South Africa has one of the highest incidences of tuberculosis (TB) in the world. Emergence of
drug-resistant tuberculosis (DR-TB) in this context can add a significant strain to the country’s
TB control efforts. DR-TB treatment is more expensive to treat, and often associated with poor
treatment outcomes. The costs of managing multi-drug resistant TB (MDR-TB) are dependent
on the model of care used. Models of MDR-TB treatment that require lengthy hospitalization
have been shown to be more expensive than predominantly ambulatory models.
This study is a cost analysis of an integrated community-based decentralized MDR-TB
management model developed by an international medical humanitarian organization,
Médecins Sans Frontières (MSF), in partnership with the City of Cape Town and Provincial
Government of the Western Cape (PGWC). Costs were assessed from a societal perspective
(including patient costs) in 2013 South African rand using standard methods. An ingredient
approach was used to estimate health systems costs, using a predominantly top-down analysis.
Data on patient costs was collected via a structured questionnaire survey in order to determine
patient’s costs’ associated with DR-TB treatment.
Results show that the MSF model was the least expensive model of MDR-TB management when
compared to the model that required all MDR-TB patients to be admitted at the central TB
hospital: R99,324 versus R120,036 (US$ 10,680 versus US$12,907). Patient costs were also
found to be slightly lower under the MSF model; R28, 799 (US$3,097 versus R29, 291
(US$3,150).
The MSF model is therefore the least expensive , including the costs of a sub-acute facility
known as Lizo Nobanda (for patients that are sick but do not require hospitalization). If
implemented throughout the province, the MSF model is likely to reduce the costs of
management of DR-TB treatment. At the time of writing, this process of decentralization has
already begun. Furthermore, given that the decentralized model is likely to be highly preferable
to patients, it thus has the added advantage of improved treatment numbers, if not improved
outcomes, though these outcomes are not assessed in this report.