Respiratory Syncytial Virus (RSV) is the most commonly identified respiratory pathogen in young children and is most often associated with bronchiolitis and in severe cases, pneumonia. Although RSV is not a notifiable disease in Australia, the health and economic burden in children is greater than that of influenza. RSV displays distinct seasonality with seasonal epidemic peaks in the winter months associated with cooler temperatures and increased rainfall, characteristics typical of temperate climates. Knowledge of annual variations in the timing of epidemic peaks is important for the planning and implementation of preventive strategies to maximise their effectiveness. Currently the only available preventive tool for RSV is costly immunoprophylaxis with palivizumab, although considerable progress is being made towards an RSV vaccine with ongoing phase 3 clinical trials. In this rapidly changing landscape of disease prevention tools, models of respiratory transmission are crucial to guide prevention and management of disease.
To improve the accuracy of dynamic transmission models, real population data are needed. Through population-based data linkage we have obtained a set of RSV positive and negative detections on a birth cohort in Western Australia from 1996 to 2012. Using these unique population-based linked data, we have developed a simple compartmental dynamic transmission model for RSV positive detections in metropolitan Perth. This model accurately mimics the seasonal peaks observed and provides a platform to investigate some of the parameter assumptions in the model and other RSV characteristics. These include the addition of age structures to the model, understanding how parameters might change for different geographical areas that display different seasonal patterns, addressing the climatic drivers of the seasonality term and how vaccination may be incorporated into the model.