Epidemiologic models for the eradication of poliomyelitis

Martin Eichner , 1992
PhD thesis, Fakultät für Biologie der Universität Tübingen, Germany

Institute for Medical Biometry, University of Tübingen, Germany

Abstract

Objectives

In 1988 the World Health Organisation pronounced the plan of the world-wide eradication of poliomyelitis until the year 2000. By means of epidemiological models it is investigated how (and whether at all) this aim can be achieved.

Methods

Different vaccination strategies are tested by mathematical and computer-aided modelling, taking into consideration the peculiarities of the two vaccines in use: Both are highly effective in preventing paralysis due to a polio virus infection. The oral vaccine (OPV) causes an infection in the vaccinee's digestive tract and can spread to other persons, but it often fails in tropical countries. The inactivated vaccine (IPV) works well both in temperate and tropical countries but does not readily prevent polio virus infections.

More than 300,000 computer simulations were performed. The results refer to populations with poor hygienic standards and low life expectancy.

Results

In most cases local extinction can be achieved within ten years, if at least 55 % of all new-borns are (effectively) vaccinated with OPV or if at least 80 % are vaccinated with IPV.

Local extinction can be reached with quite low vaccination coverage, which is not sufficient to maintain an infection-free state. Reintroductions of wild polio virus infections may result in disastrous epidemics which affect most susceptibles. Because only few infections manifest themselves as paralysis, an infection imported to a village or town can spread unrecognised for some months. Infections already occur in many other villages and towns before the first paralysis takes place.

New endemic situations caused by successfully reintroducted wild virus infections never have been observed. In all simulations infections spontaneously disappeared at the end of the epidemic.

In populations with bad hygienic standards and low life expectancy, vaccine virus infection can (at least theoretically) be endemic. If vaccination with OPV is reduced gradually or finished abruptly after wild virus extinction has been reached, vaccine virus infections soon will disappear, too. Even introduction of vaccine virus infection into partially susceptible, non-vaccinated populations only rarely results in endemic situations.

So-called national vaccination days contributed to the elimination of poliomyelitis in South-America: as many children under five years as possible are vaccinated with OPV at vaccination days, which take place once or twice a year. Following the predictions of a model with unstructured population, an effective vaccination coverage of only 25 % once a year is enough to ensure extinction of wild virus infection within 10 years.

In more realistic models the total population was subdivided into villages and towns. A great proportion of the children from villages is regarded to be reachable at vaccination days, whereas in towns only a smaller proportion can be reached. In scattered compounds nobody at all is vaccinated. Extinction within ten years is very likely, if at least half of the population, regarded to be reachable, is (effectively) vaccinated once a year.

If annual vaccination days do not take place in the country as a whole, but are designed individually for each city or village, the results remain the same.

Conclusions

Global eradication of poliomyelitis is possible. With the strategy of vaccinating new-borns it might be very difficult to reach the necessary proportion of infants, especially in inaccessible compounds and in the slums of greater tropical cities. The strategy of (national or local) vaccination days can lead to excellent vaccination coverage. Even models with more realistic assumtions about the accessibility of the differen population subgroups predict local extinction (and finally global eradication) at reasonable annual vaccination coverage.