The role of frequent HIV testing in diagnosing HIV in men who have sex with men.

OBJECTIVES: In the UK, quarterly HIV testing is recommended for high-risk men who have sex with men (MSM). In this manuscript we determined the risk of being newly diagnosed with HIV in MSM by their HIV testing history, considering both the frequency and periodicity of testing.

METHODS: Data on HIV incidence in MSM attending a sexual health clinic (SHC) in England in 2013-2014 with testing history (previous 2 years) were obtained from GUMCAD, the national sexually transmitted infection (STI) surveillance system in England. HIV testing patterns among MSM were defined using the frequency and periodicity of testing, based on 3 month intervals, in the year preceding the first attendance during the study period. Cox proportional hazards regression was used to determine the association between HIV testing pattern and time to HIV diagnosis with and without adjustment for demographic confounders. Analyses were stratified by risk stratum, with 'high risk' defined as a history of a bacterial STI in the past year. Adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs) are reported.

RESULTS: Among the 37 702 HIV-negative MSM attending an SHC in 2013-2014, 1105 (3%) were diagnosed with HIV infection within 1 year of their first attendance. The probability of HIV diagnosis was highest in MSM who were tested quarterly compared with those who were not tested in the past year (aHR 2.51; 95% CI 1.33-4.74); this increased 1.8-fold among high-risk MSM (aHR 4.48; 95% CI 0.97-21.17).

CONCLUSIONS: The probability of subsequent HIV diagnosis was greatest in high-risk MSM who were tested most frequently. Quarterly HIV testing increased the likelihood of identifying undiagnosed HIV infection and should remain a continued recommendation for high-risk MSM.