Archive | COVID-19 RSS for this section

In Need of Hypotheses

GG wrote something about this awhile back, but it feels worthy of a revisit. Just why is it that geologists still like this “multiple working hypotheses” ideas?

What reminded GG of this was reading Naomi Oreskes’s book on the rejection of continental drift (or Amazon link). In there, it sort of seems as though multiple working hypotheses comes across as something of an excuse used by twentieth century American geoscientists to dance past the evidence for continental drift. It kind of comes across as a dated approach for pre-quantitative science. GG would argue that in studying complex phenomena that it is an important tool–one perhaps worthy of keeping in mind in dealing with the current pandemic.

Read More…

Mysteries of the COVID spikes

Something that has bothered GG has been just how rapidly rising COVID cases have turned into falling COVID cases, for instance:

COVID cases for Colorado and Boulder County, Colorado, from COVIDActNow

This tendency of COVID cases to rise rapidly and then fall equally dramatically is best seen in the geographically tighter county-level data, but you can also see it in places like the United Kingdom, where rapidly rising cases turned into a freefall in a couple days in the summer and in some states in the southern U.S. that are now passing their peaks.

Read More…

Skewing COVID vaccines

From a professional standpoint, one of the most annoying aspects of the pandemic has been the demonstration of numerical illiteracy–not only from the public, which is kind of expected, but from the media and even professionals engaged in public health. The net effect at the moment is to grossly underestimate the efficacy of the current set of vaccines. How does this happen? Well, you compare apples and oranges.

Let’s consider the New York Times‘s attempt to examine the rates of breakthrough cases as a means of seeing just how good the vaccines are. They were trying to get around the biases caused by there being different numbers of vaccinated and unvaccinated people, which has muddied the waters as breakthrough cases mounted. Their analysis finds that in Colorado (for example), if you are vaccinated, you are 22 times less likely to end up in the hospital and 8 times less likely to die of COVID-19. That should give you pause, because it suggests that should you, as a vaccinated individual, happen to get a breakthrough case severe enough to put you in the hospital, you are more likely to die than an unvaccinated patient. This is not unique to Colorado: pretty much all the states’ data look the same.

Read More…

Are COVID models useful?

OK, this is *way* out of GG’s wheelhouse, so please take this as rantings of an ignorant observer…

But something GG has noticed is that the epidemiological models out there–as far as he can see, *all* of them–will not reproduce the kind of rapid rise followed by a rapid fall in cases of COVID-19. To be clear, they can do OK on the rapid rise part–frankly, it is trivial to fit an exponential to the rising side, and all you have to do is adjust Rt a bit to hit the numbers. It is the reversal that simply gets missed; usually the models predict a long drawn-out peak, which really has not been the case for the most part (you only see that for much larger regions where local peaks are offset in time from one another). And so we have mysteries like why the bit Indian spike a few months ago suddenly died out, or even why the UK’s spike suddenly reversed a couple of weeks ago.

Why should things reverse so suddenly? Maybe back in March 2020, it was everybody suddenly cowering at home; that was certainly the single most abrupt behavioral change in the pandemic. But since then, has there really been that kind of radical reversal of behaviors? Enough to stop rising infections in their tracks? This seems unlikely; the fact that the models do not produce such an outcome despite sometimes having sudden changes in “social distancing” suggests it isn’t the likely cause. It is something missing from the models; what might it be?

Read More…

The Power of Denial

As we in the U.S. stand at the brink of our fifth (!) wave of COVID-19, it is worth taking a moment to ponder what this says about society as a whole and our chances of reining in climate change.

First up, why a fifth wave? Unlike the previous four (March-April 2020, mid summer 2020, late fall 2020, spring 2021), this one is one that should have been prevented. An irony is that the Trump administration put all its eggs into one basket marked “vaccine”–and it paid off–and yet the people who are most disdaining of the vaccines are Trump’s supporters. Because we are increasingly living in politically uniform communities, the result has been some communities where nearly nobody has gotten vaccinated and others where nearly everyone has. With the delta variant rolling along (it has been responsible for most Colorado cases for well over a month), case numbers have returned to rising. In places like Boulder, this is a rise from very small numbers of cases to small numbers. In places like Branson, MO and Grand Junction CO, it is a rise from kind of OK to oh-my-gosh. It seems clear that we’re headed for a long haul of COVID kind of rumbling around, flaring up in unvaccinated areas while just steaming a bit on corners of well-vaccinated communities.

The first and obvious lesson is that even experience isn’t enough to fully educate some fraction of the populace. Lincoln supposedly said that you can fool some of the people all of the time and all of the people some of the time, but you cannot fool all the people all the time. With dying patients spitting in the eye of nurses who are telling them they have COVID and people who spent many days in ICU saying that COVID isn’t that big a deal, it is pretty clear we have identified the people who can be fooled all the time. The science is about as crystal clear as it gets: COVID-19 is a transmissible disease that can be made far less infectious and far less dangerous through the use of vaccines. Everything else is quibbling.

So it is clear that no number of hurricanes or droughts or forest files or king tide floods or historic heat waves are going to convince some people–probably close to 30-40%–that climate change is a real problem. There will not be a hosanna moment when we all unite to defeat the scourge of climate change. After all, if we can’t unite to defeat a fatal disease by getting one or two shots, how likely is it that we’ll be willing to unite behind a realignment of society’s energy system?

So we’re doomed?

Read More…

Risk and Sensibility

The Denver Post ran an opinion piece posing a useful question: how did we do in evaluating risk of COVID-19? Answers are all over the map, but so many are so breathtakingly stupid as to make the value of the piece pretty minimal (“Hey, the mortality rate was only 1.4%, so not so bad!” “My restaurant didn’t have a case, so restaurants were safe!”). The tone of the op-ed leans towards “hey, we panicked, this wasn’t such a big threat after all.” Having the cojones to put that in print after over half a million Americans perished in under a year–easily the number three cause of death in 2020, and when you toss in the excess mortality since the start of the pandemic, you are up to ~650,000 deaths either directly or indirectly caused by the pandemic. The age-adjusted death rate increased by nearly 16% in 2020. COVID-19 was the number three killer in the U.S. Sure, your odds of dying if you were between 25 and 34 only increased by about 4%, making COVID about the #6 killer in that age range, but for 44-55 year olds, the odds of dying went up by 11% or so, making COVID the number 4 killer and close enough to “unintentional injury” that it could well be #3. And then none of this considers the long term impacts of having had COVID-19–of the more than two million Americans who have spent time in the hospital, something like a half million or more are seeing long-term effects from COVID-19, effects that are often pretty debilitating.

A way of looking at this is years of life lost to the disease–in essence, if you die at 25 and were expected to live to 85, that is 60 years lost, but if you die at 80 it is only 5 years lost. A paper in Nature tackles this and makes comparisons with traffic deaths, seasonal influenza, and heart disease. In the U.S., COVID-19 has been seven times the losses from influenza, about 2.3 times the loss due to transport (mainly auto accidents) and maybe about 0.4 times the loss due to heart disease. This is with a pretty conservative estimate of COVID’s impacts, and it is worth keeping in mind that we tried very hard to limit COVID-19 but not so much many of these other causes of death. Additionally, this study found that about 20% of the loss of years of life in the U.S. was suffered by those under 55.

So were we overreacting to a disease that did cause more loss of life that auto accidents (by more than a factor of two)? Certainly had we done nothing the toll would have been far, far worse (millions dead), so the question maybe should be, what were the really effective steps, and what was theater that did little? in other words, a better question is, how well did we do in balancing the risks posed by different activities for dying of COVID-19?

Read More…

Herd Racing

So how does the pandemic end? While we’ve discussed possible and likely overall mortality, the details of the end of the pandemic were so far away that it wasn’t worth much thought. But with vaccinations underway, we can start to see the end with herd immunity in the offing. But it probably isn’t what you are thinking. Let’s play with a straightforward scenario: Vaccinations proceed roughly as has already been outlined. Does this mean that COVID cases gradually die out? GG isn’t so sure.

Here in Colorado the governor’s main worry has been overwhelming hospitals, and he has been eager to remove restrictions on businesses. So by the end of February all those 70 and over who want a vaccination will get it. By the state’s calculation, this will reduce pressure on hospitals overall by a third, deaths by nearly four-fifths, and ICUs somewhere in between. So sometime in March the reports on deaths will plummet and the numbers hospitalized will drop. Two things are likely at this point: the state will probably reduce restrictions on businesses, and the public’s perception of the hazards of personal interaction will decrease.

So what seems highly plausible is that we’ll see the pent-up demand for movie theaters and restaurants and weddings and all the rest start to go upward. With vaccinations reaching the 60-70 year old crowd, the fraction of those infected that go to the hospital will drop by more than a factor of two. For instance, in Boulder County, a quarter of those 75 and older who test positive end up in the hospital or morgue. 65-74 year olds only end up in the hospital 7.5% of the time. 55-64 6%, 25-54 less than 2%. And the fraction of 25-54 year olds who are infected but never tested is probably higher than the older population, so it is quite possible that the actual hospitalization rate is well under 1%. So by the time the 60-year-olds and up are immunized, personal acquaintanceship with dire stories of saying goodbye to grandma via the phone will be fading fast, to be replaced with more folks saying “it was like a bad cold, not so awful, really.”

So there is an excellent chance that rather than seeing numbers gradually decline, there will be one final wave of COVID-19, one tearing through young adult communities more than likely (communities that have already demonstrated the impatience of youth). Given that we are seeing rates of infection in LA County where 1% of the population is testing positive each week–and probably 2-3 times as many are actually infected–it might be a race to see if more people become immunized by having had COVID-19 or by having gotten a shot. While hospitals and morgues will be far less likely to be overwhelmed, case numbers could skyrocket in this final wave. This might end quite abruptly as we rapidly approach herd immunity levels of exposure+vaccination.

In a way this sounds hopeful: we’d actually reach herd immunity sooner in such an environment. But that is not an unmitigated good. While less than 1 in 50 of these victims of the disease will be hospitalized, that is still a huge number of people. And given the high rates of long-term effects of COVID-19 among those hospitalized, ending the pandemic by a final wave among younger people would carry some real costs in terms of disabilities and collateral illnesses.

What might prevent this? Keeping high-risk businesses closed longer might well help; this realistically requires these businesses to get financial support from the government (here in Colorado, Gov. Polis has noted that the absence of such support figures into his decisions on COVID interventions). Having a known date to get the vaccine might help, too. If your calendar reminds you how many more days until you get the vaccine, you might be more willing to tough it out than to break with your good behavior to go out with friends. Certainly the faster vaccines are distributed, the fewer cases of COVID we will see. And finally, there is the weather factor. Going into spring we might see cases decline simply because we have fresh air in houses and spend more time outdoors; maybe this is enough to counterbalance these other impulses.

So while the end is in sight, how the end comes remains in question.

The ICU Christmas

Pictures from around the country, but especially California, are giving GG déjà vu. Two years ago I had the misfortune to spend Christmas (and New Years) in an ICU room. Seeing the images of drawings of Santa or snowmen taped to walls or plastic tents brought back memories of the decorations my family brought in to the hospital. Not really warm, wonderful memories.

I discovered that ICUs are usually pretty quiet over the holidays (this year an obvious exception); this largely because elective surgeries are on hiatus for the most part. Other than that, I can’t imagine it is all that different from the rest of the year in terms of routine and inconvenience; it is of course quite different emotionally. One day you are stringing up lights, and then you’re the one strung up on tubes.

Read More…

COVID killed which learning?

Last spring the word was that because of the coronavirus, remote teaching would finally be the killer app that would make good on the promise of MOOCs and other efforts to liberate colleges from the hidebound methods of teaching that date back to the Middle Ages. College presidents looking at budgets heavy on old professors’ salaries were rubbing their hands with some glee; this could finally make colleges more financially secure. Then students weighed in with “ick.” Over the summer the institutional response was, well of course the old farts couldn’t teach online that well, you had to do it right with preparation and tools and workshops. Universities spent money on IT resources and the various teaching improvement groups held workshops and faculty played with Zoom breakout rooms and such not. So here in the fall, we were to see the conquering of education by remote teaching after the slap-dash failure in the spring.

What did we get? Um, lawsuits demanding refunds for lower than expected teaching quality. Schools cutting programs because students were not flocking to get an education through a somewhat small TV screen. So where are we going?

Right now, of course, there will be more of the same for the spring. But let’s skip past that. What will things look like in the fall of 2021?

Read More…

Tenure, the coming COVID fatality

Universities and colleges have been under tremendous financial pressure due to the coronavirus pandemic. This is particularly true of state schools that have seen general fund moneys decline markedly since the Great Recession. While students and families yell, scream, lobby or file lawsuits to try to get money back from an educational environment that they find to be substandard, the response from most colleges and universities is, we hear you, but our pockets are bare.

What does this mean for the future? Well, administrators are now saying things like “our budget is quite brittle”–which kind of means that if stressed, it doesn’t bend, it breaks. If you get beyond a certain level of cuts that can be managed with furloughs and temporary salary reductions, the next step is outright dissolution of departments and programs. Part of the reason is presented to be tenure, though the reality is that tenure protections do not extend to financial hardship of the university. Of course it can be hard to show that a tenured faculty member had to be let go while colleagues in the same department continue forward; it is for this reason that most schools view the easier solution as ending programs.

So what is the solution?

Read More…