Abstracts

Introduction

Brazil, an upper middle-income country, holds one of the largest universal public health systems, the Brazilian Unified National Health System (SUS, acronym in Portuguese), which covers over 150 million people who depend exclusively on its services. The Brazilian Ministry of Health frequently receives the highest budget of the Federal Government. This budget is dedicated to all health activities, thus incorporation of new technologies requires approval by an independent committee, which evaluates evidence on safety, effectiveness, and cost-effectiveness ¹1. Departamento de Ciência e Tecnologia, Secretaria de Ciência, Tecnologia e Insumos Estratégicos, Ministério da Saúde. Diretrizes metodológicas: diretriz de avaliação econômica. Brasília: Ministério da Saúde; 2014..

Brazil is one of the 30 highest tuberculosis (TB) burden countries ²2. United Nations. Resolution adopted by the General Assembly on 10 October 2019. https://undocs.org/en/A/RES/74/2 (accessed on 30/Mar/2022).
https://undocs.org/en/A/RES/74/2... . To achieve TB elimination, countries need to invest on TB preventive treatment ³3. Apriani L, McAllister S, Sharples K, Alisjahbana B, Ruslami R, Hill PC, et al. Latent tuberculosis infection in health care workers in low and middle-income countries: an updated systematic review. Eur Respir J 2019; 53:1801789.. Since 2010, TB preventive treatment is recommended for all age contacts with a positive tuberculin skin test (TST) and a normal chest X-ray in Brazil, but uptake of TB preventive treatment has been slow ⁴4. Ministério da Saúde. Protocolo de vigilância da infecção latente pelo Mycobacterium tuberculosis no Brasil. Brasília: Ministério da Saúde; 2022.. Healthcare workers are one of the priority contact populations, thus they should be tested for TB infection annually, and treated if conversion is detected and the chest X-ray is normal. Frequent losses occur in their cascade-of-care, particularly among healthcare workers with prior Bacilli Calmette-Guérin (BCG) vaccination, who are more likely to refuse TB preventive treatment ⁵5. Xu Y, Schwartzman K. Referrals for positive tuberculin tests in new health care workers and students: a retrospective cohort study. BMC Public Health 2010; 10:28.. Pre-treatment losses, such as access to TB infection diagnostic tests, are also an important bottleneck ⁶6. World Health Organization. WHO consolidated guidelines on tuberculosis. Module 1: prevention. Tuberculosis preventive treatment. Geneva: World Health Organization; 2021.^,77. Nationwide shortage of tuberculin skin test antigens: CDC recommendations for patient care and public health practice. MMWR Morb Mortal Wkly Rep 2019; 68:552-3..

Following the 2018 United Nations high-level meeting recommendations ²2. United Nations. Resolution adopted by the General Assembly on 10 October 2019. https://undocs.org/en/A/RES/74/2 (accessed on 30/Mar/2022).
https://undocs.org/en/A/RES/74/2... , some steps to scale up TB preventive treatment have been implemented in Brazil, including training of healthcare workers starting in 2018; implementation of a TB preventive treatment surveillance information system in the same year ⁴4. Ministério da Saúde. Protocolo de vigilância da infecção latente pelo Mycobacterium tuberculosis no Brasil. Brasília: Ministério da Saúde; 2022.; incorporation of the QuantiFERON-TB Gold Plus (QFT-Plus) test (Qiagen; https://www.qiagen.com) for people living with HIV infection (PLWH), contacts aged 2-10, transplant candidates, and people using immunosuppressive drugs in 2021 ⁸8. Ministério da Saúde. Portaria SCTIE/MS nº 50, de 11 de novembro de 2020. Torna pública a decisão de incorporar o teste de liberação de interferon-gama (interferon gamma release assay - IGRA) para detecção de tuberculose latente em pessoas vivendo com HIV, crianças em contato com casos de tuberculose ativa e pacientes candidatos a transplante de células-tronco. Diário Oficial da União 2020; 13 nov.; and incorporation of the 3HP regimen (three months of weekly doses of 900mg rifapentine and 900mg isoniazid) as the first choice for TB preventive treatment in 2022 ⁹9. Coordenação-Geral de Vigilância das Doenças de Transmissão Respiratória de Condições Crônicas, Departamento de Doenças de Condições Crônicas e Infecções Sexualmente Transmissíveis, Secretaria de Vigilância em Saúde, Ministério da Saúde. Nota Informativa nº 1/2022-CGDR/.DCCI/SVS/MS. Brasília: Ministério da Saúde; 2022..

The intermittent shortage of PPD Rt-23 (Statens Serum Institut; https://en.ssi.dk/) ⁷7. Nationwide shortage of tuberculin skin test antigens: CDC recommendations for patient care and public health practice. MMWR Morb Mortal Wkly Rep 2019; 68:552-3. and the lower specificity of TST in BCG-vaccinated populations ¹⁰10. Diel R, Lampenius N, Nienhaus A. Cost effectiveness of preventive treatment for tuberculosis in special high-risk populations. Pharmacoeconomics 2015; 33:783-809. pose significant challenges, prompting the Brazilian National Tuberculosis Control Program (PNCT, acronym in Portuguese) to evaluate newer, more specific tests. QFT-Plus and TB antigen-based skin tests (TBST) use the more specific antigens, namely early secreted antigenic target 6 (ESAT-6) and culture filtrate protein 10 (CFP-10). Currently, three tuberculosis TBST are being validated in Brazil for eventual incorporation: Diaskintest (Generium Pharmaceutical; https://www.generium.ru/en/), C-TST (Zhifei Longcom Biologic Pharmacy Co.; https://en.zhifeishengwu.com), and Cy-TB (Serum Institute of India; https://www.seruminstitute.com/). QFT-Plus require laboratory facilities and is more costly than TST ¹¹11. Loureiro RB, Maciel ELN, Caetano R, Peres RL, Fregona G, Golub JE, et al. Cost-effectiveness of QuantiFERON-TB Gold In-Tube versus tuberculin skin test for diagnosis and treatment of latent tuberculosis infection in primary health care workers in Brazil. PLoS One 2019; 14:e0225197.. Diaskintest was found to be cheaper and more effective than TST for PLWH in Brazil, but there is no data on cost-effectiveness of TBST in healthcare workers, the largest eligible population for TB preventive treatment ¹²12. Steffen RE, Pinto M, Kritski A, Trajman A. Cost-effectiveness of newer technologies for the diagnosis of Mycobacterium tuberculosis infection in Brazilian people living with HIV. Sci Rep 2020; 10:21823.. In this study, we analyzed the cost-effectiveness of TBST and QFT-Plus compared to TST for TB infection screening healthcare workers with a first negative TST test from the perspective of the SUS, over a 5-year time horizon.

Methods

Setting and current policies

The SUS provides, via the PNCT and municipal health departments, free-of-charge diagnosis and treatment for TB and TB infection. Investigation of TB infection in healthcare workers is recommended in Brazil as part of the worker’s pre-employment and annual health visits, regardless of the level of healthcare service ¹³13. Ministério da Saúde. Manual de recomendações para o controle da tuberculose no Brasil. 2nd Ed. Brasília: Ministério da Saúde; 2018.. Healthcare workers with a first negative TST are invited to repeat the test in one to three weeks to assess the booster effect (induration size increment of 10mm). Individuals with a persistent negative TST then undergo a 1-step TST annually. TB preventive treatment is recommended when conversion (a 10mm increment over the latest induration size) occurs ¹³13. Ministério da Saúde. Manual de recomendações para o controle da tuberculose no Brasil. 2nd Ed. Brasília: Ministério da Saúde; 2018.. Since March 2022, the TB preventive treatment regimen of choice is 3HP, except for children under two years of age, contacts over 50 years of age, those with liver disease, contacts of people with isoniazid- or rifampicin-monoresistant TB or intolerance to isoniazid or rifapentin ⁹9. Coordenação-Geral de Vigilância das Doenças de Transmissão Respiratória de Condições Crônicas, Departamento de Doenças de Condições Crônicas e Infecções Sexualmente Transmissíveis, Secretaria de Vigilância em Saúde, Ministério da Saúde. Nota Informativa nº 1/2022-CGDR/.DCCI/SVS/MS. Brasília: Ministério da Saúde; 2022.. Prior to TB preventive treatment, for any indication, a chest radiograph is mandatory to rule out TB.

Population, study perspective, time horizon, and discount rate

The study population consists of a hypothetical cohort of 10,000 Brazilian healthcare workers of both sexes with an average age of 35 years, working at any level of healthcare service and negative TST results in the previous year. The SUS perspective was adopted for this study ¹1. Departamento de Ciência e Tecnologia, Secretaria de Ciência, Tecnologia e Insumos Estratégicos, Ministério da Saúde. Diretrizes metodológicas: diretriz de avaliação econômica. Brasília: Ministério da Saúde; 2014.. Each patient with positive bacilloscopy for Mycobacterium tuberculosis infects an average of 10 individuals per year 1 ¹⁴14. Sepkowitz KA. How contagious is tuberculosis? Clin Infect Dis 1996; 23:954-62.; about 1% to 5% of exposed individuals develop TB shortly after exposure (primary TB), and 10% to 30% become infected (TB infection) ¹⁵15. Jones-López EC, Namugga O, Mumbowa F, Ssebidandi M, Mbabazi O, Moine S, et al. Cough aerosols of Mycobacterium tuberculosis predict new infection: a household contact study. Am J Respir Crit Care Med 2013; 187:1007-15.. Among those with TB infection, approximately 5% to 15% will develop the disease over their lifetime ¹⁶16. Sutherland I. The evolution of clinical tuberculosis in adolescents. Tubercle 1966; 47:308., with a higher risk in the first 2 to 5 years after infection ¹⁷17. Cardona PJ. Reactivation or reinfection in adult tuberculosis: is that the question? Int J Mycobacteriol 2016; 5:400-7.. A time horizon of five years was considered, as individuals are at increased risk of developing TB in the first 2 to 5 years after TB infection ¹⁷17. Cardona PJ. Reactivation or reinfection in adult tuberculosis: is that the question? Int J Mycobacteriol 2016; 5:400-7.. This study followed the recommendation of the Brazilian Ministry of Health, in which discount rates are standardized at 5% per year for effects and costs to increase the comparability of studies ¹1. Departamento de Ciência e Tecnologia, Secretaria de Ciência, Tecnologia e Insumos Estratégicos, Ministério da Saúde. Diretrizes metodológicas: diretriz de avaliação econômica. Brasília: Ministério da Saúde; 2014.. Additionally, it suggests using different discount rates (0% and 10%) in sensitivity analysis to determine how much the arbitrary selection of the rate affected the study’s conclusion ¹1. Departamento de Ciência e Tecnologia, Secretaria de Ciência, Tecnologia e Insumos Estratégicos, Ministério da Saúde. Diretrizes metodológicas: diretriz de avaliação econômica. Brasília: Ministério da Saúde; 2014..

Model structure

Given the increased risk of developing TB in the first 2 to 5 years after TB infection ¹, a Markov decision-analytic model was developed to simulate a cohort of 10,000 healthcare workers over five annual cycles for TB preventive treatment with 3HP. Then, four strategies for TB infection detection were compared to the standard of care (TST), namely the QFT-Plus and three novel skin tests: Diaskintest, C-TST, and Cy-TB. The model simulated the natural history of TB infection, comparing the clinical outcomes and economic impacts of the newer diagnostic technologies to the traditional TST. All analyses were performed with TreeAge Pro Healthcare 2022 (https://www.treeage.com/). The health states considered were (1) no TB infection; (2) TB infection; (3) no TB infection, false positive test and not treated; (4) no TB infection, false positive test and treated; (5) TB infection and not tested; (6) TB infection, positive test and not treated; (7) TB infection, positive test and treated; (8) TB; (9) cured TB; and (10) death (Supplementary Material - Figure S1; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf). The model estimated the number of TB cases that occurred in each of the strategies. The effectiveness measure was the number of TB cases avoided, estimated by subtracting the number of cases observed in the strategy with the reference test (TST) from those with the tests under evaluation (QFT-Plus and TBST). The number needed to misdiagnose was also estimated, defined as the number of patients who need to be tested in order for one to be misdiagnosed by the test (either false positive or false negative results), considering a 0.37 TB infection prevalence (95% confidence interval [95%CI]: 0.17; 0.36) ¹⁸18. Nasreen S, Shokoohi M, Malvankar-Mehta MS. Prevalence of latent tuberculosis among health care workers in high burden countries: a systematic review and meta-analysis. PLoS One 2016; 11:e0164034.^,1919. Habibzadeh F, Yadollahie M. Number needed to misdiagnose: a measure of diagnostic test effectiveness. Epidemiology 2013; 24:170.. The incremental cost-effectiveness ratio (ICER) per TB case avoided was estimated by subtracting the number of cases observed in the strategy with the reference test (TST) from those with the tests under evaluation (QFT-Plus and TBST).

The probability of progression from TB infection to TB without TB preventive treatment was based on observed rates among healthcare workers for the first two years after infection, then it considered the probability for the general population ²⁰20. Campbell JR, Winters N, Menzies D. Absolute risk of tuberculosis among untreated populations with a positive tuberculin skin test or interferon-gamma release assay result: systematic review and meta-analysis. BMJ 2020; 368:m549..

Testing procedures and interpretation

This study considered five strategies, with four being based on skin tests, three using the recombinant ESAT-6 and CFP-10 immunogens (Diaskintest, C-TST and Cy-TB), and one - the current standard of care - using the tuberculin PPD Rt-23 (TST). By replacing PPD with M. tuberculosis specific antigens, TBST combine the operational advantages of the TST with the specificity of interferon-gamma release assays (IGRA). The fifth strategy employs the QFT-Plus test, the only available IGRA in Brazil ⁸8. Ministério da Saúde. Portaria SCTIE/MS nº 50, de 11 de novembro de 2020. Torna pública a decisão de incorporar o teste de liberação de interferon-gama (interferon gamma release assay - IGRA) para detecção de tuberculose latente em pessoas vivendo com HIV, crianças em contato com casos de tuberculose ativa e pacientes candidatos a transplante de células-tronco. Diário Oficial da União 2020; 13 nov..

All skin tests involve the intradermal application of the antigen on the volar aspect of the forearm, following the Mantoux method in any healthcare facility (usually by a trained nurse) ¹³13. Ministério da Saúde. Manual de recomendações para o controle da tuberculose no Brasil. 2nd Ed. Brasília: Ministério da Saúde; 2018.. The test result is read 48 or 72 hours after application. The cutoff point considered to define the result as positive was ≥ 5mm. If the reading is lost, no further skin tests is conducted and the healthcare workers should be retested after a 1-year interval.

QFT-Plus testing is conducted in the laboratory following the manufacturer’s instructions, using 1mL aliquots of whole blood, which can be obtained in the healthcare facility and transported to the laboratory or directly in the laboratory. If the result is indeterminate, the test is repeated once. Conversion is defined when a negative test turns positive ¹³13. Ministério da Saúde. Manual de recomendações para o controle da tuberculose no Brasil. 2nd Ed. Brasília: Ministério da Saúde; 2018.. Only healthcare workers with a negative initial test undergo subsequent annual testing. When conversion of any TB infection test occurs, a chest radiograph is performed and if negative, TB preventive treatment is recommended ¹³13. Ministério da Saúde. Manual de recomendações para o controle da tuberculose no Brasil. 2nd Ed. Brasília: Ministério da Saúde; 2018..

Model parameters

Effectiveness data

Meta-analyses (Table 1) provided estimates for clinical, epidemiological and tests accuracy parameters ²¹21. Alsdurf H, Hill PC, Matteelli A, Getahun H, Menzies D. The cascade of care in diagnosis and treatment of latent tuberculosis infection: a systematic review and meta-analysis. Lancet Infect Dis 2016; 16:1269-78.^,2222. Ziakas PD, Mylonakis E. 4 months of rifampin compared with 9 months of isoniazid for the management of latent tuberculosis infection: a meta-analysis and cost-effectiveness study that focuses on compliance and liver toxicity. Clin Infect Dis 2009; 49:1883-9.^,2323. Kahwati LC, Feltner C, Halpern M, Woodell CL, Boland E, Amick HR, et al. Primary care screening and treatment for latent tuberculosis infection in adults: evidence report and systematic review for the US Preventive Services Task Force. JAMA 2016; 316:970-83.^,2424. Smieja MJ, Marchetti CA, Cook DJ, Smaill FM. Isoniazid for preventing tuberculosis in non-HIV infected persons. Cochrane Database Syst Rev 2000; (2):CD001363.^,2525. Krutikov M, Faust L, Nikolayevskyy V, Hamada Y, Gupta RK, Cirillo D, et al. The diagnostic performance of novel skin-based in-vivo tests for tuberculosis infection compared with purified protein derivative tuberculin skin tests and blood-based in vitro interferon-? release assays: a systematic review and meta-analysis. Lancet Infect Dis 2021; 22:250-64.^,2626. Starshinova A, Dovgalyk I, Malkova A, Zinchenko Y, Pavlova M, Belyaeva E, et al. Recombinant tuberculosis allergen (Diaskintest) in tuberculosis diagnostic in Russia (meta-analysis). Int J Mycobacteriol 2020; 9:335-46.^,2727. World Health Organization. Use of alternative interferon-gamma release assays for the diagnosis of TB infection: WHO policy statement. Geneva: World Health Organization; 2022.^,2828. Sharma SK, Sharma A, Kadhiravan T, Tharyan P. Rifamycins (rifampicin, rifabutin and rifapentine) compared to isoniazid for preventing tuberculosis in HIV-negative people at risk of active TB. Evid Based Child Health 2014; 9:169-294.^,2929. Melnychuk L, Perlman-Arrow S, Bastos ML, Menzies D. A systematic review and meta-analysis of tuberculous preventative therapy adverse events. Clin Infect Dis 2023; 77:287-94.^,3030. Oh CE, Ortiz-Brizuela E, Bastos ML, Menzies D. Comparing the diagnostic performance of qft-plus to other tests of latent tuberculosis infection: a systematic review and meta-analysis. Clin Infect Dis 2020; 73:e1116-25.. To estimate transition probabilities in the model, this study employed different types of measurement statistics reported in the literature using the method described by Gidwani & Russell ³¹31. Gidwani R, Russell LB. Estimating transition probabilities from published evidence: a tutorial for decision modelers. Pharmacoeconomics 2020; 38:1153-64. (Supplementary Material; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf). Age-specific all-cause mortality rates were obtained from the 2022 Brazilian National Mortality Table ³²32. Instituto Brasileiro de Geografia e Estatística. Tábuas completas de mortalidade. https://www.ibge.gov.br/estatisticas/sociais/populacao/9126-tabuas-completas-de-mortalidade.html?=&t=resultados (accessed on 05/Oct/2021).
https://www.ibge.gov.br/estatisticas/soc... .

Costing data

Brazil boasts one of the largest universal public health systems (SUS), which provides coverage for over 150 million people who rely exclusively on its services. Given this, the national guideline for conducting economic evaluation studies recommends adopting the perspective of the SUS as the purchaser of services, thereby accounting for all costs covered by the public health system. The costs of medical visits, chest radiograph, sputum smear, blood work-up, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) dosage, hospitalization, and death due to adverse events were obtained from the Brazilian Hospital Information System (SIH, acronym in Portuguese) ³³33. Ministério da Saúde. Sistema de Gerenciamento da Tabela de Procedimentos, Medicamentos e OPM do SUS. http://sigtap.datasus.gov.br/tabela-unificada/app/sec/procedimento/publicados/consultar (accessed on 01/Jul/2022).
http://sigtap.datasus.gov.br/tabela-unif... . The costs for isoniazid (300mg/pill), rifapentine (150mg/pill), QFT-Plus (kit), and TST were informed by the PNCT (2024; personal communication). The cost of treating TB with directly observed therapy (DOT) were estimated by Steffen et al. ³⁴34. Steffen R, Menzies D, Oxlade O, Pinto M, de Castro AZ, Monteiro P, et al. Patients' costs and cost-effectiveness of tuberculosis treatment in dots and non-dots facilities in Rio de Janeiro, Brazil. PLoS One 2010; 5:e14014.. A micro-costing analysis for conducting diagnostic tests for TB infection was conducted and published elsewhere ¹¹11. Loureiro RB, Maciel ELN, Caetano R, Peres RL, Fregona G, Golub JE, et al. Cost-effectiveness of QuantiFERON-TB Gold In-Tube versus tuberculin skin test for diagnosis and treatment of latent tuberculosis infection in primary health care workers in Brazil. PLoS One 2019; 14:e0225197.. Table 2 and Supplementary Material (https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf) detail the sources of costs estimates. All values were adjusted for inflation by the accumulated variation of the Brazilian Extended Consumer Price Index (IPCA, acronym in Portuguese) for the period and converted into U.S. dollars (USD) using the average annual rate according to Brazilian Central Bank (USD 1 = BRL 5.16) for 2022 ³⁵35. Banco Central do Brasil. Cotações e boletins. https://www.bcb.gov.br/estabilidadefinanceira/historicocotacoes (accessed on 10/Jan/2023).
https://www.bcb.gov.br/estabilidadefinan... , as described by Turner et al. ³⁶36. Turner HC, Lauer JA, Tran BX, Teerawattananon Y, Jit M. Adjusting for inflation and currency changes within health economic studies. Value Health 2019; 22:1026-32. (Tables 2 and 3, and Supplementary Material; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf).

The complete TB preventive treatment costs included 12 weekly doses of rifapentine (900mg) and isoniazid (900mg) and two medical monitoring visits; 3HP was self-administered ⁹9. Coordenação-Geral de Vigilância das Doenças de Transmissão Respiratória de Condições Crônicas, Departamento de Doenças de Condições Crônicas e Infecções Sexualmente Transmissíveis, Secretaria de Vigilância em Saúde, Ministério da Saúde. Nota Informativa nº 1/2022-CGDR/.DCCI/SVS/MS. Brasília: Ministério da Saúde; 2022.. Brazil does not acquire consumables from the Global Drug Facility (GDF); thus, information were obtained from the Brazilian Ministry of Health, and the GDF value considered the combined weekly dose of 3HP in the sensitivity analysis ³⁷37. Partnership ST. Global drug facility medicines catalog. February 2022. https://www.stoptb.org/sites/default/files/gdfmedicinescatalog_1.pdf (accessed on 01/Feb/2022).
https://www.stoptb.org/sites/default/fil... .

Partial treatment costs for TB infection were considered for those who developed severe adverse events and did not complete TB preventive treatment. The incomplete TB preventive treatment included costs with two months of weekly doses of rifapentine (900mg) and isoniazid (900mg) and one medical consultation.

Cases of severe adverse events with 3HP incurred in hospitalization costs, valued by the code Treatment of Complications of Surgical or Clinical Procedures from the SIH table, with a 35% increase in values related to professional care in hospitals considered as type II urgency ³³33. Ministério da Saúde. Sistema de Gerenciamento da Tabela de Procedimentos, Medicamentos e OPM do SUS. http://sigtap.datasus.gov.br/tabela-unificada/app/sec/procedimento/publicados/consultar (accessed on 01/Jul/2022).
http://sigtap.datasus.gov.br/tabela-unif... .

In cases of adverse events that evolved to death, costs equivalent to two daily hospitalizations to the intensive care unit were included ³³33. Ministério da Saúde. Sistema de Gerenciamento da Tabela de Procedimentos, Medicamentos e OPM do SUS. http://sigtap.datasus.gov.br/tabela-unificada/app/sec/procedimento/publicados/consultar (accessed on 01/Jul/2022).
http://sigtap.datasus.gov.br/tabela-unif... . For those who survived, the monitoring costs of severe adverse events were also included, encompassing three doctor visits, two blood count tests, and two tests for AST and ALT.

Sensitivity analyses

One-way sensitivity analysis was conducted on all model inputs, using the 95%CI range of transition probabilities. For the cost parameters, the lower limit of the range was half the price and the upper limit, twice the price ¹1. Departamento de Ciência e Tecnologia, Secretaria de Ciência, Tecnologia e Insumos Estratégicos, Ministério da Saúde. Diretrizes metodológicas: diretriz de avaliação econômica. Brasília: Ministério da Saúde; 2014.. A probabilistic sensitivity analysis was also performed in a Monte Carlo simulation with 10,000 iterations (Table 3). It was assumed that the uncertainty in clinical probabilities and accuracies followed a beta distribution, with the standard deviation (SD) estimated based on data extracted from systematic literature reviews (Supplementary Material; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf). A gamma distribution was assumed for costs, with the SD equal to 10% of the original value. The generated distributions of costs, TB cases avoiding diagnosis of TB infection and its preventive treatment, and ICER are shown using median and 95% uncertainty ranges (UR).

Model assumptions

The cohort undergoes standard procedures for healthcare workers, as prescribed by the Brazilian national guidelines. Several simplifying assumptions were incorporated into the model: (1) all healthcare workers were asymptomatic and without TB; (2) no one was HIV-infected; (3) healthcare workers with a positive test result underwent chest radiography and physical examination to exclude TB; (4) chest radiography and symptom screening are 100% accurate in ruling out TB; (5) individuals with a second indeterminate result on the QFT-Plus test had the same risk of developing TB as those with a negative result; (6) nonadherence to treatment due to adverse events occurred in the first two months and conferred no partial protection; (7) all individuals with TB accepted treatment; (8) patients who completed TB treatment were considered cured; (9) all TB and TB infection cases are sensitive to antituberculosis drugs; (10) people who did not adhere to TB preventive treatment did not develop adverse events; (11) the death of an healthcare workers from other causes had no financial impact on the Brazilian Ministry of Health; and (12) our assessment considered the stability of newer recombinant reagents to be comparable to that of PPD Rt-23.

Willingness-to-pay threshold

In Brazil, the government-established threshold for cost-effectiveness is USD 7,752 per quality-adjusted life-year (QALY) gained ³⁸38. Ministério da Saúde. O uso de limiares de custo-efetividade. Brasília: Ministério da Saúde; 2022.. Due to the lack of a threshold reference for case avoidance in Brazil, the QALY threshold was adopted as a reference, which has been the metric used for the incorporation of technologies into the SUS. To assess the cost-effectiveness of the intervention, this threshold was employed for each TB case prevented. Then, a cost-effectiveness acceptability curve was constructed, which illustrates the probability of the intervention being deemed cost-effective across a range of willingness-to-pay threshold values.

Price threshold analysis

Price threshold analysis was conducted to determine the maximum price at which the strategy incorporating QFT-Plus would be considered cost-effective. This analysis was carried out probabilistically, enabling the assessment of uncertainties in the model parameters and their influence on the cost-effectiveness of the intervention.

This cost-effectiveness report followed the Consolidated Health Economic Evaluation Reporting Standards 2022 (CHEERS 2022) statement ³⁹39. Husereau D, Drummond M, Augustovski F, de Bekker-Grob E, Briggs AH, Carswell C, et al. Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022. Explanation and elaboration: a report of the ISPOR CHEERS II Good Practices Task Force. Value Health 2022; 25:10-31..

Results

Table 4 presents the projected costs and health outcomes of simulated strategies. The SUS costs for screening 10,000 healthcare workers over a 5-year period, including screening and treatment costs for both TB infection using weekly 3HP doses and TB treatment with DOT, were: USD 298,236 (95%UR: 251,196; 353,603) for Diaskintest; USD 286,640 (95%UR: 241,310; 340,003) for Cy-TB test; USD 434,984 (95%UR: 357,541; 525,148) for C-TST test; USD 334,619 (95%UR: 273,173; 409,566) for TST; and USD 801,727 (95%UR: 620,926; 1,028,999) for QFT-Plus. Compared to TST, the Diaskintest and Cy-TB strategies were the most cost saving for TB infection diagnosis (USD 7,239 and USD 62,388 per TB case averted, respectively) (Supplementary Material - Figure S2 and Table S4; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf). Despite QFT-Plus demonstrating slightly higher effectiveness compared to Diaskintest and Cy-TB, the incremental cost of using QFT-Plus was USD 84,038 per TB case averted (Table 4).

With regards to number needed to misdiagnose, 29, 92, 155, 191, and 217 tests are necessary for 10 healthcare workers to be misdiagnosed by the TST, Cy-TB, C-TST, QFT-Plus, and Diaskintest, respectively.

Monte Carlo simulations showed that, at a willingness to pay threshold of USD 7,752 per TB case averted and considering all screening strategies simultaneously, the Diaskintest showed the highest net benefit in 92.9%. The Cy-TB strategy presented the highest net benefit in 7.1%. Cost-effectiveness acceptability curve, illustrating the proportion of simulations in which each strategy holds the highest net benefit at different willingness to pay thresholds is provided in Supplementary Material (Figure S3; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf).

One-way sensitivity analyses demonstrated that the incremental cost-effectiveness of TB infection screening strategies among healthcare workers was most sensitive to estimated TB infection incidence in healthcare workers (TST), the cost of TB preventive treatment, the cost of excluding TB, and the cost of Diaskintest (Supplementary Material - Table S4 and Figure S4; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf). The cost-effectiveness estimates for Cy-TB strategy was also sensitive to TST specificity (Supplementary Material - Table S4 and Figure S4; https://cadernos.ensp.fiocruz.br/static//arquivo/suppl-e00178623_5325.pdf).

Considering the GDF value for the combined weekly dose of 3HP, a cost reduction is observed for all screening strategies: USD 281,689 for Diaskintest; USD 270,768 for Cy-TB; USD 418,605 for C-TST test; USD 308,507 for TST; and USD 784,573 for QFT-Plus. When compared to TST, assuming a higher probability of progression from TB infection to TB in the first two years after infection (0.015) in healthcare workers, the Diaskintest strategy remains the most cost saving for TB infection diagnosis, with savings of USD 5,244 per TB case averted.

Price threshold analysis, at a willingness-to-pay threshold of USD 7,752 per TB case averted, showed that for the QFT-Plus strategy to be more cost-effective when compared to Diaskintest, it should cost USD 0.16. On the other hand, for the screening strategy with QFT-Plus to be more cost-effective when compared to that of Cy-TB, the QFT-Plus test should cost USD 0.91.

Discussion

In this cost-effectiveness model, we compared the Diaskintest screening strategy with the TST for Brazilian healthcare workers. The Diaskintest test was the most cost saving, followed by the Cy-TB test. The QFT-Plus test was more costly with slightly higher effectiveness. Diaskintest has also been shown to be cost saving in Brazilian PLWH ¹³13. Ministério da Saúde. Manual de recomendações para o controle da tuberculose no Brasil. 2nd Ed. Brasília: Ministério da Saúde; 2018.. The main reason for these findings is the costs of tests. All ESAT-6/CFP-10-based tests, whether in vivo (TBST) or ex vivo (QFT-Plus), exhibit similar accuracy ²⁵25. Krutikov M, Faust L, Nikolayevskyy V, Hamada Y, Gupta RK, Cirillo D, et al. The diagnostic performance of novel skin-based in-vivo tests for tuberculosis infection compared with purified protein derivative tuberculin skin tests and blood-based in vitro interferon-? release assays: a systematic review and meta-analysis. Lancet Infect Dis 2021; 22:250-64.. Consequently, their cost-effectiveness is based primarily on their costs. The QFT-Plus (and other World Health Organization [WHO]-approved IGRA tests) requires equipment, laboratory infrastructure, sample transportation from health facilities to the laboratory, and labor-intensive processes. Consumables (mainly the test kits) also contribute to the cost. In contrast, while the tuberculin unit is expensive, the test requires less labor (approximately 6.5 minutes total in low-and middle-income countries) ⁴⁰40. Alsdurf H, Oxlade O, Adjobimey M, Ahmad Khan F, Bastos M, Bedingfield N, et al. Resource implications of the latent tuberculosis cascade of care: a time and motion study in five countries. BMC Health Serv Res 2020; 20:341. and can be conducted in any clinic with a refrigerator for storing vials and properly trained personnel.

The WHO recommends TB infection tests for high-risk populations, such as PLWH and household contacts with bacteriologically confirmed pulmonary TB (regardless of age or HIV status) ⁴¹41. World Health Organization. WHO operational handbook on tuberculosis. Module 3: diagnosis. Tests for tuberculosis infection. Geneva: World Health Organization; 2022.. In Brazil, TB infection tests are mandatory for most individuals undergoing TB preventive treatment ⁴4. Ministério da Saúde. Protocolo de vigilância da infecção latente pelo Mycobacterium tuberculosis no Brasil. Brasília: Ministério da Saúde; 2022.. Our findings support the Brazilian recommendation and suggest that Diaskintest or another specific test with similar costs should be incorporated. However, there is currently more evidence on the safety and accuracy of the Cy-TB test, and decisions to incorporate new technologies into the health system follows a multicriteria logic. As for QFT-Plus to be more cost-effective in this population, it should cost less than USD 1.

While both Diaskintest and Cy-TB exhibited cost-saving advantages over alternative interventions, their overall savings are contingent upon various factors, including final costs, which are subject to fluctuations in exchange rates and the optimization of reagent supply chains. Factors such as storage requirements, dosage per vial, and reagent stability post-opening significantly influence final test costs and, consequently, their cost-effectiveness. Moreover, integrating these tests into programs targeting TB preventive treatment scale up could stimulate local innovation and foster competition to meet regional demand. Recurrent consumable stock-outs could also be overcome by the availability of different tests. An additional benefit compared to IGRAs is the absence of implementation costs, given their long-standing use as standard practice in Brazil and the current efforts to expand skin test competence in healthcare workers (ExpandTPT project, 2024; personal communication). Finally, skin tests are provided at the point of care, which enables a wider availability of these tests in rural and other remote areas.

An universal limitation of economic analyses of TB infection detection and treatment is related to the sensitivity of tests. Since no golden standard exists for TB infection diagnosis, these economic analyses are based on studies performed in populations with TB, in whom TST, TBST, and IGRA tests, which are based on immune response, may show lower sensitivity ⁴²42. Pai M, Zwerling A, Menzies D. Systematic review: T-cell-based assays for the diagnosis of latent tuberculosis infection: an update. Ann Intern Med 2008; 149:177-84.. Previous studies were also sensitive to TB infection incidence, a parameter with a high degree of uncertainty among healthcare workers ³3. Apriani L, McAllister S, Sharples K, Alisjahbana B, Ruslami R, Hill PC, et al. Latent tuberculosis infection in health care workers in low and middle-income countries: an updated systematic review. Eur Respir J 2019; 53:1801789.. We used a large range of TB infection incidence (9%-24%) in our sensitivity analysis, and there was no impact in the final conclusion.

This study shows several limitations. Firstly, decision analyses inherently depend on the quality and accuracy of their base-case modeling parameter values. We did not consider drug-resistant TB strains due to their modest prevalence in Brazil and the current recommendation not to treat contacts of persons with drug-resistant TB ⁴³43. Bastos ML, Cosme LB, Fregona G, do Prado TN, Bertolde AI, Zandonade E, et al. Treatment outcomes of MDR-tuberculosis patients in Brazil: a retrospective cohort analysis. BMC Infect Dis 2017; 17:718.. Moreover, we did not consider the entire TB spectrum (incipient and sub-clinical disease) ⁴⁴44. Drain PK, Bajema KL, Dowdy D, Dheda K, Naidoo K, Schumacher SG, et al. Incipient and subclinical tuberculosis: a clinical review of early stages and progression of infection. Clin Microbiol Rev 2018; 31:e00021-18., as no commercially available tests or treatment implications exist for other TB status. Additionally, our model is based on numerous assumptions. We tested the potential impact of major assumptions using sensitivity analyses, while others were assumed as fixed. For instance, our study did not model HIV infection, TB treatment dropout rates, or skin test booster effects. We also did not address the cost-effectiveness of baseline screening at hiring, which is necessary to identify workers who would benefit from subsequent tests and to interpret the results, regardless of the specific test and frequency adopted. We did not consider costs for repeated skin tests in cases of lost readings, but we estimated the rate of non-return at 3% ⁴⁵45. Souza FM, Prado TN, Pinheiro JS, Peres RL, Lacerda TC, Loureiro RB, et al. Comparison of interferon-c release assay to two cut-off points of tuberculin skin test to detect latent Mycobacterium tuberculosis infection in primary health care workers. PLoS One 2014; 9:e102773.. Finally, due to the lack of a threshold reference for case avoidance in Brazil, we opted to use the QALY threshold as a reference, which has been the metric used for the incorporation of technologies into the SUS, notwithstanding the limitations of this comparison ¹1. Departamento de Ciência e Tecnologia, Secretaria de Ciência, Tecnologia e Insumos Estratégicos, Ministério da Saúde. Diretrizes metodológicas: diretriz de avaliação econômica. Brasília: Ministério da Saúde; 2014.^,3838. Ministério da Saúde. O uso de limiares de custo-efetividade. Brasília: Ministério da Saúde; 2022..

Despite these limitations, this cost-effectiveness analysis can contribute to decision-making in Brazil and serve as a reference for economic evaluations in other countries. The findings suggest that alternatives to the QFT-Plus approach may be worth considering, as these new tests could potentially facilitate the scale-up of TB prevention programs and contribute to global TB control efforts. Importantly, these alternatives may not require venipuncture or the use of costly laboratory facilities.

References

Publication Dates

History