May 26, 2020

Washington State Lacks a Rational Coronavirus Testing Approach: Can We Follow Oregon's Lead?

Washington State is in trouble.

Testing of coronavirus has stagnated and our current testing is being done in an uninformative, unscientific, reactive way.


We need not only to increase our COVID-19 testing, but importantly to begin random sampling of our population.  Embarrassingly, such science-based, rational testing is exactly what Oregon is about to start.


As stated by Oregon governor Kate Brown:
“This program is a game changer,” Brown said. “It will give us a more accurate understanding of the true rate of infections in Oregon and to have ongoing precision monitoring of any new outbreaks.”

Oregon is not alone.  Other states, such as Ohio and Indiana, are also doing randomized testing to understand the true state of the infection.  But not Washington.




Random sampling is the scientific way to deal with the COVID-19 crisis

Today, Washington State is dealing with the coronavirus pandemic in ignorance.  We are only testing those that have symptoms and ask for a test.  We do not know what percentage of the population is currently infected.  We don't know what percentage has been infected and thus might be immune.  We have no way to identify infected, but asymptomatic, people that need to be quarantined.

We are driving blind...and our state leadership doesn't seem to understand what needs to be done.

Scientists of all types (yes including atmospheric scientists) are well aware of the importance of random sampling to understand a population or system we are studying.  Pollsters use random sampling  for every election cycle.    But we are not even talking about doing it in Washington State when it comes to a fatal disease with vast economic implications. Why?

Our state leadership is throwing away a tremendous opportunity to defeat the virus.  If appears that random testing of a population is a powerful tool for keeping diseases like COVID-19 in check.  Several papers have shown exactly how this can be done.  One samples a few tenths of a percent of the population each day.  Such testing provide a good idea of the trend in infection.  But even more important, you can catch asymptomatically infectious folks and take them out of circulation.  If you have the resources, you can trace their contacts and test them as well.  Slowly, but surely you defeat the virus by quarantining the infected.


This approach, plus wearing masks in crowded indoor spaces and reasonable social distancing, might well allow us to reopen society much more quickly and extensively....and do so safely.

Not only is Washington State testing in an ineffective, unscientific way, but the number of tests is essentially stagnated, something evident from the statistics on the Washington State Department of Health website (below).  No increase in May.  Not good.  Massive increases in testing is needed, the lack of which is a failure on both the state and Federal levels.



This is a lot of talk about us being involved in a "war" with the coronavirus.  In that spirit, let's consider the military analog.

Consider two approaches for dealing with an enemy that is attacking and marauding the population of a nation.

One approach is to tell your population to hide in their homes and lock the doors, with the hope that the enemy will keep away.   Only when the enemy attacks does the military react by capturing and imprisoning the enemy. 

There is never an attempt to determine the strength of the enemy with surveillance  or to weaken the enemy by proactive raids.  The enemy, unmolested, retains their strength and the war drags out.  With folks stuck inside, the economy flounders.  This is how Washington State is fighting our war with Covid-19.


Another nation has a different approach. They start comprehensive surveillance of the enemy with armed scouting parties and overflights. They know exactly the strength of the enemy and where it is surging in strength. When they find the enemy preparing an incursion, they attack and take the enemy off the field.

Citizens are not told to lock themselves in their homes, but to go about their normal productive lives, but with care and watching for any trouble. They can do so, because they have confidence they will be warned if a threat looms. Slowly but surely the raids on the enemy weaken it and eventually they give up the fight. The nation's economy remains strong.

This is what Oregon, Ohio, and Indiana, among others, is going to do. What many municipalities and states are talking about doing. What Washington State should do. And yes, what the whole country should be doing.

But it is worse than that
. Not only is our state missing the boat on testing, but there have been a number o unproductive steps that are not based on science, such as closing parks, restricting the use of outdoor spaces, and promoting excessive lockdowns of the economy. 


As the economic toll becomes catastrophic, society will have no other option but slowly reopening. Random testing offers an approach to do so safely.

Here we have a state with unequaled scientific capabilities and knowledge, deep experience in data science and medicine, and unparalleled economic clout and we fail to follow an obvious, science-based path to opening our economy and loosening restrictions on our population. 
 
When considering anything dealing with conflict, it is often valuable to consider some of the wisdom of the Chinese military genius, Sun Tzu:

If you know the enemy and know yourself, you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained you will also suffer a defeat. 

Time to know our enemy.



25 comments:

  1. Cliff,

    I am strongly in favor of testing similar to what Oregon is doing! Thanks for blogging about it. Frankly, this strategy seems so obvious. Perhaps funding is an obstacle, but at this point we can't afford not to invest in managing our future co-vid 19 issues.

    ReplyDelete
  2. Well said. Art of war. Art of science

    ReplyDelete
  3. This is a badly informed take.

    One, the paper you linked does make some interesting points about how effective testing of asymptomatic people can be in controlling a coronavirus outbreak. BUT it is describing a situation where you have a 50x increase in the number of tests that are done each day, compared to what the US is doing. Which, yes, if we ramped up to that level of testing capacity, we should -- and would have to -- use some of it on asymptomatic people. But talking about random testing is putting the cart before the horse, because the real benefit comes from having enough that volume of testing in the first place.

    And yes, I agree we should be doing all we can to increase testing capacity. Do you have any insight into why it's stagnated? Any idea of what's bottlenecking it and how to fix it? Because that's where your efforts should go, not to "random testing".

    Secondly, the Oregon program you're touting isn't even really doing random testing, and certainly not in the way that paper is describing:

    "A request for volunteers will go out mid-May. When any of the 100,000 volunteers develop COVID-19 symptoms, they will be tested."

    So a) they're not actually random, they're volunteers, and b) they don't get a test unless they become symptomatic!

    The main issue that we ACTUALLY have is that we don't have sufficient testing to randomly sample the population and follow the suggestion of that paper, nor do we have enough to do effective contact tracing and testing of those a risk because of their contacts.

    THAT is the thing to focus on. Not random testing, which is not the best way to make use of limited testing resources.

    ReplyDelete
    Replies
    1. Chris....thanks for your comment...but ill-informed? Random testing is being tried in a number of locations....Ohio and Indiana to start with. Randomized testing is valuable at even a low level...such like polling...it tells you how much disease is out there. And the impacts of random testing on removing sick individuals increases with the amount of testing--even a little helps. And yes, we need a lot more tests to establish real control. Regarding Oregon, the individuals are randomly selected and asked to volunteer. They are not forced to be tested. Would you force them?...cliff

      Delete
    2. "When any... develop COVID-19 symptoms, they will be tested."

      This part is the problem: it's not random testing when it has 'symptomatic' as a precondition. Not sure why you're focusing on the 'volunteer' part.

      Delete
    3. My point is basically this: We (unfortunately) a limited number of tests. So the question is how to most effectively spend those tests.

      First in line should be symptomatic patients, because they have the most chance of actually testing positive.

      Second in line should be people in contact with those that have tested positive. Generally, families/contacts of positive-testing patients, and healthcare workers.

      Only then do you get to random sampling, which at the scale that we're at isn't likely to find very many positive cases. And sure, it's interesting data which can be useful for modeling the rate of covid in the overall population, but absent a massive increase in testing volume, it's not actually all that useful for minimizing infections. Or at least it's way way less useful than the other two things I listed.

      Which is why I think your focus on random testing is misguided. It's for sure interesting data -- I'd love to have it to and see what it tells us about spread too! But the core issue is just testing volume. Random testing is, rightly, at the back of the line. The problem isn't that it's back there, it's that we're not getting through the line.

      Delete
  4. The number of active cases in Germany is falling a lot faster than the number of active cases in Washington State because Germany does alot of testing and tracing. The German government finds out who is sick requires them to stay home for 2 to 3 weeks and delivers food to them. Germany also has enough PPE and everyone wears a mask in public. Germany recognizes that most people get the virus in an indoor environment has left most of the trails open and encourages people to go outside so long as they social distance. We cannot stay locked down for the year or two it will take to develop a vaccine to many people and business will go bankrupt. If we re-open without extensive testing/tracing, more PPE and social distancing rules there almost certainly be a second wave. Therefore the only way forward is to follow Oregon and Germany's example. I have supported democrats in the past but this level of democratic incompetence astounds me and I might vote Republican in the next election.

    ReplyDelete
    Replies
    1. Or vote for a presidential candidate who understands that you can't effectively fight a pandemic at the state level.

      Delete
  5. FB is banning sharing because someone thinks this is abusive. Apparently it doesn't abide by Washington's leaderships perfect vision of themselves and their sheep's googly-eyed worship.

    ReplyDelete
    Replies
    1. I had the same experience. If anyone can put in a baseless claim, there is no free speech. The Republican and Democrat parties will have a field day blocking each other's candidates postings.

      Delete
  6. As long as the Culture Wars rage, the "war" against the pandemic is tenuous. Assume everyone is Covid-19 positive and take care of yourself is really the best strategy now. Even if masks/social distancing is outside of individual scope of belief, don't purposely disrespect others. We are going to be stuck with this virus for a while and being rude to others just makes it all that much more of a burden.

    Yes it would certainly be a boon for the understanding of how the virus operates to vigorously test, but if Trump has shifted focus to strictly economic/political goals, lack of Federal support for a testing scheme is going to be severe headwinds. Washington State has to be able to pay for the reagents, sundries such as swabs, employees to run the program etc. The money has to come from somewhere. Perhaps Oregon has nimble finances or their governor decided to not be publicly critical of the President?

    Washington State's finances were in jeopardy BEFORE the virus, thanks to Tim Eyman (to be fair, the voters spoke) as well as assuring education is paid for FULLY. So it might just be the funding for a testing program that is the greatest challenge. Tax revenue has definitely cratered and as the economic hardship grinds on, the budget battles will only get more heated.

    Also, there has been no unified front in the Pandemic "war" in the US. Basically its fought by poorly trained, underfunded militias with minimal logistical support and no unified command structure. The intel has been sound, but some have chosen to ignore it. It might be almost fair to say some of the generals are aiding/abetting the enemy. That about sums it up.

    ReplyDelete
    Replies
    1. President Trump is still pushing for the testing but he also has a country to run. Getting the economy back up and running is also important. He has made the test available and is the responsibility of the states to push it. He is leaving it up to states to control which is what the governors whined about. Now they have the control and how they run is on the governors shoulders. But, like Dictator Inslee they blame the president.

      Delete
    2. In March Inslee said because of the rainy day fund Washington would do find financially during the pandemic. Am holding it to that!

      Delete
  7. Great article, why is Facebook blocking its reposting??? (Your message couldn't be sent because it includes content that other people on Facebook have reported as abusive.) How is there anything in this article that's "abusive".

    ReplyDelete
    Replies
    1. Its probably a bot that might parse article content AND comments. If there was a re-posting of one of Cliff's articles in the past that was somehow flagged due to either content OR commentary, then Cliff's blog is now on their naughty database. Or it could be there is a link to malware somehow unintentionally. Again, its a bot doing it and not a real person so its logic is limited to whatever database its referencing, even if the data is wrong. Facebook and Twitter really are trying to somewhat distance themselves from anything controversial, unverifiable information or lightening rods of vitriol. To not do so would open them up for litigation.

      Delete
    2. So far so good! I posted a link 10 minutes ago on FB and it is still there.

      Delete
  8. Is this essay from March 2020? That's when we really needed to do this, though now is good as well.

    ReplyDelete
  9. https://www.nytimes.com/2020/05/15/us/coronavirus-testing-seattle-bill-gates.html

    ReplyDelete
  10. It's not this article. It looks like FB has been blocking posts since at least the Blob is back post last week. I was able to post it but it was pulled a couple days later and I think all Cliff Mass blogs are now banned.

    ReplyDelete
  11. It's not a question of EITHER social distancing, masks, and carefully opening OR testing ... BOTH should be done
    The lack of testing here is inane. What's causing it? The federal government (trying to hide the extent), or the state (others are doing a better job of testing.) IMHO: both

    On the SCAN website (a large testing program in Seattle) https://scanpublichealth.org/updates/2020-05-13 on the it says the Feds have shut them down and required them to stop.

    Compare that to the UK. Today I got an email from a friend in Manchester England, that in part reads:
    "due to one of life’s weird co-incidences I am being tested every week (just for a month) for Corvid 19! A nurse comes to my house and takes a swab of my mouth and nose. I am on the data base for the Office of National Statistics and some months ago I was interviewed for the census. (it is done very 10 years but now instead of each household filling in a printed form, they are choosing every 10th household to interview – I was one of those). He asked me if I would be willing to be on their data base so of course I agreed – and voila, I am being tested. So far I have stayed healthy. It is all part of a national survey to see where the “hot spots” etc are in the country. And I get paid for it. Got a voucher today for £50 to spend in local supermarket. Which I will do when things are a bit safer."

    ReplyDelete
  12. Washington is still doing less than 3000 tests a day - and very unlikely by choice. That is not even enough to check those with symptoms, medical workers, other exposed workers. The number of tests needed to suppress cases has been estimated at 3/4 of a million to 13 million a day across the US , for Washington the lower number is about 15,000 a day. It is a national disgrace that ramp ups in testing and PPE is still not available. The feds likely were getting the first warnings the last few days of December, certainly not more than two weeks later. The best leader ever leader was still denying until mid March there was a real problem.

    ReplyDelete
    Replies
    1. RLL wrote "The best leader ever leader was still denying until mid March there was a real problem. "

      It is not clear which of the following you mean: Tedros Adhanom Ghebreyesus, Mayor Bill de Blasio, or Andrew M. Cuomo.

      Or do I miss your meaning and your point?

      Delete
  13. "Diagnostic tests for coronavirus are unreliable, inherently worthless, wrong-headed & deceptive. Why should you take at face value ['experts'] claim they’ve found a single virus causing a major outbreak of disease? Professional liars should not be accorded such a level of respect.

    ReplyDelete
  14. Cliff,
    There is abundant evidence that the economic toll is already catastrophic and society cannot wait another day to re-open as much as possible immediately.
    Whatever value random testing offers there can be no more delay in re-starting the economy.

    ReplyDelete
  15. Cliff,
    You may be interested in this OSU "Trace-Covid-19" study that uses door to door approach.

    https://trace.oregonstate.edu/

    https://www.oregonlive.com/coronavirus/2020/05/latest-from-osu-coronavirus-study-672-people-tested-0-infections-found.html

    https://ktvz.com/news/oregon-northwest/2020/05/07/first-corvallis-covid-19-door-to-door-test-results-suggest-2-per-1000-infected/

    ReplyDelete

The Fascinating History of Weather Forecasting and the Perfect Weekend Ahead: New Podcast Cover Both

My new podcast is out (see below).    My second segment tells you about the fascinating history of weather forecasting, including a surprisi...