ART for people living with TB and HIV: practice still lags behind policy

The impact of HIV status and antiretroviral treatment on TB treatment outcomes of new tuberculosis patients attending co-located TB and ART services in South Africa: a retrospective cohort study.

Nglazi MD, Bekker LG, Wood R, Kaplan R. BMC Infect Dis. 2015 Nov 19;15(1):536. doi: 10.1186/s12879-015-1275-3.

Background: The implementation of collaborative TB-HIV services is challenging. We, therefore, assessed TB treatment outcomes in relation to HIV infection and antiretroviral therapy (ART) among TB patients attending a primary care service with co-located ART and TB clinics in Cape Town, South Africa.

Methods: In this retrospective cohort study, all new TB patients aged ≥ 15 years who registered and initiated TB treatment between 1 October 2009 and 30 June 2011 were identified from an electronic database. The effects of HIV-infection and ART on TB treatment outcomes were analysed using a multinomial logistic regression model, in which treatment success was the reference outcome.

Results: The 797 new TB patients included in the analysis were categorized as follows: HIV- negative, in 325 patients (40.8 %); HIV-positive on ART, in 339 patients (42.5 %) and HIV-positive not on ART, in 133 patients (16.7 %). Overall, bivariate analyses showed no significant difference in death and default rates between HIV-positive TB patients on ART and HIV-negative patients. Statistically significant higher mortality rates were found among HIV-positive patients not on ART compared to HIV-negative patients (unadjusted odds ratio (OR) 3.25; 95 % confidence interval (CI) 1.53-6.91). When multivariate analyses were conducted, the only significant difference between the patient categories on TB treatment outcomes was that HIV-positive TB patients not on ART had significantly higher mortality rates than HIV-negative patients (adjusted OR 4.12; 95 % CI 1.76-9.66). Among HIV-positive TB patients (n = 472), 28.2 % deemed eligible did not initiate ART in spite of the co-location of TB and ART services. When multivariate analyses were restricted to HIV-positive patients in the cohort, we found that being HIV-positive not on ART was associated with higher mortality (adjusted OR 7.12; 95 % CI 2.95-18.47) and higher default rates (adjusted OR 2.27; 95 % CI 1.15-4.47).

Conclusions: There was no significant difference in death and default rates between HIV-positive TB patients on ART and HIV negative TB patients. Despite the co-location of services 28.2% of 472 HIV-positive TB patients deemed eligible did not initiate ART. These patients had a significantly higher death and default rates.

Abstract [1] Full-text [free] access [2]

Editor’s notes: There is clear evidence that for people with TB and HIV, particularly individuals with low CD4+ cell counts (<350 cells/µL), being on antiretroviral therapy (ART) during TB treatment reduces the risk of mortality. However, practice still lags far behind policy in this area, as in 2013, globally, only around a third of known HIV-positive people with TB were treated with ART. This paper from a single health centre in South Africa highlights the impact of this treatment gap, and emphasizes the fact that co-location of TB and HIV services does not always translate to integrated patient-centred care.

The people included in this analysis were treated for TB between 2009 and 2011, which was before South Africa adopted guidelines recommending ART for all people with TB testing positive for HIV. Nevertheless, the majority of the people living with HIV had CD4+ cell counts that would have made them eligible for ART at the time of the study. Although overall outcomes were relatively good, one in six people starting TB treatment died or were lost to follow-up. Mortality among HIV-positive people not on ART was substantially higher than individuals on ART and people who were HIV-negative. One in four people who were ART-eligible did not start ART. It was not clear whether some did not start ART because they had already died or had been lost to follow-up. In this analysis, there was no differentiation between people already on ART at the time of starting TB treatment and people who started ART during TB treatment.

This study illustrates that co-location of HIV and TB services does not necessarily meet peoples’ needs if care remains fragmented. Care was provided by different people, and the HIV and TB programmes had separate organizational structures, as is still common. Workable models of integrated, patient-centred care for HIV and TB are necessary. Furthermore, to achieve targets of ending TB deaths, we still need a deeper understanding of why people die after starting TB treatment.  

Avoid TB deaths [4]
Africa [9]
South Africa [10]
  • [11]