treatment or needle exchange and prevalent or incident HCV infection. These studies have produced varying results. While some have shown that not having used a drug treatment program (Rezza et al., 1996) and non-use of needle/syringe exchange programs (Hagan et al., 1995, Estrada, 2002, Burt et al., 2007) are associated with increased risk of HCV infection, some studies have found otherwise (Crofts et al., 1997, Selvey et al., 1997, Hagan et al., 1999a, Thiede et al., 2000, Diaz et al., 2001).

However, it is not participation in such programs per se that may prevent infection, but rather their influence on reducing specific risk behaviors. Indeed, several studies have shown that participants who have used drug treatment or needle exchange programs (NEPs) were less likely to engage in behaviors such as sharing needles (Hart et al., 1989, Watters et al., 1994, Monterroso et al., 2000, Bluthenthal et al., 2000, Huo and Ouellet, 2007), injecting with used needles (Hagan and Thiede, 2000, Thiede et al., 2000, Huo and Ouellet, 2007), injecting more frequently (Thiede et al., 2000), daily injection (Vlahov et al., 1997), sharing drugs mixed in a single syringe, sharing injection paraphernalia (Vlahov et al., 1997, Thiede et al., 2000, Bailey et al., 2003), and distributive syringe sharing (Golub et al., 2007, Huo and Ouellet, 2007). Yet, there are also studies that have shown no effects of program use on injection risk behaviors, as well as some that have shown mixed results (Hahn et al., 1997, Monterroso et al., 2000, Hagan and Thiede, 2000, Burt et al., 2007).

So are these programs effective in preventing infection with HCV among injection drug users (IDUs)? Many previous investigations of these issues have been limited in terms of study characteristics such as sample size (small numbers), study design (geographically limited), or relevant measures (incidence estimates). In the current study, we were able to examine program effectiveness, in particular NEP use, by combining data from three longitudinal cohort studies of IDUs carried out in four major U.S. cities that employed similar protocols over a ten-year period. We therefore had data from a large and geographically diverse sample of IDUs, in addition to a substantial subset for whom we could assess incident infection, and among whom we could test our hypotheses.

We expect the effect of participation in NEPs to operate indirectly on incident HCV infection, that is, through their influence on injection risk behaviors, and expect no independent effect of NEPs on incident infection when injection risk behaviors are taken into account. However, in a conservative test of the hypothesis, our analytic approach allows for an independent program effect beyond that which may be exerted by these measured risk behaviors. Thus, we hypothesize that IDUs who have participated in an NEP would be less likely to engage in injection risk behaviors and those who are less likely to engage in injection risk behaviors would be less likely to have or acquire HCV infection.