Overall, our results support the WHO recommendation that countries introduce rubella vaccine into their regular vaccination program if they can maintain coverage above 80% through a combination of routine vaccination and SIAs 
. Under realistic assumptions about mixing between age groups the number of settings where we expect an increase in CRS goes up, however the 80% rule still seems adequate to avoid an increase in CRS. Using the charts presented here stake holders can get a sense of the likely outcome of adding rubella vaccine to a countries program without performing a sophisticated country specific modeling exercise, which may be time consuming and require extensive additional data collection. As more data on rubella incidence becomes available, these estimates can be updated to reflect our growing knowledge of rubella transmission dynamics.
Of the 40 countries included in our R0
estimation, 19 had routine MCV vaccination rates of 80% or greater, and would be able to introduce rubella vaccination by the WHO criteria with no supplemental campaigns (). Cape Verde, the only country considered here to have introduced rubella vaccine as of 2011 (vaccine was introduced in 2010) 
, has both high vaccination coverage and a low birthrate, hence is firmly in the “safe-zone”. Over the last year, Rwanda has added rubella to their immunization schedule, based on a funding window opened by GAVI; and Ghana and Senegal are expected to follow shortly 
. While all three have high enough MCV1 coverage to justify introducing rubella vaccine, they have relative high birth-rates as well and should be careful to maintain high coverage.
Figure 3 WHO-UNICEF estimated first dose measles (MCV1) vaccination coverage (from ) for 2011, 2010 birth rate (from ), R0 point estimate and population age structure (0–80 years of age, from ) for 40 African countries used in the analysis. (more ...)
By attempting to be general across a wide range of countries and settings, this work necessarily makes many generalizations and has several limitations. Data on rubella incidence is based on the analysis of suspected measles cases, and relies on surveillance systems that differ markedly by country and may be biased towards detecting rubella in particular age groups. Estimating the age distribution of rubella cases from measles surveillance data may bias the estimated age of infection downwards. However, this will tend to increase the estimate of R0
, thereby increasing the predicted required coverage and resulting in more conservative predictions. Likewise, while most countries considered in this analysis have a pyramidal age structure, some do not (); which may lead to a slight overestimation in the overall distribution of R0
across countries. While differences in age structure may affect estimates of R0
, the critical threshold appears to be insensitive to drivers of age structure other than birth (i.e., mortality 
) Assumptions about age specific mixing are based upon studies conducted in Europe, where MMR vaccine is already used widely, and may not apply to African and Asian countries considering introducing MR vaccine. However, the observation of assortative mixing by age has been replicated in many settings 
Previous work has, for the most part, either ignored the possible effect of assortative mixing between age groups on the introduction of rubella vaccine 
, or used simple matrix structures that many not fully capture increased mixing between parents and children 
. If assortative mixing is ignored, one would conclude that in low birth rate countries any level of rubella vaccination would lead to a reduction in CRS cases. However, under more realistic assumptions about how ages interact, low levels of vaccine coverage can lead to an increase in the incidence of CRS even when the birthrate is low. This is particularly troubling as increasing numbers of individuals in the United States and Europe decide to forgo MMR vaccination, dramatically decreasing coverage levels in some areas. Vaccine refusal tends to cluster geographically 
, and has already led to outbreaks of several previously eliminated childhood infections (e.g., measles, pertussis 
, rubella 
). If this trend continues we may see a resurgence in CRS cases in developed countries.
Even if a country is within the range where introducing rubella vaccine is predicted to result in a reduction in CRS, countries must carefully consider their individual situation. Administrative coverage estimates may overestimate actual vaccine coverage 
, whether by underestimating the size of the target population or overestimating the number vaccinated. More importantly, if countries are achieving the coverage necessary to safely introduce rubella vaccine by supplementing routine coverage with SIAs, it is imperative they keep performing regular SIAs until their routine vaccination program achieves sufficient coverage. Regardless of how coverage is achieved, if levels of rubella vaccination drop off then CRS cases may increase. Conversely, vaccination in the private sector may increase the risk of CRS and change the risk-benefit tradeoff with the introduction of vaccine through state sponsored programs.
The threshold R0
identified for each birth rate/vaccination coverage combination is such that vaccination reduces the cumulative burden of CRS over 30 years, but makes no predictions about transient increases in the CRS burden. Previous reports of CRS increases in Greece and Costa Rica likely reflect this pattern 
, with single years where a large number of CRS cases occur, despite an overall decline in cumulative CRS burden. Policy makers should consider whether such events can be dealt with by the health system and the degree to which such events may adversely affect attitudes towards vaccination.
Our projections also ignore the effect of local disease dynamics on CRS burden. Local extinction of rubella may lead to an increase in the CRS burden by allowing individuals to age into childbearing years without exposure to the infection, up until the point where rubella is re-introduced 
. Local extinction may be particularly likely in areas where population mixing of more remote communities with central population centers is rare. Analysis of detailed data from South Africa makes it clear such extinctions regularly occur in that country 
, but how widespread this is and the interaction with birth rates remains unclear. Heterogeneity in vaccination coverage then becomes an important issue, as the CRS burden may increase in under-served districts. This is particularly likely to be an issue when vaccination policy is defined at a scale that differs from that of transmission 
; and should be another consideration for policy makers.
Rubella vaccination is part of a renewed focus on vaccination through Decade of Vaccines and other initiatives by the funding community and public health agencies. Like the other vaccines in these initiatives, rubella vaccine has the potential save lives and prevent serious morbidity. However, unlike many vaccines, the introduction of rubella vaccines carries some risk. Because of uncertainties in rubella epidemiology, the case for vaccine introduction will not always be cut and dry. The charts and methods presented here aim to help funders, policy makers and other stakeholders make decisions about rubella vaccination while accounting for this uncertainty. These decisions can be made easier by continued research into the epidemiology of rubella, alternate approaches to predicting changes in CRS risk and, critically, careful monitoring of CRS incidence after the introduction of rubella vaccine.