An adaptive method of emergency department syndromic surveillance to nowcast the frequency of presentations that will have a severe 28-day outcome following influenza or COVID-19 infection: a retrospective analytical record linkage study.
A retrospective analytical study demonstrates improved prediction of severe outcomes in respiratory virus cases, suggesting better epidemic response strategies.
Key Points
The aim is to better predict the frequency of severe outcomes in influenza and COVID-19 cases using emergency department data.
Developed a retrospective record linkage study using emergency department data and infection outcomes from Australia.
Applied daily time series modelling with generalised additive models to estimate severe outcome frequencies.
Nowcasted severe outcomes with a focus on a 28-day period after infection.”],
results
key points
cards
events
open-descriptions
highlights
length-check
highlights
objective
length-check
проводить
highlight
anything-here
highlight
anything-here
highlight
The mean daily nowcast error was 2.7 (34.2%) when predicting severe outcomes at day -4.
Using outcomes known at day -1 resulted in a higher mean error of 3.5 (43.8%).
The adaptive method could enhance real-time epidemic assessments for public health.