Test of HIV incidence shows continuing HIV transmission in homosexual/bisexual men in England and Wales

It has been suggested that HIV incidence will decrease with the increased use of antiretroviral Therapy (ART) in HIV infected homosexual/bisexual men. HIV incidence was measured using a sensitive/less sensitive assay technique, at a time when combination ART was widespread. The Serological Testing Algorithm for Recent HIV Seroconversion (STARHS)13 technique was applied to syphilis test specimens collected from homosexual/bisexual men attending 15 sexually transmitted infections (STI) clinics which participated in an unlinked anonymous serosurvey of HIV infection during 1998. The HIV incidence rate was adjusted to compensate for patients who had a repeat syphilis test within the same year. Leftover syphilis test sera from 6202 men had been unlinked and anonymised, of which 415 were HIV positive. Sera from 412 (99.3%) patients were available. The STARHS assay showed 62 to have been recently infected with HIV (approximately in the last four months), giving an incidence of 3.33% per annum (95% CI: 2.06%-5.27%). The highest incidence was seen in those aged 35-44 years. About 46% of all HIV-infected homosexual/bisexual men were probably receiving combination ART at this time. If 10% of those on treatment were misclassified as recent infections the incidence would have been 2.58% per annum (95% CI: 1.53%-4.24%). In homosexual/bisexual men having syphilis tests at STI clinics in the UK during 1998 the incidence of HIV infection was between two and three per hundred per year. Treatment with combination ART of almost a half of homosexual/bisexual men who are HIV infected in the population is compatible with appreciable continuing HIV transmission among those at high behavioural risk. Public health surveillance systems for those at high risk for HIV infection should, as soon as possible, incorporate the STARHS methodology for monitoring recent HIV incidence.