Abstract

Aim: Universal testing for HIV in tuberculosis (TB) patients has been advocated for over a decade. Our aim was to describe HIV prevalence and testing practices among tuberculosis centres in London.

Methods: A cohort study was undertaken of all TB patients in Greater London in 2003 to 2004 (n=1941). Logistic regression was used to assess factors affecting being offered and accepting testing and having a positive HIV result.

Results: The overall known prevalence of HIV was 9.9% (193/1941). In those with a test result (including those diagnosed previously) it was 25.6%. Overall 50.8% of patients ¡Ý20 yrs without previous testing were offered HIV testing and of these 73% accepted. In multivariable analysis, factors associated with being HIV positive were aged 20-49 years, of black ethnicity and born overseas. Those with smear negative disease and with a poor understanding of English were significantly less likely to be offered HIV testing. Factors associated with refusal of an offered test were; being female or aged >49 years. HIV status was not associated with smear status, drug resistance or death, but was associated with CNS disease (OR 1.8, 95% CI: 1.0-3.0, p=0.003).

Conclusions: Nearly half of TB patients in London in 2003/04 were not offered HIV testing. In those offered testing, uptake was high. Patients in higher risk groups were more likely to be offered testing, but even within the highest-risk groups testing was not universally offered. This represents a missed opportunity for diagnosing HIV in TB patients in London.