The economic burden of TB diagnosis and treatment in South Africa

Abstract

Social protection against the cost of illness is a central policy objective of Universal Health Coverage and the post-2015 Global strategy for Tuberculosis (TB). Understanding the economic burden associated with TB illness and care is key to identifying appropriate interventions towards achieving this target. The aims of this study were to identify points in patient pathways from start of TB symptoms to treatment completion where interventions could be targeted to reduce the economic impact on patients and households, and to identify those most vulnerable to these costs. Two cohorts of patients accessing TB services from ten clinics in four provinces in South Africa were surveyed between July 2012 and June 2013. One cohort of 351 people with suspected TB were interviewed at the point of receiving a TB diagnostic and followed up six months later. Another cohort of 168 patients on TB treatment, at the same ten facilities, was interviewed at two-months and five-months on treatment. Patients were asked about their health-seeking behaviour, associated costs, income loss, and coping strategies used. Patients incurred the greatest share of TB episode costs (41%) prior to starting treatment, with the largest portion of these costs being due to income loss. Poorer patients incurred higher direct costs during treatment than those who were less poor but only 5% of those interviewed were accessing cash-transfers during treatment. Indirect costs accounted for 52% of total episode cost. Despite free TB diagnosis and care in South Africa, patients incur substantial direct and indirect costs particularly prior to starting treatment. The poorest group of patients were incurring higher costs, with fewer resources to pay for it. Both the direct and indirect cost of illness should be taken into account when setting levels of financial protection and social support, to prevent TB illness from pushing the poor further into poverty.

Catastrophic costs of tuberculosis care: a mixed methods study from Puducherry, India

“…Another study from Chennai, India reported a mean total cost of US$48 [9]. Mean total costs for TB care in South Africa and Nigeria were US$207 and $157 respectively [15,16]. …”

Cost-effectiveness of Xpert MTB/RIF for tuberculosis diagnosis in South Africa: A real-world cost analysis and economic evaluation

“…We used two methods: applying mean per-event costs from our subsample and applying multiple imputation to assign per-event costs from our subsample of 351 to the total sample of 4656 XTEND participants. 21 The multiple imputation covariates included sex (binary), number of years at school (ordinal), self-reported HIV status (categorical), initial tuberculosis test result (binary), geographical location (binary: urban or rural), income (categorical), number of days with at least one symptom (continuous), anyone else in household with a regular job (binary), household socioeconomic status (categorical: derived through principal component analysis), number of health-care visits (continuous), country of birth (categorical), age (continuous), province (categorical), and distance from residence to clinic (ordinal). In view of the high number of zero observations in our subsample for some per-event costs, we applied a two-part model, which first predicts non-zero costs for each type of …”

“…We estimated the direct and indirect costs incurred by study participants while accessing care from the reported start of tuberculosis symptoms. 21,22 In summary, three samples of patients were interviewed. 21 We consecutively subsampled one in three of the patients treated at the ten XTEND study clinics selected for health service costing and succeeded in enrolling a cohort of 351 individuals being investigated for tuberculosis, who were interviewed at XTEND enrolment and then at followup 6 months later.…”

“…21,22 In summary, three samples of patients were interviewed. 21 We consecutively subsampled one in three of the patients treated at the ten XTEND study clinics selected for health service costing and succeeded in enrolling a cohort of 351 individuals being investigated for tuberculosis, who were interviewed at XTEND enrolment and then at followup 6 months later. We interviewed a further cohort of 168 patients on first-line tuberculosis treatment, who were not part of the XTEND cohort (to protect from interview overload), at 2 months and 5 months from start of treatment.…”

Health Insecurity and Social Protection: Pathways, Gaps, and Their Implications on Health Outcomes and Poverty

“…This paper has three main aims: first, to provide a debate on the potential of social protection contribution to addressing health insecurity, poverty, and vulnerability brought by healthcare expenditure in low-income countries. [19][20][21] Second, to explore the gaps in current and proposed social protection measures in healthcare, and third, to provide suggestions on how social protection intervention aimed at addressing health insecurity, poverty, and vulnerability may be effectively implemented. Unlike a standard and structured review paper, in this paper we adopted the approach in debate articles, where a convenience sample of the literature is reviewed and cited accordingly to inform the debate.…”