Ebola:

Still a slow grind upwards in new cases.  At 118 confirmed now.  Biggest development is some of these are in health areas that border the Republic of the Congo, a separate Congo to the northwest of the Democratic Republic of the Congo.  Travel checkpoint screening could be a little tighter, but they are getting the vast majority of people going through them.  Otherwise, more of the same.  Many known cases are still in the community.  Community resistance to disease control measures is high (again, no surprise post-COVID).  Contact tracing is inadequate.  There are now concerns about adequate lab supplies for Ebola testing this week.  Risk of cross border spread to at least the other Congo is low overall, but exists and is rising inexorably with the trickle of new cases each week.  The vaccine and Ebola specific treatments continue to be employed though, which is likely keeping new cases from a parabolic explosion.

Coronavirus:

–Overall zeitgeist:  New cases again continue to ramp all across Europe and discussions of re-lockdown on at least regional and city levels continue.  The UK, for instance, was definitely rattling that saber this week.  There are reasonable questions of “if the first lockdown didn’t work, and it clearly didn’t if cases are back and rising, what will the second actually accomplish?”  Nationwide, hard to argue that they make sense (unless your nation is the size of a locality).  Local/regional lockdowns may still be useful in select instances though, as it may keep the hospital down the street from having a “bed’s taken” problem if the case velocity is sufficiently high. I don’t have enough local insight to know where the case velocity is sufficiently high all over the world though.  So you will just have to make the incredibly dangerous assumption that the people in charge of your local area have some semblance of an idea of what they are doing.  Who knows–they actually might.  

Stateside, the Great Plains and upper Midwest are seeing a lot of action this week.  Wyoming leads the Rt tables (in a bad way–no word yet if tumbleweeds were a contributing risk factor for the spread of SARS-CoV-2 from the one person in Wyoming to the only other person who is in Wyoming).  Wisconsin is seeing a bit of another spike, and Texas had a massive rip up earlier in the week that looks more like a bunch of tests or old results got reported–mostly because there were no terror headlines about it, even in Texas media that I can see with a total of 3 minutes Googling.  New York City was vocally concerned about hot spots in Brooklyn and another neighborhood I cannot recall of the top of my head.  New cases for the state have not budged from the 650-850 or so per day that New York has been stuck at since June, but hospitalizations did hit a small increase late this week.  This may (hopefully) all still be noise in the data.  My guess is the hot spots in the City are very localized, and amenable to pretty quick containment–plus, a lot of people in New York got exposed already, so the virus probably doesn’t have as much room to spread as you would otherwise think.  Still, something to keep an eye on for a couple weeks and see how the trends play out. 

Indiana is moving to re-open all bars, restaurants, gyms etc. this Saturday, although daily new positive cases are just coming off the summer plateau.  Mask rule is still in effect though.  We’ll see where this goes over the next two weeks.  If there is no bump, that will be a VERY positive sign.  Following up on previous stories, the mandatory mask rule in Marion County, Indiana made no immediately obvious difference in positive case rate versus the suburb counties.  Formal statistical analysis might find something there, but again, masks seem best at controlling your individual risk.

–Getting into some more details…  Pre-publication data out from the University of Houston which sequenced a little over 5,000 samples over the course of the outbreak since March.  They report that the D614G mutant of SARS-CoV-2, whom you may recall from such DOOOOOOOOOOOOM headlines as the mutant that binds the human ACE2 receptor better with fears that it would cause more significant disease, has pretty much taken over as THE main viral strain.  They report, with numbers that I doubt will change much on the way to publication, that 71% of the SARS-CoV-2 samples in March/April were the D614G variant.  Now, that variant is 99.9% of new cases in Texas, including the huge spike from June through July.  As we have mentioned, it appears this variant is indeed a bit more contagious than its cousins–but fears that it would lead to greater incidence of severe disease were a little premature.  As we have said all summer, even as nominal cases went up, hospitalization rates have stayed largely steady or even down.  That may be due to the biology of this strain (more contagious, less severe disease) and/or improved recognition of COVID and how to best treat cases requiring hospitalization.

In fact, it’s worth mentioning that in the US at least, nation wide hospitalization and death rates continue to decline.

–More adventures in poor reporting this week.  The CDC released a study assessing relative risk of contracting SARS-CoV-2 based on US patients in July, compared to a group of US residents who did NOT get SARS-CoV-2 in that period.  A national reporting example can be found here:   https://www.yahoo.com/lifestyle/dr-fauci-says-3-riskiest-205900256.html
The first tip off is the “Dr. Fauci says” clickbait headline, but the 3 riskiest activities are listed as gyms, bars and restaurants.  The article cherry picks poor Dr. Fauci a bit, because he DID say that closing those facilities or restricting them helped three states with big spikes earlier in the summer contain the rate of spike a bit.  That is true.  Dr. Fauci then goes on, as quoted in the article to say that restaurant dining, especially when people are not following social distancing recommendations, is the highest risk activity based on the data.

That is also true–and the correct interpretation of the data.

The problem I have with this reporting is the article also dinged gyms and restaurants, and did not emphasize that restaurants are only a risk if social distancing is not being enforced.

Here’s the actual data from the CDC’s report:

This is an odds ratio, so what you are seeing is the chance that someone with a behavior listed on the left (y-axis) got COVID relative to the control patients who did not, but also engaged in that same activity.  You also have a breakdown in dark blue versus light blue for if they had a known close COVID-19 contact at home.  If the light blue and dark blue lines cross the vertical dotted line, there is NO statistically significant increase in risk of getting COVID for that activity, at least by looking at people who DID that activity AND got COVID compared to those who DID that activity but did NOT get COVID.  The shorter the lines, the more confident the prediction.

The ONLY significant risk is “restaurant” (NOT bars and gyms), and the CDC specifically caveated that, as the risk is confined to those restaurants where the study participants remembered social distancing guidelines not being strictly enforced.  Bars/coffee shops appear to be a risk for people without a known close contact at home, but there is a -wide- variance in the estimate of “how risky”–that one may not hold up if a larger study like this is done and looks at bars/coffee shops again.

Without providing this important context, the lay article makes bars/coffee shops/gyms sound more dangerous than the actual data suggests they are.  Also, restaurants appear to be suspect ONLY if the people in them are acting suspect in terms of their social distancing adherence.

Continue to be prudent, but the good news is there are a LOT of “activities of daily living” that do NOT appear to increase your chances of contracting COVID-19 over baseline, especially when good social distancing practices are used.

–In terms of travel, multiple headlines this week that the CDC has documented 11,000 people who may have been exposed to SARS-CoV-2 while traveling on an airplane with one or more of 1600 known positives who were traveling around the time they tested positive.  There are a few case series from planes, mostly in Asia, in February/March suggesting it is possible to get SARS-CoV-2 on a plane.  In at least a few cases, getting exposed on the plane seems probable.  That said, the CDC has not confirmed that any of the 11,000 exposures on a plane got SARS-CoV-2 from that possible exposure during air travel.  That’s not to say it didn’t happen–they have incomplete follow up, and proving that the plane exposure did it would be difficult in the best of circumstances given how widespread the virus is.  But “exposed” and “contracted” are two very different things.  Currently, the CDC on its travel recommendations merely cautions that you may be sitting less than 6 feet from someone who has the virus if you travel by air, and recommends wearing a mask for the duration of the flight.  Which, to the best of my knowledge, is something all airlines insist on at the moment anyways.  The CDC also states that it believes spread of the virus on a plane is low risk because of the way the air in the cabin is circulated and filtered.  So assess your comfort level with those facts. 

I have no plans to change planned air travel coming up.  I do plan on wearing a mask for the entire flight (and through the air port).

–Good read on the situation on the ground from Sebastian Rushworth, a physician in Stockholm, Sweden here:  https://sebastianrushworth.com/2020/09/19/covid-19-does-sweden-have-herd-immunity/

–He even found a paper that suggests that T-cell based immunity for SARS-CoV-1 (the original SARS virus) was still present in patients who got the original SARS in the early 2000s seventeen years later.  Bodes well for long term immunity for SARS-CoV-2 following an infectious exposure, if T-cells continue to appear to be a major, or THE major, source of immunity.  You can find the link in his blog post above.  My man, Sebastian! 

–Speaking of T-cells, higher than expected innate immunity to SARS-CoV-2 (possibly because of those T-cells), and the estimated number for herd immunity, this was published recently in the BMJ:  https://www.bmj.com/content/370/bmj.m3563

A lot of it is going to sound familiar, but I swear I did not write it under a pen name!  Instead, because it agrees with several themes I’ve been running here lately, I -absolutely- link to it : )  Not only does it help my confirmation bias, but it shows you it’s not just me following some of these threads!

–On the flip side, a new meta-analysis this week suggests most people who catch COVID actually do develop symptoms when followed over time:  https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1003346

We have talked before that the three possibilities (other than true sample identification or lab error) of a positive test with no symptoms break down as follows:

1)  You truly have subclinical infection, and will not develop symptoms

2)  You are at the end of a COVID infection, and either didn’t mention your symptoms, or they were so mild you didn’t even notice them

3)  You are early in a COVID infection, and just haven’t gotten symptoms yet.

The authors’ argument is that the handful of studies that have followed patients without symptoms at the time of testing after they got tested, only about 20% of them never developed symptoms and were true asymptomatic positives.  Scenario 3 may be more common than we realize. 

If that’s true, my back of the envelope for how close the US, at least, is to herd immunity is off.

So, since that disagrees with me, we will immediately poke holes in the method.  Cognitive biases are so fun, aren’t they? 

At any rate, even the authors concede the study is limited by the fact that no one defines “asymptomatic” the same.  Since SARS-CoV-2 infection is usually flu like, but can include things like diarrhea and then of course there are some that would take any symptom or abnormality in the context of a positive test as clear evidence of SARS-CoV-2. 

“Well, I developed a brain tumor the two days after my test…” 

“COVID!!!!  COVID CAUSES BRAIN TUMORS NOW!!!”

We’ll call those atypical and rare symptoms of SARS-CoV-2 infection. 

Anyways, in the meta-analysis, which again is a collection of similar published papers to make the total number of subjects much larger than any one study and see which way the literature as the whole is trending, there was no good uniformity to what was a “symptom” beyond the flu and diarrhea to know what was truly being called “asymptomatic.”  If the authors of the component papers were accepting ANY symptom, then it may over-estimate the true number of cases symptomatic with COVID by attributing symptoms that coincided with, but were not caused by, SARS-CoV-2.  On the other hand, if they were strictly defining SARS-CoV-2 symptoms, they may actually undercount the number of true infections, since they may miss some cases with atypical symptoms.

There also may be a bias in asking people about their symptoms after a positive test, especially if they have to think hard to remember those symptoms.  Was that one night of diarrhea four days after your positive test the ‘Rona, or because you read the reviews for Haribo sugar free gummy bears and you doubted? (one of many “best ofs” here) I mean, you coughed in the last week, right?  Was that the ol’ corona?  But it was also the night you thought vaping would be super cool.  (It was not).  So again, there could be bias in over-reporting mild symptoms, especially if they were due to something else going on. 

The finding that only 20% of cases of COVID are asymptomatic is also at odds with serology testing, which, again, estimates 7-12x more cases out there (at the low end) than PCR confirmed cases.  Not all of those were symptomatic (thank God), and a breakdown by how many positive cases, when followed over time, become hospitalizable cases is probably more informative.  It would also be really nice to see big study or meta-analysis of that now versus March/April, as the D614G strain has basically taken over and we should be able to see if there really is a change in the hospitalization rate that could be attributable to this strain.

Your chances of catching Ebola this week are equivalent to the chances you read the Haribo reviews, and were not tempted.

Your chances of catching coronavirus this week are equivalent to the chances you regret that you gave in to Haribo sugar less gummy bear temptation.

<Paladin>