The impact of a partially effective HIV vaccine on a population of intravenous drug users in Bangkok, Thailand: a dynamic model.

Because of the variability of HIV, the first AIDS vaccine is likely to be only partially effective. There is some concern among scientists that a low-efficacy vaccine could worsen the HIV epidemic if vaccinated individuals increase their risk behavior under the false assumption of immunity. To address this concern, we constructed a dynamic compartmental model that simulated the course of the HIV/AIDS epidemic in a population of injection drug users in Bangkok, Thailand. The model calculated long-term HIV prevalence, number of AIDS cases, and total population size for two scenarios: vaccination program versus no vaccination program. We used sensitivity analyses to evaluate the impact of postvaccination risk behavior change on HIV prevalence. A 75% effective vaccine led to a 40-year HIV prevalence of 37% with vaccination and 50% without vaccination. Postvaccination behavior change had only a limited effect on the results with a 75% effective vaccine but a significant effect with a 30% effective vaccine. If 90% of low-risk individuals responded to a 30% effective vaccine with increased high-risk behavior, the benefit of vaccination disappeared. These results agree with analyses of the epidemic among gay men. If injection drug behavior is indeed modifiable, our findings have significant policy and planning implications.