151
Despite HAART, HIV-1 Is Once again Spreading Epidemically among Men Having Sex with Men in the Netherlands
Daniela Bezemer*1, F de Wolf1,2, F de Wolf1,2, M Boerlijst3, A van Sighem1, D Hollingsworth2, M Prins4,5, M Prins4,5, R Geskus4,6, R Geskus4,6, L Gras1, R Coutinho6,7, R Coutinho6,7, and C Fraser2
1HIV Monitoring Fndn, Amsterdam, The Netherlands; 2Imperial Coll London, UK; 3Inst for Biodiversity and Ecosystem Dynamics, Univ of Amsterdam, The Netherlands; 4Hlth Svc Amsterdam, The Netherlands; 5Ctr for Infection and Immunity Amsterdam, Academic Med Ctr, Univ of Amsterdam, The Netherlands; 6Academic Med Ctr, Univ of Amsterdam, The Netherlands; and 7Ctr for Infectious Disease Control, Natl Inst of Publ Hlth and the Environment, Bitlhoven, The Netherlands
Background: Although HAART limits onward HIV transmission from
treated patients, the annual number of new HIV-1 diagnoses among men having sex
with men (MSM) in the Netherlands
has increased by 37% since the introduction of HAART in 1996. We analyzed the transmission
dynamics of HIV-1 among MSM over the past 25 years to achieve insight in the effect
of HAART on transmission.
Methods: The reproduction number R(t) defined as the average number of MSM infected by each HIV-1+
MSM over their infectious lifespan, was calculated using a mathematical model. This
measure summarizes the effect of both public health interventions and risk
behavior on transmission, and epidemic spread arises when R(t) is above the threshold 1. Data on HIV-1 disease progression,
diagnosis, and use of HAART were obtained from the Netherlands national observational
HIV cohort ATHENA and the Amsterdam Cohort Studies. By fitting the model to the
observed annual number of new HIV-1 and AIDS diagnoses, risk behaviour and
diagnosis parameters were estimated for 4 distinct historical periods. The risk
behavior parameter indicates the extent of unsafe sex practices between
infectious HIV+ and HIV– MSM in each period relative to
the initial period.
Results: R(t) was estimated to be 1.7
during 1980-1984, the period when the first AIDS cases were diagnosed.
Thereafter between 1984-1995, the risk behavior parameter declined 40%, and R(t) was reduced to 0.9, ie, just below
the epidemic threshold. With the introduction of HAART in 1995, R(t) declined to 0.8, despite a 17%
increase in the risk behaviour parameter. However, for the period 2000-2004 R(t) increased to 1.1, due to further
increase in risk behaviour of MSM. This return of the HIV-1 epidemic is despite
reductions in the estimated mean time from infection to diagnosis (from 3.5
years in 1984-1995 to 2.7 years in 2000-2004) and widespread treatment with
HAART. We estimated 70 to 93% of HIV-1 transmission originating from MSM who
are unaware of their HIV+ status. Scenario analysis revealed that in
the absence of HAART, R(t) would have
been 1.5 in the 2000-2004 period. If, on the other hand, HAART had been
introduced without increases in the risk behaviour parameter, R(t) would have been 0.6 in 2000-2004.
Conclusions: The effect of widespread use of HAART on overall
transmission of HIV-1 was offset by an increase in risk behavior. Consequently,
HIV-1 is once again spreading epidemically among MSM in the Netherlands.
|