Integrating information from historical data into mechanistic models for influenza forecasting

Seasonal influenza causes significant annual morbidity and mortality worldwide. In France, it is estimated that, on average, 2 million individuals consult their GP for influenza-like-illness (ILI) every year. Traditionally, mathematical models used for epidemic forecasting can either include parameters capturing the infection process (mechanistic or compartmental models) or rely on time series analysis approaches that do not make mechanistic assumptions (statistical or phenomenological models). While the latter make extensive use of past epidemic data, mechanistic models are usually independently initialized in each season. As a result, forecasts from such models can contain trajectories that are vastly different from past epidemics. We developed a mechanistic model that takes into account epidemic data from training seasons when producing forecasts. The parameters of the model are estimated via a first particle filter running on the observed data. A second particle filter is then used to produce forecasts compatible with epidemic trajectories from the training set. The model was calibrated and tested on 35 years’ worth of surveillance data from the French Sentinelles Network, representing the weekly number of patients consulting for ILI over the period 1985–2019. Our results show that the new method improves upon standard mechanistic approaches. In particular, when retrospectively tested on the available data, our model provides increased accuracy for short-term forecasts (from one to four weeks into the future) and peak timing and intensity. Our new approach for epidemic forecasting allows the integration of key strengths of the statistical approach into the mechanistic modelling framework and represents an attempt to provide accurate forecasts by making full use of the rich surveillance dataset collected in France since 1985.Author summary: Seasonal influenza causes significant morbidity and mortality worldwide. In France, on average, 2 million individuals consult their GP for influenza-like-illness (ILI) every year. Forecasting the future trajectory of an epidemic in real-time can inform public health responses. Traditionally, two types of mathematical models are used to forecast infectious diseases outbreaks: (1) mechanistic models that explicitly capture transmission mechanisms and (2) statistical models that rely on the similarity between epidemic dynamics over the years. In contrast to statistical models, mechanistic models usually do not use information on past seasons to inform forecasts. This may result in poor performance when forecasted trajectories are very different from past epidemics. Here, we propose a framework that combines these two approaches by allowing mechanistic models to learn from trends observed in past epidemics. We evaluate this approach in the context of seasonal influenza in France. Our results show that the new method improves upon standard mechanistic approaches. In particular, our model provides increased accuracy for predicting the epidemic trajectory one to four weeks into the future, as well as the timing and size of the peak. Our new approach for epidemic forecasting allows combining statistical and mechanistic models to improve forecasts performance.