HIV infection: studies from New Haven and Tacoma. These two sets of studies were selected on the basis of the wealth of published information available about the programs they analyze. That is, not only do they provide a sizable amount of information on various endpoints of interest (i.e., incident infection and risk behaviors), but they also have carefully addressed potential alternative explanations for their reported findings.
At the outset, it is important to recognize that the effects of needle exchange programs can be viewed from a number of different perspectives. Some of these perspectives involve outcomes relevant to improving the health status of injection drug users, and others reflect community-level concerns regarding potential negative effects that may be associated with the implementation of such programs. The following section identifies the outcome domains that are relevant to those distinct perspectives and are most germane to the panel's task of assessing the effects of needle exchange programs.
Needle exchanges are established in order to: (1) increase the availability of sterile injection equipment and (2) at the same time, remove contaminated needles from circulation among the program participants. Operation of the exchange, then, is expected to result in a supply of needles with reduced potential for infecting program participants with HIV and also to reduce sharing between individuals because of easier access to clean needles for any program participant. Typical exchanges also maintain such services as education concerning risk behaviors, referral to drug treatment programs (a step toward eliminating the route for all infection), and distribution of condoms. These measures offer independent prospects for reducing the spread of HIV. Appraisal of the success of a needle exchange program may involve measuring, for example, the numbers of needles exchanged; the cleanliness of circulating needles; the prevalence and incidence of HIV and other needle-borne diseases; referrals to drug treatment programs; enrollments in treatment programs; and changes in the risk behaviors of needle exchange participants. An observed pattern of favorable outcomes would reflect health benefits from the operation of the program.
The possibility of negative results from needle exchange program operations also demands attention. One possible negative outcome is an increase