Projected population level impact and cost-effectiveness of clinic and community-based tuberculosis screening approaches

Technical Summary: Projected Population Level Impact and Cost-Effectiveness of Clinic and Community-Based Tuberculosis Screening Approaches

Problem Statement
South Africa faces a critical challenge in meeting its End TB Strategy 2025 milestones, with TB mortality reductions lagging significantly behind targets and an estimated 26% of TB cases remaining undiagnosed. While the National Department of Health has set ambitious goals to scale up testing, primarily through clinic-based strategies, this expansion coincides with substantial reductions in global TB funding (estimated at a net loss of US$34 million). Consequently, there is an urgent need to identify the most cost-effective screening approaches to maximize health gains within constrained budgets. Current evidence on Active Case Finding (ACF) is heterogeneous, and there is no clear guidance on the optimal choice of screening approach or target population in high-burden settings.

Methodology
The authors developed a sophisticated individual-based mathematical model of TB transmission, diagnosis, and treatment, calibrated to national prevalence survey data and World Health Organization (WHO) estimates for South Africa. Key methodological features include:

• Stratification: The model explicitly stratifies the population by sex and HIV/ART status, incorporating empirical data on clinic-visiting behaviors and care-seeking patterns.
• Disease States: Infectious TB is divided into bacteriologically positive cases with and without TB symptoms (defined by the WHO four-symptom screen). The model accounts for self-cure in asymptomatic individuals and mortality only in symptomatic cases.
• Intervention Scenarios: Six screening approaches were simulated over the period 2026–2035, both individually and in combination:
  1. Community-based: Symptom screening, radiographic screening (mobile vans), and universal Xpert Ultra testing.
  2. Clinic-based: Intensified Targeted Universal TB Testing (TUTT) for key risk groups (PLHIV, recent contacts, recent TB), intensified TUTT allowing saliva samples, and universal clinic symptom screening.
• Economic Analysis: A mechanistic cost function was employed to estimate provider and patient-incurred costs from a societal perspective. This function distinguishes between fixed costs (programmatic, facility-level) and variable costs, allowing for the modeling of economies of scale and scope. Costs were parameterized using primary data collected from South African facilities and a cross-sectional survey of patients.
• Outcomes and Thresholds: The primary outcome was Disability-Adjusted Life Years (DALYs) averted. Cost-effectiveness was evaluated using three thresholds reflecting South Africa's health spending elasticity and willingness to pay. Expansion paths were constructed to identify optimal combinations of interventions as funding increases.

Key Results

• Impact on Incidence: Clinic-based symptom screening was the most effective standalone approach for reducing TB incidence, achieving a mean reduction of 25.1% (95% UI 19–31%) between 2026 and 2035. This was driven by higher population-level coverage of Xpert testing and reduced diagnostic delays.
• Cost-Effectiveness: Community radiographic screening at 10% annual population coverage emerged as the most cost-effective standalone approach, with an Incremental Cost-Effectiveness Ratio (ICER) of $421 per DALY averted (98% of model runs fell under the lowest threshold of $618/DALY).
• Expansion Path: The analysis identified that combining complementary approaches yields superior cost-effectiveness compared to standalone strategies. Ten of the eleven scenarios on the cost-effectiveness expansion path involved community radiographic screening at progressively higher coverage levels, combined with a clinic-based approach (either intensified TUTT or clinic symptom screening).
• Target Populations: The results highlight that the optimal strategy depends on the specific sub-population and existing coverage. For instance, while intensified TUTT is crucial for People Living with HIV (PLHIV) on ART who already attend clinics frequently, universal clinic symptom screening is more effective for HIV-negative individuals who have lower clinic attendance rates. However, when combined with community screening, the marginal value of universal clinic symptom screening diminishes relative to intensified TUTT.
• Sensitivity Analysis: The choice of discount rate and time horizon had the largest impact on ICERs. Changes in Xpert Ultra pricing or the addition of CAD4TB software had minimal impact on the relative ordering of approaches, suggesting that the overall design (who is screened and how) drives cost-effectiveness more than the absolute cost of the diagnostic technology.

Significance and Claims
The paper claims that a two-pronged approach, integrating clinic-level case finding with community-level chest radiography, represents the most efficient use of limited TB resources in South Africa. The study emphasizes that screening strategies must be tailored to both the TB disease risk of specific sub-populations and their existing patterns of health facility attendance. By utilizing a unified modeling framework that combines behavioral, epidemiological, and economic data, the authors demonstrate that aligning programs with these factors is essential for maximizing health benefits. The findings provide a benchmark for policymakers to understand the relative value of different screening modalities and suggest that expanding community radiographic screening, particularly when paired with targeted clinic interventions, offers the best value for money under current funding constraints. The authors note that while their model accounts for economies of scale and scope—a strength over previous studies—it does not account for operational health system constraints such as staff shortages or equipment failures, which could limit the feasibility of high-volume testing scenarios.