The idea here is that if visits for ILI increase at the same time that flu positivity rates are declining, this might be a sign that there is another pathogen, perhaps #COVID19, causing influenza-like illness.
2/
2/
The challenge is that these data are very sensitive to external factors. If there are changes in care-seeking behavior, these could show up as abnormal patterns in the ILI data. Similarly if there are changes in protocols around who is tested for flu.
3/
3/
So our conclusions here must be interpreted very carefully and and with the understanding that the data we are monitoring could be quite sensitive to increasing public awareness of #COVID19, as we have seen in the last few weeks.
4/
4/
All this said, for the second week in a row, our analysis, which includes data through Saturday, March 21, is showing an uptick in non-flu-related influenza-like illness in many states in the US.
github.com/reichlab/ncov/…
5/
github.com/reichlab/ncov/…
5/
It is really important to NOT interpret these results as definitive evidence that #COVID19 is worse in some places than others. However, we do think that these results could be used to guide investigations using direct data-collection mechanisms, such as testing.
6/
6/
We have yet to tease apart detailed results from specific states, or look at how these results might be impacted by ongoing intervention efforts. These data are already a week old (released today on schedule). We are looking into extending this analysis to more real-time data.
7/
7/
Many thanks to everyone working in hospitals and clinics, and federal, state, and local health departments who help collect these data and, more importantly, continue to tirelessly push towards a future where we can be less worried about #COVID19. 🙏🏽❤️
8/
8/
