Paper # 1001
Estimation of the Proportion of New HIV Cases among Men Who Have Sex with Men Transmitted from Persons in Different Stages of HIV Infection, Diagnosis, and Treatment
Matthew Golden*1,2, S Goodreau1, J Hughes1, J Dombrowski1, J Kent2, and J Stekler1,2
1Univ of Washington, Seattle, US and 2Publ Hlth-Seattle & King County, WA, US
Background: The likelihood that persons living with
HIV/AIDS (PLWHA) will transmit the virus varies based on their stage, awareness
of infection, and whether they are on antiretroviral therapy (ART). At a population level, the relative contributions of groups defined by these factors to overall
HIV transmission will vary based on the population’s testing frequency and ART use.
Methods: We developed a simple, cross-sectional
mathematical model of HIV transmission in MSM that includes ART use, mean time from infection to HIV diagnosis, relative measures of behavioral risk, and
transmissibility. Base case epidemiologic parameters were derived from
surveillance data from King County, WA. Biological parameters were taken from
the scientific literature. In our base case scenario, we assumed that PLWHA
decrease their behavioral risk 70% following diagnosis, that primary HIV is 8
times as transmissible as chronic infection, and that ART decreases
transmissibility by 90%.
Results: Based on King County data, we assumed that
the mean time from HIV infection to diagnosis is 15 months; HIV incidence and
prevalence are 1% and 15%, respectively; and 70% of diagnosed persons are on ART; 7% of HIV infections would be undiagnosed. In addition, 31% of infections would be transmitted
by persons with primary HIV; 26% by persons with chronic, undiagnosed HIV; 35%
by diagnosed persons off-ART, and 8% by diagnosed persons on ART. The proportion of infections attributable to chronic, undiagnosed infection increases with
longer periods from infection to diagnosis; increasing this parameter from 15
to 24 months increases the proportion of infection transmitted from persons
with undiagnosed chronic infection from 26% to 37%, and decreases the
proportion transmitted from diagnosed persons off ART from 35% to 29%. Decreasing
the proportion of diagnosed persons on ART from 70% to 50%, increases the
proportion of all infections transmitted by diagnosed persons off ART from 35% to 48%.
Conclusions: A majority of new HIV infections among MSM are probably transmitted by persons who are unaware of their HIV status, but approximately half
of these cases are likely attributable to primary HIV. Estimates are sensitive
to parameters that are not known with precision and probably vary by population.
Under plausible scenarios, a sizable proportion of infections are transmitted
from diagnosed persons off ART, infections that could be reduced through more
widespread treatment.
|