Liv Nymark 2017

Nymark, Liv. 2017. Economic evaluation of the indirect effects of vaccines. The herd immunity effects of vaccines against human transmissible diseases and the non-specific effects of the measles-mumps-rubella and Bacillus Calmette-Guérin vaccines. Bandim Health Project, CVIVA, LSHTM and Aarhus University, Department of Public Health. 202 p. 


The topic of this thesis is positioned within a broader discussion in health economics regarding the value of externalities (‘broader effects’) of vaccines in economic evaluations. 
The economic literature suggests that the indirect (versus direct) benefits of vaccines are often not captured in economic evaluations. This may imply potential underestimation of the value of vaccination. The dissertation focuses on the economic value of indirect effects of vaccines. More specifically, the economic value of herd immunity of human transmission vaccines (study l) and the beneficial non-specific effects (NSEs) of the live vaccines measles-mumps-rubella (MMR) (study ll) and Bacillus Calmette-Guérin (BCG) (study lll) are addressed. 


The primary objective of the PhD thesis was to analyse and estimate the economic role of indirect effects of vaccines (i.e. herd immunity effects and beneficial non-specific effects) The PhD thesis consists of three original research papers. The specific aims were:

  • To develop a comprehensive review of the cost-effectiveness studies of human transmission vaccines which include herd immunity effects and describe which methods were used to account for herd immunity effects. 
  • Additionally, to explore if NSE (i.e. all-cause mortality or morbidity) have been included in cost-effectiveness analyses of human transmission vaccines. 
  • To estimate the average cost per any infectious disease hospitalization in Danish birth cohorts from 2008-2010 based on MMR vaccination status
  • To provide economic estimates of the savings associated with avoiding any infectious disease hospitalizations (non-specific effects) in alternative MMR vaccination programs compared with vaccination program costs and varying coverage rates under incremental changes to the recommended vaccination program
  • To determine the cost-effectiveness of Bacillus Calmette–Guérin immunization in Danish neonates with a hereditary predisposition to atopic dermatitis (non-specific effects) compared to no Bacillus Calmette–Guérin immunization expressed as an incremental cost-effectiveness ratio.  

The main findings of this thesis were:

  • The systematic review (study l) showed that of 625 economic evaluations of human transmission vaccines, published between 1976-2015, only 172 (28%) included herd immunity effects
  • The majority of the studies which considered herd immunity were published between 2007-2015
  • While 4% of studies included herd immunity in 2001, 53% of those published in 2015 did this. 
  • Pneumococcal, human papilloma and rotavirus vaccines represented the majority of studies (63%) considering herd immunity. 
  • Ninety-five of the 172 studies utilised a static model, 59 applied a dynamic model, eight a hybrid model and ten did not clearly state which method was used. 
  • Relatively crude methods and assumptions were used in the majority of the static model studies.
  • In study ll, the results of the register-based cost analysis among the 2008-2010 birth cohorts suggested that the children who did not have MMR vaccine as their most recent vaccine had a higher average cost per any infectious disease hospitalization at every age in months (15-48) compared to those who had MMR vaccine as their most recent vaccine
  • The main results from the decision model predicted that schedule D (MMR at 6 months and 13 months) returned the largest savings (15 million DKK) compared to schedule A (recommended) in the base case; when actual coverage rates were applied and MMR was provided at an earlier age. 
  • In the sensitivity analysis in which the coverage rate was increased to 95% schedule D also returned the largest savings compared to schedule A (recommended).  
  • In study lll, the cost-utility analysis of BCG vaccination for the prevention of atopic dermatitis in Danish neonates with a predisposition suggested that during the 13-months follow-up period of the Danish Calmette study, BCG was more effective but also costlier than no BCG. 
  • In additional scenario analyses in which BCG efficacy was assumed to extend beyond the 13-months follow-up, BCG appeared to be the dominant strategy.


The proportion of economic evaluations using a dynamic model has increased in recent years (study l). However, 55% of the included studies used a static model for estimating herd immunity. Values from a static model can only be considered appropriate if high quality surveillance data are incorporated into the analysis. Without this, the results are questionable and they should only be included in sensitivity analysis. Future studies could explore the implications of different model designs (static versus dynamic) on the results.  Study ll demonstrated that the decision model type was capable of estimating the economic impact of NSE in terms of avoided any infectious disease hospitalizations. Further economic studies are warranted to understand the impact of NSE as well as build the evidence base to determine if the inclusion of NSE in cost-effectiveness analyses should be considered routinely. In the base case of study lll (i.e. with BCG efficacy during the 13 months follow-up period) BCG vaccination was more effective but costlier. Future research is needed to understand the implication of this and replicating the cost-effectiveness study in other high income contexts could enhance our understanding of the benefit of offering the BCG vaccination to infants with a high risk of developing atopic dermatitis.