|
We used a dynamic compartmental model that tracks HIV transmission and disease progression in a population of IDUs and non-IDUs (i.e., the general population). We modeled HIV transmission by considering injecting and sexual behavior, and adjusting for viral load and HAART status. Based on data from St. Petersburg, Russia, we assumed 4% of the adult population aged 15 to 49 was an IDU, and HIV prevalence among IDUs and non-IDUs was 40% and 0.7%, respectively. Currently, no HIV-infected IDUs and approximately 1% of infected non-IDUs receive HAART. We simulated the model over a 20-year time horizon and estimated the resulting HIV prevalence among both IDUs and non-IDUs.
If use of HAART did not expand beyond current practice, the HIV epidemic continued to grow dramatically, with prevalence of 78% among IDUs and 7% among the general population at the end of 20 years. However, if 100% of IDUs and 1% of non-IDUs received HAART, then HIV prevalence increased to 67% among IDUs and only to 2.7% among the general population, preventing 21,700 and 136,000 infections among these populations, respectively. In comparison, if no IDUs and 100% of non-IDUs received HAART, then HIV prevalence remained at 78% among IDUs (300 infections averted) but still reached 5.3% among non-IDUs (86,400 infections averted) at 20 years. Assuming 50% of IDUs and 1% of non-IDUs received HAART, HIV prevalence reached 73% and 4.2% among IDUs and non-IDUs, respectively. Our results were sensitive to the transmission reduction due to HAART, and to needle-sharing and sexual behavior.
Providing HAART only to IDUs has a greater impact on the HIV epidemic among the general population than does providing HAART only to non-IDUs. This finding holds even if all non-IDUs are adherent and only 50% of IDUs are adherent. Treating IDUs with HAART is critical for controlling the HIV epidemic in Russia.
See more of Oral Concurrent Session C - Public Health
See more of The 27th Annual Meeting of the Society for Medical Decision Making (October 21-24, 2005)