HIV coinfection among persons diagnosed with hepatitis B in England in 2008-2014.

Abstract

The aim of the study was to estimate HIV prevalence among persons with hepatitis B virus (HBV) infection in England and to examine associated risk factors. Persons aged ≥15years with an HBV surface antigen (HBsAg) test reported to Public Health England (PHE) sentinel surveillance during 2008-2014 were linked to the PHE national HIV/AIDS database. Coinfection was defined as an HIV diagnosis prior to, or within 6months following, a positive HBsAg test. During 2008-2014, 2149933 persons were tested for HBsAg and 3.9% (1129 of 28789) of HBsAg-positive persons were HIV positive. The probable route of HIV infection was heterosexual exposure for 95.3% of female patients and 32.3% of male patients, with 61.5% of male patients reporting sex between men. Among African-born coinfected persons, 84% probably acquired HIV there. Predictors of HIV positivity included older age [adjusted odds ratio (aOR) 1.1] and being of black ethnicity (aOR 15.5 for males; aOR 16.4 for females) or being male and of white ethnicity (aOR 8.2) compared with being female and of white ethnicity. HIV coinfection was more likely when HBV was diagnosed in sexual health (aOR 55.0), specialist liver (aOR 6.7), emergency department (aOR 5.3) and renal services (aOR 2.8) compared with general practice. Most (60.4%; 682 of 1129) coinfected persons were diagnosed with HIV infection > 6months before HBV diagnosis. Persons testing positive for HBsAg had a low HIV infection rate and fell largely into two groups: those of black ethnicity with probable Africa-acquired infections and white men who have sex with men (MSM) with probable UK-acquired infections. Findings reinforce existing recommendations to sustain and improve both HBV testing of migrants from HBV-prevalent countries and vaccination among HIV-positive MSM. Findings also support blood-borne virus testing in sexual health services and emergency departments.