Open Letter to The Atlantic

Dear Robinson and Alexis,

Thank you for your article today on the COVID positivity rate over time. [1] Good data is hard to come by, but proper interpretation of the meager data we have is almost non-existent.

I believe we will study the 2020 COVID-19 epidemic for decades, but perhaps not for the reason you suspect. We will study it as an abject failure of epidemiology; a failure to define and execute a scientifically proven method to accurately estimate a virus’s crucial properties and characteristics in the face of totally obvious and predictable testing shortfalls and biases.

It seems like everyone today is an armchair epidemiologist. Perhaps because the basic concepts are truly quite approachable. Whether we are discussing IFR, CFR, Ro, or Rt… these variables give us a fairly easy to grasp handle on how quickly should we expect a virus to spread, how much we should worry if we catch it, and what measures would be appropriate to take in curbing the spread. And yet, experts have constantly lead us far afield of reality in estimating these absolutely crucial factors.

This is why I took keen interest in your article today discussing attempts to establish prevalence, and another piece of news, being the antibody study out of Santa Clara. [2]

It should be obvious to everyone that a virus for which no pre-existing surveillance networks exist, for which no pre-existing testing assays exist, for which no human has innate immunity, which spreads easily between humans, and for which asymptomatic cases are numerous if not the overwhelming majority… it should be immediately obvious that the number of reported cases for such a virus will dramatically lag the number of actual cases. The Santa Clara study out today estimates that testing has failed to detect on the order of 98 – 99.9% of positive cases (undercounting by 50x – 85x).

This is actually tremendously good news. While the WHO irresponsibly and incorrectly pushed a “CFR” of 3.4% in China [3], what we are dealing with is, in fact, a virus that can be very dangerous to a slim minority of people (80%+ of fatalities reporting 3 or more co-morbitities) but where in the vast majority of cases is mild or even totally asymptomatic. Every day now we see new data which point to the overall IFR of COVID-19 to be closer to 0.3%, making it, despite uproarious objections to the contrary, roughly on par with a bad flu season.

The implication for the public health response is impossible to under-state. As we finally discover and come to terms with the fact that the prevalence of COVID-19 in metropolitan populations is likely 3-5%, and quickly approaching 10%, we can finally admit that containment is impossible, and mitigation is fruitless. The only reachable end-point is, as it is every flu season, herd immunity.

What we have control of is one key variable, which is how quickly we approach the point of herd immunity, the effect of which will be to provide innate suppression of the Ro or viral spread. In this case, the herd immunity threshold that will confer this benefit is roughly 65% exposed, given the baseline estimates Ro of SARS-CoV-2 (for lower Ro, a lower herd immunity threshold is needed to innately slow transmission in the general population). So while we appear to be roughly 1/6th the way there numerically, luckily, on a time-scale, this whole ordeal is more than half over.

All that is left, then, is to dispel this peculiar internet-meme turned national health policy that is “flattening the curve”. The premise of spreading out the time scale of infections to allow more health-care-hours available per hospitalized patient can sometimes make sense depending on the hospitalization rate, the available resources, and the effective treatment modalities.

But here’s the rub. After hearing so much about ventilators and PPE, the stark truth is this; The vast majority of healthcare worker are already exposed and immune, as antibody testing of frontline workers will soon demonstrate, making extensive PPE use essentially unnecessary. More sobering, is that 80 – 90% of patients which are ventilated have died. [4] The primary intervention has proven not only to be ineffective, but actively lethal. It’s almost maddening to consider that flattening the curve in such an environment actually kills more patients than it saves.

It will take some time for everyone to come to grips with this new data and viewpoint, but I hope it will start happening quickly, based on articles like yours.

A virus for which the only proven effective treatments are extremely scalable and moderate (non-invasive oxygen therapy and frequent repositioning), for which the vast majority of people infected will never even know they had it unless tested… calls for the immediate and complete re-opening of our economy. Literally trillions of dollars are on the line, and 20% of all workers stand unemployed, until the narrative on this disease can come to terms with the facts of COVID-19 as they have finally become readily apparent.

I hope that you will continue to report on these crucial epidemiological factors and continue to explain how to interpret them as new data becomes available, as it will help shift the narrative and ultimately help re-open our economy sooner, saving countless lives and livelihoods