In 1998, we reported that anti-HCV prevalence among injectors from Glasgow had declined between 1990 and 1995. We set out to ascertain if the anti-HCV prevalence among injectors from Edinburgh had declined similarly during this period and if there had been any trend in prevalence among injectors from both cities since 1995. Residual sera from both cities' injecting drug users who had undergone named HIV testing were identified, linked to age band and gender information and tested anonymously for anti-HCV. Among Edinburgh's injectors, significant (p < 0.0001) decreases in anti-HCV prevalence from 69% (1989/90) to 13% (1997) and from 80% (1989/90) to 54% (1997) were seen in those aged < 25 y and > or = 25 y, respectively. Among Glasgow's injectors, a significant (p < 0.0001) decrease in prevalence from 91% (1990) to 43% (1997) was seen only among those aged < 25 y. Of both cities' 15-19 y olds, sampled during 1995-97, 17% (24/139) were anti-HCV-positive. The findings suggest that the incidence of HCV among young injectors continued to decrease during the 1990s--the era of needle/syringe exchange and other interventions--but is still too high. Further investigative and preventive work is required.