Maternal immunisation against Group B Streptococcus: a global analysis of health impact and cost-effectiveness

1 1 Title: Maternal immunisation against Group B Streptococcus: a global analysis of health impact and 2 cost-effectiveness 3 4 Authors 5 Simon R. Procter1,2*, Bronner P. Gonçalves1,2, Proma Paul1,2, Jaya Chandna1,2, Farah Seedat1,2, Artemis 6 Koukounari1,2, Raymond Hutubessy3, Caroline Trotter4, Joy E Lawn1,2, Mark Jit1,5* 7 8 *Corresponding authors: [email protected]; [email protected] 9 10 Affiliations 11 1. Department of Infectious Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, United 12 Kingdom 13 2. Maternal, Adolescent, Reproductive & Child Health (MARCH) Centre, London School of Hygiene & Tropical Medicine, 14 London, United Kingdom 15 3. Department of Immunization, Vaccines and Biologicals (IVB), World Health Organization, Geneva, Switzerland 16 4. Disease Dynamics Unit, Department of Veterinary Medicine, University of Cambridge, Cambridge, United Kingdom 17 5. School of Public Health, University of Hong Kong, Hong Kong SAR, China 18 19 20 . CC-BY 4.0 International licenseIt 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 (which was not certified by peer review) The copyright holder for this preprint this version posted July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 2 21 Abstract 22 Background 23 Group B Streptococcus (GBS) can cause invasive disease (iGBS) in young infants, typically presenting 24 as sepsis or meningitis, and is also associated with stillbirth and preterm birth. GBS vaccines are 25 under development, but their potential health impact and cost-effectiveness have not been assessed 26 globally. 27 28 Methods 29 We assessed the health impact and value (using net monetary benefit, NMB, which measures both 30 health and economic effects of vaccination into monetary units), of GBS maternal vaccination across 31 183 countries in 2020. Our analysis uses a decision-tree model, combining risks of GBS-related 32 outcomes from a Bayesian disease burden model with estimates of GBS related costs and Quality- 33 Adjusted Life Years (QALYs) lost. We assumed 80% vaccine efficacy against iGBS and stillbirth, 34 following the WHO Preferred Product Characteristics, and coverage based on the proportion of 35 pregnant women receiving at least four antenatal visits. One dose was assumed to cost $50 in high- 36 income countries, $15 in upper-middle income countries, and $3.50 in low-/lower-middle income 37 countries. We estimated NMB using alternative normative assumptions that may be adopted by 38 policy makers. 39 40 Findings 41 Vaccinating pregnant women could avert 214,000 (95% uncertainty range 151,000 – 457,000) infant 42 iGBS cases, 31,000 deaths (14,000 – 67,000), 21,000 (9,000 – 52,000) cases of neurodevelopmental 43 impairment, and 23,000 (10,000 – 58,000) stillbirths. A vaccine effective against GBS-associated 44 prematurity might also avert 172,000 (13,000 – 378,000) preterm births. Globally, a 1-dose vaccine 45 programme could cost $1.7 billion but save $385 million in healthcare costs. Estimated global NMB 46 ranged from $1.1 billion ($-0.2 – 3.8 billion) to $17 billion ($9.1 – 31 billion). 47 48 Interpretation 49 Maternal GBS vaccination could have a large impact on infant morbidity and mortality globally and 50 at reasonable prices is likely to be cost-effective. 51 52 . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 3 53 INTRODUCTION 54 Streptococcus agalactiae, commonly known as Group B Streptococcus (GBS), is an important 55 bacterial pathogen causing morbidity and mortality in pregnant women and their babies and is also 56 increasingly recognised as a cause of disease in non-pregnant adults.1–3 Invasive GBS (iGBS) disease 57 in neonates and young infants can result from maternal colonisation and vertical transmission or 58 environmental exposure after birth. It is classified by age at onset with early-onset GBS (EOGBS) 59 occurring in the first 6 days of life, and late-onset disease (LOGBS) occurring between ages 7 and 89 60 days, and typically presents as sepsis, meningitis, or pneumonia. In 2020, an estimated 20 million 61 pregnant women globally were colonised with GBS resulting in 231,000 (114,000 – 455,000) cases of 62 EOGBS and a further 162,000 (70,000 – 394,000) LOGBS cases.1 Together these were estimated to 63 have caused 58,000 to 91,000 infant deaths depending on the assumptions made about mortality in 64 cases without access to healthcare. Furthermore, survivors of iGBS are at risk of long-term 65 neurological sequelae with an estimated 37,100 (14,600 – 96,200) surviving infants developing 66 moderate or severe neuro-developmental impairment (NDI).1,4 Maternal colonisation with GBS is 67 also an important cause of adverse pregnancy outcomes with an estimated 46,000 (20,000 – 68 111,000) GBS stillbirths and is potentially linked with 518,000 (36,000 – 1,142,000) excess preterm 69 births. 70 71 Currently the main strategies for preventing iGBS are based on intrapartum antibiotic prophylaxis 72 (IAP). Many higher-income countries have reduced EOGBS incidence through IAP with eligible 73 pregnant women identified either through risk-factor based screening or routine testing based on 74 microbiological culture.5 Despite this success IAP has several limitations, notably it is not effective 75 against LOGBS or GBS-associated stillbirths. In addition, the need for access to laboratory testing for 76 microbiological screening based strategies, and the requirement to deliver antibiotics intravenously 77 substantially limits the prospect of attaining high IAP coverage in many low-resource settings where 78 the burden of iGBS is highest.5 There are also concerns that routine administration of antibiotics 79 could contribute to antimicrobial resistance and might also have unintended impacts on the gut 80 microbiota of newborns.6 Hence, there is substantial interest in alternative approaches to 81 prevention. 82 83 Maternal immunisation is a potential alternative strategy whereby vertical transfer of antibodies in 84 utero from a woman vaccinated during pregnancy affords protection to the mother, unborn foetus 85 and newborn infant.7 Maternal immunisation with Tetanus Toxoid has been successfully used to 86 reduce the burden of neonatal tetanus since the 1970s and, in the last decade, countries have been 87 increasingly recommending routine vaccination of pregnant women against influenza and pertussis.8 . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 4 88 In 2015 development of a maternal vaccine against GBS was identified as a priority by the WHO 89 Product Development for Vaccines Advisory Committee (PDVAC),9 and three GBS maternal vaccine 90 candidates have progressed to Phase-2 clinical trials.10 In 2021, the licensure of an affordable GBS 91 vaccine by 2026 was identified as a key milestone in the WHO global roadmap for Defeating 92 Meningitis by 2030.11 93 94 There have been previous economic evaluations of maternal GBS vaccination in the United States,12– 95 14 Europe,15–17 and Sub-Saharan Africa.18–20 However, none of these studies have estimated the value 96 of GBS vaccination in all world regions. A global economic evaluation of GBS vaccination is important 97 to drive investment into vaccine development by indicating the vaccine’s potential value in different 98 markets. It would also enable financing and pricing mechanisms to be put in place for equitable 99 access to the vaccine once it is available. Such an evaluation is central to a Full Value of Vaccines 100 Assessment (FVVA), which WHO has identified as key to catalysing vaccine development and 101 subsequent equitable access.21,22 To inform the WHO GBS vaccine FVVA,23 we conducted the first 102 global economic evaluation of maternal GBS vaccination in 183 countries, drawing on recently 103 updated global disease burden estimates for GBS.1 104 105 METHODS 106 Model overview 107 We developed a decision-tree model (Fig. 1) to assess the health impact and cost-effectiveness of 108 maternal vaccination against GBS in an annual cohort of pregnant women and their babies for the 109 year 2020 compared with current practice of no vaccination. The size of the cohort of women in 110 each country was calculated by combining country-specific estimates of the number of births from 111 the United Nations World Populations Prospects (UNWPP)24 together with the stillbirth risk from the 112 WHO Global Health Observatory.25 Our analysis included the 183 countries out of 195 UN member 113 states for which UNWPP birth data was available, which excludes countries with estimated 114 populations below 90,000. 115 116 The health impact model structure was designed to reflect the natural history of pregnancy related 117 GBS infections and was aligned with the modelling framework used in recently reported global 118 estimates of GBS burden.1,26 The model first stratifies pregnant women based on GBS colonisation 119 status, and then by whether pregnancy results in a live birth or stillbirth. Live births are further sub- 120 divided into preterm and term births, with infants then at risk of developing either EOGBS or LOGBS; 121 the risk of EOGBS amongst babies born to non-colonised mothers was assumed zero. Invasive GBS . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 5 122 disease (EOGBS or LOGBS) may then result in death or, amongst survivors of GBS sepsis or 123 meningitis, either full recovery or long-term neurodevelopmental impairment. 124 125 The analysis used the lifetime of babies as the analytical time-horizon with health costs and Quality 126 Adjusted Life Years (QALYs) calculated over the lifetime of infant iGBS survivors using country- 127 specific life expectancy at birth.24 The model was used to compare scenarios with vaccination plus 128 current practice against current practice without vaccination (i.e. assuming no change in each 129 country’s IAP policy following vaccine introduction). All analyses were performed using R version 130 4.0.2. A Consolidated Heath Economic Evaluation Reporting Standards (CHEERS) checklist is provided 131 in supplementary appendix 1. 132 133 Figure 1: Decision tree for GBS-related outcomes in children for an annual birth cohort in 183 countries 134 comparing maternal vaccination against no vaccination (current standard of care) Numbered boxes 135 represent repeated model structure, however the risks for some outcomes vary across repeated branches. 136 137 Disease risk 138 Model inputs are summarised in supplementary appendix A2. We parameterised the probability of 139 different GBS-related outcomes in our model using posterior samples of key epidemiological 140 parameters from the global burden estimates reported by Gonçalves et al.1 We used country-specific 141 estimates of the prevalence of maternal GBS colonisation and of the risk of EOGBS in infants born to 142 colonised mothers. The risks of LOGBS were then calculated using region-specific estimates of the Mother not colonised EOGBS LOGBS Stillbirth Mother colonised iGBS death iGBS survivor Sepsis Meningitis Livebirth Term Preterm Pregnant women per country No iGBS No NDI Mild NDI Mod. NDI Severe NDI No NDI Mild NDI Mod. NDI Severe NDI Maternal vaccination No vaccination 1 21 2 3 3 4 4 . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 6 143 fraction of iGBS cases that are EOGBS vs LOGBS. Regional classifications were based on the United 144 Nations (UN) Sustainable Development Goal (SDG) region definitions.27 145 146 Case fatality risks (CFR) for EOGBS and LOGBS were also based on regional level estimates from 147 Gonçalves et al. There are no data on CFRs for infants with EOGBS without access to care, so the 148 authors considered two scenarios where they had either 90% CFR (following the approach of Seale 149 and co-workers3) or the same CFR as other infants with EOGBS. In our analysis, we assumed in the 150 base case that these infants had the same CFR as other infants with EOGBS, to be conservative about 151 this highly uncertain parameter and because mothers of these children might also be less likely to 152 receive maternal vaccines. 153 154 Among iGBS survivors the proportion of sepsis and meningitis, and the excess risk of mild, moderate, 155 and severe NDI outcomes after meningitis were based on pooled global estimates, while NDI risks 156 after sepsis were based on separate estimates for high-income and for low- and middle-income 157 countries. The excess risk attributable to iGBS exposure was calculated assuming a counterfactual 158 risk of mild or moderate and severe NDI amongst unexposed children from a large Danish cohort 159 study.28 We based the proportion of moderate and severe NDI that was severe on the same study. 160 Following the approach used in the burden estimation, our base case analysis included only the 161 excess risk of moderate or severe NDI, which is likely to be more consistent across settings, but 162 include mild NDI as a sensitivity analysis.1,4 163 164 To estimate country-specific GBS-associated stillbirth risk, national stillbirth estimates from the WHO 165 Global Health Observatory25 were combined with regional estimates of the proportion of stillbirths 166 caused by GBS.1 For the risk of GBS-associated prematurity we used national data on the proportion 167 of preterm births29 together with the global odds ratio for the association between GBS maternal 168 colonisation and preterm births.1 Further details on these calculations are provided in 169 supplementary appendix A2.3 and A2.4. 170 171 Health Outcomes 172 To calculate QALYs we assumed country-specific life-expectancy at birth for both normal births and 173 survivors of iGBS. For term births we assumed no reduction in Health-Related Quality-of-Life 174 (HRQoL), but for preterm births we applied a utility decrement over the child’s lifetime based on a 175 systematic review and meta-analysis by Petrou et al.30 For the acute iGBS episode we approximated 176 QALY loss assuming 29 days duration based on the average length-of-stay among studies in a recent . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 7 177 systematic review of the acute costs of infant sepsis and meningitis,31 and applied health state utility 178 decrements for hospitalisation with acute sepsis or meningitis from a US study in young children.32 179 For survivors with long-term sequelae, we applied utility decrements for mild, moderate, and severe 180 NDI to each year of life and conservatively, given previous studies provide evidence of post-acute 181 mortality after bacterial meningitis,33,34 assumed no change in life-expectancy. These utility values 182 were based on a UK study, which assessed HRQoL in a cohort of children with NDI followed up at age 183 11.35 184 185 Vaccination 186 Although clinical studies have demonstrated immunogenicity of candidate GBS vaccines, to date 187 there have been no phase-3 efficacy trials.10 We therefore based our assumptions about vaccine 188 efficacy (VE) and other characteristics of a GBS vaccine on the WHO preferred product 189 characteristics (PPCs).36 In our base case we assumed a single-dose vaccine with 80% efficacy against 190 both infant iGBS disease and GBS-stillbirth across all GBS serotypes. We also assumed no effect on 191 GBS-associated prematurity because (i) the WHO PPC does not specify that GBS vaccines must 192 reduce colonisation, which is most likely pathway for preventing GBS-associated prematurity, and (ii) 193 the association between GBS maternal colonisation and higher risk of prematurity may be 194 confounded.37 It is likely that delivery of GBS vaccines will need to be timed in either the late second 195 trimester or third trimester and could be delivered through existing routine antenatal care (ANC) 196 services. Hence, we assumed vaccine coverage based on the proportion of pregnant women in each 197 country who attend at least four ANC visits (ANC4).25 198 199 We also considered a range of alternative scenarios (supplementary table 4): higher vaccine 200 coverage based on the proportion of women attending at least one ANC visit (ANC1); a two-dose 201 regimen; lower and higher VE (60% and 90%); and a vaccine that is also effective against GBS- 202 associated prematurity. For the latter scenario we estimated the proportion of preterm births that 203 are potentially protected through vaccination by combining the distribution of preterm births by 204 gestational age38 with the timing of vaccine visits based on country-specific ANC data39 205 (supplementary appendix A2.5). 206 207 Costs 208 Our analysis was undertaken from a healthcare payer economic perspective, and all costs are 209 reported in 2020 United States Dollars (USD). Where cost inputs were for different years they were 210 inflated using the World Bank Gross Domestic Product (GDP) deflator.40 Costs reported in different . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 8 211 currencies were then converted to 2020 USD using historical foreign exchange rates.41 To estimate 212 acute healthcare costs, we combined one GBS-specific cost estimate from a study in the United 213 Kingdom,42 with the findings from a systematic review on the acute costs of infant sepsis and 214 meningitis43, and result of a recent study reporting the acute costs of neonatal bacterial sepsis and 215 meningitis in Mozambique and South Africa.44 We used linear regression to extrapolate country- 216 specific cost estimates using total per capita healthcare expenditure as a predictor (supplementary 217 appendix A2.6.) For long-term costs, no direct GBS-specific estimates exist in the literature. nnual 218 costs amongst survivors with moderate and severe NDI were parameterised as a fixed proportion of 219 between 4% and 28% of the acute cost estimate in each country, based on the range between a UK 220 study of costs in children with NDI35 and a US study of costs in adults with disabilities.45 221 222 For the vaccine programme costs we extrapolated results from a systematic review of maternal 223 vaccination delivery costs using regression against GDP per capita (supplementary appendix A2.7).31 224 We used previously estimated vaccine prices by World Bank country income group, which were 225 based on a combination of price benchmarking against other vaccines and cost of goods analysis: 226 $50 for high-income countries; $15 for upper-middle-income countries; and $3.50 for lower-middle- 227 income and low-income countries.46 228 229 Normative assumptions 230 A health intervention may be considered cost-effective if the cost per QALY gained falls below that 231 country’s cost-effectiveness threshold. Here we use two commonly cited thresholds: (i) country 232 gross domestic product per capita,47 (ii) published thresholds based on empirical estimates of the 233 health opportunity cost of health care spending (supplementary appendix A2.8).48,49 234 235 A second normative assumption is the QALY loss attributed to a stillbirth. In many settings these are 236 not assigned any health or disability weight, but it has been argued that they should be assigned a 237 QALY loss close or the same as that of the death of a newborn.50 Here we consider two scenarios, 238 one in which stillbirths are not assigned any QALY loss, and a second in which they are assigned the 239 same QALY loss as the death of a newborn. 240 241 Following WHO guidelines, we discount costs at 3% and health effects at both 0% and 3% in 242 alternative scenarios.51 Table 1 summarises the normative scenarios used. 243 244 . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 9 245 Table 1: Parameter values used for least and most favourable normative assumptions Parameter Least favourable Assumptions Most favourable assumptions Discount rate 3% for costs and benefits 3% for costs; 0% for benefits Inclusion of stillbirth quality- adjusted life-years Not included Included Cost-effectiveness threshold Based on empirical estimates* 1 x GDP per capita 246 *Cost-effectiveness thresholds were based either on estimates from Ochalek et al. and Woods et al. See 247 supplementary appendix A2.8 for more detail. 248 249 Economic analysis 250 To assess the cost-effectiveness of GBS maternal vaccination compared to current practice, we 251 follow a Net Monetary Benefit (NMB) approach in which both the health and fiscal benefits of 252 vaccination are expressed in monetary units.52 To calculate the NMB the incremental benefits in 253 QALYs are multiplied by a country-specific cost-effectiveness threshold (CET; either empirical or 1 x 254 GDP per capita) in USD and then the incremental costs are subtracted. An intervention may be 255 considered cost-effective if the NMB is positive, since this is mathematically equivalent to the 256 incremental cost-effectiveness ratio (ICER) being less than the CET. In addition to NMB, we also 257 estimated the threshold price per dose at which a GBS vaccine would be cost-effective in each 258 country. 259 260 An advantage of adopting an NMB framework is that our estimates for individual countries can be 261 directly combined to estimate the aggregate value of vaccination both regionally and globally. To 262 account for parameter uncertainty, for each scenario we ran 4000 simulations per country and 263 calculated the median and the 95% uncertainty range (UR) based on 2.5 and 97.5 percentiles of the 264 simulations. At the country level we also calculated the probability maternal GBS vaccination was 265 cost-effective (i.e., the proportion of simulations with NMB > 0). 266 267 Role of the funding source 268 The funder of the study had no role in study design, data collection, data analysis, data 269 interpretation, or writing of the report. 270 271 RESULTS . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 10 272 We estimate that vaccinating 99.8 million pregnant women across 183 countries could cost $1.7 273 billion but could save around $300 million in acute healthcare costs and $85 million in long-term 274 healthcare costs, although these estimates have wide uncertainty. Overall, the incremental cost of 275 GBS vaccination is about $1.3 billion, with the biggest cost increase in Europe and Northern America 276 (Table 2). 277 278 Globally, the vaccine programme could avert an estimated 127,000 (UR: 63,300 – 248,000) EOGBS 279 cases and 87,300 (UR: 38,100 – 209,000) LOGBS cases, thus avoiding 31,100 (UR: 14,400 – 66,400) 280 infant deaths and 17,900 (UR: 6,380 – 49,900) cases of moderate and severe NDI. Additionally, 281 23,000 (UR: 10,000 – 56,400) GBS stillbirths could be prevented and, if a vaccine also proves 282 effective against GBS associated prematurity, 185,000 (UR: 13,500 – 407,000) preterm births might 283 be avoided. The highest burden of iGBS cases and deaths, around two-fifths of the total, is averted in 284 Sub-Saharan Africa which accounts for about one-fifth of the women vaccinated. In contrast, only 285 about 1% of the deaths occur in Europe and Northern America despite a tenth of vaccinated women 286 being in this region. 287 288 Overall, iGBS cases averted through vaccination resulted in a projected gain of 2.5 million (UR: 1.2 – 289 5.4 million) undiscounted QALYs, and a further 1.5 million (UR: 0.6 – 3.6 million) QALYs when 290 avoided stillbirths are included. A vaccine that prevents GBS-associated prematurity could add 291 another 0.8 million (UR: 0.1 – 2.1 million) QALYs. The relative contribution of preventing iGBS, 292 stillbirths and prematurity to the overall QALY gain varies by region. For example, in Sub-Saharan 293 Africa, and Northern Africa and Western Asia preventing iGBS contributes the majority of the QALY 294 gain, but in Europe and Northern America, and Central and Southern Asia avoided stillbirths make a 295 larger contribution. In Europe and Northern America preventing preterm births might result in larger 296 QALY gains than iGBS cases and stillbirths combined. 297 Description Central & Southern Asia Eastern & South-Eastern Asia Europe & Northern America Latin America & Caribbean Northern Africa & Western Asia Oceania Sub-Saharan Africa Global^ Number of women vaccinated (millions) 22.8 25.5 11.7 9.51 7.8 0.546 21.9 99.8 Vaccine programme costs (discounted; $ millions) 124 (117, 136) 470 (452, 495) 648 (621, 687) 173 (169, 178) 127 (124, 131) 24.4 (22.9, 26.6) 107 (104, 112) 1,680 (1,640, 1,720) Acute healthcare costs (discounted; $ millions) -7.93 (-17.4, -3.74) -54.6 (-119, -24.9) -155 (-352, -60.9) -21.4 (-47.5, -10.7) -27.8 (-56.9, -13.5) -3.89 (-8.87, -1.57) -14.3 (-30.5, -6.67) -298 (-534, -155) . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 11 Long-term healthcare costs (discounted; $ millions) -2.95 (-11.3, -0.581) -19.8 (-79.2, -3.78) -33.2 (-117, -7.02) -8.13 (-31.1, -1.63) -10 (-36.6, -2.08) -0.871 (-2.97, -0.185) -4.85 (-18.6, -1.01) -86.5 (-252, -20.6) Total incremental costs (discounted; $ millions) 113 (96.6, 127) 394 (286, 446) 456 (200, 581) 143 (99.5, 160) 88.9 (39.5, 110) 19.6 (13, 23.2) 87.9 (60.6, 99.7) 1,290 (948, 1,490) EOGBS cases (thousands) -22.8 (-43.6, -11.6) -30.6 (-62.4, -14.5) -3.37 (-5.94, -1.57) -8.91 (-16.8, -4.72) -17.9 (-36, -8.6) -0.421 (-0.86, -0.207) -42.3 (-86.4, -20.1) -127 (-248, -63.3) LOGBS cases (thousands) -11.3 (-33.1, -2.94) -15.2 (-45.7, -3.86) -1.99 (-4.2, -0.84) -5.9 (-20.4, -1.94) -12.9 (-33, -5.24) -0.275 (-1.39, -0.0888) -36.4 (-101, -14) -87.3 (-209, -38.1) Moderate & severe NDI cases (thousands) -2.91 (-8.52, -0.933) -3.86 (-11.8, -1.21) -0.257 (-0.572, -0.0975) -1.31 (-3.99, -0.42) -2.63 (-7.58, -0.9) -0.0541 (-0.223, -0.0182) -6.66 (-19.4, -2.2) -17.9 (-49.9, -6.38) GBS deaths (thousands) -4.22 (-9.6, -1.78) -5.58 (-13.4, -2.32) -0.335 (-0.668, -0.148) -1.95 (-4.7, -0.804) -4.76 (-10.8, -2.06) -0.0836 (-0.336, -0.0242) -13.6 (-32.2, -5.65) -31.1 (-66.4, -14.4) GBS stillbirths (thousands) -6.59 (-23, -1.63) -3.13 (-10.9, -0.794) -0.586 (-1.48, -0.206) -1.34 (-8.35, -0.222) -1.33 (-3.12, -0.591) -0.0646 (-0.336, -0.0174) -9.31 (-18.5, -4.12) -23 (-56.4, -10) GBS associated preterm births* (thousands) -34.7 (-78.5, -2.33) -28.7 (-66.1, -2.05) -33.2 (-72, -2.4) -18.5 (-40.9, -1.37) -23.8 (-52.4, -1.81) -1.45 (-3.27, -0.103) -44.4 (-97.4, -3.23) -185 (-407, -13.5) QALYs from averted GBS disease (discounted; thousands) 150 (66.3, 339) 203 (86.9, 485) 12.5 (5.64, 24.8) 70.2 (30.7, 168) 164 (73, 371) 2.96 (1.01, 10.6) 431 (186, 1,010) 1,060 (486, 2,270) QALYs from averted stillbirths (discounted; thousands) 180 (43.7, 627) 87.2 (21.7, 301) 16.5 (5.77, 41.3) 37.3 (6.18, 233) 36.9 (16.3, 85.9) 1.75 (0.474, 9.06) 243 (108, 486) 622 (271, 1,550) QALYs from averted preterm births* (discounted; thousands) 63.6 (5.01, 162) 53.8 (4.24, 143) 62.4 (4.99, 160) 34.6 (2.77, 88.8) 44.2 (3.58, 113) 2.77 (0.213, 7.23) 77.4 (6.26, 198) 338 (27.6, 857) QALYs from averted GBS disease (undiscounted; thousands) 358 (158, 807) 513 (220, 1,220) 32.7 (14.6, 64.6) 177 (77, 422) 408 (181, 923) 6.94 (2.46, 24.5) 947 (407, 2,230) 2,490 (1,160, 5,370) QALYs from averted stillbirths (undiscounted; thousands) 429 (104, 1,490) 218 (54.4, 752) 42.7 (15, 107) 94.1 (15.6, 586) 91.4 (40.4, 213) 4.19 (1.17, 20.9) 532 (237, 1,060) 1,460 (630, 3,670) QALYs from averted preterm births* (undiscounted; thousands) 152 (12, 387) 135 (10.7, 359) 163 (13, 417) 87.8 (7.03, 225) 110 (8.91, 282) 7.3 (0.56, 19) 171 (13.8, 436) 825 (67.4, 2,090) Table 2: Annual global and regional impact of GBS maternal vaccination compared with no vaccination for the year 2020 All values are reported to 3 significant figures. Values in brackets are 95% uncertainty ranges. GBS = Group B Streptococcus; EOGBS = Early-Onset GBS; LOGBS = Late-Onset GBS; NDI = Neurodevelopmental Impairment; QALY = Quality Adjust Life Year. *in scenario analysis where vaccine is assumed to have 80% VE against GBS associated prematurity ^global median values do not exactly equal the sum of the regional median values 298 299 Using our base case assumptions about the vaccine characteristics the estimated global NMB of 300 vaccination ranged from $1.1 billion (UR: $-0.2 – 3.9 billion) to $17 billion (UR: $9.1 – 31 billion) 301 depending upon the normative assumptions (Fig. 2A). Including stillbirth QALYs increases the NMB 302 by between $1.4 billion and $7.1 billion depending on the other normative assumptions made. 303 304 Under the most-favourable normative assumptions vaccination had a positive NMB in all regions 305 (Fig. 2B). However, for least-favourable assumptions the NMB was negative for Central and Southern 306 Asia, Europe and Northern America, and Oceania. Nevertheless, if stillbirth QALYs were included, the 307 NMB for these regions was again positive (Supplementary Fig. 1). 308 . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 12 309 310 Figure 2: Net Monetary Benefit of GBS maternal vaccination (A) globally under different normative 311 assumptions (see Table 1); and (B) by region for the most and least favourable normative assumptions. 312 Least-favourable normative assumptions were the use of an empirical CET, 3% discounting of QALYs, and 313 exclusion of stillbirth QALYs. Most-favourable assumptions were the use of 1 x GDP per capita CETs, 0% . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 13 314 discounting of QALYs, and inclusion of stillbirth QALYs. M = Millions; B = Billions; CET = Cost-Effectiveness 315 Threshold; GDP = Gross Domestic Product; GDPPC = GDP per capita; QALY = Quality Adjusted Life Year; SB = 316 Stillbirth; SDG = Sustainable Development Goal. 317 318 For the most-favourable normative assumptions vaccination is likely cost-effective in almost all 319 countries (Fig 3.), but for least-favourable assumptions this was reduced to just over half (103/183) 320 of countries. Notably, vaccination was less likely to be cost-effective amongst countries with lower 321 GDP per capita within the Sub-Saharan Africa, Central & Southern Asia, and Europe & Northern 322 America regions. 323 324325 Figure 3: Probability that GBS maternal vaccination is cost-effective in each country under most favourable 326 and least favourable normative assumptions (see Table 1). Least-favourable normative assumptions were the 327 use of an empirical CET, 3% discounting of QALYs, and exclusion of stillbirth QALYs. Most-favourable 328 assumptions were the use of 1 x GDP per capita CETs, 0% discounting of QALYs, and inclusion of stillbirth 329 QALYs. CET = Cost-Effectiveness Threshold; GDP = Gross Domestic Product; GDPPC = GDP per capita; QALY = 330 Quality Adjusted Life Year; SB = Stillbirth; SDG = Sustainable Development Goal. 331 332 Figure 4 shows how the global NMB of vaccination varies under different scenarios. Inclusion of mild 333 NDI, assuming GBS births without skilled birth attendants have 90% case-fatality or increasing 334 vaccine efficacy from 80% to 90% slightly increase the global NMB of vaccination, while assuming 335 zero long-term costs for NDI slightly decreases the NMB. However, none of these assumptions have 336 a dramatic effect. If vaccine efficacy is decreased to 60%, global NMB remains positive under least- . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 14 337 favourable assumptions, but the number of countries for which vaccination is no longer cost- 338 effective increases to 96 (Supplementary Fig. 4). A vaccine that requires two doses to achieve 80% 339 efficacy would have a negative global NMB, and vaccination would not be cost-effective in 111 340 countries. However, a vaccine with protection against preterm birth substantially increases the 341 global NMB and is especially influential in the Europe & Northern America region (see 342 Supplementary Fig. 2). 343 344345 Figure 4: Annual global Net Monetary Benefit of GBS maternal vaccination under most favourable and least 346 favourable normative assumptions (see Table 1) for different vaccination scenarios. Points show median 347 estimates and lines show 95% uncertainty ranges. Least-favourable normative assumptions were the use of an 348 empirical CET, 3% discounting of QALYs, and exclusion of stillbirth QALYs. Most-favourable assumptions were 349 the use of 1 x GDP per capita CETs, 0% discounting of QALYs, and inclusion of stillbirth QALYs. B = Billions; CET 350 = Cost-Effectiveness Threshold; CFR = Case Fatality Risk; GDP = Gross Domestic Product; NDI = 351 Neurodevelopmental Impairment; SBA = Skilled Birth Attendant; QALY = Quality Adjusted Life Year; VE = 352 Vaccine Effectiveness. 353 354 The distribution of vaccine threshold prices amongst countries within each World Bank income 355 group are shown in Figure 5 (results by SDG region are shown in Supplementary Fig. 5, and for other 356 vaccine scenarios in Supplementary Fig. 6). The threshold price is usually positive (i.e., there is some 357 price at which purchasing the vaccine would be cost-effective), and generally higher in high-income 358 and upper-middle-income countries. However, under least-favourable normative assumptions . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 15 359 threshold price is negative in eight countries, indicating that even with a free vaccine the delivery 360 costs outweigh the health benefits in this analysis. 361362 Figure 5: Distribution of GBS vaccine threshold prices amongst countries within each World Bank income 363 group under most and least favourable normative assumptions (see Table 1). Threshold vaccine prices above 364 $800 per dose are not shown. Least-favourable normative assumptions were the use of an empirical CET, 3% 365 discounting of QALYs, and exclusion of stillbirth QALYs. Most-favourable assumptions were the use of 1 x GDP 366 per capita CETs, 0% discounting of QALYs, and inclusion of stillbirth QALYs. CET = Cost-Effectiveness Threshold; 367 GDP = Gross Domestic Product; QALY = Quality Adjusted Life Year. 368 369 DISCUSSION 370 A high-coverage global maternal immunisation programme against GBS could avert hundreds of 371 thousands of GBS cases, alongside tens of thousands of deaths, stillbirths, and cases of long-term 372 disability. We estimate that such a programme may have a net cost of around $1.4 billion, with most 373 costs occurring in Europe and Northern America. Nevertheless, it would be cost-effective in most 374 countries under favourable assumptions, particularly if it can reduce preterm births. 375 376 Even under less favourable assumptions, a single-dose GBS vaccine could still be cost-effective due 377 to additional factors we did not explore. In some high-income countries, GBS vaccination plus . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 16 378 current practice may be less cost-effective compared to current practice alone because of lower GBS 379 incidence in babies due to IAP. However, GBS vaccination might allow high-income countries to 380 achieve additional cost savings by revising IAP algorithms for vaccinated mothers. In low- and lower- 381 middle income countries, iGBS incidence may be higher, but so is the health opportunity cost of 382 healthcare spending due to budget constraints leading to lower thresholds at which interventions 383 may be considered cost-effective. 384 385 More competitive pricing may enable vaccination to be cost-effective, even under least favourable 386 assumptions. Competitive and finely tiered vaccine prices could also be beneficial for 387 manufacturers, with financial analyses suggesting that high global demand is needed to ensure the 388 development costs of a GBS vaccine to be recouped.46 Our economic evaluation can inform both 389 manufacturers and donors investigating the financial viability of investing in GBS vaccine 390 development, as well as countries identifying the price they should be willing to pay for such a 391 vaccine. 392 393 Our study is the first to estimate the value of maternal GBS vaccination across all regions and 394 country income groups. Previous analyses have estimated cost-effectiveness in the United States,12– 395 14 the Netherlands,17 United Kingdom,15,16 South Africa,19 The Gambia,18 and 37 Gavi countries in 396 Africa.20 These prior estimates suggested cost-effectiveness of vaccination ranged from $320-573 397 per DALY averted in Gavi-eligible countries,20 to $3550 per DALY averted in South Africa,19 to over 398 $50,000 per QALY in the United States,12,13 which is broadly consistent with our results. Like our 399 analysis, Kim et al. also found that the ability to avert GBS-associated prematurity greatly improved 400 vaccine cost-effectiveness.19 401 402 This was the first cost-effectiveness study to use new global estimates of the health burden due to 403 GBS including infant morbidity and mortality, long-term neurodevelopmental impairment, stillbirth, 404 and GBS-associated prematurity. This burden study propagated parametric uncertainty 405 comprehensively by using a Bayesian framework to synthesise existing data sources. Posterior 406 distributions from the study then informed a probabilistic sensitivity analysis for our cost- 407 effectiveness model. Similarly, for cost data, parameters with multiple sources of data from previous 408 systematic reviews were synthesised using regression models. Conversely, the main limitations of 409 our analysis reflected parameters with limited data such as those governing health-related quality of 410 life and long-term costs from disability, where estimates were based on only 1-2 relevant studies. 411 Our analysis also excluded the potential impact of vaccination on maternal morbidity and the costs . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 17 412 of GBS-related disability beyond the health-sector. However, both these factors would likely 413 reinforce our main findings on cost-effectiveness. 414 415 A further set of uncertainties govern GBS vaccine characteristics such as efficacy, number of doses 416 needed and impact on GBS-associated prematurity. Since there is currently no licensed vaccine, 417 these parameters were informed by the WHO PPC which is based on expert assumptions. We 418 therefore used scenario sensitivity analyses to identify which of these characteristics are the most 419 important drivers of vaccine value. Further data from carefully designed vaccine trials and other field 420 studies are needed to inform these data gaps. Vaccine value is also driven by normative health 421 economic assumptions around discounting, cost-effectiveness thresholds and the value of 422 preventing stillbirths, which reflect uncertainty about the values of society rather than about 423 empirical data. 424 425 Overall, our results suggest high coverage of a competitively priced maternal GBS vaccine has the 426 potential to save tens of thousands of lives globally and is likely to be a cost-effective investment, 427 particularly if the vaccine can reduce GBS-associated prematurity. 428 429 Contributors’ statement 430 Conceptualisation – SRP, AK, JEL, and MJ; Methodology – SRP, BG, CT and MJ; Investigation – SRP; 431 Formal Analysis – SRP; Software – SRP; Validation – SRP; Data Curation – SRP and BG; Writing – 432 Original Draft – SRP and MJ; Writing – Review & Editing – SRP, BG, PP, JC, FS, AK, RH, CT, JEL and MJ; 433 Visualisation – SRP; Supervision – MJ and JEL; Funding Acquisition – MJ and JEL. 434 435 Declaration of interests 436 FS is employed by the UK NSC which developed the policy recommendation for maternal GBS 437 screening. 438 439 Acknowledgements 440 We would like to thank the authors of the GBS burden paper for sharing data on the posterior 441 estimates of parameters in the burden model. We thank the GBS Full Value of Vaccine Assessment 442 project Scientific Advisory Group for helpful discussion. We also thank Clint Pecenka and Ranju Baral 443 for sharing estimates of antenatal care coverage by gestational age. This work was supported by a 444 grant (INV-009018) to the London School of Hygiene & Tropical Medicine (PI Joy Lawn) from the Bill 445 & Melinda Gates Foundation. RH is member of the WHO. The views expressed in this article are . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 18 446 those of the authors and do not necessarily represent the decisions, official policy or opinions of the 447 WHO. 448 449 Data sharing statement 450 Code and data used in this analysis are available at https://github.com/mert0248/GBS-vax-econ- 451 model . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 19 452 References 453 1. Gonçalves BP, Procter SR, Paul P, Chandna J, Lewin A, Seedat F, et al. Group B 454 streptococcus infection during pregnancy and infancy: estimates of regional and global 455 burden. The Lancet Global Health. 2022 Apr;S2214109X22000936. 456 2. Navarro-Torné A, Curcio D, Moïsi JC, Jodar L. Burden of invasive group B Streptococcus 457 disease in non-pregnant adults: A systematic review and meta-analysis. Melo-Cristino 458 J, editor. PLoS ONE. 2021 Sep 30;16(9):e0258030. 459 3. Seale AC, Bianchi-Jassir F, Russell NJ, Kohli-Lynch M, Tann CJ, Hall J, et al. Estimates of 460 the Burden of Group B Streptococcal Disease Worldwide for Pregnant Women, 461 Stillbirths, and Children. Clinical Infectious Diseases. 2017 Nov 6;65(suppl_2):S200–19. 462 4. Paul P, Chandna J, Procter SR, Dangor Z, Leahy S, Santhanam S, et al. 463 Neurodevelopmental and growth outcomes after invasive Group B Streptococcus in 464 early infancy: A multi-country matched cohort study in South Africa, Mozambique, 465 India, Kenya, and Argentina. eClinicalMedicine. 2022 Apr;101358. 466 5. Le Doare K, O’Driscoll M, Turner K, Seedat F, Russell NJ, Seale AC, et al. Intrapartum 467 Antibiotic Chemoprophylaxis Policies for the Prevention of Group B Streptococcal 468 Disease Worldwide: Systematic Review. Clinical Infectious Diseases. 2017 Nov 469 6;65(suppl_2):S143–51. 470 6. Patras KA, Nizet V. Group B Streptococcal Maternal Colonization and Neonatal Disease: 471 Molecular Mechanisms and Preventative Approaches. Front Pediatr. 2018 Feb 22;6:27. 472 7. Heath PT, Culley FJ, Jones CE, Kampmann B, Le Doare K, Nunes MC, et al. Group B 473 streptococcus and respiratory syncytial virus immunisation during pregnancy: a 474 landscape analysis. The Lancet Infectious Diseases. 2017 Jul;17(7):e223–34. 475 8. Abu Raya B, Edwards KM, Scheifele DW, Halperin SA. Pertussis and influenza 476 immunisation during pregnancy: a landscape review. The Lancet Infectious Diseases. 477 2017 Jul;17(7):e209–22. 478 9. Vekemans J, Moorthy V, Friede M, Alderson MR, Sobanjo-Ter Meulen A, Baker CJ, et al. 479 Maternal immunization against Group B streptococcus: World Health Organization 480 research and development technological roadmap and preferred product 481 characteristics. Vaccine. 2019 Nov;37(50):7391–3. 482 10. Carreras-Abad C, Ramkhelawon L, Heath PT, Le Doare K. A Vaccine Against Group B 483 Streptococcus: Recent Advances. IDR. 2020 Apr;13:1263–72. 484 11. Venkatesan P. Defeating meningitis by 2030: the WHO roadmap. The Lancet Infectious 485 Diseases. 2021 Dec;21(12):1635. 486 12. Kim SY, Nguyen C, Russell LB, Tomczyk S, Abdul-Hakeem F, Schrag SJ, et al. Cost- 487 effectiveness of a potential group B streptococcal vaccine for pregnant women in the 488 United States. Vaccine. 2017 Oct;35(45):6238–47. . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 20 489 13. Oster G, Edelsberg J, Hennegan K, Lewin C, Narasimhan V, Slobod K, et al. Prevention of 490 group B streptococcal disease in the first 3 months of life: Would routine maternal 491 immunization during pregnancy be cost-effective? Vaccine. 2014 Aug;32(37):4778–85. 492 14. Mohle-Boetani JC, Schuchat A, Plikaytis BD, Smith JD, Broome CV. Comparison of 493 prevention strategies for neonatal group B streptococcal infection. A population-based 494 economic analysis. JAMA. 1993 Sep 22;270(12):1442–8. 495 15. Colbourn TE, Asseburg C, Bojke L, Philips Z, Welton NJ, Claxton K, et al. Preventive 496 strategies for group B streptococcal and other bacterial infections in early infancy: cost 497 effectiveness and value of information analyses. BMJ. 2007 Sep 29;335(7621):655. 498 16. Giorgakoudi K, O’Sullivan C, Heath PT, Ladhani S, Lamagni T, Ramsay M, et al. Cost- 499 effectiveness analysis of maternal immunisation against group B Streptococcus (GBS) 500 disease: A modelling study. Vaccine. 2018 Nov;36(46):7033–42. 501 17. Hahn BA, de Gier B, van Kassel MN, Bijlsma MW, van Leeuwen E, Wouters MGAJ, et al. 502 Cost-effectiveness of maternal immunization against neonatal invasive Group B 503 Streptococcus in the Netherlands. Vaccine. 2021 May;39(21):2876–85. 504 18. Ahmed N, Giorgakoudi K, Usuf E, Okomo U, Clarke E, Kampmann B, et al. Potential 505 cost-effectiveness of a maternal Group B streptococcal vaccine in The Gambia. Vaccine. 506 2020 Mar;38(15):3096–104. 507 19. Kim SY, Russell LB, Park J, Verani JR, Madhi SA, Cutland CL, et al. Cost-effectiveness of a 508 potential group B streptococcal vaccine program for pregnant women in South Africa. 509 Vaccine. 2014 Apr;32(17):1954–63. 510 20. Russell LB, Kim SY, Cosgriff B, Pentakota SR, Schrag SJ, Sobanjo-ter Meulen A, et al. 511 Cost-effectiveness of maternal GBS immunization in low-income sub-Saharan Africa. 512 Vaccine. 2017 Dec;35(49):6905–14. 513 21. Hutubessy RCW, Lauer JA, Giersing B, Sim SY, Jit M, Kaslow D, et al. The Full Value of 514 Vaccine Assessments (FVVA): A Framework to Assess and Communicate the Value of 515 Vaccines for Investment and Introduction Decision Making. SSRN Journal [Internet]. 516 2021 [cited 2022 Mar 2]; Available from: https://www.ssrn.com/abstract=3841999 517 22. Gessner BD, Kaslow D, Louis J, Neuzil K, O’Brien KL, Picot V, et al. Estimating the full 518 public health value of vaccination. Vaccine. 2017 Nov;35(46):6255–63. 519 23. World Health Organization, London School of Hygiene and Tropical Medicine. Group B 520 streptococcus vaccine: full value of vaccine assessment [Internet]. Geneva: World 521 Health Organization; 2021 [cited 2022 Apr 29]. Available from: 522 https://apps.who.int/iris/handle/10665/347595 523 24. United Nations, Department of Economic and Social Affairs, Population Division. World 524 Population Prospects 2019. [Internet]. Available from: 525 https://population.un.org/wpp/Download/Standard/Population/ . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 21 526 25. World Health Organization. Global Health Observatory. [Internet]. [cited 2022 Mar 2]. 527 Available from: https://www.who.int/data/gho/ 528 26. Gonçalves BP, Procter SR, Clifford S, Koukounari A, Paul P, Lewin A, et al. Estimation of 529 country-level incidence of early-onset invasive Group B Streptococcus disease in 530 infants using Bayesian methods. Althouse BM, editor. PLoS Comput Biol. 2021 Jun 531 14;17(6):e1009001. 532 27. United Nations, Department of Economic and Social Affairs, Statistics Division. SDG 533 Indicators — Regional groupings used in Report and Statistical Annex [Internet]. [cited 534 2022 Mar 3]. Available from: https://unstats.un.org/sdgs/indicators/regional-groups/ 535 28. Horváth-Puhó E, van Kassel MN, Gonçalves BP, de Gier B, Procter SR, Paul P, et al. 536 Mortality, neurodevelopmental impairments, and economic outcomes after invasive 537 group B streptococcal disease in early infancy in Denmark and the Netherlands: a 538 national matched cohort study. The Lancet Child & Adolescent Health. 2021 539 Jun;5(6):398–407. 540 29. Chawanpaiboon S, Vogel JP, Moller AB, Lumbiganon P, Petzold M, Hogan D, et al. 541 Global, regional, and national estimates of levels of preterm birth in 2014: a systematic 542 review and modelling analysis. The Lancet Global Health. 2019 Jan;7(1):e37–46. 543 30. Petrou S, Krabuanrat N, Khan K. Preference-Based Health-Related Quality of Life 544 Outcomes Associated with Preterm Birth: A Systematic Review and Meta-analysis. 545 PharmacoEconomics. 2020 Apr;38(4):357–73. 546 31. Procter SR, Salman O, Pecenka C, Gonçalves BP, Paul P, Hutubessy R, et al. A review of 547 the costs of delivering maternal immunisation during pregnancy. Vaccine. 2020 548 Sep;38(40):6199–204. 549 32. Bennett JE, Sumner W, Downs SM, Jaffe DM. Parents’ utilities for outcomes of occult 550 bacteremia. Arch Pediatr Adolesc Med. 2000 Jan;154(1):43–8. 551 33. Chhibber AV, Hill PC, Jafali J, Jasseh M, Hossain MI, Ndiaye M, et al. Child Mortality 552 after Discharge from a Health Facility following Suspected Pneumonia, Meningitis or 553 Septicaemia in Rural Gambia: A Cohort Study. Kollmann TR, editor. PLoS ONE. 2015 Sep 554 9;10(9):e0137095. 555 34. Roed C, Omland LH, Engsig FN, Skinhoj P, Obel N. Long-Term Mortality in Patients 556 Diagnosed with Meningococcal Disease: A Danish Nationwide Cohort Study. Ng LFP, 557 editor. PLoS ONE. 2010 Mar 12;5(3):e9662. 558 35. Petrou S, Johnson S, Wolke D, Marlow N. The association between 559 neurodevelopmental disability and economic outcomes during mid-childhood: 560 Neurodevelopmental disability and economic outcomes. Child: Care, Health and 561 Development. 2013 May;39(3):345–57. 562 36. World Health Organization. WHO Preferred Product Characteristics for Group B 563 Streptococcus Vaccines. Geneva; 2017. . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 22 564 37. Bianchi-Jassir F, Seale AC, Kohli-Lynch M, Lawn JE, Baker CJ, Bartlett L, et al. Preterm 565 Birth Associated With Group B Streptococcus Maternal Colonization Worldwide: 566 Systematic Review and Meta-analyses. Clinical Infectious Diseases. 2017 Nov 567 6;65(suppl_2):S133–42. 568 38. Blencowe H, Cousens S, Chou D, Oestergaard M, Say L, Moller AB, et al. Born Too Soon: 569 The global epidemiology of 15 million preterm births. Reprod Health. 2013 570 Nov;10(S1):S2. 571 39. Baral R, Fleming J, Khan S, Higgins D, Hendrix N, Pecenka C. Inferring antenatal care 572 visit timing in low- and middle-income countries: Methods to inform potential 573 maternal vaccine coverage. Larson BA, editor. PLoS ONE. 2020 Aug 20;15(8):e0237718. 574 40. The World Bank. GDP deflator (base year varies by country) | Data [Internet]. [cited 575 2022 Mar 3]. Available from: https://data.worldbank.org/indicator/NY.GDP.DEFL.ZS 576 41. The World Bank. Official exchange rate (LCU per US$, period average) | Data [Internet]. 577 [cited 2022 Mar 3]. Available from: https://data.worldbank.org/indicator/PA.NUS.FCRF 578 42. Schroeder EA, Petrou S, Balfour G, Edamma O, Heath PT. The economic costs of Group 579 B Streptococcus (GBS) disease: prospective cohort study of infants with GBS disease in 580 England. Eur J Health Econ. 2009 Jul;10(3):275–85. 581 43. Salman O, Procter SR, McGregor C, Paul P, Hutubessy R, Lawn JE, et al. Systematic 582 Review on the Acute Cost-of-illness of Sepsis and Meningitis in Neonates and Infants. 583 Pediatric Infectious Disease Journal. 2020 Jan;39(1):35–40. 584 44. Aerts C, Leahy S, Mucasse H, Lala S, Bramugy J, Tann CJ, et al. Quantifying the Acute 585 Care Costs of Neonatal Bacterial Sepsis and Meningitis in Mozambique and South 586 Africa. Clinical Infectious Diseases. 2022 Jan 20;74(Supplement_1):S64–9. 587 45. Khavjou OA, Anderson WL, Honeycutt AA, Bates LG, Razzaghi H, Hollis ND, et al. 588 National Health Care Expenditures Associated With Disability. Medical Care. 2020 589 Sep;58(9):826–32. 590 46. Malvolti S, Pecenka C, Mantel CF, Malhame M, Lambach P. A Financial and Global 591 Demand Analysis to Inform Decisions for Funding and Clinical Development of Group B 592 Streptococcus Vaccines for Pregnant Women. Clinical Infectious Diseases. 2022 Jan 593 20;74(Supplement_1):S70–9. 594 47. Chi YL, Blecher M, Chalkidou K, Culyer A, Claxton K, Edoka I, et al. What next after GDP- 595 based cost-effectiveness thresholds? Gates Open Res. 2020 Nov 30;4:176. 596 48. Woods B, Revill P, Sculpher M, Claxton K. Country-Level Cost-Effectiveness Thresholds: 597 Initial Estimates and the Need for Further Research. Value in Health. 2016 598 Dec;19(8):929–35. . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint 23 599 49. Ochalek J, Lomas J, Claxton K. Estimating health opportunity costs in low-income and 600 middle-income countries: a novel approach and evidence from cross-country data. BMJ 601 Glob Health. 2018 Nov;3(6):e000964. 602 50. Phillips J, Millum J. Valuing Stillbirths: Valuing Stillbirths. Bioethics. 2015 Jul;29(6):413– 603 23. 604 51. WHO guide for standardization of economic evaluations of immunization programmes. 605 2nd ed. Geneva: World Health Organization; 2019. 606 52. Drummond M. Methods for the economic evaluation of health care programmes. 607 Fourth edition. Oxford, United Kingdom; New York, NY, USA: Oxford University Press; 608 2015. 609 . CC-BY 4.0 International licenseIt 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 July 12, 2022. ; https://doi.org/10.1101/2022.07.11.22277482doi: medRxiv preprint