When it comes to predicting the spread of disease, modern infectious disease epidemiologists must keep track of the interactions between infected and uninfected individuals, as well as the impact of environmental factors on the disease spread.
[Photo: Dr. Andreas Handel]
This is known as a dynamic systems approach, which recognizes the complex interaction between people, pathogens and the environment.
Yet in the classroom, infectious disease epidemiology is still often taught from a mainly classical epidemiological perspective, which assumes each person in a population has an independent chance of getting a disease.
“This approach is not sufficient to fully understand infectious diseases,” said Dr. Andreas Handel, associate professor of epidemiology and biostatistics at University of Georgia College of Public Health.
Though there are textbooks and other resources available that teach the modern, model-based approach, Handel says the difficulty of the material, which requires some background in advanced mathematics and writing computer code, is a roadblock for many graduate public health students.
“I often experience that the difficulties students have with writing or modifying computer code prevents them from understanding the underlying scientific concepts,” he said. “I wanted to find a way that uses computer models but allows students to focus on the scientific concepts.”
In response to this dilemma, Dr. Handel, with contributions from students, built a new software tool designed for public health students and others who may want to learn more about the dynamical systems approach to infectious diseases, but aren’t ready to start writing computer code.
The Dynamical Systems Approaches to Infectious Disease Epidemiology (DSAIDE) software is an R package that features twelve applications covering topics like patterns of infectious disease incidence, models of direct transmission, and critical community size. Each app includes an interactive model, detailed documentation, and learning tasks.
A unique feature of the package is its modularity, which allows students to easily move between the graphical interface of the model and the underlying code powering it. Students can interact with the code as far as they feel comfortable, all the way to full modification of the provided simulations. Handel hopes that this design will allow students at all levels of interest and skill to engage with the package.
“We’d want to have more people be able to apply this type of approach because in some ways, I’d say it’s just the future,” said Dr. Handel.
Computational modeling is increasingly used to drive policy and decision making related to disease management, he says, so it’s becoming equally important that the next generation of public health professionals is familiar with modeling concepts.
The software package description and installation guide are detailed in his paper, “Learning infectious disease epidemiology in a modern framework,” was published in PLOS Computational Biology. A PDF version of the paper is available online.