Why We Don't Need Politics But Science.
Why we don't need politics but science.
People seem to want to make everything about politics these days. But if we are going to figure out what to do with coronavirus and all the other viruses that aren’t going anywhere soon, or which will start up in the future, because that’s what viruses do, we need to gather large amounts of accurate data, and then have it analyzed by non-political experts in the fields of virology and epidemiology, so we can minimize viral impact with the least amount of collateral damage.
Imagine a scenario where you go to the hospital with a bad toe infection. And the surgeon there cuts your leg off at the hip. So obviously, that fixed your toe problem, but there was a lot of collateral damage. Was it worth it? Maybe. Or maybe you could have just had the toe taken off, or your leg below the knee, or maybe you just needed antibiotics. Or maybe even worse, the surgeon cut off the wrong leg, and now you have an infected toe and only one leg. Or even worse than that, the surgeon cut off the wrong leg, and it turns out it wasn’t a bad infection, just a hang nail.
Hopefully, if you found yourself in that situation, the surgeon would have some data to back up his or her decision. He or she could show you statistics, or scholarly articles that provided data about what would happen and why the amputation was the best course of action.
We aren’t there yet with this virus.
We really don’t know what works, what is necessary to work and what is overkill, or what measures specifically are most effective. We aren’t even sure of our endgame. Actually, we aren’t even sure of our beginning game, or what the point of the game even is.
Things we need to know:
What is the current incidence of this disease? That is, how many actual active cases are going on now? (Remember, those trackers on your search engine only track cases since the beginning of time, so they don’t tell you how many cases are going on now.)
How prevalent is this disease really? How many unknown cases are there?
What is the real death rate, once we know the prevalence?
How is this the same or different from the pathogen threats we have faced our entire lives?
These are fact questions. Once we know this we also have to ask ourselves:
Who do we need to protect?
What interventions actual help control spread, what ones didn’t do anything and what ones were actively harmful?
How can we design a plan that maximizes benefit while reducing collateral damage?
These are all big questions, so we will start with talking about prevalence, and its effect on death rate, in this post.
The other day we wrote about COVID-19 estimated death rates using New York data. But that data doesn’t tell the whole US story. In fact, most of our data doesn’t tell the whole story, of course.
According to the Bing search engine “tracker” there have been (as of May 12, 2020) 1,382,630 cases of COVID-19 infection in the US since actual tracking began sometime in early March. As we have stated before that does not mean that there are currently 1.4 million people with active disease. It is difficult to say exactly how many people are currently symptomatic, given the changing curve of illness and the variability in illness length. The three states with the least cases; Alaska, Montana and Hawaii, who really do seem to be tracking recoveries, report that out of 1,476 cases, 119 may still be active. That’s 8 percent. But most of the other states, with more cases are not tracking this number. An estimate would only be able to say somewhere between 120,000 and 250,000 people is likely.
FYI, for a deep dive into perspective land, 1.4 million is 0.42 of the US population, or 4 in a 1000. Currently ill would be about 4 to 8 in 10,000. That’s about the same number as the number of men per 10,000 that have been convicted of rape in the US. Your chance of walking outside your door and meeting someone with active COVID-19 is overall the same as meeting a convicted rapist.
The data does show that we shouldn’t have used New York the other day for a projection of US death rates, because New York State is an outlier. As of May 12, 2020 the Bing tracker reports 315,415 cases in New York since the beginning of time, and 21,640 deaths. That’s 7%. But most states report death rates about one-half that, in the 3-4% range.
Why are these different? It’s impossible to say for sure, but three likely culprits; the willingness of doctors to put coronavirus or COVID-19 as a cause of death on a death certificate, the amount of testing made available to people with less serious symptoms or less reported contacts, and the base population involved.
Firstly, about those death certificates. It was highly disillusioning to learn in medical school and residency that these are often filled out in a very haphazard manner. Once the patient is dead, it’s not usually seen as that important what killed them. Was the primary cause of death pneumonia, or the late stage cancer they also had? You’re a busy resident with living patients, so just pick one and move on. No one is paying you to list every illness, either. Maybe the patient actually had a heart attack on top of everything else, but you never knew. In general, you cannot assume that a cause of death on a death certificate is gospel. (Also, we make up those times that we put on the birth certificates for when a baby is born. We are too busy with the actual event to time it to the precise minute.)
Now to make one thing clear, when we talk about the willingness of doctors to put COVID-19 as a primary cause of death, we do not mean, in the US, any political agenda or conspiracy. Americans are, after all, incapable of keeping their mouths shut on any topic, making secrets impossible. It’s more that in some areas the custom is now to err on the side of certainty, and in others more presumption is allowed. In some areas, in order for a person to be listed as dying of coronavirus illness, a positive test and a certain pattern of disease progression may be required and in others, it might be enough to have an old person with a viral-type pneumonia.
Testing impacts the death rate because the more mild or asymptomatic people are tested, the more mild, hidden infections you find and thus the lower the rate of serious illness and complications, including death, because you now have a bigger pool of illness for comparison. When testing is expanded by relaxing the criteria for getting a test, then the overall severity of people found with illness is likely to decrease.
As for the base population, let us look at the example of Iowa. Iowa currently reports 289 fatalities out of 12,912 reported cases. That’s 2 percent. Why is it so low? Because unlike some states where elderly folks in communal living situations have been hit hard, some of Iowa’s biggest clusters have been in meat packing plants, plants staffed by much younger, healthier people. One thing we do know with certainty about COVID-19 is, like all respiratory viruses, it hits older people harder and 85 percent plus of all deaths have been in people over 60. So, Iowa, through no fault of its own, gets a lower death rate.
Nothing about this evolving situation or this discussion has changed our position here at You&Me. We need vastly expanded testing in the United States, both for active disease and also for antibodies, so we can get an accurate picture of just what we are dealing with in terms of dangerous complications, including death.
Until we have a more than a wild guess at how many people have actually had some kind of episode with this coronavirus, we cannot move forward to determine how dangerous it really is for most people, or where we should put our efforts in controlling it.