Return to

“To vaccinate or not to vaccinate………..”


Paul Sutmoller

European Livestock Alliance,




We can be assured that this will be the big question when the next Foot and Mouth Disease (FMD) outbreak occurs in Europe.


The main arguments against vaccination during the last 2001 FMD epidemic were:

“If you vaccinate you will have to live with the disease”

“You cannot distinguish between vaccinated and infected animals”

“Vaccination causes carriers”

“Carriers are a great risk!”

“Consumers will not buy meat or other products of vaccinated animals”


These arguments have largely evaporated and, moreover, the general public is increasingly against the killing and destruction of large numbers of healthy animals for disease control purposes. FMD vaccination policies should therefore be seriously considered as an alternative or an adjunct to stamping-out methods for FMD control and eradication.


Risk analysis and cost benefit analysis will likely be an important approach on which “decisions to vaccinate or not” will be based. Such a cost benefit analysis was published recently; ( In that model control options were simulated based on the culling of animals on infected farms and possible epidemiologically linked holdings such as direct contacts or contiguous farms. Various vaccination options were also explored based on the Defra FMD contingency plan to vaccinate cattle to a radius of 10 km from an infected farm.


There are some outcomes of the simulation that most FMD experts would agree with.  For instance, the report noted the advantages of a stand still of livestock movements, an immediate national movement ban and swift action when FMD would be first detected. However, with regard to vaccination policies the report suggested that vaccination would reduce the size and extent of medium and large outbreaks only between approximately 15% and 50% and also reduce the number of animals culled for larger outbreaks by approximately the same percentages. Thus, according to these simulation results, the overall epidemiological effects of vaccination would be mediocre.


These simulation results are in disagreement with field evidence showing that FMD vaccines are able to extinguish or abort FMD outbreaks. Well-documented examples of the successful control and eradication of FMD through vaccination are: the extensive Mexican outbreak in the early fifties, the limited outbreak in the Netherlands in 2001 and the very extensive epidemic in Uruguay during that same year.


Other results of the simulations suggest that vaccine of low or high potency vaccines would make little difference on the course of an epidemic. These results again are in disagreement with field observations. There is no doubt that the success of control of a FMD outbreak is directly related to the use of high potency of the vaccines.


If simulation models do not behave according to field observation then either field observations are inaccurate, the model has flaws or the input is incorrect. 


Virus transmission on farms and between farms is an extremely complex biological/ epidemiological process, which is even more so when animals on those farms are vaccinated. The outcomes of a computer model fully depend on the algorithms used to simulate such a complex situation. If there is too much simplification, the model may not at all represent the dynamics of the disease under natural conditions.


The transparency of the model is therefore very important if one is to understand the simulation results. Of particular importance is a clear definition of the terms “transmissibility” and “susceptibility” of vaccinated farms exposed to FMD virus in order to understand and model the transmission mechanism between vaccinated farms. Both factors determine the rate of spread or dissemination of the disease and therefore are crucial for the simulation outcomes. The report is not transparent in this regard. It is not clear whether the “probability of virus transmission” (a function of “transmissibility” and “susceptibility”) between vaccinated farms actually is reduced by a factor of 10 or 100, with the use of a vaccine with a protection potency of 90% of the population. However, most epidemiologists would agree that any epidemic would quickly run out of steam if the probability of FMD transmission between farms would be reduced by a factor of 10 and that a reduction factor of 100 would halt an epidemic in its tracts.


It appears that the results obtained by the model are, at least partly, a reflection of the inefficiency of the 10 km ring vaccination according to Defra’s contingency plan (lack of resources!!) used as input for the simulation. With a good organization the 10 km rings can likely be vaccinated within 7-10 days and a solid immunity can be obtained in 90% of the population 7 days thereafter, particularly if high potency vaccines are used. The poor vaccination results as indicated by the simulation outcomes might also be caused by adhering to a 10 km vaccination radius from an infected farm, instead of considering natural zones based on geographical and demographical conditions. For modelling purposes 10 km rings may be attractive, but it is not the way FMD behaves in real live. Also, the vaccination of sheep or other susceptible species might have to be considered.


During the outbreak in the Netherlands in 2001 over 100,000 cattle on thousands of farms were vaccinated in a triangle with sides of about 25 km within a period of 3 days. Only a few FMD cases occurred on vaccinated farms with the last case occurring 5 days after the vaccination round. When these post-vaccination cases occurred the surrounding farms in the triangle were already solidly protected and no secondary cases occurred.


In conclusion there are at least two important unexplained discrepancies between the simulation results of the model and field observations. However, the conclusions of the model are supported by an impressive number of tables and graphs. Decision makers therefore might accept these conclusions and recommendations without sufficiently considering that the outcome of simulation exercises are completely dependent on the model logic and structure, the assumptions made and the inputs fed into the model. If during the next FMD outbreak, the conclusions of the simulation model will provide too much guidance for the decision-making process the eradication effort may turn into another catastrophe. 







































Archive July and August 2005

Archive Jan 2005 - July 2005

Archive Oct 2004 - Dec 2004

Archive August 2004 -October

OTHER WARMWELL ARCHIVES(opens in new window)