This study addresses the highest ranking health research priority in South Africa, which is, to develop and test optimal models of HIV-TB service delivery that will enhance retention, adherence and coverage of HIV-TB co-infected patients. HIV and TB are highest in sub-Saharan Africa, a region with limited health budgets, infrastructure, human resources, and suboptimal TB infection control practices. There is compelling clinical evidence suggesting that integrating HIV and TB services saves lives and presents an effective and efficient use of resources directed at optimizing health outcomes. Quality improvement (QI) methods are increasingly being used to systematically test and incorporate local ideas into strategies for reliable implementation and scale up. This trial is designed to test a practical, implementable and affordable strategy aimed at improving HIV-TB service integration to reduce TB and HIV associated deaths. This is a cluster randomized controlled trial, which evaluates and tests the effectiveness of implementing a QI model to integrate HIV-TB service delivery in primary health care clinics, on reducing morbidity and mortality in TB-HIV co-infected patients. This study will be conducted in 2 districts, Ugu and uThungulu, in KwaZulu-Natal, South Africa. The model of integrated care delivery for TB and HIV using the QI method offers a systems approach to care delivery to directly enhance treatment outcomes by enabling comprehensive effective care designed around the patients journey from entry to the clinic, through screening treatment initiation, treatment completion, and retention in care that is directed at the goals of cure for TB, effective sustainable HIV viral suppression and reduced HIV associated TB mortality as the main health impact. The scalability of the model, once proven effective, is the critical element that makes it increase population coverage of quality diagnosis and treatment of HIV-TB co-infection. QI methods promote front line staff engagement in identification and rapid testing of local implementation solutions to gaps in performance of processes of care along the steps of the patient journey. Gaps in care are identified through continuous feedback on a core set of indicators collected monthly as routine collection of data.

The primary aim of this study is to test the effectiveness of a peer mentor-led, quality-improvement model of service delivery of integrated HIV-TB treatment on mortality in HIV-TB co-infected patients treated in rural primary health care clinics in KwaZulu-Natal, South Africa

Specific Objectives (i) To determine the impact of a QI-mediated HIV-TB service integration on patient mortality. All patients that access services in intervention and control clinics, via either the TB entry point or via the HIV entry point will be tracked during clinic follow-up visits or, through a community care giver, and will have their vital status ascertained 12 months after clinic randomization.

(ii) To determine the effectiveness of peer-led Quality Improvement (QI) to integrate HIV-TB services. The effect, on HIV-TB integrated processes of care, of the deployment of a QI approach (systems view, data driven decision making, culture of continuous improvement, trained peer mentors) to ensure uniform implementation of an essential package of evidence based HIV-TB interventions that support HIV-TB integration. The impact on clinical outcomes of using QI methods to implement integrated HIV and TB management will be assessed using the following indicators: Time to ART initiation among HIV infected TB suspects and cases; HIV testing rates in TB patients; Number of HIV-TB co-infected patients receiving co-treatment for TB and HIV at the same facility; Number of patients infected with HIV or TB that are retained in care at 12 months; Indicators of treatment adherence such as - number of HIV patients that are virologically suppressed at 12 months and TB treatment outcomes; Hospitalisation rates among patients receiving co-treatment for TB and HIV.

(iii) To identify clinic-level factors that impact on integrated HIV-TB services. Understanding the context (environmental, social and political factors) in which we are working is essential to identifying factors that promote or inhibit the implementation of the intervention. We will use the COACH tool (Context Assessment for Community Health) [8] to collect data and assess the organizational context and the influence of factors such as organizational culture, leadership, resources and HCWs remuneration etc. on the intervention

(iv) To determine the cost-effectiveness of implementing HIV-TB services using Quality Improvement methodology (Intervention Clinics) versus the base-case of implementing HIV-TB services independently, through a within-trial approach using both health service (e.g. training, remuneration) and patient costs (e.g. travel, opportunity costs) as inputs. We will also calculate total intervention costs to assess its affordability and explore cost-effectiveness under various scenarios (e.g. different TB-HIV co-infection rates) using decision analytical modeling.

(v) To identify a set of interventions, change ideas, tools and approaches that can be used to scale up adoption, implementation and sustainability of integrated HIV-TB services across South Africa and in other resource constrained settings.

(vi) To strengthen the capacity of CAPRISA to independently perform implementation research in PRDs, including community-randomized trials and health economic analysis, through expert mentoring and supervision of PhD programmes.

Inclusion Criteria:

• South African Department of Health Primary Health Care Clinics
• ART site
• Supported by BroadReach Healthcare

Exclusion Criteria:

• Mobile clinics
• Clinics that do not offer ART
• Clinics with only 1 nurse
• Hospitals and Gateway clinics