The feasibility of using an 'opt-out' approach to achieve universal HIV testing of tuberculosis patients in Alberta

Universal HIV testing of tuberculosis (TB) patients, defined as testing greater than 80% of incident cases, has been recommended but not achieved in Canada. The objectives of this study were: i) to assess the success of an 'opt-out' approach, whereby HIV testing is routine unless the patient specifically chooses otherwise, and ii) to determine the risk factors for HIV in patients tested before and after this approach was implemented.

TB and HIV databases in the province of Alberta were cross-matched before HAART (highly active anti-retroviral therapy) was available (1991-1997), after HAART but before 'opt-out' testing was implemented (1998-2002), and after 'opt-out' testing was implemented (2003-2006), and the HIV status of TB patients in each time period was described. The demographic and clinical characteristics of HIV-positive and -negative TB patients aged 15-64 years were compared.

HIV testing of TB patients increased from 11.5% before HAART, to 44.9% after HAART but before 'opt-out' testing, to 81.9% after 'opt-out' testing was implemented. Between 1991 and 2006, 50 TB patients were diagnosed with HIV co-infection, all in the age group 15-64 years. Among TB patients aged 15-64 years who were HIV tested, those testing positive were significantly less likely to be female and to have respiratory TB and significantly more likely to have both respiratory and non-respiratory TB. The prevalence of HIV positivity in HIV-tested TB patients aged 15-64 years was 7.4% in 2003-2006.

Universal HIV testing of TB patients is achievable through 'opt-out' HIV testing.