average number of years of injection drug use. Using demographic data obtained from needle exchange program enrollees over the course of the program operation (November 1990 to December 1991), Heimer and colleagues (Heimer et al., 1993) show that the mean age and mean duration of injection did not change over time. Throughout the study period, male and female enrollees were, on average, approximately 33 to 34 years old. Male enrollees reported using injection drugs for about 10 years; female self-reports on duration showed more month-to-month variability (averaging about 5 to 10 years). From this set of indicators, it would appear that the needle exchange program did not increase the number of new initiates.
The authors also provided an additional observation that counters the argument that needle exchange programs encourage the initiation of injection drug use. They argued that, if the presence of a needle exchange program did enhance use, an increase in the number of new initiates to injection drug use would be most prominent following public disclosure of the first report on the effects of the program. But, at the time of the report's release (and publicity), there was "no increase in the percentage of enrolling clients with very short durations of intravenous drug use" (p. 219).
Participants requested drug treatment at a nearly constant rate (25 percent) throughout the study period. The percentage entering treatment increased from 15 percent (first 7 months) to 18 percent (end of 1991). Another noteworthy observation is that a substantial proportion of people who visit the program's van are not needle exchange participants but are visiting the van because they are seeking treatment for drug abuse (Heimer, 1994).
As we have noted in our critique of the model-based estimates of HIV incidence, the change in outcomes (especially the infectivity of needles) might possibly be due to changes in the population served. It is possible that the reduction in the infectivity of needles and syringes could be due to changes in the risk characteristics of participants. But little of the available evidence indicates such client population shifts (Kaplan and Heimer, 1994a). The mean age of enrollees and the mean duration of drug use did not change over the course of the program. The percentage of women remained constant at 20 percent over the course of the study (November 1990 to June 1992). Self-reported drug behaviors were not associated with enrollment date. Analyses of changes over time in frequency of injection, use of shooting galleries, injections shared, using cocaine, and risky injection frequency