Over the past few weeks, the number of tests given has increased by 60%.
As best I can tell, Oregon previously calculated the percentage positive value as:.
- Positive test results / number of people tested.
Because many people are tested more than once and receive multiple positive test results, this had the effect of dramatically increasing the state’s positive test percentage by about a factor of 2!
Now, OHA will calculate the percentage using:
- Positive test results / number of tests given.
Of course, this produces a lower result. Oregon will now calculate percentage positive in the same way that the CDC and other states have done this.
Over the past month or two, Oregon’s “percent positive” rate climbed to nearly 16%, causing internal panic – and leading them to discover the error. This error has propagated across the Internet into official state rankings such as this one at Johns Hopkins. After doing the proper calculation, Oregon moves way down the list.
The numbers have been off for 7 months. OHA says the numbers only recently began to deviate off in to space. They were using significantly incorrect calculations to set pandemic policies – hence, another reason that the effects of public health pandemic policies are random and largely ineffective.
Mind blown – it took 7 months to identify a standardized way of calculating percent positive tests? That is unbelievable incompetence. If this were a bank or finance department in a corporation, people would be fired.
But there’s more from today’s press conference.
The following quote is not true in claiming all these deaths were preventable in Oregon. 8 of the 14 people who died of Covid-19 in my County were in-facility hospice patients at the time of their death. While Covid-19 may have accelerated their death, it is a lie to say their deaths were preventable, as OHA says in the 2nd tweet, below.
The following statement is not true either:
Death “rates” are decreasing. He means to say more people, numerically, are dying but the rate of death from Covid-19 continues to drop. In the spring, 4 to 5 people died for every 100 new cases. This has now fallen to just under 1 death for every 100 new cases (still too high though).
This chart illustrates – multiply the Y-axis by 100 to get the number of deaths per 100 new cases (this is only a crude estimate – the trend is the important aspect of this chart, not the specific number).
The drop in the death rate is likely due to a combination of factors per the “experts”. In much of the spring, you had to be quite ill to get a test, so we did not count anyone with no or few symptoms in the total. Second, the demographics have shifted from the elderly to the young and the young are much less likely to die. I saw a number yesterday that the death rate for those under age 20 is 3 in 100,000.
While the state’s testing numbers are increasing, they may not mean testing more distinct people. According to the Oregonian, it seems likely that up to 29,000 long term care facility workers will be tested once or twice per week and this could add 29,000 up to twice that number to the total test count, every week – but without testing additional people. Using a diagnostic test for mass screening will also contribute to the false positive if no steps are taken to control for the false positives.
Further, over the next few weeks the State will offer free testing in 12 counties – no referral, symptoms or known contact with a Covid-19 patient is required. Again, if steps are not taken to control for false positive, this will be a mess (I have asked OHA and CDC how they are controlling for false positives; OHA never responded and CDC sent an answer to a question I never asked.)
At this point, inadequate testing and inability to do contact tracing once the disease is widespread means there isn’t much that can be done except Lock down 2.0, which the state is now doing. OHA has in the past blamed everyone else for their errors and inability to do their jobs.
And those news stories point to another problem with the new “percent positive” calculation. The state does not actually know how many tests are being run each week.
Collecting data is fundamentally the job of public health epidemiologists – from the very beginning, they were unable to collect, process and release data to the public. 7 months later they find a fundamental error in their calculations. It’s incompetence from the get-go. And this is another reason the effect of public health pandemic control policies are random: its been garbage in, garbage out from the beginning.
The Oregon Health Authority Director is quoted in the Bend Bulletin (11/21/2020) as saying:
We all want this pandemic to be over. We all have the power to end this nightmare.(Paid subscription required) https://bendbulletin-or.newsmemory.com/?token=000BB1FAC8CA6-9CD2-4BE3-8428-23C7F1894FB5
He believes we end the pandemic by wearing face masks and living in lock up 2.0 forever. That is not how any pandemic in history has ended. Pandemics end through population immunity and/or vaccines (which are a relatively new way), the virus mutates to a less virulent form, or people get on with their lives.
High compliance (84% 2 months ago, 89%-93% now) face masks have been mandated since July 1 – while cases have exploded in number. There is no explanation as to why face masks do not work (I’ve asked about that too but no one is willing to discuss that.) – instead they have made face masks the cornerstone of the response. Am waiting for someone to suggest that only if we wore 2 or 3 face masks …
Update – according to COVIDcast, from Carnegie-Mellon University, Oregon now has a 90% mask wearing compliance rate (11/21/2020). According to CMU research, the overall average mask wearing compliance in the United States is currently 89%. My county has 94% compliance rate. When you hear a public health official claim, without evidence, that if 95% of the country wore a face mask, this pandemic would be over with – they are telling you a straight up lie. What they are saying is that just 6% of the population, and those not wearing face masks, are responsible for most cases in the country.
Incidentally – for another post – a TV station used Freedom of Information Act requests to obtain emails about the state’s use of disease models. They learned that “officials” were aware for months that the disease model they used was inaccurate and way off target – for projections just weeks into the future. They quietly switched to a simpler model weeks ago.
Public health is a mess.