Dynamic Transmission Models

Dynamic Transmission Models

Introduction

Dynamic models are generally used for communicable diseases. Unlike static models which assume that infection risk is constant, in dynamic models the infection risk parameter is variable. These models can be deterministic or stochastic, individual or cohort based.

See also: https://www.ispor.org/docs/default-source/resources/outcomes-research-guidelines-index/dynamic_transmission_modeling-5.pdf?sfvrsn=86c71849_0.

When might I use this?

The key feature of these models is that risk of infection varies with the number of infectious agents present over the time modelled. This means that indirect effects, such as herd immunity due to mass vaccination, are accounted for. Dynamic transmission models are useful for examining interventions and their impact on pathogen ecology, as well as exploring an intervention’s impact on transmission. For example, these models can be used to assess the effect of: * decreasing the susceptible proportion of the population (e.g. through vaccination), * decreasing contact between individuals (e.g. via isolation measures), * reducing the duration of the infectious period (e.g. by delivering drug interventions), * reducing transmissibility of the pathogen (e.g. by delivering drug interventions).

In particular, these models are commonly used to measure the impact of a vaccine in a population, and to investigate the use of antivirals or isolation measures in disease outbreaks. For example, dynamic transmission models have been used to assess the potential malaria risk to the UK (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2845590/).

Building a dynamic transmission model often requires more than one package. However packages for learning to build dynamic transmission models, which may be useful to the beginner are available. An example is described below.

Transmisison Model Packages:

Package: DSAIDE

DSAIDE - Dynamical Systems Approach to Infectious Disease Epidemiology (Ecology/Evolution)

Maintained by

Andreas Handel (ahandel@uga.edu)

URLs

https://CRAN.R-project.org/package=DSAIDE https://ahgroup.github.io/DSAIDE https://github.com/ahgroup/DSAIDE https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005642

What does this package do?

The package contains a set of models focused on infectious disease epidemiology. Each model comes with documentation and instructions which highlight important concepts in infectious disease epidemiology and explain how models can be used to understand these concepts. All models can be explored through a graphical user interface, with no reading or writing of code required. However, the package is designed to help the user to move to direct interaction with pre-written models and modification of the underlying models.

How do I input my data to it/what inputs does it take?

Inputs will be information on the population and the infection that the user wants to model, such as the number of people who will be susceptible initially and the rate of recovery from the infection.

What outputs do I get?

Outputs will be information on the progression of the infection in the population, such as the total number of individuals infected over the simulation, which can be used for calculation of the economic cost and benefits associated with an intervention.

Sample code section

A range of sample code sections are available at https://ahgroup.github.io/DSAIDE.

Other packages

There are a wide variety of other packages availables for learning to build dynamic transmission models, as well as a variety of packages that are commonly used by those more familiar with dynamic transmission modelling. Examples that demonstrate the use of some of these packages can be found on the EpiRecipies website (http://epirecip.es/epicookbook/).

Other helpful resources

The EpiRecipies project (http://epirecip.es/epicookbook/) aims to collate mathematical models of infectious disease transmission, with implementations in R, Python, and Julia. The website has a diverse selection of examples including simple deterministic models using ordinary differential equations, simple stochastic models, models with time-varying parameters, spatial models and network models. There are also examples of applications to specific disease systems.