key: cord-0324323-boal4kyr
authors: Portnoy, A.; Clark, R. A.; Quaife, M.; Weerasuriya, C. K.; Mukandavire, C.; Bakker, R.; Deol, A. K.; Malhotra, S.; Gebreselassie, N.; Zignol, M.; Sim, S. Y.; Hutubessy, R.; Baena, I. G.; Nishikiori, N.; Jit, M.; White, R. G.; Menzies, N. A.
title: The cost and cost-effectiveness of novel tuberculosis vaccines in low- and middle-income countries: a modelling study
date: 2022-05-05
journal: nan
DOI: 10.1101/2022.05.04.22274654
sha: b6f94bc84d69d18ecc774eb2b8a291746ffafcbd
doc_id: 324323
cord_uid: boal4kyr

Background: Tuberculosis (TB) is preventable and curable but eliminating it has proven challenging. Safe and effective TB vaccines that can rapidly reduce disease burden are essential for achieving TB elimination. We assessed future costs, cost-savings, and cost-effectiveness of introducing novel TB vaccines in low- and middle-income countries (LMICs) for a range of product characteristics and delivery strategies. Methods: We developed a system of epidemiological and economic models, calibrated to demographic, epidemiological, and health service data in 105 LMICs. For each country, we assessed the likely future course of TB-related outcomes under several vaccine introduction scenarios, compared to a 'no-new-vaccine' counterfactual. We estimated the incremental impact of vaccine introduction for a range of health and economic outcomes. In the base-case, we assumed a vaccine price of $4.60, and used a 1x per-capita GDP cost-effectiveness threshold (both varied in sensitivity analyses). Findings: Vaccine introduction was estimated to require substantial near-term resources, offset by future cost-savings from averted TB burden. From a health system perspective, vaccination was cost-effective in 65 of 105 LMICs. From a societal perspective (including productivity gains and averted patient costs), vaccination was projected to be cost-effective in 75 of 105 LMICs and cost-saving in 57 of 105 LMICs, including 96% of countries with higher TB burden. When considering the monetized value of health gains, we estimated that introduction of an adolescent/adult vaccine could produce $258-447 billion in economic benefits by 2050. Interpretation: TB vaccination would be highly impactful and cost-effective in most LMICs.

Evidence before this study Previous studies have highlighted the economic impact of tuberculosis disease and mortality, and the potential economic impact that novel tuberculosis vaccines could have on reducing this burden in specific low-and middle-income countries (LMICs). The cost and cost-effectiveness of novel tuberculosis vaccines will depend on vaccine price and delivery strategy, which may vary by country. No modelling studies have estimated cost and cost-effectiveness with countryspecific assumptions for medical and non-medical costs, indirect costs, vaccine delivery costs, and delivery strategies across a wide range of LMICs.

We estimated the costs, cost-effectiveness, and incremental net monetary benefit of tuberculosis vaccine introduction from both the health system and societal perspective in order to inform global-level decision-making for novel tuberculosis vaccine investment and introduction. Using mathematical and economic models, we assessed scenarios for the introduction of novel vaccines with a wide range of characteristics and a diverse set of health and economic outcomes. We assumed country-specific introduction years from 2028-2047, determined based on indicators for disease burden, immunization capacity, classification of the country as an "early adopter/leader," lack of regulatory barriers, and commercial prioritization. We varied vaccination coverage, vaccine delivery strategy, vaccine price, duration of protection, and future trends in tuberculosis incidence in scenario analysis.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ;  https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint Our analysis projected that an effective new tuberculosis vaccine could offer large potential health and economic benefits over 2028-2050. From a societal perspective, vaccination was projected to be cost-effective in 75 LMICs compared to a 1x per-capita gross domestic product threshold. When considering the monetized value of health gains, we estimated that introduction of an adolescent/adult vaccine could produce $258-447 billion in economic benefits by 2050, with greater benefits in LMICs with elevated tuberculosis incidence.

Introduction of a new tuberculosis vaccine was found to be impactful and cost-effective for a range of assumptions on vaccine price and delivery strategies, with aggregate health and economic benefits of similar scale to the most influential health interventions in LMIC settings in recent years. The results of these analyses can be used by global and country stakeholders to inform vaccine policy and introduction preparedness, as well as decision-making around future development, adoption, and implementation of novel tuberculosis vaccines.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint 6 Background Tuberculosis (TB) is the world's second greatest cause of infectious disease death after COVID-19. 1 It remains the leading cause of death for people living with HIV and a major contributor to antimicrobial-resistance-related deaths. The COVID-19 pandemic has reversed years of progress in providing TB services and, consequently, the number of people who died from TB increased to 1•5 million in 2020. 1 The World Health Organization (WHO)'s End TB Strategy targets a 90% reduction in TB mortality and 80% decline in TB incidence by 2030, compared to 2015. 2 Achieving these targets will require a comprehensive multisectoral response, along with transformational new tools. The cost of not meeting the End TB Targets by 2030 and facing the excess deaths resulting from COVID-19-related disruptions to TB services may translate into 31•8 million TB deaths globally corresponding to an economic loss of $17•5 trillion between 2020 and 2050. 3 Developing new safe, affordable, and effective TB vaccines is critical for achieving these targets. While promising candidates exist (for example, the M72/AS01E candidate vaccine 4 ), limited market incentives to invest in TB prevention has delayed the development of novel TB vaccines.

The WHO promotes the Full Value of Vaccines Assessment framework to improve decisionmaking around vaccine development and use. 5, 6 Using this framework, we estimated the costs, cost-effectiveness, and net monetary benefit of TB vaccine introduction, from health system and societal perspectives, to inform global-level decision-making for novel TB vaccine investment and introduction. 5, 6 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint

We estimated a range of outcomes quantifying the health and economic impact of new vaccine introduction for affected countries. To do so, we used linked epidemiological and economic models to project changes in healthcare utilization, health outcomes, and healthcare costs for various vaccine introduction scenarios compared to a 'no-new-vaccine' counterfactual (full epidemiological model details have been previously described by Clark and colleagues 7 ). We estimated outcomes for each of 105 low-and middle-income countries (LMICs) over a 2028-2050 evaluation period (Appendix S1). We summarized results as the incremental costs, costeffectiveness, and incremental net monetary benefits of vaccine introduction. Results are presented for a range of analytic assumptions and introduction scenarios, organized by major country groupings (WHO region, World Bank income level, 8 and WHO high-TB burden grouping 9 ).

We constructed a 'no-new-vaccine' baseline with current TB interventions continuing into the future at current levels. Compared to this baseline, we evaluated two different vaccine product profiles: an infant 'pre-infection' vaccine (i.e., efficacious for individuals uninfected at time of vaccination) with 80% efficacy targeting neonates, and an adolescent/adult 'pre-and postinfection' vaccine (i.e., efficacious in all individuals aside from those with active TB disease at time of vaccination) with 50% efficacy, based on WHO preferred product characteristics. 10 For both vaccine product profiles, we assumed an average ten-year duration of protection, with exponential waning. We assumed the infant vaccine would be delivered through the routine vaccination program, and the adolescent/adult vaccine delivered through routine vaccination of . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint nine-year-olds plus a one-time vaccination campaign for ages 10+. In the base-case scenario, we assumed countries would achieve linear scale-up to a specified coverage over five years. Based on consultation with global stakeholders, we assumed a coverage target of 85% for the infant vaccine (average coverage of diphtheria-tetanus-pertussis third dose for LMICs 11 ), 80% for routine delivery of adolescent/adult vaccine, and 70% of the adolescent/adult vaccination campaign. 12 We assumed country-specific introduction years from 2028-2047, determined based on indicators for disease burden, immunization capacity, classification of the country as an "early adopter/leader," lack of regulatory barriers, and commercial prioritization. 7

We projected future TB epidemiology and health service utilization using an age-structured TB transmission model calibrated to reported demography, TB burden estimates, and TB service utilization in each modelled country. 7 Out of 135 LMICs, 8 we excluded 20 due to lack of critical calibration data and 10 due to unsuccessful calibration results. We analyzed the remaining 105 countries (Appendix S1), representing 93•3% of global TB burden. 13 In countries with a significant burden of HIV-associated TB, the model included the effects of HIV and antiretroviral therapy (ART) on TB infection and progression risks. Using this model, we estimated changes in TB epidemiology and related service utilization for each modelled scenario.

We estimated disability-adjusted life-years (DALYs) averted to quantify the health gains achieved by vaccination. To calculate years lost due to disability (YLDs), we assigned each modelled health state a disability weight from the Global Burden of Disease classification system (Appendix S2). 14 For each scenario and year, total YLDs were calculated by summing life-years . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint lived across all health states, weighted by the disability weight for each state. For each scenario and year, years of life lost were calculated by multiplying deaths at each year of age by reference life expectancy at that age 15 , and summing across all ages.

We estimated the costs of vaccine introduction, as well as changes in the costs of other health services (TB care, HIV care) by multiplying health service volume indicators (vaccines delivered, TB cases diagnosed and treated, ART patient-years) by country-specific unit costs.

Diagnostics costs for drug-susceptible (DS) and rifampicin-resistant (RR) TB were obtained from published literature, 16 and extrapolated to all LMICs by country income level. 8 Unit costs for TB treatment were calculated as an average of DS-TB 17 and RR-TB 16,18 treatment costs, weighted by country-level RR-TB prevalence. 1 For ART costs, direct non-medical costs (travel, accommodation, food, nutritional supplements) to the patient, and productivity costs (income loss experienced by patients during TB care), we derived unit costs by extrapolating estimates reported by the Global Health Cost Consortium 19 (sample size = 39) and WHO patient cost surveys (sample size = 20). 20, 21 Productivity costs due to premature death were estimated as the incremental number of life-years gained under a given vaccination scenario, multiplied by 2020 per-capita GDP as an approximation of income.

As the per-dose cost for novel TB vaccines is unclear while products are still under development, the base-case used an LMIC price of human papillomavirus (HPV) vaccine ($4•60) for a novel vaccine proxy with an injection supply cost per dose of $0•11 and 5% wastage. 22, 23 Countryspecific vaccine delivery costs were based on a meta-analysis of childhood 24 and HPV vaccine delivery unit costs for the infant and adolescent/adult vaccines, respectively, plus additional one-. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint time vaccine introduction costs ($0•65 and $2•40 per targeted individual in the first year of introduction for infant and adolescent/adult vaccines, respectively). 25 Costs are reported in 2020 US dollars.

Incremental cost-effectiveness ratios (ICERs) were calculated from health system and societal perspectives, with a 3% discount rate, across the 2028-2050 evaluation period. We also reported a specification in which costs are discounted but not health outcomes. The health system perspective considered costs of vaccine introduction, plus the costs of TB and HIV services indirectly affected by vaccine introduction. The societal perspective additionally included patient non-medical and productivity costs. ICERs were compared to a range of country-specific costeffectiveness thresholds, including multiples of per-capita GDP 26 and recent estimates of the opportunity cost of healthcare spending. 27,28

We quantified the return on investment as the incremental net monetary benefit (iNMB) from the societal perspective of each vaccine scenario compared to baseline for a range of willingness-topay thresholds. [26] [27] [28] iNMB was calculated as the sum of monetized health gains (DALYs averted multiplied by the estimated willingness-to-pay per DALY averted) minus incremental costs. We estimated the market size for each vaccine product profile, summing all individuals across 2028-2050 who were vaccinated in the model in the base-case scenario in countries in which the vaccine was cost-effective (ICER less than per-capita GDP). We also estimated market size based on countries in which vaccination was cost-saving under the societal perspective.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022.

We explored estimation uncertainty using a 2 nd -order Monte Carlo simulation. 29 We constructed probability distributions representing uncertainty in economic inputs and disability weights and drew 1000 random values for each uncertain parameter. We represented uncertainty in healthcare utilization and epidemiological outcomes (counts of each outcome by scenario, year, and population stratum) using 1000 results sets from the transmission-dynamic model. This analysis generated 1000 estimates for each outcome of interest, which we summarized as equal-tailed 95% uncertainty intervals.

Compared to the base-case coverage targets (85%, 80%, 70% for routine infant vaccine delivery, routine adolescent vaccine delivery, and campaign adolescent/adult vaccine delivery, respectively), we examined a low-coverage scenario (75%, 70%, and 50%, respectively) and a high-coverage scenario (95%, 90%, and 90%, respectively). We examined three alternative vaccine price scenarios, including scenarios in which the basecase vaccine price of $4•60 was both halved ($2•30) and doubled ($9•20), respectively. A third scenario examined high-middle-tier vaccine pricing, with higher prices for middle-income countries based on UNICEF vaccine pricing data ($10•25 for non-Gavi countries with gross . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint national income (GNI) per capita less than $3,995 and $14•14 for non-Gavi countries with GNI per capita greater than $3,995; Appendix S1). 22 We also estimated results with an alternative set of assumptions about TB incidence trends in the no-new-vaccine baseline, with incidence assumed to decline more rapidly through the scale-up of existing preventive treatment and case detection, meeting the 2025 'End TB' incidence reduction target without introduction of a new vaccine. 2 Compared to the base-case assumption of ten-year duration of protection, we also examined lifelong duration of protection conferred by vaccination.

The funder was involved in developing the research question, study design, and provided comments on the manuscript draft, but had no role in the collection, analysis, and interpretation of the data, or writing of the report. All authors had the opportunity to access and verify the data, and all authors were responsible for the decision to submit the manuscript for publication. There was greater, and more rapid, impact from an adolescent/adult vaccine compared to an infant vaccine over the study period (Appendices S8-S9). Across 2028-2050, infant vaccine costs were projected to increase smoothly from the year of vaccine introduction, whereas the adolescent/adult vaccine scenario required major upfront investments during vaccine introduction and 5-year campaign roll-out, then decreased substantially after campaigns were completed.

In the base-case analysis, from the health system perspective, we found that infant vaccination would be cost-effective (ICER below 1-times per-capita GDP) compared to no vaccination in 48 of 105 modelled LMICs (46%) and 26 of 27 with high-TB burden (96%). Using the same assumptions, we found that adolescent/adult vaccination would be cost-effective in 65 out of 105 countries (62%) and all 27 with high-TB burden. Neither vaccine strategy would be cost-saving in any country. Figure 1 displays the distribution of country-level cost-effectiveness results from the health system perspective for infant and adolescent/adult vaccines, stratified by TB incidence . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint level. Vaccine introduction was more likely to be cost-effective in countries with higher TB incidence.

From the societal perspective, the infant vaccine was cost-effective in 55 out of 105 countries (52%), including all with high-TB burden, and cost-saving in 47 countries (45%). Similarly, the adolescent/adult vaccine was cost-effective in 75 out of 105 countries (71%), remaining costeffective in all with high-TB burden, and cost-saving in 57 countries (54%). Figure 2 displays the percentage of the modelled population that live in countries where vaccination was costeffective based on different cost-effectiveness thresholds (Appendix S10 shows the percentage of countries where vaccination was cost-effective; Appendices S11-S12 present tabular results).

Tables 1 and 2 report summary health outcomes, costs, and cost-effectiveness of the base-case vaccination scenarios. Across all 105 analyzed countries, the majority of TB cost-savings accrued in high-TB-burden settings, particularly in lower middle-income settings and WHO African region (AFR) and South-East Asian region (SEAR). Assuming 0% discounting on health outcomes decreased ICERs (indicating greater cost-effectiveness) for the infant vaccine by approximately 76% and for the adolescent/adult vaccine by approximately 69% from the health system perspective (Appendices S13-S14).

With each averted DALY valued at per-capita GDP and costs assessed from the societal perspective, we estimated a cumulative $50•8 (range: $32•6-74•7 across examined thresholds) billion incremental net monetary benefit (iNMB) globally for infant vaccine introduction in countries where introduction was cost-effective at 1-times per-capita GDP (Figure 3 ; tabular results in Appendix S15). For the adolescent/adult vaccine, we estimated iNMB of $343 billion . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. (Appendices S16-S17).

From both health system and societal perspectives, DALYs averted and costs decreased in the low-coverage scenario, and increased in the high-coverage scenario, for both the infant and the adolescent/adult vaccine, with evidence of diminishing returns as coverage increases (Appendices S18-S21).

Compared to the base-case vaccination introduction and delivery scenario, the accelerated scaleup scenario had greater health impact (DALYs averted) and better cost-effectiveness (assuming per-unit vaccination costs were unchanged), with vaccination being cost-effective in a greater number of countries compared to a per-capita GDP threshold (i.e., 70 compared to 48 for the infant vaccine and 98 compared to 65 for the adolescent/adult vaccine from the health system perspective; Appendices S22-S25). Conversely, the routine-only scenario had a much smaller health impact and modestly worse cost-effectiveness profile, as compared to the base-case analysis (Appendix S26).

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint DALYs (36% greater than the base-case; Appendix S40). Both vaccine products remained costsaving from the societal perspective; assuming lifelong duration of protection decreased the ICER by approximately 40% and 35% for the infant and adolescent/adult vaccine, respectively (health system perspective).

An effective novel TB vaccine would offer large potential health and economic benefits over 2028-2050. These results demonstrate that, when available, TB vaccines could be cost-effective in a majority of LMICs, particularly from the societal perspective, and essentially in all highburden countries. Introducing novel TB vaccines could also offer high value in terms of incremental net monetary benefit to patients, the health system, and society, particularly in countries with high burden of TB, HIV-associated TB, and/or multidrug-resistant/RR-TB.

For both vaccine product profiles, vaccination was more likely to be cost-effective in lower middle-income countries (relative to low-income and upper middle-income countries), as countries in this income group are more likely to have both significant TB burden and sufficient economic resources to justify additional TB investments without displacing other important health interventions. Vaccination was more frequently cost-effective in AFR and SEAR regions and both vaccines were estimated to be cost-effective in all countries in the 27 modelled high-TB burden countries that accounted for 81•8% of global incident TB cases and 80•9% of global TB deaths in 2020. 9, 13 Although vaccines can be economically less viable for manufacturers, we estimated large potential markets for vaccinees in high-burden, middle-income settings.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint There was greater, and more rapid, impact from an adolescent/adult vaccine over the 2028-2050 time horizon compared to an infant vaccine, as this vaccine is targeted to a population with the highest burden of TB, and the delay between vaccination and TB prevention impact is shorter with the adolescent/adult vaccine. For the adolescent/adult vaccine, we estimated major shortterm costs from introduction and one-time vaccination campaigns, with the highest costs incurred during the ten years following vaccine introduction. In contrast, the cost-savings from averted TB disease were realized gradually over 2028-2050, growing in magnitude towards the end of the time horizon. By assuming the no-new-vaccine baseline meeting the End TB targets, the remaining TB burden that could be averted by vaccination was estimated to be smaller, yielding results that were less cost-effective.

This analysis had several limitations. We were constrained by data availability, with only 105 countries successfully parameterized and calibrated. However, these 105 countries represent 93•3% of LMIC TB incidence and 93•6% of LMIC TB mortality globally in 2020. 13 As a "preand post-infection" vaccine, the adolescent/adult vaccine was assumed to be equally effective regardless of previous infection status, which may have led to an overestimation of averted TB cases and deaths if the vaccine is less effective in infected vaccinees. We also extrapolated from published literature [16] [17] [18] for several major unit cost inputs, potentially omitting important countrylevel heterogeneity in these costs. The sample size of patient cost surveys used for non-medical and productivity costs was small (20) ; therefore, the extrapolation to other country settings may not capture the level of potential variation in these costs. We did not investigate targeting highrisk subgroups for vaccination; vaccination could still be cost-effective when targeted to subgroups in settings where vaccination was not estimated to be cost-effective in a national roll-. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint out. Finally, we did not consider all possible product and introduction scenarios, but demonstrated the potential value of novel TB vaccines according to specified characteristics.

Across this analysis, introduction of a novel TB vaccine was found to be impactful and costeffective for a range of assumptions on vaccine price and delivery strategies, with aggregate health and economic benefits of similar scale to the most influential health interventions in LMIC settings in recent years. 30 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) cost-saving c a Costs from the health system perspective include vaccination costs, tuberculosis testing and treatment costs, and antiretroviral treatment costs. b Costs from the societal perspective include health system perspective costs, as well as patient non-medical costs and productivity losses. c Both the point estimate and the interval estimates are cost-saving. d High-TB, high-TB/HIV (HIV-associated TB), and high-MDR/RR-TB (multidrug/rifampicin-resistant TB) burden countries as defined by the World Health Organization. 9 e LIC: Gross national income (GNI) per capita of $1,025 or less; LMIC: GNI per capita of $1,026 to $3,995; UMIC: GNI per capita of $3,996 to $12,375 (World Bank 2019). Note: All countries include 105 low-and middle-income countries analyzed. AFR = African region; AMR = Region of the Americas; EMR = Eastern Mediterranean region; EUR = European region; GDP = gross domestic product; LIC = low-income; LMIC = lower middle-income; SEAR = Southeast Asian region; UMIC = upper middle-income; WPR = Western Pacific region. Table 2 . Discounted costs, disability-adjusted life-years (DALYs) averted, and cost-effectiveness of adolescent/adult tuberculosis vaccines Figure 1 . Cost-effectiveness results from the health system perspective for novel tuberculosis vaccines by country and vaccine.

Note: Points represent each of 105 low-and middle-income countries analyzed in the base-case scenario, stratified by tuberculosis incidence per 100,000. Line represents a cost-effectiveness threshold of 1x per-capita GDP in 2020. Vaccine introduction would be considered cost-effective for countries falling underneath this line. DALY = disability-adjusted life year; GDP = gross domestic product.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint Figure 2 . Percentage of population that live in countries where vaccination was costeffective compared to percentage of GDP per capita thresholds, comparing health system and societal perspectives. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint Figure 3 . Incremental net monetary benefit of novel tuberculosis vaccines assessed from the societal perspective, for several willingness-to-pay thresholds.

Note: Estimates include the incremental net monetary benefit from the countries that are cost-effective at the respective threshold. 26 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 5, 2022. ; https://doi.org/10.1101/2022.05.04.22274654 doi: medRxiv preprint

Geneva: World Health Organization

World Health Organization. The End TB Strategy. Geneva: World Health Organization

Economic impact of tuberculosis mortality in 120 countries and the cost of not achieving the Sustainable Development Goals tuberculosis targets: a full-income analysis

Final Analysis of a Trial of M72/AS01(E) Vaccine to Prevent Tuberculosis

The Full Value of Vaccine Assessments (FVVA): A Framework to Assess and Communicate the Value of Vaccines for Investment and Introduction Decision Making

The case for assessing the full value of new tuberculosis vaccines

The impact of alternative delivery strategies for novel tuberculosis vaccines in low-and middle-income countries: a modelling study

World Bank Country and Lending Groups

World Health Organization. WHO releases new global lists of high-burden countries for TB, HIV-associated TB and drug-resistant TB

Geneva: World Health Organization

World Health Organization. WHO-UNICEF estimates of DTP3 coverage: monitoring system 2019 global summary

Potential impact of tuberculosis vaccines in China, South Africa, and India

World Health Organization. WHO TB Burden Estimates. Geneva: World Health Organization

Disability weights for the Global Burden of Disease 2013 study

United States of America: Institute for Health Metrics and Evaluation (IHME)

Costs to Health Services and the Patient of Treating Tuberculosis: A Systematic Literature Review

Cost of tuberculosis treatment in low-and middleincome countries: systematic review and meta-regression

The socioeconomic impact of multidrug resistant tuberculosis on patients: results from Ethiopia, Indonesia and Kazakhstan

Tuberculosis patient cost surveys: a hand book. Geneva: World Health Organization

World Health Organization. National surveys of costs faced by TB patients and their households

Vaccines pricing data

Producing Standardized Country-Level Immunization Delivery Unit Cost Estimates

Gavi The Vaccine Alliance. GAVI Alliance Vaccine Introduction Grant and Operational Support for Campaigns Policy. Version No. 1.0. Geneva: Gavi, The Vaccine Alliance

World development indicators

Estimating health opportunity costs in low-income and middle-income countries: a novel approach and evidence from cross-country data

Country-Level Cost-Effectiveness Thresholds: Initial Estimates and the Need for Further Research

Model parameter estimation and uncertainty: a report of the ISPOR-SMDM Modeling Good Research Practices Task Force--6

The Global Fund. Fight for What Counts: The Global Fund Investment Case

We are grateful for the support of the World Health Organization for funding this research (2020/985800-0). We thank all the attendees at the WHO meetings on the Full Value Assessment of TB Vaccines for insightful advice and direction.

Not applicable. All authors read and approved the final manuscript.

Scholarship Fund. CKW is funded by UKRI/MRC (MR/N013638/1). RGW is funded by the

cost-saving c a Costs from the health system perspective include vaccination costs, tuberculosis testing and treatment costs, and antiretroviral treatment costs. b Costs from the societal perspective include health system perspective costs, as well as patient non-medical costs and productivity losses. c Both the point estimate and the interval estimates are cost-saving. d High-TB, high-TB/HIV (HIV-associated TB), and high-MDR/RR-TB (multidrug/rifampicin-resistant TB) burden countries as defined by the World Health Organization. 9 e LIC: Gross national income (GNI) per capita of $1,025 or less; LMIC: GNI per capita of $1,026 to $3,995; UMIC: GNI per capita of $3,996 to $12,375 (World Bank 2019). Note: All countries include 105 low-and middle-income countries analyzed. Values in parentheses represent equal-tailed 95% credible intervals. AFR = African region; AMR = Region of the Americas; EMR = Eastern Mediterranean region; EUR = European region; GDP = gross domestic product; LIC = low-income; LMIC = lower middle-income; SEAR = Southeast Asian region; UMIC = upper middle-income; WPR = Western Pacific region.