Cost-effectiveness of tuberculosis infection prevention and control interventions in South African clinics: a model-based economic evaluation informed by complexity science methods

Abstract

IntroductionNosocomialMycobacterium tuberculosis(Mtb) transmission substantially impacts health workers, patients and communities. Guidelines for tuberculosis infection prevention and control (TB IPC) exist but implementation in many settings remains suboptimal. Evidence is needed on cost-effective investments to preventMtbtransmission that are feasible in routine clinic environments.MethodsA set of TB IPC interventions was codesigned with local stakeholders using system dynamics modelling techniques that addressed both core activities and enabling actions to support implementation. An economic evaluation of these interventions was conducted at two clinics in KwaZulu-Natal, employing agent-based models ofMtbtransmission within the clinics and in their catchment populations. Intervention costs included the costs of the enablers (eg, strengthened supervision, community sensitisation) identified by stakeholders to ensure uptake and adherence.ResultsAll intervention scenarios modelled, inclusive of the relevant enablers, cost less than US$200 per disability-adjusted life-year (DALY) averted and were very cost-effective in comparison to South Africa’s opportunity cost-based threshold (US$3200 per DALY averted). Two interventions, building modifications to improve ventilation and maximising use of the existing Central Chronic Medicines Dispensing and Distribution system to reduce the number of clinic attendees, were found to be cost saving over the 10-year model time horizon. Incremental cost-effectiveness ratios were sensitive to assumptions on baseline clinic ventilation rates, the prevalence of infectious TB in clinic attendees and future HIV incidence but remained highly cost-effective under all uncertainty analysis scenarios.ConclusionTB IPC interventions in clinics, including the enabling actions to ensure their feasibility, afford very good value for money and should be prioritised for implementation within the South African health system.

Time to change the way we think about tuberculosis infection prevention and control in health facilities: insights from recent research

“… 55 The wearing of surgical masks by patients and N95 respirators by clinic staff, the intervention with the highest incremental cost-effectiveness ratio (ICER) in our model, costs only US$151 per DALY averted. 54 The intervention with the highest impact compared to base case was the introduction of queue management systems with outdoor waiting areas (US$28 per DALY averted), followed by the installation, running, and maintenance of upper room UVGI (US$57 per DALY averted). 54 Two interventions were estimated to be cost-saving compared to base case: optimizing the use of community dispensing for chronic conditions, which we modeled as an increase in the number of HIV-positive people with stable disease collecting their medications from external pickup points and modifications (retrofits) to clinic buildings to improve ventilation.…”

“… 54 Two interventions were estimated to be cost-saving compared to base case: optimizing the use of community dispensing for chronic conditions, which we modeled as an increase in the number of HIV-positive people with stable disease collecting their medications from external pickup points and modifications (retrofits) to clinic buildings to improve ventilation. 54 …”

“… 53 ICERs falling substantially below the cost-effectiveness threshold indicate that both the interventions and enablers are affordable and should be implemented. 54 …”

“…6 Cost-effectiveness analyses suggest that all of the TB-IPC interventions modeled, including combinations of these interventions, are highly cost-effective. 54 We used the current opportunity cost-based threshold for South Africa, estimated at US$3,200 per disability-adjusted life-year (DALY) averted. 55 The wearing of surgical masks by patients and N95 respirators by clinic staff, the intervention with the highest incremental cost-effectiveness ratio (ICER) in our model, costs only US$151 per DALY averted.…”

“…54 The intervention with the highest impact compared to base case was the introduction of queue management systems with outdoor waiting areas (US$28 per DALY averted), followed by the installation, running, and maintenance of upper room UVGI (US$57 per DALY averted). 54 Two interventions were estimated to be cost-saving compared to base case: optimizing the use of community dispensing for chronic conditions, which we modeled as an increase in the number of HIV-positive people with stable disease collecting their medications from external pickup points and modifications (retrofits) to clinic buildings to improve ventilation. 54 These cost-effectiveness analyses included not only the costs of each intervention but also the costs of the necessary enablers to relieve constraints to intervention uptake and implementation, for example, additional targeted training and supervision at facility level.…”