Background Tuberculosis (TB)-associated mortality in South Africa remains high. This review aimed to systematically assess risk factors associated with death during TB treatment in South African patients. Methods We conducted a systematic review of TB research articles published between 2010 and 2018. We searched BioMed Central (BMC), PubMed®, EBSCOhost, Cochrane, and SCOPUS for publications between January 2010 and December 2018. Searches were conducted between August 2019 and October 2019. We included randomised control trials (RCTs), case control, cross sectional, retrospective, and prospective cohort studies where TB mortality was a primary endpoint and effect measure estimates were provided for risk factors for TB mortality during TB treatment. Due to heterogeneity in effect measures and risk factors evaluated, a formal meta-analysis of risk factors for TB mortality was not appropriate. A random effects meta-analysis was used to estimate case fatality ratios (CFRs) for all studies and for specific subgroups so that these could be compared. Quality assessments were performed using the Newcastle-Ottawa scale or the Cochrane Risk of Bias Tool. Results We identified 1995 titles for screening, 24 publications met our inclusion criteria (one cross-sectional study, 2 RCTs, and 21 cohort studies). Twenty-two studies reported on adults (n = 12561) and two were restricted to children < 15 years of age (n = 696). The CFR estimated for all studies was 26.4% (CI 18.1–34.7, n = 13257 ); 37.5% (CI 24.8-50.3, n = 5149) for drug-resistant (DR) TB; 12.5% (CI 1.1–23.9, n = 1935) for drug-susceptible (DS) TB; 15.6% (CI 8.1–23.2, n = 6173) for studies in which drug susceptibility was mixed or not specified; 21.3% (CI 15.3-27.3, n = 7375) for people living with HIV/AIDS (PLHIV); 19.2% (CI 7.7–30.7, n = 1691) in HIV-negative TB patients; and 6.8% (CI 4.9–8.7, n = 696) in paediatric studies. The main risk factors associated with TB mortality were HIV infection, prior TB treatment, DR-TB, and lower body weight at TB diagnosis. Conclusions In South Africa, overall mortality during TB treatment remains high, people with DR-TB have an elevated risk of mortality during TB treatment and interventions to mitigate high mortality are needed. In addition, better prospective data on TB mortality are needed, especially amongst vulnerable sub-populations including young children, adolescents, pregnant women, and people with co-morbidities other than HIV. Limitations included a lack of prospective studies and RCTs and a high degree of heterogeneity in risk factors and comparator variables. Systematic review registration The systematic review protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42018108622. This study was funded by the Bill and Melinda Gates Foundation (Investment ID OPP1173131) via the South African TB Think Tank.

Background In the high disease burden and resource-constrained contexts of SSA, health workers experience a range of psychosocial stressors that leave them vulnerable to developing burnout, which can reduce service quality and negatively impact their own health and wellbeing. As universal testing and treatment (UTT) for HIV scales up across SSA, it is important to understand the implications of this human resource intensive approach to HIV prevention to inform decision making about health workforce staffing and support needs. MethodsUsing the Maslach Burnout Inventory, we assessed the prevalence of three domains of burnout—emotional exhaustion, depersonalization, and personal accomplishment— among three cadres of health workers delivering health services in areas receiving a UTT intervention in Zambia and South Africa. These cadres included health facility workers (n=478), community health workers (n=159), and a study-specific cadre of community HIV care providers (CHiPs) (n=529). We used linear regression to assess risk factors associated with emotional exhaustion, the only domain with sufficient variation in our sample.ResultsAll groups of health workers reported lower levels of emotional exhaustion than found in previous studies of this type (median scores ranged from 8 to 14 out of 54 across health cadres). Higher emotional exhaustion was associated with higher educational attainment (βadj= 2.24, 95% CI 0.76 to 3.72), greater years providing HIV services (βadj=0.20, 95% CI 0.03 to 0.36), and testing negative for HIV at last HIV test (βadj=-3.88 -95% CI 5.69 to -2.07). Working as a CHW was significantly associated with lower emotional exhaustion (β = -2.52, 95% CI -4.69 to -0.35). Amongst all health workers, irrespective of HIV status, witnessing stigmatizing behaviors towards people living with HIV among their co-workers was associated with significantly increased emotional exhaustion (βadj = 3.38, 95% CI 1.99 to 4.76). ConclusionsThe low levels of emotional exhaustion detected among health workers are reassuring. However, it remains important to assess how UTT may affect levels of emotional exhaustion among health workers, which may impact the quality of HIV services they are able to provide. Interventions to reduce HIV stigma in health facilities may protect against emotional exhaustion among health workers. ClinicalTrials.gov number: NCT01900977.