Save the Children UK GVAP & GIVS target vaccine analysis

Question:

Impact of scaling up DTP, Hib, Pneumococcal,Rotavirusand Measles vaccines as a combination and individually on U5 deaths, U5 pertussis, pneumonia, meningitis, diarrhea, and measles deaths, and U5 deaths due to each vaccine in 73 GAVI-eligible countries

Packages/Interventions of interest:

Package of 5vaccines (DTP, Hib, Pneumococcal, Rotavirus and Measles)

Scenarios of interest:

  1. Scaling each vaccine (DTP, Hib, Pneumo, Rota, Measles) individually from their current coverage to 90% by year 2020.
  2. Scaling DTP, Hib, Pneumo, Rota and Measles vaccine as a combination from their current coverage to 90% by year 2020
  3. Scaling DTP vaccine from its current coverage to 90% by year 2015.

Note that when any vaccine’s base year coverage exceeds 90%, the coverage of that vaccine stays constant between 2011 to 2020.

Baseline Values:

Population trendis obtained from the The2010 Revisionof theWorld Population Prospects published by the Population Division of the United Nations Department of Economic and Social Affairs of the United Nations Secretariat.

Child mortality estimates are obtained from The estimates from year 2010.5 were used.

Causes of death in children under 5 are obtained from the following article

  • Black, Robert E , Simon Cousens, Hope L Johnson, Joy E Lawn, Igor Rudan, Diego G Bassani, PrabhatJha, Harry Campbell, Christa Fischer Walker, Richard Cibulskis, Thomas Eisele, Li Liu, Colin Mathers, for the Child Health Epidemiology Reference Group of WHO and UNICEF (2010).Global, regional, and national causes of child mortality in 2008: a systematic analysis.Lancet. 375 (9730): 1969 - 1987.

Vaccine coverage estimates

Vaccine coverage values 2004-2011used is from WHO/UNICEF estimates, last updated Aug 3, 2011

Vaccine effectiveness is based on their impact on specific causes of death:

An affected fraction refers to the percent of a specific cause of death which can be averted by an intervention. This refers to etiology when available.

Standard vaccine effect sizes and affected fraction used in LiST:

  • DPT vaccine’s effect on pertussis.
  • DPT vaccine (3 doses) has an effect of 0.85, meaning that full vaccination can prevent 85% of pertussis deaths.
  • It is assumed that all (100%) pertussis deaths can benefit by full coverage of the vaccine.
  • The effect size is based on expert opinion
  • Measles vaccine’s effect on measles.
  • Measles vaccine (1 dose) has an effect of 0.85, meaning that full vaccination can prevent 85% of measles deaths.
  • It is assumed that all (100%) measles deaths can benefit by full coverage of the vaccine.
  • Pneumococcal and Hib vaccine’s effect on pneumonia and meningitis
  • Pneumococcal vaccine (full coverage) has an effect of 0.26, meaning that full vaccination can prevent 26% of pneumonia deaths, and since the review is not yet finished, we are assuming the same effect (26%) on meningitis deaths.
  • Hib vaccine (full coverage) has an effect of 0.18. This means that full vaccination can prevent 18% of pneumonia deaths and, since the review is not yet finished, we are also assuming the same effect (18%) on meningitis deaths.
  • The affected fraction for each of these vaccines is considered to be the entire population of pneumonia deaths because that is how the reviews were completed. Thus the presented effect sizes are in essence, combinations of both effectiveness of the vaccines and the affected fraction, or proportion of pneumonia due to these bacteria.
  • Reference: Theodoratou, Evropi, Sue Johnson, ArnoupeJhass, Shabir A Madhi, Andrew Clark, Cynthia Boschi-Pinto, Sunil Bhopal, Igor Rudan, Harry Campbell (2010).The effect of Haemophilusinfluenzae type b and pneumococcal conjugate vaccines on childhood pneumonia incidence, severe morbidity and mortality.Int. J. Epidemiol. 39: i172-i185.
  • Rotavirus vaccine’s effect on diarrhea
  • 39% of diarrhea deaths can benefit by full coverage of the vaccine.
  • 5 different Rota vaccine effect sizes were used. Countries were categorized as 'Africa', 'Low mortality Asia', 'High mortality Asia', 'Latin America', and 'Developed', and their effect sizes as follow
  • Africa: 49.7%
  • Low mortality Asia: 87.9%
  • High mortality Asia: 42.7%
  • Latin America: 81%
  • Developed: 90.6%
  • Reference: Fischer Walker CL, Black RE. Rotavirus vaccine and diarrhea mortality: quantifying regional variation in effect size. BMC Public Health 2011;11Suppl 3:S16.

Time frame of interest:

Baseline year: 2011

Target year: 2015 for GIVS target, 2020 for GVAP target

LiST Version Used:

Spectrum 4.48 Beta 14

Outputs of Interest:

•Yearly U5 deaths (and deaths averted) by country between 2011-2020, total U5 deaths (and deaths averted) by countryin 2011-2020 for the three scenarios

•Yearly U5 deaths (and deaths averted)by disease cause by country between 2011-2020, total U5 deaths (and deaths averted) by disease cause by countryin 2011-2020 for the three scenarios

•Yearly U5 deaths averted by first dose of vaccine by country between 2011-2020, total U5 deaths averted by first dose of vaccine by countryin 2011-2020 for the three scenarios

Note on spreadsheet

All three spreadsheets have a similar layout. The first group of tabs shows U5 death counts. In each individual tab, there are results of U5 deaths count in the no scale up scenario, U5 deaths count in the scaled up scenario, and U5 deaths averted (which is the different between the previous two scenarios).

The second group of tabs shows U5 death counts by disease cause. In each individual tab there are results of U5 deaths count by each disease cause in the no scale up scenario, U5 deaths count by each disease cause in the scaled up scenario, and the U5 deaths averted by each disease cause.

The third group of tabs shows U5 deaths averted by the year that the first dose of vaccine was received by child, compared to the no scale up scenario. This means that the deaths averted listed under any given year, say 2015, is the full benefit of vaccinating that 2015 cohort with the given vaccine. Since vaccinating a cohort with a given vaccine will have an impact for the next 4 years, the deaths averted number under year 2015 will consist of vaccinating that cohort's impact that goes from 2015-2019. This explains why the deaths averted by year of first dose number is always bigger than the deaths averted by disease cause when you scale each vaccine up individually.

To summarize all the key results, I have included a summary sheet (2nd tab on'Save GVAP individual vaccines LiST results'). Each spreadsheet also starts with a tab with the base year vaccine coverage, which are the 2010 estimates from WHO/UNICEF.