As South Africa's national antiretroviral therapy (ART) treatment program matures, it faces significant challenges in expanding access to harder to reach populations while also maintaining quality of care for those for whom HIV has become a manageable chronic disease. In light of these challenges, monitoring and evaluation of the treatment program is critical to its sustained improvement. However as policy makers face decisions on how to improve their program, key national parameters on which to base decisions remain unknown. The answers to these questions are critical for planning for changes to national treatment guidelines, yet to date there has been no routine way to answer these questions at a national level and without being heavily influenced by losses from care, which can be significant. We propose using South Africa's entire national public-sector laboratory monitoring database to create a national patient-level cohort. We will use this cohort to evaluate the national program while also building capacity in South Africa to analyze big data to answer important questions about the HIV treatment scale-up. The study is designed as a prospective cohort study of over 20 million laboratory observations and over 3 million patients designed to answer four questions: 1) What is the effect of the national HIV testing campaign on linkage to HIV care and treatment?; 2) What are national retention rates before and after ART initiation?; 3) What are the long run treatment outcomes for the national patient cohort?; and 4) What are the effects of immediate vs. deferred ART eligibility at a CD4 count of 350 on retention and clinical outcomes? To answer these questions we will use novel methods for causal inference, including regression discontinuity design and quantitative bias analysis, and will work to increase South African capacity for implementing these methods. This study is significant because South Africa has the largest treatment program in the world, and is investing 1.4 billion dollars annually, yet it is not clear what the benefits are on a national scale. Individual cohort studies that have tried to assess local impacts have all been clouded by loss to follow up from HIV care and transfer between sites. Our study is innovative because it will generate a unique dataset that will be able to evaluate the impact of the South African HIV treatment program on a national scale. It is also innovative because it will allow the first ever comprehensive look at migration between clinics within South Africa and the first evaluation of treatment impact without being affected by programmatic losses. Finally it is innovative because it will combine the data with innovative designs and analytic approaches that will allow us to assess causal effects.
The proposed study aims to address one of the biggest limitations in evaluating South Africa's HIV care and treatment programs: how to generate nationally representative estimates of retention and treatment outcomes that are not strongly influenced by patient loss from care. We will use data from the National Health Laboratory Service, which conducts all laboratory testing in South Africa's public-sector to create a patient level cohort describing the HIV treatment program since scale up began in 2004. This will allow us to evaluate the outcomes of the program (including viral suppression, CD4 change, retention, etc.) as well as the impact of policy changes (e.g. mass HIV testing, changing ART eligibility thresholds) in order to make recommendations to improve the program.
| Carmona, Sergio; Bor, Jacob; Nattey, Cornelius et al. (2018) Persistent High Burden of Advanced HIV Disease Among Patients Seeking Care in South Africa's National HIV Program: Data From a Nationwide Laboratory Cohort. Clin Infect Dis 66:S111-S117 |
| Fox, Matthew P; Bor, Jacob; Brennan, Alana T et al. (2018) Estimating retention in HIV care accounting for patient transfers: A national laboratory cohort study in South Africa. PLoS Med 15:e1002589 |
| Maskew, Mhairi; Bor, Jacob; Hendrickson, Cheryl et al. (2017) Imputing HIV treatment start dates from routine laboratory data in South Africa: a validation study. BMC Health Serv Res 17:41 |
| Clouse, Kate; Vermund, Sten H; Maskew, Mhairi et al. (2017) Mobility and Clinic Switching Among Postpartum Women Considered Lost to HIV Care in South Africa. J Acquir Immune Defic Syndr 74:383-389 |
