Modelling the potential outcome of maternal immunisation campaigns

It is the objective of this work to propose and stimulate the discussion of a preliminary set of population level epidemiological models for the study of any potential outcomes and consequences of a public health intervention by means of mass maternal immunisation. Candidate vaccines of this nature are currently being considered for a number of highly prevalent viral infections (such as human Respiratory Syncytial Virus) that cause significant morbidity and mortality among neonate and young infant age classes. Immunisation is typically administered to pregnant women at around 30 weeks of gestation with the primary intention being a significant rise in disease specific neutralising antibody in the newborn infant, subsequently providing additional immune protection in the first 3-6 months of life, hence reducing the incidence of disease below a target age level and raising the average age at primary infection. The model variations discussed in this work can be derived around a general MSIR type PDE system, where the rate at which individuals lose maternally acquired protection and become fully susceptible is taken to be directly dependent on the population cord antibody distribution that wanes with increasing age. The models have been used for a preliminary, qualitative study of the static (age profile), dynamic (time series) and seasonal (temporally forced) behaviour of the system in response to various applications of a potential campaign. In addition a number of representative age dependent transmission functions and re-infection mechanisms have also been considered in an attempt to identify cases where such an intervention may give rise to any significantly beneficial or undesirable outcomes. Preliminary analysis of the proposed systems has indicated that in most cases maternal immunisation is unlikely to ultimately reduce (or increase) the overall prevalence of infection within a host population, given that any additional immunity induced is inevitably lost, and typically within a relatively short period of time with respect to average life expectancy. Therefore it is also unlikely to provide significant protection to unvaccinated individuals through herd immunity as is the case with many active childhood vaccine programmes such as that for MMR. The analysis has however suggested some exceptional circumstances where this may not be the case, for example if there exists a rapid change in age related transmission characteristics encompassed by the average increase in duration of protection, perhaps due to some aspect of physiological development. Given that epidemic behaviour is largely driven by susceptibility it has also been shown that in some cases rapid application of a maternal immunisation campaign can cause sufficient perturbation to the inflow of susceptible individuals to induce large epidemic dynamics throughout the population. In the case of seasonally forced systems it can also be shown that the timing of administration throughout the (annual) inter-epidemic cycle can be equally critical in mitigating potentially disruptive dynamics and optimising the overall efficacy of the intervention