Article Text

Download PDFPDF

Potential impact of introducing the pneumococcal conjugate vaccine into national immunisation programmes: an economic-epidemiological analysis using data from India
  1. Itamar Megiddo1,2,
  2. Eili Klein2,3,
  3. Ramanan Laxminarayan1,2,4
  1. 1 Department of Management Science, University of Strathclyde, Glasgow, UK
  2. 2 Center for Disease Dynamics Economics and Policy, Washington, District of Columbia, USA
  3. 3 Department of Emergency Medicine, Johns Hopkins University, Baltimore, Maryland, USA
  4. 4 Princeton Environmental Institute, Princeton University, Princeton, New Jersey, USA
  1. Correspondence to Dr Itamar Megiddo; itamar.megiddo{at}


Pneumococcal pneumonia causes an estimated 105 000 child deaths in India annually. The planned introduction of the serotype-based pneumococcal conjugate vaccine (PCV) is expected to avert child deaths, but the high cost of PCV relative to current vaccines provided under the Universal Immunization Programme has been a concern. Cost-effectiveness studies from high-income countries are not readily comparable because of differences in the distribution of prevalent serotypes, population and health systems. We extended IndiaSim, our agent-based simulation model representative of the Indian population and health system, to model the dynamics of Streptococcus pneumoniae. This enabled us to evaluate serotype and overall disease dynamics in the context of the local population and health system, an aspect that is missing in prospective evaluations of the vaccine. We estimate that PCV13 introduction would cost approximately US$240 million and avert US$48.7 million in out-of-pocket expenditures and 34 800 (95% CI 29 600 to 40 800) deaths annually assuming coverage levels and distribution similar to DPT (diphtheria, pertussis and tetanus) vaccination (~77%). Introducing the vaccine protects the population, especially the poorest wealth quintile, from potentially catastrophic expenditure. The net-present value of predicted money-metric value of insurance for 20 years of vaccination is US$160 000 (95% CI US$151 000 to US$168 000) per 100 000 under-fives, and almost half of this protection is for the bottom wealth quintile (US$78 000; 95% CI 70 800 to 84 400). Extending vaccination to 90% coverage averts additional lives and provides additional financial risk protection. Our estimates are sensitive to immunity parameters in our model; however, our assumptions are conservative, and if willingness to pay per years of life lost averted is US$228 or greater, then introducing the vaccine is more cost-effective than our baseline (no vaccination) in more than 95% of simulations.

  • pneumonia
  • pneumococcal disease
  • vaccines
  • mathematical modelling
  • health economics

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See:

View Full Text

Statistics from


  • Handling editor Alberto L Garcia-Basteiro

  • Contributors All authors contributed extensively to the work. IM, EK and RL conceived and contributed to the design. IM developed the models and conducted the analyses. IM, EK and RL all verified the analytical methods, discussed the results and contributed to writing the final manuscript.

  • Funding This work was funded by the Bill & Melinda Gates Foundation through the Global Antibiotic Resistance Partnership (GARP) Phase I and II (grant no. OPP1034526) and Building Regional Networks Using the GARP (grant no. OPP1135911), and RL was also supported by Princeton University through its Grand Challenges Program.

  • Competing interests RL and IM were supported by research grants from the Bill & Melinda Gates Foundation, and RL reports grants from Princeton University Grand Challenges Program for the submitted work; EK received research grants from MedImmune that are unrelated to the current paper; no other relationships or activities that could appear to have influenced the submitted work.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data from model simulations are several gigabits and are therefore not currently in an online repository, but the data will be made available upon request.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.