(NY Times) 72 Hours Inside a Hospital Battling Coronavirus interviews Dr. Colleen Smith at Elmhurst Hospital in Queens, NYC. The Times is providing COVID coverage free to non-subscribers; this video should be universally viewable. At 4:00 in the video, she says, “What’s getting a little scary now is that the patients we’re getting are much sicker”,… “young people without comorbidity” , that is, underlying health conditions.”
Dr. Smith’s observation is unavoidably subject to confounding factors, a.k.a. “noise.” The swell of sicker patients could be individuals who thought they could tough it out, or are not very aware of how they feel. Still, it’s appropriate to consider that COVID-19 may be gaining virulence in NYC. COVID-19: A Warning.
Plagues of the past were characterized by virulence that increased in waves. Many mechanisms, including but not limited to, mutation-evolution, are involved. Some of these are understood. Virulence factors are identified features of a pathogen which affect the severity of a disease.
There are huge holes of understanding. Knowledge of how these factors work at the viral level varies from complete to partial to slight. How they turn on or off is partly, or in a few cases, completely understood. There is no quantitative understanding that connects biochemical understanding of virulence with epidemiology. Yet the macroscopic viewpoint of epidemiology, with centuries of observation, has much to say about it.
There is no way to predict the virulence of a novel virus from viral genes. Understanding has come from the reverse direction: the mechanism of virulence is observed, and then identified in the viral genetics. This is because, despite unraveling of the genetic code, and detailed knowledge of a minority of cellular enzyme systems, most enzyme systems remain a mystery. There are just too many of them.
If you’ve been following the news, you know there is no way to predict R_o, or lethality of COVID, at all. It’s purely observational, depending not just on the pathogen, but the behavior of people, and how they live. It may depend so much on local circumstances to be an invalid concept. Later on, we’ll understand more.
The demand to know the future fuels predictions that have no basis. (CNN) Coronavirus deaths in the US could reach peak in three weeks, epidemiologist says. Of the four authorities, three are doubtful, or subject to misinterpretation in the context of the article. Dr. Stefan Flasche is credible. Quoting,
Dr. Stefan Flasche, a disease modeler at The London School of Hygiene & Tropical Medicine, told CNN by email the peak was influenced by the efficiency of lockdown measures, and “may be anywhere between some time very soon and not for another few months.”
“One scenario is we can indeed reverse the spread as done in China and South Korea, then reach a point to lift the distancing measures,” Flasche said. “But (we may) have to repeat this cycle for a few times because of an inevitable resurgence of cases in the absence of population immunity. In that scenario, we would see multiple peaks in the upcoming 12 months.”
In this vagueness, we have found an honest man. An honest team may be found at (pdf 16 March 2020, Imperial College COVID-19 Response Team) Impact of non-pharmaceutical interventions to reduce COVID-19 mortality and healthcare demand. They are honest because they lay out all of their assumptions, which are unavoidably incorrect. They are predictably wrong, just as the CEP (circular error probable for a gravity bomb) is predictable.
But what is an epidemiologist to do? WHO couldn’t even predict COVID-19 would get out of China, and we now demand detail? Ignored in normal times, this honorable and troubled profession is now belabored with impossible demands. In the press, this includes a hopeful confusion of “peak deaths”, which is a rate, versus cumulative mortality, which most article writers ignore, and looms bleakly in our future.
In China, two variants of COVID-19 have already been identified, Type L and Type S. COVID-19: A Warning. This factor is absent from all the predictions. The epidemiologists have too much on their plates to deal with it.
How do you find an epidemiologist you can trust? Throw him in the snake pit:
My apologies to all epidemiologists. I couldn’t resist it.
To be continued shortly.