Abstract:
Methods for reviewing epidemiological signals are necessary to building and maintaining data-driven public health capabilities. We have developed a novel approach for assessing the plausibility of infectious disease forecasts and surveillance data. The PLANES (PLausibility ANalysis of Epidemiological Signals) methodology is designed to be multi-dimensional and flexible, yielding an overall score based on individual component assessments that can be applied at various temporal and spatial granularities. Here we describe PLANES, provide a demonstration analysis, and discuss how to use the open-source rplanes R package. PLANES aims to enable modelers and public health end-users to evaluate forecast plausibility and surveillance data integrity, ultimately improving early warning systems and informing evidence-based decision-making. The rplanes package is available on GitHub here.
Read preprint here.
Authors
VP Nagraj1, Amy E. Benefield1, Desiree Williams1, Stephen D. Turner1
1Signature Science, LLC, 1670 Discovery Drive, Charlottesville, VA