States added 663 deaths today to the US cumulative total (vs 918 last Fri). The 7-day avg for report-date deaths resumes its decline, falling to 710. Reminder: CTP reports include deaths that occurred anywhere from a few days to several weeks ago.
9/12: COVID-19 Positive Test Results (PTRs) vs Reported Deaths. Studies now find that PCR tests are calibrated at levels too sensitive to accurately assess spread, with up to 90% of PTRs from those with viral loads so low that they aren't - or, perhaps, never where - contagious.
Latest CDC Deaths by Date of Death (orange) vs CTP Report-Date Deaths (blue), now with WE Sep 12 added. The recent CDC Deaths by DoD "sunbelt" peak still holds at WE July 25 vs CTP's peak at WE Aug 15. The decline in weekly CTP reported deaths continues apace.
Same CDC vs CTP deaths data presented cumulatively below. Note that weekly CTP reported deaths must decline rapidly in the coming weeks if the 2 tallies are to converge at some point at roughly the same number.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Todd Lowdon

Todd Lowdon Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @tlowdon

17 Sep
The US has conducted more CoV2 tests per capita than any other large country. We test people without symptoms, students entering school, workers returning to their jobs, etc. The more you test, the more you find - and we find more cases than other countries. 1/
It's also now widely reported that the most common test is improperly calibrated. It's much too sensitive & finds trace amounts of inactive virus in people who stopped being infectious weeks ago. Consequently, we over-count positives by as much as 10X. 2/
Then, we have false positives caused by contamination or human error. FP rates are ~2% of tests conducted. This isn't a big problem when we test small numbers or only those with symptoms, but at 600K/day, 10K-15K or more are probably false positives. 3/
Read 7 tweets
29 Aug
Stunning new research finds C19 PCR tests are much too sensitive to be useful in assessing spread. "In 3 sets of testing data...compiled by officials in MA, NY & NV, up to 90% of people testing positive carried barely any virus." The implications are huge.
nytimes.com/2020/08/29/hea…
The CDC noted earlier that symptomatic cases are infectious for only a few days, but PCR tests can detect inactive RNA for up to 12 wks. It's now understood that the highly sensitive tests are also returning positives for very small amounts of active virus.
PCR testing amplifies genetic matter to determine whether it is present or not, usually returning only a "positive" or "negative" result without any indication of viral load or level of infectiousness.
Read 6 tweets
12 Jul
States reported 757 US deaths today, raising the 7-day avg for "report-date" deaths to 678. State reports today & earlier this week include unusually large numbers of backdated deaths which actually occurred in earlier weeks & months.
7/11 Update: US C19 CASES vs DEATHS. While actual deaths in southern latitude states may be up (although it's hard to determine the actual increase w/ certainty), this week's upturn in CTP's reported deaths curve is unduly influenced by reports including deaths from weeks before.
Deaths occur mean ~3 wks after infection. Infection to symptom onset = mean 5-6 days. Symptom onset to death = ~2 wks (CDC says 13-15 days, Verily/Lancet said 16-18, others in that range). So if most get tested ~1 wk post infection, a death outcome likely occurs ~2 wks later.
Read 4 tweets
9 Jul
Update 7.9.20: CDC's latest Deaths by Date of Death reports (orange bars) vs CTP's tally (blue bars). These are provisional & will change - recent weeks more, older weeks less. But, note that actual weekly deaths peaked earlier & thousands higher than states reported in April.
The "Hypothetical" line (yellow) approximates what we might see when the counting is completed *if* the CDC's ultimate DoD totals finish 5.5K ahead of CTP's for the period ending July 4. CDC's count apparently includes NYC's ~5.5K "probables" whereas CTP's does not.
Same updated CDC data presented cumulatively below. Note that by Apr 25, the CDC's cumulative provisional count is ~13K (!) greater than what states had reported at the time. COV2 was subsequently coded for many thousands more deaths than we could track back in March & April.
Read 5 tweets
5 Jul
You could be forgiven for being confused about the virus if you didn't do a bit of your own analysis or critical thinking. Infectious disease & public health experts - even the C19 data - can be incoherent. Here's one example dealing with the question of masks. ImageImage
In Feb & Mar, experts stated - even demonstrated - that masks weren't required. You wouldn't need one, even if supplies were unlimited. But by June, you we're scolded for questioning their effectiveness (#WearADamnMask!) and Gov Newsom was praised for his statewide mask mandate. ImageImageImage
The doctor's bio says she teaches at Stanford Med School & serves on committees at the American Society of Tropical Medicine & Hygiene and Infectious Diseases Society of America. She's also an NPR commentator. She'd rightly be regarded as knowledgeable on the issue of masks.
Read 4 tweets
19 Jun
The CDC updates its C19 Deaths by Date of Death reports daily & is trending ~5.5K ahead of @COVID19Tracking's total. With the latest updates, we can extrapolate to an est. avg deaths figure for last week of ~470/day vs. CTP's 754/day. This might change w/ future CDC updates...
...but probably not by much. And, given that CTP will probably finish this week at ~4K total, the current week's true daily avg will be lower still. The graph above is derived from this one, which compares CDC Provisional & Predicted Deaths by DoD with CTP's Report-Date Deaths.
The point: *IF* the CDC's final tally is ~5K greater than @COVID19Tracking's, we know w/ near certainty that recent CTP death reports are too high - probably by more than 250/day for last week. We'll see what this week's story is when the CDC puts out new figures next week.
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!