A decline in HIV-infected needles returned to New Haven's needle exchange program: client shift or needle exchange?

Abstract
The New Haven needle exchange program experienced a significant decline in the fraction of returned needles containing human immunodeficiency virus 1 (HIV-1) proviral DNA. Is this decline due to the operations of the needle exchange or to a shift in clients? Analysis of demographic and behavioral data revealed that only one variable, the race of participating clients, changed significantly over time. However, HIV-1 prevalences in needles given to Whites and to non-Whites were not statistically different. Thus, client shift cannot be responsible for the decline in the observed HIV prevalence in needles. Instead, needle circulation times were a significant predictor of HIV prevalence.