The riskintroanalysis R package provides functions to analyse the risk of introduction of animal diseases within a geographic area. It is intended to be useful as a stand-alone package, but also to integrate directly into the riskintro Rshiny app. The motivation behind these projects is to easily conduct geospatial risk analysis using existing data from WAHIS (World Animal Health Information System).
You can install the development version of riskintroanalysis like so:
# Install pak if you don't already have it
install.packages("pak")
pak::pak("git::https://gitlab.cirad.fr/astre/riskintro-app/riskintroanalysis.git")
# Accept updates to
# + riskintrodata
# + riskintroanalysis
# + ideally any other dependenciesThe package provides functions for certain risk of introduction analysis, but requires external data to be provided. The functions are inteded to be flexible, but each analytical method requires the corresponding data.
The RShiny app riskintro provides a graphical interface for the following analyses.
The central datasets to each method are:
- The emission risk score table, which provides a risk of emission diseases (that is to say the risk of a disease spreading from that country). This score is based on summary of risk factors, many of which are available from WAHIS (see more here). This data exists for many countries around the world, and is augmented with geospatial data corresponding to each country’s administrative boundaries.
- The area of interest, Geospatial data defining the borders of a region. This is the area for which we want to know the risk of introduction of a disease. The most common data is that of a country and its administrative boundaries, including internal boundaries. For example, French departments, or German states.
These two datasets are used along with other data to calculate risk scores for the following analyses:
- Border lengths: using the length of shared borders the risk of introduction is weighed comared to each neighbouring country and their risk score.
- Border entry points: using an additional dataset that defines broder entry points, risk of introduction is weighted by legality of entry point.
- Animal mobility: using an additional dataset defining legal animal commerce flows, risk of introduction is weighed based on the number of animals entering the area.
The forth analysis method does not use emission risk data, just data from a road accesibility raster and the epidemiological units:
- Road access risk: using raster data of the world, applied to the area of interest, risk of introdcution is infered through road acces. An area more accessible by road is considered higher risk.
These methods of analysis are intended to be used together or individually to montor is risk of introduction. This is done by ensuring risks are using the same scale (for example between 0 and 100) and they can easily added to a table of risk.
Additionally, some tools are provided to add other risk (such as the ones not show above). Currently supported are pre-calculated risks or risks from raster files. Which can be rescaled and added to the risk table as well. This allows of the risk analyst the flexibility to add other risks outside of the default risks.
For detailed guides on using riskintroanalysis, see the package vignettes:
- Emission Scores - Understanding and calculating emission risk factors
- Introduction Risk - Combining multiple risk sources into comprehensive assessments
- Border Risk - Risk from shared land borders
- Entry Points - Risk from border crossing points
- Animal Mobility - Risk from animal trade flows
- Road Access - Risk based on road accessibility
- Additional Risks - Adding custom risk factors from raster data
- Risk Scaling - Understanding and applying risk score transformations