Discriminating needle exchange attenders from non-attenders.

In 1990, approximately 1500 of Glasgow's estimated 9400 injecting drug users attended the city's four needle and syringe exchanges. In the same year, a multisite and citywide strategy was used to interview an estimated 5% sample of the city's injectors using an internationally agreed WHO structured schedule. Of the 503 injectors who participated in the study 54% (n = 272) had made use of exchanges in the last 6 months (including 165 injectors recruited from exchanges), while the remaining 46% (n = 231) had not. Twenty-seven variables were entered into a stepwise discriminant analysis to predict group membership (attenders vs. non-attenders). Fourteen variables emerged in the final discriminant function, which explained 17.3% of the variance between the two groups. Comparison of the model to random classification of cases resulted in a third fewer cases being wrongly assigned. Needle exchange attenders report injecting fewer drugs, less injecting with and passing on of used equipment and greater use of condoms with casual partners, compared to non-attenders. They also had greater knowledge of AIDS and had made and maintained more harm reduction changes, although they were less likely to have received treatment than non-attenders. The finding that attenders were less likely to inject with or pass on used needles and syringes is discussed in the context of availability of clean equipment and perception of risk. Variables reflecting lifestyle such as housing, employment and prison experience were not found to have any discriminatory power.