1. incomplete, inadequate, or not relevant to the research question;

  2. unacceptable: contains flaws in design or reporting that make interpretation unreliable;

  3. acceptable: provides credible evidence but has limited detail, precision, or generalizability;

  4. well done: provides detailed, precise, and persuasive evidence; and

  5. excellent: compelling and complete.

The final ranking of an article was determined by agreement of at least two project members. Only studies ranked 3 or higher were used in the synthesis.

In addition to the review of existing research, the University of California team conducted site visits to 15 cities, 10 of which were in the United States, 3 in Canada, and 2 in Europe.2 The sites were selected on the basis of a published list of programs and reflected the range of existing needle exchange programs with respect to size, legal status, geographical location, injection drug users' HIV seroprevalence, and extent of prior evaluation research. (CDC was consulted during the selection process.) At each site, the research team used multiple data collection methods with multiple iterations, consisting of interviews, focus groups, and observation using a formal qualitative research strategy.

The methodology was codified in a manual. Standardized training of the research staff was provided. In the 15 cities, 33 needle exchange sites were visited and a total of 239 interviews with needle exchange directors and staff, public health officials, injection drug use researchers, community leaders, program participants (11 focus groups), and injection drug users not enrolled in programs (7 focus groups) were completed. Observation guidelines were pretested at two sites and the results were compared qualitatively for interrater reliability before adopting the final guidelines.


Of the nine outcomes and expectations for successful needle exchange programs listed in Table 7.1, the University of California report addressed eight. That is, research findings concerning four of the five possible positive outcome domains were reviewed: reduction in drug-related and sexual risk behaviors, increase in referrals to drug abuse treatment, and reduction in HIV and other infection rates. The report addressed all four possible negative outcomes: increases in (1) drug use by program participants; (2) new initiates to injection drug use; (3) drug use in the community in general; and (4) the number of contaminated needles discarded.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement