Projections of the impact of circumcision on the HIV epidemic in sub-Saharan Africa based on clinical trials may underestimate the number of infections that can be averted by around 40%, according to an international group of epidemiological modellers.
The findings, published in advance online by the journal Sexually Transmitted Infections, come from new epidemiological modelling work that incorporates findings from a pooled analysis of two recent studies that evaluated the impact of circumcision on HIV transmission from men to women.
The epidemiological modellers, from Imperial College, London, Weill Cornell Medical College, New York, and Fred Hutchinson Cancer Research Center, Seattle, took two existing models of the impact of circumcision on HIV incidence and applied data from a pooled analysis of two recent studies.
The models projected HIV incidence in Zimbabwe and Kisumu, Kenya, using data from a number of locally relevant studies to inform the assumptions about sexual behaviour.
However, the original models lacked information about the rate of HIV transmission from circumcised men to women, and about the rate of HIV transmission during the period of wound healing after men were circumcised.
Neither model took into account the interaction between sexually transmitted infections (some of which increase the risk of HIV transmission) and circumcision (which may reduce the risk of men acquiring sexually transmitted infections).
In order to update the models, the epidemiologists took data from a pooled analysis of two studies which had each evaluated the rate of male-to-female transmission in circumcised men.
This pooled analysis found that from two years after the operation (when the effect begins to become apparent in trials and cohort studies) the rate of HIV transmission from circumcised men to women was reduced by 46%.
Assuming that only 50% of men remained uncircumcised after ten years and no men resume sex during the wound healing period, HIV incidence would be reduced by 20.5% in Zimbabwe, where no men had been circumcised before the intervention. In Kisumu, Kenya, where 25% of men were already circumcised, ensuring that half of all men are circumcised would lead to a 7.4% reduction in HIV incidence after ten years.
One concern about circumcision programmes is the potential for men to acquire or transmit HIV during the four to five week wound-healing period after the operation. The modelling found that even if all men remained sexually active throughout the wound healing period – the most pessimistic assumption possible – HIV incidence would still fall by 19% in Zimbabwe and 6.2% in Kisumu after a decade.
Adding in the information about the rate of male-to-female transmission resulted in a greater projected reduction in HIV incidence. Over 20 years, HIV incidence would fall by 28% in Zimbabwe and 16.8% in Kisumu, and specifically among women, it would fall by 23.7% in Zimbabwe and 13.9% in Kisumu after 20 years.
The reduction in incidence could be as great as 43% after 20 years in Zimbabwe, and a reduction of at least 23% could be expected. the researchers calculated.
Fewer circumcisions would be required to avert each infection – 28% fewer in Zimbabwe and 41% fewer in Kisumu – implying that circumcision could be more cost-effective than previously calculated.
The new figures show that assuming no change in sexual behaviour as a result of circumcision, previous projections underestimated the effect of circumcision on HIV incidence by at least 40% in Zimbabwe and by 79% in Kisumu.
The researchers say that “projections for the impact of circumcision interventions on population-level HIV may need to be dramatically revised: the impact of male circumcision implementation could be realised much sooner and with greater cost-efficiency than had previously been thought”.
References
Hallett TB, et al. Will circumcision provide even more protection from HIV to women and men? New estimates of the population impact of circumcision interventions. Sex Transm Infect, advance online publication, October 21 2010. (Link to full text article)