exchange policies. Some programs insist on one-for-one exchange, and others do not. Some programs limit the number of needles that may be exchanged at one time, and others do not. Each of these policies has pros and cons, but we know too little about them, in terms of what balance is optimal in maximizing gains, minimizing harm, and contributing to the ultimate goal of the program—prevention of HIV transmission.

Also, as Chapter 2 describes in detail, the aims of needle exchange and bleach distribution programs are broader than simply providing sterile needles to injection drug users. Other goals include linking users to needed health care and social services, providing drug abuse counseling, and facilitating entry into drug treatment. Research on how to improve these ancillary services is virtually nonexistent and should be pursued. The most effective ways to pursue these goals presumably depends in part on the characteristics of the participants, such as age, ethnicity, education, drug-using career, and socioeconomic status.

Identification of program characteristics that enhance or inhibit effectiveness is needed. The limited available research on the organizational characteristics of needle exchange programs and case studies presented in Chapter 3 hints at specific operational characteristics that may facilitate program effectiveness—for example, user-friendliness (see Stimson et al., 1988).

Evaluation research in this area of study would also benefit from adopting programmatic research strategies that acknowledge the need for feeding back information to program administrators to allow rational modification of procedures and operational characteristics. This type of iterative research process attempts to identify optimal procedures for maximizing the overall effectiveness of programs or identifying problematic components resistant to effectiveness modification. Quality assurance programs are witness to the value of systematic internal research with feedback as a method for improving the operations of almost any kind of organization.

In addition to examining the relative effects of different operational characteristics of programs (such as staffing, location, hours of operation, program policies) and various combinations of ancillary services on traditional outcomes (e.g., risk behaviors, infection rates)—more attention must be given to understanding how these characteristics may also impact the recruitment and retention of program participants.

The relationships between program characteristics and success at both the individual and the community levels should be examined. For example, it may be possible for a needle exchange program to be effective when limited to a small number of injection drug users, yet it may show no noticeable effect on either risk behaviors or HIV infection rates among the broader local population of injection drug users. By the same token, programs that reach larger populations may show a more modest effect among

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement