Epidemiologists and others attempting to make predictions about the COVID-19 pandemic have a challenging task. There is a great deal of uncertainty when attempting to forecast the short-term and/or long-term dynamics and impact of the disease.
In a Perspective article published in the New England Journal of Medicine, scientists discuss the different types of models used to forecast future pandemic scenarios, the limitations of these methods, and the types of questions these models can and cannot be used to answer.
In their article, scientists discuss five key questions to ask when evaluating models to guide COVID-19 policies:
All of these questions must be asked to evaluate the forecasting ability or mechanistic nature of the models. As the authors explain, models are only as useful as the data used in the model, assumptions about the model, and our understanding of the limitations of the data and the model.