Wednesday, August 10, 2022

BTRTN: On Covid Data and Magical Thinking

Tom explodes the Covid myths that too many people, many of whom should know better, cling to in rationalizing their risky behaviors.

On March 14, 2020, I posted a piece on early Covid case trends. At the time there were a mere 2,144 cases in America.  But I was alarmed at what I saw and sought to convince readers, including friends and family, of the nature of the threat.  While the piece seems quaint now, and hardly begins to capture the ultimate nature of the threat as we know it, it was educational in stating that Covid was not going to be a short-term blip. (I have included a link to that post at the end of this article.)  At that time, many thought it would be over in a few weeks or months, and were hardly prepared for the behavior changes that would be required.  

We are nearly two-and-a-half years from that point now, well along the Covid road – but I find myself having similar fears today, 92 million reported cases later. 

In the days that followed, still in March, 2020, I began collecting and distributing an email “report” to family and friends.  The one-pager, basically a spreadsheet with some headline interpretations, summarized Covid new case trends by state and about a dozen selected countries.  One of the reasons I did it was, in those days, there were a number of different data sources, but none of them were terribly user-friendly or comprehensive geographically.  Apart from trying to offer that modest public service, my goals were also self-centered; my wife and I used the report to help guide our own behavior, including, ultimately, where we might travel. 

Wendy has long been interested in health care, and in time became a contact tracer for our state.  Between us, we became quite conversant, for lay people, in Covid, and with every twist and turn, that seemed to serve us well.  Between her contact tracing training and experience, her voracious appetite for Covid articles, and my data, we navigated Covid as best as we could and gave advice to others who sought it.  We kept up with an increasingly complex environment, with the politicization of Covid, new variants and offshoots; different types of testing; vaccines and boosters; and finally treatments. 

Last week I decided to stop distributing the data.  I had long thought the reported data was far from perfect, underreporting cases early on when testing was hard to come by, and then again more recently when home testing became routine.  But I felt it gave a pretty accurate sense of the trend-line of COVID in various geographies.  But this current U.S. surge, driven by the Omicron Ba5 variant, which experts believe (and wastewater data confirms) is extremely high, is not being reflected much at all in the reported data.  Obviously, the rise of home testing and the ever-diminishing frequency of state and country updates has taken its toll. 

I think that the best potential for tracking COVID now lies in wastewater data.  There are now 900+ reporting sites across the country, and growing, probably enough for good national trending, but not for more local than that (apart from the sites themselves).  Wastewater is an objective way of tracking the presence of the virus that requires no human decision-making.  Unfortunately the CDC wastewater reporting leaves a lot to be desired.  The reports on the site are difficult to interpret.  And while I have not spent a lot of time on it, I have done enough exploring to know that I cannot easily (if at all) get to the granular data that would enable me to create a more user-friendly report (which was my goal with the initial "data.")  

Without good data, we are flying blind.  We have little sense of whether Covid is prevalent in our area or not, whether the trend is going up or down, and thus no clear measures to calibrate the riskiness of activities or travels we may undertake.  And without good data reporting, those who are inclined to wish Covid away cannot be persuaded by compelling numbers. 

Thus to the real point of this article – not to bemoan the demise of solid data, but rather a far worse phenomenon that the lack of data enables:  the rise of “magical thinking” among people who ought to know better. 

Once Covid became politicized, our country quickly sorted itself out into the familiar red/blue divide.  On the one side were those who were Covid skeptics, anti-vaxxers and those who quickly shed, if they ever had it, any sense of shared sacrifice and civic duty to their fellow citizens, people who willfully ignored safe practices of any kind.  On the other side were those who respected the threat posed by the virus, followed the science, and dutifully vaxxed up, masked up and avoided the most risky of settings. 

But now the latter group is splintering, and many previously "like-minded" people are ignoring basic safety practices, thereby endangering themselves, others, and helping the virus find new hosts which give the virus a more ample breeding ground to create and sustain new variants.  The CDC has not been helpful at all in the rise of this splinter group, and has undermined itself with a truly abysmal performance in providing clear, compelling guidance.  On top of this, the leaders of the anti-COVID effort, such as Dr. Fauci (who has been a critic, at times, of the CDC), simply do not have the platform or prominence they once had.  If you want a metric that gives an indication of the impact of all these forces, consider that of the 223 million Americans who are considered "fully vaccinated" (that is, two shots, or one J&J shot), less than half (108 million) received a first booster shot.  Somewhere along the way 100 million Americans more or less gave up. 

With all this has come a rise of “magical thinking” on the part of these formerly responsible people.  It’s not that they simply ignore Covid like their red counterparts; rather, they come up with various “Covid myths” to support doing the riskier activities they now want to do, be it going to the wedding, or the play, or the concert, or the dinner – in short, resuming their pre-Covid lives.  This is a classic case of knowing the answer – “I want to do this” – and then finding the bullet points that provide the rationale.  We hear this all the time, and it is incredibly frustrating and dangerous. 

So let me shred some enabling Covid myths.

·     “Covid is going away.”  It is actually the opposite.  Covid is actually getting worse.  Every new omicron variant appears to be more transmissible, if not more dangerous, than the last.  Tons of people are getting it; the anecdotal evidence among our immediate friends and family is overwhelming and inescapable.  And the higher transmissions are leading to more hospitalizations and deaths. 

·     “Everyone is inevitably going to get it anyway, so you might as well just get it over with.”  Actually, everyone is not getting it, and if you behave reasonably responsibly, armed with the latest information, you can lower your odds markedly (though you can’t eliminate them).  And you don’t want to get this:  if you get it more than once, you are potentially weakening your body more and more each time.  It is far better to avoid getting it, and if you get it, try not to get it again, especially if you are older. 

·     “If you get Covid, you are protected against ever getting it again.”  This, too, is false.  At best you have a month, give or take, with Ba.5. 

·     “Everyone I know is getting Covid so clearly the vaccines don’t work.”  Current vaccines do not protect against getting Covid; rather they protect against the worst effects of it, including hospitalization and death.  But they are quite good at preventing those, and you should stay updated on boosters to maximize your chances of avoiding very bad outcomes. 

·     “OK, if that is true, then the worst that can happen is basically just like a bad cold, and I’m not going to sacrifice for that.”  For some people, a case of COVID is truly quite mild (or even asymptomatic).  But for others, it can be hellish (trust me, we know).  If you have some sort of compromised health status, it can put you in the hospital, even if you are double boosted. And even if you don’t have any underlying health issues, it can put you flat on your back for a week with utterly miserable symptoms (the worst headache or sore throat you have ever had, lost sense of smell, fever, nausea, day after day), and weaken you for weeks thereafter.  And that’s even if you take Paxlovid.  I can assure you this from the personal experience of a number of people I know.  Then there’s long Covid.

·     “Oh c’mon, there’s no such thing as long Covid.”  Wrong.  We still don’t know much about long Covid, and will learn more about it in the coming years.  But some material percentage of people experience long Covid symptoms, with estimates ranging from 5-50%.  These people suffer from brain fog or all-consuming fatigue months after they tested negative after a bout with Covid, and even worse things can happen to organs that have been infected with the virus.

·     “OK, OK, but as long as I’m outside, I’m protected, right?”  Not quite; it is certainly safer outdoors, but being outdoors is not a guarantee for avoiding COVID.  If you are in a reasonably crowded setting outdoors, such as a stadium or arena, or even a crowded outdoor restaurant or wedding reception, the Ba5 variant and its already identified successors (such as the new Ba2.75 from India) will find you.  Better to avoid such places, or mask up.  For outdoor restaurants, better to find one that is less crowded or has excellent spacing, and mask up when dealing with the waiter.

·     “Well, I have Covid now, but all I have to do, according to the CDC, is wait five days, and then I can go out without risk of infecting anyone else.”  Wrong!  Part of the CDC’s madness is that this statement accurately describes their advice, but their advice willfully ignores the fact that 30% of people are still testing positive after five days.  Better to follow President Biden’s example and isolate until you have two consecutive negative tests, and stop counting days. 

·     “But positive tests can linger for 90 days!  You can’t expect me to sit it out for 90 days!”  It is only the PCR tests that can linger that long; the rapid tests that you do at your home does not linger. 

Ask yourself, if you are reading this:  do I lean on these type of arguments to justify risky behavior?  If so, then heal thyself, and help others.  Recommit to the discipline we need to prevent this scourge from continually reinventing itself, and killing tens of thousands along the way. 

The future?  Who knows, but the optimist in me says that between better vaccines and better treatments, ultimately Covid will be controllable at some level that will enable a reasonably safe return to most normal pre-Covid behaviors.  But until that time, please:  avoid indoor dining, wear KN-95 masks when you are indoors with others; wear N-95 masks in airports and on airplanes; avoid stadiums, arenas and any crowded indoor or outdoor spaces.  If for some reason you decide to attend something risky, we'll wear a mask the entire time. 

It would be wonderful if we could recapture something we perhaps had for only a few fleeting moments back in early 2020, that we were all in this together, that we need to sacrifice for one another.  But if we can’t do that, then please, let’s all behave in our own self-interest.  Stay safe, be well. 

Here’s the article from March 14, 2020.  It was called, “Why the Coronavirus Numbers Scare Me”:




  1. Since you like data, here’s some from the UK that I saw on Reddit. Can you tell me if this guy is wrong?

    Comparison of both covid and all-cause mortality in the UK between Spring of 2021 and 2022. (It’s not looking good for the vaccinated, and it’s looking even worse for the boosted)

    It took a bit of time to sort through the data and compile these numbers, but if you must have a TLDR, here it is:

    Data from the spring of this year shows that the percentage of deaths from the vaxxed population is HIGHER than the percentage of the population that they actually represent, meaning that vaccine efficacy against death is officially negative.

    The following represents a comparison of the percentage of vaxxed versus unvaxxed deaths from covid and all-causes in the UK from March through May of 2021 against March-May of 2022. A lot of us have been following this data for a long time, but the UK HSA and ONS stopped quantifying theses numbers in their reports earlier this year. I specifically focused on the 40-49 and 50-59 age groups in attempt to significantly reduce age-bias in the deaths counts, eliminating that as an easy brush-off explanation for these data.

    Furthermore, and perhaps most importantly, by April of this year, 70% of the entire UK population was estimated to have had covid, which means that the vast majority of vaccinated people in the 2022 dataset had both vaccine-induced AND natural immunity, which we’ve been told is as good as it gets:

    Here are the results:

    Covid mortality by age group ever-vaccinated versus unvaccinated Spring 2021:

    40-49: 13.9% vaxxed vs. 86.17% unvaxxed deaths, with approximately 65% of this population vaccinated

    50-59: 24.6% vaxxed vs. 75.4% unvaxxed deaths, with approximately 85% of this population vaccinated

    All-cause mortality by age group ever-vaccinated versus unvaccinated Spring 2021:

    40-49: 47.7% vaxxed vs. 52.3% unvaxxed deaths, with approximately 65% of this population vaccinated

    50-59: 62.2% vaxxed vs. 37.8% unvaxxed deaths, with approximately 85% of this population vaccinated

    Covid mortality by age group ever-vaccinated versus unvaccinated Spring 2022:

    40-49: 68.4% vaxxed vs. 31.6% unvaxxed deaths, with approximately 76.9% of this population vaccinated and 61.2% boosted

    50-59: 86.6% vaxxed vs. 13.4% unvaxxed deaths, with approximately 86.1% of this population vaccinated and 76% boosted

    All-cause mortality by age group ever-vaccinated versus unvaccinated Spring 2022:

    40-49: 83.1% vaxxed vs. 16.9% unvaxxed deaths, with approximately 76.9% of this population vaccinated and 61.2% boosted

    50-59: 88.3% vaxxed vs. 11.7% unvaxxed deaths, with approximately 86.1% of this population vaccinated and 76% boosted

    Deaths figures come from Table 8

    Present vaccine/booster uptake figures

    Past vaccine uptake figures

  2. Two things spring to mind with respect to the UK data - lurking variables and survivorship bias.

    I can think of many variables besides age and vaccination status that could impact the ultimate outcome of catching Covid. To name just a few: smoker vs. nonsmoker, asthmatic vs. non-asthmatic, incidence of hypertension, diabetes, and heart disease, health of liver and kidneys, whether the patient was obese, sedentary vs. active life style, whether the patient was previously hospitalized for Covid, etc.. These are all lurking variables because we would suspect they could have an impact on the outcome but they are not included in any analysis of the data.

    Survivorship bias is also important because you would expect that a significant portion of the elderly or chronically ill patients that were unvaccinated are already dead. So for any given age cohort, I would expect that the unvaccinated portion that survived to Jan 1 2022 were on average healthier than than the vaccinated portion of that same age cohort. Plus knowing you have chronic health conditions could influence you to get vaccined in the first place.


Leave a comment