dice wrote:states seeing severe spikes in new cases over the past week or so: TX, FL, AZ, SC, GA, OK, NV, OR, ID, MT, WY, WV, HA
states pretty steadily climbing over a longer period: CA, NC, TN, AL, WA, MO, UT, AK
nationwide case counts have been on upward trend again for over a week. so much for the summer dissipation. can't have a 2nd wave when the first never ended
trump having a rally in AZ tonight, where ICUs are at 83% of capacity. keeping hospitals below capacity is a bare minimum standard as we respond to this crisis
and the EU is strongly considering blocking americans from entering
just embarrassing
Eh... be careful with repeating those assertions without looking at the quality of the source data. A lot of the data suggesting that there is a spike is, effectively, garbage. For example, the biggest spike appears to be from Florida (and specifically in the least open areas near Miami)... but you go to
https://covidtracking.com/data and look at the disclaimer for Florida (with emphasis added):
We report using the raw data provided by FDOH but we include both Florida residents and non-residents. On March 26, the dashboard stopped including non-residents in its visible counts, but they're still tracked behind the scenes.
On March 21, negatives jumped due to a methodology change: tests of people not investigated as PUIs are now included. As of May 15, Florida is reporting both specimens (PCR and antibody) and people tested. We report positives and negatives based on the number of people tested. However, the "people tested" number reported on Florida's dashboard has come into question as of May 31, because this number is not fully de-duplicated. Florida's report states that: "People tested on multiple days will be included for each day a new result was received."
Meaning that it appears that everyone who is retested (which happens when you go to the hospital at least every couple of days, and also happens as soon as you test positive once, even if not hospitalized) and gets a second, third, fourth, fifth positive test comes back as 2, 3, 4, 5 "new cases."
If you look at the disclaimers of all the states, you see hints of similar problems across the board - to include partial de-duplication correction notes (but no explanation of how that happened or whether it is being avoided), etc.
Mind you, these are the exact numbers being reported (without the huge disclaimers) in the media to claim "spikes."
The better data to look at is current (not cumulative) hospitalization numbers (where you can find them) and see if you can find a trend (it isn't in their historical data charts, but can sometimes tease it out from other data sources).