Opportunities to Improve Outcomes for Children With Tuberculosis

In this issue of Pediatrics, Brown et al use an innovative data source, the Provincial Health Data Centre (PHDC), to present a comprehensive description of childhood tuberculosis (TB), which is defined as TB in people aged younger than 15 years, in South Africa’s Western Cape province.1 This region has one of the highest incidence rates of childhood TB in the world.Unlike studies based on surveillance data from national TB programs, which can systematically underreport cases of childhood TB, the PHDC does not rely on health care workers submitting TB case report forms. Instead, the PHDC collates data from all public sector services in the province into a single patient record. Laboratories, pharmacies, and clinical settings at all levels of care submit information to the PHDC.2 More than 20.6% of childhood TB episodes identified by the PHDC were never reported to the NTP. These reporting gaps are not simply a data problem. Underreporting translates to invisibility—children who are never counted cannot be prioritized, provided resources, or protected.Brown et al observed that the NTP notification gap was higher in children aged 5 to 9 years (29.5%) and 10 to 14 years (23.1%) compared with young children aged under 5 years (18.0%). This result is unexpected because the diagnosis of TB disease is generally more challenging in the youngest children.3,4 The authors do not elaborate on the reasons behind this finding, but it would be interesting to explore whether it is related to more intensive efforts to improve TB programmatic indicators in children aged younger than 5 years. The gap between PHDC and NTP reporting was also much higher for childhood TB cases diagnosed in hospitals compared with primary health centers. Identifying other factors associated with data gaps, as well as their underlying causes, can inform strategies to improve TB surveillance.The Brown et al study underscores the need for stronger public health infrastructure to prevent and successfully treat childhood TB. In children aged younger than 5 years, TB is largely driven by recent exposure to household contacts with infectious TB and is often preventable through timely administration of tuberculosis preventive therapy (TPT). In the Western Cape, the TB incidence among children aged 0 to 4 years was strikingly high, exceeding 700 per 100 000 in 2023. In the Cape Winelands, an area with the lowest HIV prevalence in the country, incidence in young children was double that in adults. However, even the PHDC likely underrepresented the true burden of TB in young children because it would not capture children in whom TB was never considered. These high incidences likely reveal gaps in timely contact tracing and TPT uptake.Brown et al also observed interdistrict heterogeneity in childhood TB burden, with children accounting for 10% of all TB cases in some districts but over 20% in others. As discussed by the authors, the Cape Winelands had the highest proportion of all TB cases that occurred in children; at 20.7%, this proportion was 5.7% higher than that of the second highest district. Such variation may indicate a localized epidemic, uneven implementation of household contact screening, or differences in capacity to diagnose childhood TB. Globally, there are similar patterns of geographic clustering, including among informal settlements in India and other parts of sub-Saharan Africa.5 These findings highlight the necessity of district-level analyses and planning. National or provincial averages obscure hotspots where children are disproportionately at risk. Characterization of epidemics on smaller spatial scales can inform the adaptation of interventions to local realities.6Perhaps the most concerning finding from the study by Brown et al were the losses along the TB care cascade. These losses have devastating consequences because untreated or incompletely treated children face rapid disease progression and high mortality risk.7 In South Africa, many children are diagnosed in hospitals but referred to primary care centers for treatment initiation. Nearly 20% of children in this study experienced initial loss to follow-up, meaning they were diagnosed but never initiated treatment, and an additional 12% were lost to follow-up after treatment began. Understanding the magnitude of loss to follow-up was possible only because of the PHDC’s integrated data. Such losses would have remained invisible if relying solely on NTP data, which do not systematically capture interfacility transfers for treatment initiation. Every handoff between facilities is a chance for children to fall off the treatment cascade. This challenge is not unique to South Africa. Across high TB-burden countries, weak linkages between hospitals and community-based care, limited tracking systems, and socioeconomic barriers lead to loss to follow-up.8 The South African experience shows both the scale of the problem and potential solutions. The LINKEDin Project—which was carried out in the Western Cape, KwaZulu-Natal, and Gauteng—implemented systems to track and follow-up with people who were transferred to another facility for TB treatment initiation.2 In the Western Cape and KwaZulu-Natal, the LINKEDin Project led to a reduction in initial loss to follow-up. There is an urgent need to expand such efforts in South Africa and adapt this type of intervention to other settings.The lessons derived from this study by Brown et al have relevance beyond South Africa. By using a province-wide integrated data system, the authors provide a far more complete picture than traditional NTP surveillance allows. Across the globe, TB reporting systems routinely underestimate childhood TB.9 This study from the Western Cape shows the feasibility and utility of integration of health data from different registries to improve surveillance and guide targeted interventions.Although critical for understanding disease burden and outcomes, data alone are insufficient. As the authors note, funding cuts to TB programs worldwide threaten fragile progress.10,11 Without political commitment and sustained investment to strengthening health infrastructure, even the best data systems cannot prevent children from being lost in the cascade of care. Brown et al have highlighted several key challenges to improving childhood TB care in the Western Cape: high incidence among the youngest children, large geographic disparities, major cascade losses, and substantial underreporting to the NTP. The study also points the way forward: strengthening and integrating surveillance data, expanding contact tracing and TPT uptake, closing gaps in the TB treatment cascade, tailoring interventions to local contexts, and ensuring that children are never invisible in the fight against TB.