specifics types. All the respondents who used the needle exchange fell into the first two categories, while non-users fell into the latter two.
Examining the characteristics of NSE participants associated with health care and drug treatment (n=269) in Baltimore, Riley et al. (2002) found that 58 percent reported using primary health care in the previous 3 years. Being age 40 years or older, having health insurance, and exchanging more than seven syringes per visit were positively associated with use of primary health care.
Some studies have illustrated the range of unique health services provided with NSEs. For example, a study by Grau et al. (2002) described a wound and abscess clinic incorporated into an NSE in Oakland, California. In New York City, an NSE administered influenza and pneumococcal vaccines to IDUs (Stancliff et al., 2000); while in Baltimore an NSE provided tuberculosis services (Riley et al., 2002).
Pollack et al. (2002) examined whether a mobile NSE-based health care delivery system reduced the use of hospital emergency rooms by out-of-treatment IDUs in New Haven. Of 373 IDUs, 117 were NSE clients and 256 were not. After the system was implemented, use of the emergency room fell among clients and rose among non-clients.
Based on this evidence, the Committee concludes:
Conclusion 3-8: Few empirical studies have evaluated whether HIV prevention programs that include needle and syringe exchange effectively link IDUs to ancillary health and social services. The few studies examining this issue show moderate uptake of these services among NSE attendees. However, none of the studies had comparison or control groups, so the overall use of such services among drug users who do not use NSE is unknown.
Summary Conclusion: Moderate evidence from developed countries points to a beneficial effect of multi-component HIV prevention programs that include needle and syringe exchange on injection-related HIV risk behavior, such as self-reported needle sharing and frequency of injection. Modest evidence also points to decreasing trends in HIV prevalence in selected cities studied over time. Although many of the studies have design limitations, the consistency of these results across a large number of studies supports these conclusions.