High and Uneven Costs of Drug-Resistant TB Treatment in Children

Children who develop multidrug-resistant or rifampicin-resistant tuberculosis (MDR/RR-TB) are an important yet often overlooked group when health systems try to understand the cost of care. In a study led by Thomas Wilkinson, researchers used a modern digital health approach to measure what it actually costs health services to treat these children in the Western Cape, South Africa. Instead of relying on estimates or fragmented records, they turned to the Provincial Health Data Centre (PHDC), which brings together hospital inpatient and outpatient data, medication records, laboratory tests, and primary healthcare use. By linking those routine records for children diagnosed with MDR/RR-TB between 1 January 2018 and 31 December 2021, the team could follow real-world care pathways and tally the resources consumed. This approach focuses on health-service spending rather than household or societal costs, aiming to give policy makers and programme managers a clear picture of where money is being spent when children receive treatment for drug-resistant TB.

The analysis used anonymised, integrated PHDC data for a sample of n=271 children diagnosed with MDR/RR-TB during the study period. Costs were examined across patient and disease characteristics including age, sex, drug susceptibility type, site of disease, and HIV status. The study found large variation in total per-patient costs: the median was US$7,576 with an interquartile range (IQR) of $2,725 - $22,986. The overall distribution of costs fit a gamma distribution with a mean of US$13,435, α = 0.93, β = 14,496, indicating a highly skewed pattern where a small number of children incurred very high costs. Regression analysis showed that age, site of disease, living with HIV, and treatment duration had significant impacts on costs, while resistance profile and sex did not show a significant effect. The authors also highlight that treating children in hospital is a major cost driver, suggesting that reducing hospitalisation could cut overall spending.

The study’s main contribution is methodological and practical: by using routinely collected, real-world data from the PHDC, researchers produced an accurate and representative picture of how much health services actually spend on children with MDR/RR-TB. The finding that costs are both high and highly skewed matters for planning and budgeting. A relatively small group of children generate very large expenditures, driven in part by hospital admission and longer treatment durations, and costs are higher for children living with HIV and for particular disease sites. For programme developers and policy makers, these results point to two clear priorities: targeting interventions that shorten treatment when safe, and reducing the need for hospitalisation where outpatient care or alternative models are possible. Finally, the data and results can be used as inputs to cost-effectiveness and budget impact analyses, helping guide local and international decisions about how to allocate resources for paediatric MDR/RR-TB care.