This is one of the largest studies to examine the effects of HIV and ART status on TB treatment outcomes among new smear-positive PTB patients in sub-Saharan Africa. In our study, overall treatment success was 86%, indicating an overall positive programme outcome. Fifty six percent of smear-positive PTB patients were HIV co-infected and HIV co-infection was associated with a slightly poorer TB treatment outcome, even after adjusting for gender, age and year of TB registration. Only 38% of the TB/HIV new smear-positive co-infected patients were on ART. Those on ART had successful TB treatment outcomes compared to those not on ART.
Similar to other studies 
, we found a high proportion of smear-positive PTB among individuals aged 15–44 years; the same ages are at highest risk of HIV in Malawi 
. However, HIV prevalence among our smear-positive PTB patients (56%) was lower than the national HIV prevalence of 68% among all TB forms 
. This finding is likely because the national estimation of HIV prevalence in TB-infected individuals includes those with smear-negative PTB, which is quite common among HIV-infected individuals 
As expected, both HIV and ART status influenced TB treatment outcomes. In general, HIV-infected individuals were less likely to have successful treatment outcomes and were twice as likely to die as compared to HIV-negative individuals. Among HIV-infected smear-positive PTB adults, those on ART had slightly higher likelihood of successful treatment outcomes compared to those not on ART. ART enhances the immune response among HIV-infected TB patients, and this consequently leads to improved survival 
. More dramatically, we also found a 50% reduction in mortality among TB patients on ART compared to those not on ART, an indication of the positive effect of TB/ART co-infection treatment, which echoes similar findings from previous studies 
Despite the well-known benefit of ART and the integrated TB/HIV care model at MPC, only 38% of HIV-infected new smear-positive individuals received ART during TB treatment. The low ART uptake is likely due to several factors. First, some patients might be reluctant to take ART and TB drugs simultaneously due to concerns about pill burden and drug interactions. Second, the national ART guidelines recommend a guardian before ART initiation; the absence of a guardian may delay or discourage ART uptake among TB patients. Lastly, smear-positive PTB patients are expected to be reviewed by TB clinical officers – a cadre of mid-level healthcare professionals trained in both TB and ART management. However, in practice, health surveillance assistants (the lowest level of healthcare professionals) who are not trained in ART provision often conduct TB reviews for HIV-infected individuals, sometimes overlooking the need for ART initiation 
. Currently shifting management of smear-positive PTB patients from health surveillance assistants to TB clinical officers is under discussion within the Malawi NTP with the aim of improving routine programme management.
Our results should be viewed in light of the following limitations. TB treatment outcomes were not available for 5% of eligible patients and their exclusion might have introduced bias. However, there were no significant differences in terms of age, gender and HIV status between patients with and without TB treatment outcomes. Second, we used only routine programme data; these do not include information such as viral load and CD4 cell count which may have significant effects on patients’ TB outcomes. Third, since 165 (34%) of HIV-infected individuals on ART did not have a recorded ART start date, we could not explore the effect of duration on ART or timing of ART uptake on TB outcomes. Previous studies found decreased treatment success for sequential ART initiation 
. Despite these limitations, the study findings are useful to inform policy and programmes that aim to improve management of new smear-positive TB/HIV patients in Malawi and other comparable settings.
The results of this study suggest several changes that would likely improve TB outcomes and reduce the duration of infectiousness especially among TB/HIV co-infected patients in an an integrated TB/HIV clinic setting. First, although our study found satisfactory treatment outcomes for smear-positive PTB patients, this rate decreased from 90% in 2008 to 84% in 2009 and to 86% in 2010, possibly due to changing LTFU rates. In an effort to address LTFU, the Malawi HIV programme recommends active tracing of TB/HIV patients in facilities providing HIV services, a step that MPC will consider. Second, appropriate and intensified sensitization on the benefits of ART for both patients and care providers might increase ART uptake. Lastly, improved monitoring and evaluation of TB/HIV co-infected patient care, especially as provided by health surveillance assistants, would better identify and address obstacles in ART initiation during follow-up visits.