Routine HIV testing in acute hospital admissions

UK HIV testing guidelines recommend routine HIV testing in hospital general medical admissions in areas of HIV prevalence exceeding 2/1000. However, there are limited data on the feasibility, acceptability and effectiveness of implementing such a programme. Brighton & Hove has an HIV prevalence of 7.96/1000 1. For 6 months from August 2009, physicians offered routine HIV testing to 16-79-year-old patients admitted by general medical teams. In parallel, we ran anonymous sero-prevalence HIV testing for all acute hospital admissions in Brighton's Royal Sussex County Hospital. In general medical admissions, 39.7% (1553 of 3913) were offered HIV tests, with an acceptance rate of 91% (1413 of 1553). A multivariate analysis showed that patients were statistically more likely to be offered an HIV test if they were younger than 60 years and had a clinical indicator disease. Offer rates differed significantly. Junior doctors and medical students were statistically more likely to offer HIV tests than middle grades or consultants (P < 0.001). Seventy-three of 6300 sero-prevalence samples were positive from all acute hospital admissions, of which 57 samples were from patients already known to be HIV positive. This gave an undiagnosed HIV prevalence of 2.5/1000. The rate of undiagnosed HIV infection was highest in acute orthopaedic admissions (7.3/1000). The acute physician HIV testing offer rate was low, with junior medical staff more likely to offer routine testing. Doctors also continued to target HIV testing even within a routine testing programme. Once offered, HIV test uptake was high, suggesting that health care workers are a barrier to HIV testing. The anonymous sero-prevalence HIV testing showed that, in contrast to the national testing guidelines, routine HIV testing should be offered in all acute admissions in areas of high HIV prevalence. We acknowledge the Brighton HIV Testing Steering Group. The authors wish to express their deep appreciation of the late Professor Martin Fisher who was chief investigator on this project. This work was supported by a grant from the Department of Health. Ethical approval was gained from the NHS National Research Ethics Service, reference H09/H1107/69.