Data breaking bad: COVID-19 corruption in an election year
America may be “exceptional,” but Americans are often fools. That is especially true when you feed them scraps of wildly wrong and useless data. Suddenly everyone is an instant expert on climate change, education or health care.
I cannot count the number of dinner parties I’ve disrupted by asking “committed environmentalists” a simple question: “Have you even read the IPCC reports?” Those are the U.N. Climate Change reports, often referred to as the “settled science.” They do not say the world will end in 10 years, 100 years or 1,000 years, and do not provide a compelling argument for the immediate destruction of the world economy, with endless misery for the poor, to achieve “zero emissions.” Ditto Stanford’s CREDO studies, which strongly support charter schools for the poor. Ditto the National Health Insurance Experiment, which suggests “Medicare for All” might lead to Medicare for None.
But the most egregious recent example of Americans swallowing scraps of nonsense data is COVID-19 “cases data.” Suddenly, everyone in the neighborhood is an epidemiologist with religious fervor. But, self-righteous though they may be, they are largely wrong.
While real epidemiologists at Stanford, UCLA and the University of Southern California, among others, have tried to set the record straight about the huge overstatement of COVID-19 risk, Americans remain focussed on trends in “new cases” and “hot spots.” The data have led to suggestions for renewed lockdowns. They have prompted national teachers’ unions to threaten school shutdowns.
The problem with all of this is that “cases data” and “trends” are statistically meaningless. Reported cases’ trends are the result of increased, but limited, testing. It has obvious selection bias and is in no way a random sample. It has zero relationship to any reality, re: COVID-19 “increasing trends,” and draconian policies based upon “cases” are purely political, fraudulent and irresponsible. Slanted news stories about “outbreaks” and “rising trends” appear to be designed to scare the public — and they are succeeding. Even hospitalization data have selection bias and are not a reliable “trend.”
The only useful trend data are fatalities — it is presumably a 100 percent sample and there are fewer ways to fudge it, and that trend is stable or declining in most parts of the country.
Basically, the important question on most people’s minds is the relative danger of serious illness or death across various parts of the country. The only useful statistic for that is “deaths per 100,000 population,” because that number provides a standardized basis for comparing the relative likelihood of getting seriously ill or dying in different parts of the country. The short answer is that the danger is low, although there is much higher danger in New York and the large Northeastern and Mid-Atlantic states, and very low danger elsewhere in the country. New York, New Jersey, Connecticut, Massachusetts, Rhode Island and the District of Columbia represent the highest risk region in the country. States such as Arizona, Georgia, South Carolina, Minnesota, Florida and Virginia represent far less risk. States such as California, Washington, Texas, North Carolina, Oklahoma and Oregon represent very low relative risk.
That’s an important political story.
President Trump may have picked a good state — Oklahoma — to hold his recent rally; the serious danger was small. Although some who attended tested positive for the virus after the rally, it says nothing about the incidence of virus in Tulsa. The only reliable data is deaths per 100,000 and Oklahoma is very low. Similarly, Trump rallies would be relatively risk-free in most states. While Florida would have been very low-risk for the Republican Convention, North Carolina presents minuscule risk.
Further, riots and large demonstrations evidently do not necessarily trigger fatal spikes. Although New York and D.C. lead in fatalities, Minnesota, Washington and Oregon still remain at the low-risk end of the list.
Teaching restrictions, which could permanently handicap a generation of American school kids, make no sense anywhere in the country. The same can be said for broad business shutdowns or sustained draconian measures.
In most of the country, in-person November voting could take place with sensible precautions — social distancing, face masks and hand sanitizer — and relatively no risk. From an international perspective, deaths per 100,000 across the U.S. are 45, much lower than most European countries, most of which are opening up.
Which brings us to the avuncular Dr. Anthony Fauci. Although he has changed his positions drastically about the virus and protections over the course of the pandemic, he has been immune to humility. As a trained virologist, he certainly understands that the “cases data” and “trends” are nonsense, but Fauci has finely-tuned political skills after running the National Institute of Allergy and Infectious Diseases for nearly 35 years. He has a broad constituency in America, giving out billions in annual grants to major labs and universities. He knows that shutdowns are used by some as a political weapon against the president, and he may sense where the wind is blowing.
When Fauci threw the first pitch at the opening game for the Nationals, it was a wild pitch that went hard left. That’s perhaps an apt metaphor for coronavirus “cases data” and this year’s presidential politics.
Grady Means is a writer (GradyMeans.com) and former corporate strategy consultant. He served in the White House as a policy assistant to Vice President Nelson Rockefeller. He developed national health care and welfare programs at the U.S. Department of Health, Education and Welfare. His degrees are from Stanford, where he studied engineering and mathematics. Follow him on Twitter @gradymeans1.
Copyright 2023 Nexstar Media Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed. Regular the hill posts