Previous studies suggest that most injection drug users (IDUs) become infected with hepatitis C virus (HCV) and hepatitis B virus (HBV) soon after initiating drug use. The Urban Health Study (UHS) recruited serial cross-sections of IDUs in the San Francisco Bay area from 1986 to 2005. In the current study, we determined the prevalence of antibody to HCV and HBV (core) among UHS participants during 1998 to 2000. To examine whether the time from onset of injection to acquisition of viral hepatitis has increased, we also compared the findings among recent (<10 years) initiates to drug use who participated during 1998-2000 with those who participated in 1987. Of 2,296 IDUs who participated during 1998-2000, 91.1% had antibody to HCV and 80.5% to HBV. The number of years a person had injected drugs strongly predicted infection with either virus (P(trend) < 0.0001). HCV seroprevalence among recent initiates in 1998-2000, by years of injection drug use, was: <or=2, 46.8%; 3 to 5, 72.4%; 6 to 9, 71.3%. By comparison, HCV seroprevalence among 1987 participants was: <or=2 years, 75.9%; 3 to 5, 85.7%; 6 to 9, 91.1% (P < 0.0001). A consistent pattern was observed for HBV (P < 0.0001), and these findings were not explained by demographic differences between 1987 and 1998-2000 participants. During 1987, however, 58.7% of recent initiates had shared syringes within the past 30 days compared with 33.6% during 1998-2000 (P < 0.0001).
Conclusion: HCV and HBV seroprevalence among newer initiates to injection drug use in the San Francisco Bay area decreased markedly between 1987 and 1998-2000. This decrease coincided with the implementation of prevention activities among this population.