Next Article in Journal
Bridging Gaps in Vaccine Access and Equity: A Middle Eastern Perspective
Previous Article in Journal
Perspectives of Healthcare Students on Childhood Vaccines: Insights from a Cross-Sectional Study in Bulgaria
Previous Article in Special Issue
The Platform Readiness Dashboard: A Tool for Evaluating Vaccine Platform Suitability for a Rapid Response to Epidemic and Pandemic Threats
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Vaccines
Volume 13
Issue 8
10.3390/vaccines13080805
vaccines-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Novel Approach to Estimate the Impact of PCV20 Immunization in Children by Incorporating Indirect Effects to Generate the Number Needed to Vaccinate

by
Mark H. Rozenbaum
1,*,
Maria J. Tort
1,
Blair Capitano
1,
Ruth Chapman
2,
Desmond Dillon-Murphy
2,
Benjamin M. Althouse
1,3,4 and
Alejandro Cane
1
1
Pfizer Inc., Collegeville, PA 19426, USA
2
PPD™ Evidera™ Health Economics & Market Access, Thermo Fisher Scientific, London W6 8BJ, UK
3
Information School, University of Washington, Seattle, WA 98195, USA
4
Department of Biology, New Mexico State University, Las Cruces, NM 88003, USA
*
Author to whom correspondence should be addressed.
Vaccines 2025, 13(8), 805; https://doi.org/10.3390/vaccines13080805
Submission received: 10 June 2025 / Revised: 3 July 2025 / Accepted: 25 July 2025 / Published: 29 July 2025
(This article belongs to the Special Issue Estimating Vaccines' Value and Impact)
Browse Figures
Versions Notes

Abstract

Background/Objectives: The number needed to vaccinate (NNV) is a metric commonly used to evaluate the public health impact of a vaccine as it represents the number of individuals that must be vaccinated to prevent one case of disease. Traditional calculations may underestimate vaccine benefits by neglecting indirect effects and duration of protection (DOP), resulting in NNV overestimation. This study evaluated the NNV for the pediatric 20-valent pneumococcal conjugate (PCV20) US immunization program, as compared to PCV13, with a unique approach to NNV. Methods: A multi-cohort, population-based Markov model accounting for indirect effects was employed to calculate the NNV of PCV20 to avert a case of pneumococcal disease, invasive pneumococcal disease (IPD), hospitalized non-bacteremic pneumonia (NBP), ambulatory NBP, and otitis media (OM), as well as to prevent antibiotic-resistant cases and antibiotic prescriptions. Results: The mean NNV over a 25-year time horizon to prevent one case of pneumococcal disease was 6, with NNVs of 854 for IPD, 106 for hospitalized NBP, 25 for outpatient NBP, and 9 for OM, 11 for a course of antibiotic, and 4 for resistant disease. The mean NNV per year decreased over time, reflecting the DOP and increasing indirect effects over time. Conclusions: This study presents a novel approach to NNVs and shows that relatively few vaccinations are required to prevent disease. The decrease in NNV over time highlights the necessity of including DOP and indirect effects in NNV calculations, ensuring a more realistic assessment of a vaccine’s impact.

1. Introduction

Pneumococcal infection, caused by Streptococcus pneumoniae, continues to be a significant public health concern, leading to diseases, such as pneumonia, meningitis, and otitis media (OM) [1,2]. A dramatic reduction in the incidence of pneumococcal diseases has been observed over the past 20 years since the introduction of pneumococcal conjugate vaccines (PCVs) in children [3,4].
Introduced in the year 2000, the first pneumococcal conjugate vaccine (PCV7) targeted seven pneumococcal serotypes, leading to a reduction in pneumococcal disease incidence among children under five years of age, as well as among older adults due to indirect effects [3,5,6,7]. A decade later, the introduction of PCV13, which includes an additional six serotypes (1, 3, 5, 6A, 7F, and 19A) [5,6,8], brought about a further decline in invasive pneumococcal disease (IPD) rates in young children and significantly lowered the number of pneumococcal pneumonia cases and associated hospitalizations in the adult population [9,10]. Later generation, higher valency PCVs are currently available. Although the effectiveness data of higher-valency PCVs are still under assessment, these PCVs are expected to further reduce pneumococcal disease incidence [11].
To achieve widespread immunization, vaccines are implemented into national immunization programs (NIPs). In many countries, the incorporation of vaccines into an NIP is based on several criteria. For example, in the US, the Centers for Disease Control and Prevention (CDC) uses the evidence to recommendation (EtR) framework, which considers seven criteria [12]. The number needed to vaccinate (NNV), representing the number of individuals that must be vaccinated to prevent one case of disease or another outcome, can serve as a measure for the evaluation of two of these criteria: ‘benefits and harm’ and ‘resource use’. The NNV has been historically used by the Advisory Committee on Immunization Practices (ACIP) in evaluations of and recommendations for vaccines, such as Hepatitis B [13] and pneumococcal [14,15] vaccines. Furthermore, the NNV has been frequently used in the literature to quantify the effectiveness and comparative strategies of vaccination programs [16,17,18] and has been applied to various health outcomes, vaccine types, and populations [19,20].
Originating from the number needed to treat (NNT), which measures the number of patients needed to be treated to prevent one additional adverse outcome compared to a control in therapeutic contexts, the NNV is specifically designed for evaluating the impact of vaccination programs in preventative health [17]. Despite their shared origins, the NNV introduces unique complexities and differs in its calculation compared to the NNT. While the NNT is calculated as the reciprocal of the absolute risk reduction, the NNV is traditionally calculated as the inverse of the product of the annual disease incidence rate among unvaccinated individuals and the vaccine’s effectiveness [16,17,19], thereby combining vaccine effectiveness with age-specific disease incidence [19].
Traditional NNV calculations typically rely on annual risk estimates based on short-term clinical data and, thus, have a limited time horizon. This approach is appropriate for vaccines conferring protection for ≤1 year. However, many vaccines offer protection for an extended duration, including PCVs. For example, in young children, PCV13 has been shown to provide significant protection for up to 4 years following the booster dose [21]. For vaccines with a longer DOP, the traditional approach to calculating the NNV may underestimate the true value. McLaughlin et al. [18] demonstrated that, in adults, accounting for the cumulative protection of PCV13 over five years significantly reduces the NNV for preventing community-acquired pneumonia (CAP). This underscores how limiting the assumed DOP in NNV calculations can lead to an underestimation of a vaccine’s overall impact.
Furthermore, traditional NNV calculations do not account for potential indirect effects (i.e., herd effects), leading to biased estimates up to three orders of magnitude [20]. Including indirect effects in NNV calculations is crucial because vaccines may prevent cases in unvaccinated populations, which should be considered in evaluating the NNV. A significant benefit of pediatric pneumococcal vaccination is the protection it provides to older adults, contributing substantially to the overall health impact of these vaccination strategies [22,23,24,25,26]. Although the ACIP recognizes the NNV as an appropriate and intuitive summary measure it acknowledges its limitations as not all aspects are captured, such as indirect effects [27].
Incorporating both direct and indirect effects along with the DOP into NNV calculations allows for a more comprehensive understanding of the true impact of vaccination programs to support informed public health decisions and more efficient allocation of resources [16,17,18,19,20]. Therefore, this study aimed to evaluate the NNV for PCV20 in the US population across the lifespan while incorporating indirect and cumulative effects of the pediatric immunization program. In addition, scenario analyses were conducted to better understand the impact of adjusting the model time horizon and including and excluding indirect effects on the NNV calculation.

2. Materials and Methods

Traditional NNV calculations incorporate the annual incidence rate of a vaccine-preventable disease among unvaccinated individuals and a vaccine’s effectiveness using the following formula:
T r a d i t i o n a l   N N V = 1 I n c i d e n c e   r a t e × V a c c i n e   e f f e c t i v e n e s s .
In this study, we used a model-based approach, enabling the incorporation of both the direct and indirect effects of vaccination over an extended time horizon to compare the NNV of the vaccine of interest (PCV20) to the standard of care (PCV13) when PCV20 was introduced rather than no vaccination. The updated NNV calculation used in this analysis was based on the approach outlined by Brission et al. [19] where
U p d a t e d   N N V = N u m b e r   v a c c i n a t e d C a s e s   p r e v e n t e d   o v e r   t i m e   h o r i z o n .

2.1. Model Structure

The model used in this study to estimate the NNV has been adapted to several countries to estimate the cost-effectiveness of PCV20 [28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47]. Briefly, this population-based, multi-cohort, Markov model simulates the US population based on predicted births over a time horizon of up to 25 years and accounts for the gradual build-up of indirect effects and the sustained protection provided by the vaccine. Direct effects were modeled based on reductions in the incidence of IPD, pneumonia, and OM in vaccinated individuals. Although pneumococcal carriage and transmission are not directly modeled, indirect effects are captured by incorporating age-specific and disease-specific reductions in disease incidence, benefiting even those who were not vaccinated. The model operated on an annual cycle, introducing new cohorts of children each year vaccinated with either PCV13 or PCV20 in a 3 + 1 dosing schedule, with four doses administered at 2, 4, 6, and 12–15 months. DOP in the model was conservatively assumed to be constant for four years after the final vaccine dose [21], and thereafter would exponentially decay reducing 10% relative to the preceding year; the maximum DOP was assumed to be 15 years.
Individuals’ transitions across the health states within each cycle were determined by vaccine strategy-specific and age-specific transition probabilities. Detailed information of the model structure and inputs is presented in Rozenbaum et al. [33].

2.2. Primary Analysis

The model outcomes included the NNV of PCV20 compared to PCV13 to prevent IPD and OM cases, hospitalization, and hospitalized and outpatient CAP. Additionally, the model assessed the NNV to prevent mortality and antibiotic-resistant cases prevented and antibiotic prescriptions avoided. The outcomes were analyzed over a 25-year time horizon.
Uncertainty around the base-case NNV estimates was investigated through deterministic and probabilistic sensitivity analyses. A one-way deterministic sensitivity analysis (DSA) was conducted to identify the drivers of NNV and examine uncertainty by adjusting the value of each input to the high and low range based on published data or ±20% of the base-case values in the absence of published data. A second-order stochastic probabilistic sensitivity analysis (PSA) was conducted with 1000 iterations to assess joint uncertainty in the parameter estimates using typical probability distributions following Briggs et al. [48], to generate confidence intervals around the point estimates of NNV. Confidence intervals (CIs) and standard errors (SEs) were used to inform parameter uncertainty, while upper and lower bounds (±20% of the point estimate) were applied in the absence of CIs or SEs. The values used in the DSA and PSA can be found in Rozenbaum et al. [33].

2.3. Scenario Analyses

Scenario analyses were conducted to show the impact of omitting certain critical factors. Firstly, the time horizon of the analysis varied between 1 and 25 years. The one-year time horizon reflects the basic NNV calculation, which only considers annual cases. Additionally, the impact of indirect effects was assessed, thereby allowing a comparison of the NNV with direct effects only versus the NNV capturing the population-level benefit.

3. Results

3.1. Primary Analysis Results

As shown in Table 1 and Figure 1, when comparing PCV20 to PCV13 over a 25-year time horizon, the mean NNV varied by outcome as follows: 85 vaccinations were needed to prevent one case of IPD, 106 for hospitalized pneumonia, 25 for non-hospitalized pneumonia, 94 for any hospitalization, 9 for OM, and 6 for any disease. The analysis also showed that 11 vaccinations were needed to prevent a course of antibiotics, and 7 to prevent a case of antibiotic-resistant disease.

3.2. Scenario Analysis

The NNVs for all outcomes decreased over time, with the highest NNV observed in the first year of the time horizon, followed by a steep decline and a plateau (Figure 2). These results reflect the pattern of the accumulation of indirect effects in the population. In every outcome, the NNV is highest in year 1, reflecting benefits limited to the directly vaccinated cohort; values decline steeply during years 2–10 as indirect protection accumulates in older, previously unvaccinated cohorts, and then plateau. Omitting indirect effects keeps the NNV elevated throughout the horizon, with the gap widening over time (e.g., total-disease NNV of 6 with indirect effects versus 23 without at year 25).
In the scenarios excluding indirect effects, the NNV is consistently higher than that in the primary analysis, although the difference is small with shorter time horizons (<10 years). For total disease, the NNV increased from 6 to 23 when indirect effects were excluded in the scenario with the 25-year time horizon while there was no difference in the scenario with a 1-year time horizon (NNV = 53) reflecting that indirect effects would gradually increase after the first year of the introduction of PCV20 in the national immunization program. The results of the scenarios assessing different time horizons and excluding indirect effects are shown in Table 2.

3.3. Sensitivity Analysis

The DSA revealed that the model conclusions are highly robust (Figure 3): across all plausible parameter values, the NNV varied only between 5.4 and 6.7 doses. The age-specific serotype distribution of PCV20-exclusive serotypes was the dominant driver, and higher prevalence of these serotypes reduced the NNV by ~9%, whereas lower prevalence increased it by ~13%. Baseline incidence of complex OM and the maximum indirect effect for OM and pneumonia were the next most influential factors; higher disease burden or stronger herd protection each lowered the NNV by 5–8%. Conversely, the slower accrual of indirect effects or faster waning of direct protection had modest upward pressure, but each shifted the NNV by less than one-tenth of a dose. No single epidemiological or vaccine-effect assumption was capable of overturning the overall finding that upgrading from PCV13 to PCV20 requires vaccinating about six individuals to avert one additional case of pneumococcal disease.

4. Discussion

Accurately calculating the NNV is crucial for understanding the true impact of vaccination programs. Traditional NNV calculations often overlook the DOP and indirect effects, leading to an underestimation of a vaccine’s long-term benefits. Incorporating these factors provides a more accurate and comprehensive assessment of vaccine effectiveness, which is essential for informed public health decisions and efficient allocation of resources. This study presents a novel approach to calculating the NNV by incorporating both direct protection and indirect effects, as well as the cumulative benefits over an extended time horizon. In this analysis, PCVs were used as a representative example to illustrate the application of this methodological approach.
Given that PCV7 and PCV13 have already significantly reduced the incidence of pneumococcal disease in the US [3,4], this analysis compared PCV20 to PCV13 rather than no vaccination to provide a more relevant assessment of the additional benefits provided by PCV20. Over a 25-year period, the NNVs to prevent one case of IPD, hospitalized and non-hospitalized pneumonia, any hospitalization, OM, any pneumococcal disease, and death were 854, 106, 25, 94, 9, 6, and 2488, respectively. Furthermore, the NNV to prevent one case of antibiotic resistance was 7, and the NNV to prevent one antibiotic prescription was 11. Sensitivity analysis results demonstrate that the NNV estimates are robust with relatively little variability in input scenarios. The outcomes included represent the key public health impact of pneumococcal vaccination and allow for the consideration of optimal vaccination strategy, and the inclusion of antibiotic-resistant cases and prescriptions goes beyond the standard outcomes for pneumococcal vaccines demonstrating the larger benefit to society.
The NNV for preventing various health outcomes with PCV20 substantially decreased with an extended time horizon and when including indirect effects. The decrease in the NNV over time highlights the necessity of including the duration of vaccine effectiveness and indirect effects in NNV calculations, ensuring a more realistic assessment of a vaccine’s impact. The importance of including an extended time horizon in NNV calculations was previously also shown by McLaughlin et al. [18], who compared two methodologies for calculating the NNV with PCV13 to prevent one case of community-acquired pneumonia (CAP) among US adults aged 65 years. However, their study is not directly comparable to our analysis. While McLaughlin et al. [18] focused on a cohort of 100,000 older adults receiving PCV13, we evaluated the broader impact of PCV20 across the entire US population using a population based model. Two other studies also examined the NNV for PCV20 in adults ≥18 years and older adults aged ≥60 or ≥65 years. However, these studies applied shorter time horizons—5 years [49] and 15 years [50]—compared to the 25-year horizon used in our study. Additionally, while our study compared the impact of PCV20 versus PCV13, previous studies used no vaccination as comparator [50]. Finally, our methodology for calculating the NNV also differed: Lewnard et al. [49] employed a per capita approach based in case reductions, whereas we used a model capturing cumulative vaccine impact over time.
Like all studies, ours has several limitations that warrant discussion. While the model captures indirect effects, it does not directly model pneumococcal carriage and transmission dynamics. Consequently, the true extent of indirect effects is unknown, which could potentially lead to an underestimation or overestimation of the population-level impacts. This limitation could affect the accuracy of our NNV estimates, particularly in reflecting the broader community benefits of vaccination. Conservatively, we assumed that a large proportion of the adult population (i.e., elderly persons previously vaccinated with PCV13, younger adults with underlying conditions previously vaccinated with PCV13 or PPV23) would not benefit from the indirect effects of childhood PCV20 vaccination at any time during the modeling horizon. In the absence of evidence on the DOP conferred by PCV20, data for PCV13 were employed [21]. Savulesu et al. [21] showed that up to 4 years after the booster dose the vaccine effectiveness of PCV13 was stable. In our model we assumed that after four years, the effectiveness would wane by 10% per year until year 15. After 15 years the effectiveness was conservatively assumed to be 0%. Currently there are no data on PCV20 DOP, which could potentially impact the conclusion on the influence of DOP time horizon. Additionally, the model approach does not consider serotype replacement as this was not observed in the US, serotype replacement could alter the profile of NNV over time [51,52]. While the model relies on US-specific epidemiological data, which may limit generalizability to other settings, the relative consistency of pneumococcal disease patterns and vaccine effects across developed countries suggests that our findings would remain directionally applicable, however, more studies are needed to estimate NNV in other countries. Although we did not explicitly model all possible disease sequelae, our inclusion of major outcomes, like IPD, pneumonia, and OM, captures the most significant public health impacts. The exclusion of rarer complications would likely make our NNV estimates conservative.

5. Conclusions

The results of our study strongly support the notion that the standard definition of NNV should be reconsidered for vaccines providing protection for more than one year and those conferring indirect effects. Traditional NNV calculations, which often use annual risk estimates, may underestimate the true impact of long-term vaccination programs. By incorporating cumulative effects and extending the time horizon, this study provides a more accurate and meaningful measure of vaccine effectiveness, highlighting the significant public health benefits of PCV20. Our use of a decision-analytic Markov model allows for a detailed simulation of population dynamics across multiple health outcomes and enhances the reliability and real-world applicability of the results, while the inclusion of extensive sensitivity analyses (both deterministic and probabilistic) demonstrates the robustness of the findings. Accurately estimating the NNV by incorporating cumulative, long-term public health benefits is crucial for the decision-making processes of governmental vaccine technical committees and policymakers worldwide. Our findings underscore the necessity of considering extended time horizons and indirect effects to capture the full benefits of vaccination, thus providing a more robust foundation for public health decisions.

Author Contributions

Conceptualization, M.H.R., R.C. and D.D.-M.; methodology, M.H.R., R.C. and D.D.-M.; validation, M.H.R., R.C. and D.D.-M.; formal analysis, M.H.R., R.C. and D.D.-M.; writing—original draft preparation, M.H.R., M.J.T., B.C., R.C., D.D.-M., B.M.A. and A.C.; writing—review and editing, M.H.R., M.J.T., B.C., R.C., D.D.-M., B.M.A. and A.C. All authors have read and agreed to the published version of the manuscript.

Funding

This study was sponsored by Pfizer.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Acknowledgments

Editorial support was provided by Ruth Sharf at PPD™ Evidera™ and was funded by Pfizer.

Conflicts of Interest

Ruth Chapman, Desmond Dillon-Murphy are employees of PPD™ Evidera™, which received funding from Pfizer in connection with the development of this manuscript and study. Mark H. Rozenbaum, Maria J. Tort, Blair Capitano, Benjamin M. Althouse, Alejandro Cane are employees and shareholders of Pfizer.

References

  1. Musher, D.M. Infections caused by Streptococcus pneumoniae: Clinical spectrum, pathogenesis, immunity, and treatment. Clin. Infect. Dis. 1992, 14, 801–807. [Google Scholar] [CrossRef]
  2. Dao, T.H.; Rosch, J.W. JMM Profile: Streptococcus pneumoniae: Sugar-coated captain of the men of death. J. Med. Microbiol. 2021, 70, 001446. [Google Scholar] [CrossRef]
  3. Pilishvili, T.; Lexau, C.; Farley, M.M.; Hadler, J.; Harrison, L.H.; Bennett, N.M.; Reingold, A.; Thomas, A.; Schaffner, W.; Craig, A.S.; et al. Sustained reductions in invasive pneumococcal disease in the era of conjugate vaccine. J. Infect. Dis. 2010, 201, 32–41. [Google Scholar] [CrossRef]
  4. Moore, M.R.; Link-Gelles, R.; Schaffner, W.; Lynfield, R.; Holtzman, C.; Harrison, L.H.; Zansky, S.M.; Rosen, J.B.; Reingold, A.; Scherzinger, K.; et al. Effectiveness of 13-valent pneumococcal conjugate vaccine for prevention of invasive pneumococcal disease in children in the USA: A matched case-control study. Lancet Respir. Med. 2016, 4, 399–406. [Google Scholar] [CrossRef]
  5. Centers for Disease Control and Prevention. Use of 13-valent pneumococcal conjugate vaccine and 23-valent pneumococcal polysaccharide vaccine among children aged 6–18 years with immunocompromising conditions: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb. Mortal. Wkly. Rep. 2013, 62, 521–524. [Google Scholar]
  6. Nuorti, J.P.; Whitney, C.G.; Centers for Disease Control and Prevention. Prevention of pneumococcal disease among infants and children—Use of 13-valent pneumococcal conjugate vaccine and 23-valent pneumococcal polysaccharide vaccine—Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm. Rep. 2010, 59, 1–18. [Google Scholar] [PubMed]
  7. Whitney, C.G.; Pilishvili, T.; Farley, M.M.; Schaffner, W.; Craig, A.S.; Lynfield, R.; Nyquist, A.-C.; Gershman, K.A.; Vazquez, M.; Bennett, N.M.; et al. Effectiveness of seven-valent pneumococcal conjugate vaccine against invasive pneumococcal disease: A matched case-control study. Lancet 2006, 368, 1495–1502. [Google Scholar] [CrossRef]
  8. Tomczyk, S.; Bennett, N.M.; Stoecker, C.; Gierke, R.; Moore, M.R.; Whitney, C.G.; Hadler, S.; Pilishvili, T.; Centers for Disease Control and Prevention (CDC). Use of 13-valent pneumococcal conjugate vaccine and 23-valent pneumococcal polysaccharide vaccine among adults aged ≥65 years: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb. Mortal. Wkly. Rep. 2014, 63, 822–825. [Google Scholar] [PubMed]
  9. Simonsen, L.; Taylor, R.J.; Young-Xu, Y.; Haber, M.; May, L.; Klugman, K.P. Impact of pneumococcal conjugate vaccination of infants on pneumonia and influenza hospitalization and mortality in all age groups in the United States. MBio 2011, 2, e00309-10. [Google Scholar] [CrossRef]
  10. Ramirez, J.A.; Wiemken, T.L.; Peyrani, P.; Arnold, F.W.; Kelley, R.; Mattingly, W.A.; Nakamatsu, R.; Pena, S.; Guinn, B.E.; Furmanek, S.P.; et al. Adults Hospitalized with Pneumonia in the United States: Incidence, Epidemiology, and Mortality. Clin. Infect. Dis. 2017, 65, 1806–1812. [Google Scholar] [CrossRef]
  11. Centers for Disease and Control Prevention. Pneumococcal Disease: Antibiotic-Resistant Streptococcus Pneumoniae. Updated February 6 2024. Available online: https://www.cdc.gov/pneumococcal/php/drug-resistance/index.html (accessed on 3 July 2024).
  12. Centers for Disease and Control Prevention. ACIP Evidence to Recommendation User’s Guide. Available online: https://www.cdc.gov/acip/media/pdfs/2024/09/ACIP-EtR-Users-Guide_October-1-2020.pdf (accessed on 15 January 2025).
  13. Centers for Disease Control Prevention. CDC Recommendation for Hepatitis B Vaccination Among Adults with Diabetes: Grading of Scientific Evidence in Support of Key Recommendations; CDC: Atlanta, GA, USA, 2013.
  14. Centers for Disease and Control Prevention. Advisory Committee on Immunization Practices: Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) for Pneumococcal Vaccines for Adults Aged ≥65 Years. Available online: https://www.cdc.gov/acip/grade/pneumo-vac-adult.html (accessed on 15 January 2025).
  15. Stoecker, C. Economic Assessment of PCV21 in US Adults. Available online: https://www.cdc.gov/acip/downloads/slides-2024-06-26-28/02-Pneumococcal-Stoecker-508.pdf. (accessed on 15 January 2025).
  16. Kelly, H.; Attia, J.; Andrews, R.; Heller, R.F. The number needed to vaccinate (NNV) and population extensions of the NNV: Comparison of influenza and pneumococcal vaccine programmes for people aged 65 years and over. Vaccine 2004, 22, 2192–2198. [Google Scholar] [CrossRef]
  17. Hashim, A.; Dang, V.; Bolotin, S.; Crowcroft, N.S. How and why researchers use the number needed to vaccinate to inform decision making—A systematic review. Vaccine 2015, 33, 753–758. [Google Scholar] [CrossRef]
  18. McLaughlin, J.M.; Swerdlow, D.L.; Isturiz, R.E.; Jodar, L. Rethinking number-needed-to-vaccinate for pneumococcal conjugate vaccines in older adults: Current and future implications. Vaccine 2017, 35, 5360–5365. [Google Scholar] [CrossRef]
  19. Brisson, M.; Van de Velde, N.; De Wals, P.; Boily, M.-C. Estimating the number needed to vaccinate to prevent diseases and death related to human papillomavirus infection. Can. Med. Assoc. J. 2007, 177, 464–468. [Google Scholar] [CrossRef]
  20. Tuite, A.R.; Fisman, D.N. Number-needed-to-vaccinate calculations: Fallacies associated with exclusion of transmission. Vaccine 2013, 31, 973–978. [Google Scholar] [CrossRef] [PubMed]
  21. Savulescu, C.; Krizova, P.; Valentiner-Branth, P.; Ladhani, S.; Rinta-Kokko, H.; Levy, C.; Mereckiene, J.; Knol, M.; Winje, B.A.; Ciruela, P.; et al. Effectiveness of 10 and 13-valent pneumococcal conjugate vaccines against invasive pneumococcal disease in European children: SpIDnet observational multicentre study. Vaccine 2022, 40, 3963–3974. [Google Scholar] [CrossRef] [PubMed]
  22. Flem, E.; Mouawad, C.; Palmu, A.A.; Platt, H.; Johnson, K.D.; McIntosh, E.D.; Abadi, J.; Buchwald, U.K.; Feemster, K. Indirect protection in adults ≥18 years of age from pediatric pneumococcal vaccination: A review. Expert Rev. Vaccines 2024, 23, 997–1010. [Google Scholar] [CrossRef]
  23. Ahmed, S.S.; Pondo, T.; Xing, W.; McGee, L.; Farley, M.; Schaffner, W.; Thomas, A.; Reingold, A.; Harrison, L.H.; Lynfield, R.; et al. Early impact of 13-valent pneumococcal conjugate vaccine use on invasive pneumococcal disease among adults with and without underlying medical conditions—United States. Clin. Infect. Dis. 2020, 70, 2484–2492. [Google Scholar] [CrossRef]
  24. Regev-Yochay, G.; Paran, Y.; Bishara, J.; Oren, I.; Chowers, M.; Tziba, Y.; Istomin, V.; Weinberger, M.; Miron, D.; Temper, V.; et al. Early impact of PCV7/PCV13 sequential introduction to the national pediatric immunization plan, on adult invasive pneumococcal disease: A nationwide surveillance study. Vaccine 2015, 33, 1135–1142. [Google Scholar] [CrossRef]
  25. Harboe, Z.B.; Dalby, T.; Weinberger, D.M.; Benfield, T.; Mølbak, K.; Slotved, H.C.; Suppli, C.H.; Konradsen, H.B.; Valentiner-Branth, P. Impact of 13-valent pneumococcal conjugate vaccination in invasive pneumococcal disease incidence and mortality. Clin. Infect. Dis. 2014, 59, 1066–1073. [Google Scholar] [CrossRef] [PubMed]
  26. Centers for Disease Control Prevention. Direct and indirect effects of routine vaccination of children with 7-valent pneumococcal conjugate vaccine on incidence of invasive pneumococcal disease--United States, 1998–2003. MMWR Morb. Mortal. Wkly. Rep. 2005, 54, 893–897. [Google Scholar]
  27. Centers for Disease Control and Prevention. Guidance for Health Economics Studies Presented to the Advisory Committee on Immunization Practices (ACIP), 2024 Update. Updated 18 December 2014. Available online: https://www.cdc.gov/acip/about/health-economics-studies.html (accessed on 15 January 2025).
  28. Huerta, J.L.; Ta, A.; Vinand, E.; Torres, G.I.; Wannaadisai, W.; Huang, L. PCV20 for the prevention of invasive pneumococcal disease in the Mexican pediatric population: A cost-effectiveness analysis. Hum. Vaccines Immunother. 2025, 21, 2475594. [Google Scholar] [CrossRef]
  29. Çakar, E.; Ta, A.; Peters, M.; Vinand, E.; Waterval-Overbeek, A.; Ilic, A.; Perdrizet, J. Economic Evaluation of Transitioning to the 20-Valent Pneumococcal Conjugate Vaccine in the Dutch Paediatric National Immunisation Programme. Infect. Dis. Ther. 2025, 14, 527–547. [Google Scholar] [CrossRef] [PubMed]
  30. Rey-Ares, L.; Ta, A.; Freigofaite, D.; Warren, S.; Mullen, M.M.; Carballo, C.; Huang, L. Cost-effectiveness analysis of the pediatric 20-valent pneumococcal conjugate vaccine compared with lower-valent alternatives in Argentina. Vaccine 2024, 42, 126043. [Google Scholar] [CrossRef]
  31. Ta, A.; Kühne, F.; Laurenz, M.; von Eiff, C.; Warren, S.; Perdrizet, J. Cost-effectiveness of PCV20 to prevent pneumococcal disease in the pediatric population: A German societal perspective analysis. Infect. Dis. Ther. 2024, 13, 1333–1358. [Google Scholar] [CrossRef] [PubMed]
  32. Rozenbaum, M.H.; Chilson, E.; Farkouh, R.; Huang, L.; Cane, A.; Arguedas, A.; Tort, M.J.; Snow, V.; Averin, A.; Weycker, D.; et al. Cost-Effectiveness of 20-Valent Pneumococcal Conjugate Vaccine Among US Children with Underlying Medical Conditions. Infect. Dis. Ther. 2024, 13, 745–760. [Google Scholar] [CrossRef] [PubMed]
  33. Rozenbaum, M.H.; Huang, L.; Cane, A.; Arguedas, A.; Chapman, R.; Dillon-Murphy, D.; Tort, M.J.; Snow, V.; Chilson, E.; Farkouh, R. Cost-effectiveness and impact on infections and associated antimicrobial resistance of 20-valent pneumococcal conjugate vaccine in US children previously immunized with PCV13. J. Med. Econ. 2024, 27, 644–652. [Google Scholar] [CrossRef]
  34. Fridh, A.-C.; Palmborg, A.; Ta, A.; Freigofaite, D.; Warren, S.; Perdrizet, J. An economic evaluation of pneumococcal conjugate vaccines, PCV20 versus PCV15, for the prevention of pneumococcal disease in the Swedish pediatric population. Hum. Vaccines Immunother. 2024, 20, 2400751. [Google Scholar] [CrossRef]
  35. Shinjoh, M.; Togo, K.; Hayamizu, T.; Yonemoto, N.; Morii, J.; Perdrizet, J.; Kamei, K. Cost-effectiveness analysis of 20-valent pneumococcal conjugate vaccine for routine pediatric vaccination programs in Japan. Expert Rev. Vaccines 2024, 23, 485–497. [Google Scholar] [CrossRef]
  36. Rey-Ares, L.; Averin, A.; Mac Mullen, M.; Hariharan, D.; Atwood, M.; Carballo, C.; Huang, L. Cost-effectiveness of 20-valent pneumococcal conjugate vaccine in Argentinean adults. Infect. Dis. Ther. 2024, 13, 1235–1251. [Google Scholar] [CrossRef]
  37. Kang, D.-W.; Choe, Y.J.; Lee, J.-Y.; Suk, I.-A.; Kim, Y.-S.; Kim, H.-Y.; Byun, B.-K.; Park, S.-K. Cost-effectiveness analysis of the 20-valent pneumococcal conjugate vaccine for the pediatric population in South Korea. Vaccine 2024, 42, 126000. [Google Scholar] [CrossRef] [PubMed]
  38. Lytle, D.; Grajales Beltrán, A.G.; Perdrizet, J.; Yahia, N.A.; Cane, A.; Yarnoff, B.; Chapman, R. Cost-effectiveness analysis of PCV20 to prevent pneumococcal disease in the Canadian pediatric population. Hum. Vaccines Immunother. 2023, 19, 2257426. [Google Scholar] [CrossRef]
  39. Warren, S.; Barmpouni, M.; Kossyvaki, V.; Gourzoulidis, G.; Perdrizet, J. Estimating the clinical and economic impact of switching from the 13-valent pneumococcal conjugate vaccine (PCV13) to higher-valent options in Greek infants. Vaccines 2023, 11, 1369. [Google Scholar] [CrossRef]
  40. Wilson, M.; Lucas, A.; Mendes, D.; Vyse, A.; Mikudina, B.; Czudek, C.; Ellsbury, G.F.; Perdrizet, J. Estimating the cost-effectiveness of switching to higher-valency pediatric pneumococcal conjugate vaccines in the United Kingdom. Vaccines 2023, 11, 1168. [Google Scholar] [CrossRef]
  41. Huang, L.; McDade, C.L.; Perdrizet, J.E.; Wilson, M.R.; Warren, S.A.; Nzenze, S.; Sewdas, R. Cost-effectiveness analysis of the South African infant national immunization program for the prevention of pneumococcal disease. Infect. Dis. Ther. 2023, 12, 933–950. [Google Scholar] [CrossRef]
  42. Hoshi, S.-L.; Shono, A.; Seposo, X.; Okubo, R.; Kondo, M. Cost-effectiveness analyses of 15- and 20-valent pneumococcal conjugate vaccines for Japanese elderly. Vaccine 2022, 40, 7057–7064. [Google Scholar] [CrossRef] [PubMed]
  43. Kang, D.-W.; Kim, C.-R.; Song, J.Y.; Park, S.-K. Cost-effectiveness of the 20-valent pneumococcal conjugate vaccine versus the 23-valent pneumococcal polysaccharide vaccine for older adults in South Korea. Vaccine 2024, 42, 871–878. [Google Scholar] [CrossRef] [PubMed]
  44. Gourzoulidis, G.; Barmpouni, M.; Kossyvaki, V.; Vietri, J.; Tzanetakos, C. Health and economic outcomes of 20-valent pneumococcal conjugate vaccine compared to 15-valent pneumococcal conjugate vaccine strategies for adults in Greece. Front. Public Health 2023, 11, 1229524. [Google Scholar] [CrossRef]
  45. Mendes, D.; Averin, A.; Atwood, M.; Sato, R.; Vyse, A.; Campling, J.; Weycker, D.; Slack, M.; Ellsbury, G.; Mugwagwa, T. Cost-effectiveness of using a 20-valent pneumococcal conjugate vaccine to directly protect adults in England at elevated risk of pneumococcal disease. Expert Rev. Pharmacoeconomics Outcomes Res. 2022, 22, 1285–1295. [Google Scholar] [CrossRef]
  46. Wilson, M.; McDade, C.; Beby-Heijtel, A.T.; Waterval-Overbeek, A.; Sundaram, V.; Perdrizet, J. Assessing public health impact of four pediatric pneumococcal conjugate vaccination strategies in the Netherlands. Infect. Dis. Ther. 2023, 12, 1809–1821. [Google Scholar] [CrossRef]
  47. Senders, S.; Klein, N.P.; Tamimi, N.; Thompson, A.; Baugher, G.; Trammel, J.; Peng, Y.; Giardina, P.; Scully, I.L.; Pride, M.; et al. A phase three study of the safety and immunogenicity of a four-dose series of 20-valent pneumococcal conjugate vaccine in healthy infants. Pediatr. Infect. Dis. J. 2024, 43, 596–603. [Google Scholar] [CrossRef] [PubMed]
  48. Briggs, A.; Claxton, K.; Sculpher, M. Decision Modelling for Health Economic Evaluation; Oxford University Press: Oxford, UK, 2006. [Google Scholar]
  49. Lewnard, J.A.; Hong, V.; Bruxvoort, K.J.; Grant, L.R.; Jódar, L.; Cané, A.; Arguedas, A.; Pomichowski, M.E.; Gessner, B.D.; Tartof, S.Y. Burden of lower respiratory tract infections preventable by adult immunization with 15- and 20-valent pneumococcal conjugate vaccines in the United States. Clin. Infect. Dis. 2023, 77, 1340–1352. [Google Scholar] [CrossRef]
  50. de Boer, P.T.; van Werkhoven, C.H.; van Hoek, A.J.; Knol, M.J.; Sanders, E.A.M.; Wallinga, J.; de Melker, H.E.; Steens, A. Higher-valency pneumococcal conjugate vaccines in older adults, taking into account indirect effects from childhood vaccination: A cost-effectiveness study for the Netherlands. BMC Med. 2024, 22, 69. [Google Scholar] [CrossRef]
  51. Lo, S.W.; Gladstone, R.A.; Van Tonder, A.J.; Lees, J.A.; du Plessis, M.; Benisty, R.; Givon-Lavi, N.; Hawkins, P.A.; Cornick, J.E.; Kwambana-Adams, B.; et al. Pneumococcal lineages associated with serotype replacement and antibiotic resistance in childhood invasive pneumococcal disease in the post-PCV13 era: An international whole-genome sequencing study. Lancet Infect. Dis. 2019, 19, 759–769. [Google Scholar] [CrossRef] [PubMed]
  52. Lewnard, J.A.; Hanage, W.P. Making sense of differences in pneumococcal serotype replacement. Lancet Infect. Dis. 2019, 19, e213–e220. [Google Scholar] [CrossRef] [PubMed]
Vaccines 13 00805 g001
Figure 1. NNV for outcomes in the base case analysis. Abbreviations: IPD = invasive pneumococcal disease; OM = otitis media.
Figure 1. NNV for outcomes in the base case analysis. Abbreviations: IPD = invasive pneumococcal disease; OM = otitis media.
Vaccines 13 00805 g001
Vaccines 13 00805 g002
Figure 2. Annual number-needed-to-vaccinate (NNV) for PCV20 versus PCV13 under scenarios that include versus exclude indirect (herd) effects, by disease outcome, over a 25-year horizon. The x-axis shows calendar years following PCV20 introduction; the y-axis shows the NNV required to avert one additional case. For each outcome—(A) invasive pneumococcal disease (IPD), (B) hospitalized non-bacteremic pneumonia, (C) non-hospitalized pneumonia, and (D) otitis media (OM)—solid lines trace the base-case scenario that incorporates indirect protection, while dashed lines represent a counterfactual in which herd effects are set to zero. Abbreviations: IPD = invasive pneumococcal disease; OM = otitis media; NNV = number needed to vaccinate.
Figure 2. Annual number-needed-to-vaccinate (NNV) for PCV20 versus PCV13 under scenarios that include versus exclude indirect (herd) effects, by disease outcome, over a 25-year horizon. The x-axis shows calendar years following PCV20 introduction; the y-axis shows the NNV required to avert one additional case. For each outcome—(A) invasive pneumococcal disease (IPD), (B) hospitalized non-bacteremic pneumonia, (C) non-hospitalized pneumonia, and (D) otitis media (OM)—solid lines trace the base-case scenario that incorporates indirect protection, while dashed lines represent a counterfactual in which herd effects are set to zero. Abbreviations: IPD = invasive pneumococcal disease; OM = otitis media; NNV = number needed to vaccinate.
Vaccines 13 00805 g002
Vaccines 13 00805 g003
Figure 3. One-way deterministic sensitivity analysis (DSA) for the total-disease number-needed-to-vaccinate (NNV) when replacing PCV13 with PCV20 over a 25-year time horizon. The vertical gray line marks the base-case NNV (≈5.9). For each input, the model was rerun twice—once with the parameter set to its lower bound (salmon bars) and once with the upper bound (teal bars)—while all other inputs remained at their base-case values. Horizontal bar length therefore represents the absolute change in NNV attributable to that single parameter; longer bars denote greater influence. Parameters are ordered from top to bottom by decreasing impact. The x-axis shows the resulting NNV, ranging from 5.39 (most favorable) to 6.67 (least favorable). Abbreviations: NNV = number needed to vaccinate; PCV20 = 20-valent pneumococcal conjugate vaccine; SoC = standard of care.
Figure 3. One-way deterministic sensitivity analysis (DSA) for the total-disease number-needed-to-vaccinate (NNV) when replacing PCV13 with PCV20 over a 25-year time horizon. The vertical gray line marks the base-case NNV (≈5.9). For each input, the model was rerun twice—once with the parameter set to its lower bound (salmon bars) and once with the upper bound (teal bars)—while all other inputs remained at their base-case values. Horizontal bar length therefore represents the absolute change in NNV attributable to that single parameter; longer bars denote greater influence. Parameters are ordered from top to bottom by decreasing impact. The x-axis shows the resulting NNV, ranging from 5.39 (most favorable) to 6.67 (least favorable). Abbreviations: NNV = number needed to vaccinate; PCV20 = 20-valent pneumococcal conjugate vaccine; SoC = standard of care.
Vaccines 13 00805 g003
Table 1. NNV for outcomes in the base case.
Table 1. NNV for outcomes in the base case.
OutcomeMean NNV (95% CI a)
IPD854.32 (773.62–952.85)
Hospitalized pneumonia105.72 (98.18–114.19)
Non-hospitalized pneumonia 24.92 (23.18–26.87)
Any hospitalization b94.08 (87.89–100.82)
OM8.55 (7.80–9.37)
Total disease c5.96 (5.60–6.36)
Antibiotic-resistant cases7.08 (6.64–7.55)
Antibiotic prescription 10.67 (10.02–11.38)
Death2488.07 (2290.50–2726.50)
Abbreviations: CI = confidence interval; IPD = invasive pneumococcal disease; LY = life year; NNV = number needed to vaccinate; OM = otitis media; QALY = quality-adjusted life year; a 95% CI estimates were derived from the probabilistic sensitivity analysis; b includes all IPD cases and hospitalization pneumonia; c includes all disease outcomes listed above (IPD, hospitalized and non-hospitalized pneumonia and OM).
Table 2. Mean NNV for all outcomes of PCV20 versus PCV1 for scenario analysis with and without heard effects and a time horizon of 1, 10, and 25 years.
Table 2. Mean NNV for all outcomes of PCV20 versus PCV1 for scenario analysis with and without heard effects and a time horizon of 1, 10, and 25 years.
1-Year Time Horizon10-Year Time Horizon25-Year Time Horizon
With Indirect EffectsW/o Indirect Effects<18 with Indirect EffectsWith Indirect EffectsW/o Indirect Effects<18 with Indirect EffectsWith Indirect EffectsW/o Indirect Effects<18 with Indirect Effects
IPD28,78828,78828,788121219,18011,01685415,7269040
Any hospitalization a 15341534153412498045594791370
Hospitalized pneumonia1620162016201381033475106833386
Non-hospitalized pneumonia 19961996199631614702542856
OM5757571131119259
Total disease b53535372996237
Resistant cases636363934117269
Antibiotic prescriptions959595135116114013
Death104,207104,207104,207366692,70340,974248872,05432,853
Abbreviations: IPD = invasive pneumococcal disease; LY = life year; NA = not available; OM = otitis media; QALY = quality-adjusted life year; a includes all IPD cases and hospitalization pneumonia; b includes all disease outcomes listed above (IPD, hospitalized and non-hospitalized pneumonia and OM).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rozenbaum, M.H.; Tort, M.J.; Capitano, B.; Chapman, R.; Dillon-Murphy, D.; Althouse, B.M.; Cane, A. A Novel Approach to Estimate the Impact of PCV20 Immunization in Children by Incorporating Indirect Effects to Generate the Number Needed to Vaccinate. Vaccines 2025, 13, 805. https://doi.org/10.3390/vaccines13080805

AMA Style

Rozenbaum MH, Tort MJ, Capitano B, Chapman R, Dillon-Murphy D, Althouse BM, Cane A. A Novel Approach to Estimate the Impact of PCV20 Immunization in Children by Incorporating Indirect Effects to Generate the Number Needed to Vaccinate. Vaccines. 2025; 13(8):805. https://doi.org/10.3390/vaccines13080805

Chicago/Turabian Style

Rozenbaum, Mark H., Maria J. Tort, Blair Capitano, Ruth Chapman, Desmond Dillon-Murphy, Benjamin M. Althouse, and Alejandro Cane. 2025. "A Novel Approach to Estimate the Impact of PCV20 Immunization in Children by Incorporating Indirect Effects to Generate the Number Needed to Vaccinate" Vaccines 13, no. 8: 805. https://doi.org/10.3390/vaccines13080805

APA Style

Rozenbaum, M. H., Tort, M. J., Capitano, B., Chapman, R., Dillon-Murphy, D., Althouse, B. M., & Cane, A. (2025). A Novel Approach to Estimate the Impact of PCV20 Immunization in Children by Incorporating Indirect Effects to Generate the Number Needed to Vaccinate. Vaccines, 13(8), 805. https://doi.org/10.3390/vaccines13080805

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop