Herd immunity, bad data and critical thinking

August 14, 2020. Let's start with an excellent example of Covid lies exposed. Click and watch the video. As you watch, remember to stick to the facts, not to the narrative. Set aside your politics and your feelings. Think critically. As Twain noted, the problem is not what you don't know, it's what you 'know' that 'just ain't so' that will plunge you into trouble.

That noted, here are some specific tips on understanding the jumble.

Start by avoiding articles or columns that start with terms like 'surge,' 'spike,' 'tick' and other such completely meaningless click bait terms. And when anyone cites a mean and makes claims as to a change from day to day or week to week, see if the 'journalist' says anything about standard error or confidence interval.

The Covid tests themselves (over 61 million so far in the United States) have often been shown to be unreliable (i.e., untrustworthy, inaccurate), as are many of the labs that provide the test results, and the agencies who 'post' the data. A friend received a positive test result in the mail from a lab where he had registered for a test but left after a long wait and never took the test.

Beware of policies based on 'expert' opinion or a letter signed by a hundred of anybody, i.e., lawyers, doctors, epidemiologists, climate 'scientists. Question the bias, source, reliability and validity of any claims. Don't let journalists conflate correlation with causation, or compare apples to oranges. For example, comparing deaths across countries is useless without a 'deaths per 100,000' qualifier. Same for 'confirmed cases.' Comparing the number of positive tests in the United States with Belgium's count is ridiculous, but widespread testing in the United States has been used to create a misleading tale of woe. Data for the United States need to be parsed out by geographic region, state, county or city, and comparisons between relatively equal populations are more informative than between non-comparable countries. Discount raw counts of 'confirmed cases,' especially when there is no mention of how many people were tested and tested negative.

Pay attention to what matters, especially hospitalizations and Covid-exclusive deaths. Almost all deaths attributed to Covid have an alternative possible explanation as to cause of death; 45% to 60% of 'Covid' deaths occurred in nursing homes or assisted living facilities, where life expectancy is under twelve months.

Absent a truly effective vaccine and universal, accurate Covid testing, the best hope for humanity is herd immunity, where a significant percentage of the population has been vaccinated OR contracted the virus and survived, often without ever demonstrating any symptoms. These non-symptomatic cases might include up to eighty percent of those who contract the virus. And even for those who have symptoms, a relatively small percent of them require hospitalization (maybe 20%; other stay at home, some of them quarantined), and a small percent of that small percent actually die with Covid as their exclusive cause of death (less than three percent of those confirmed positive if you are under the age of 65). See the accompanying graphic for herd immunity progress across the fifty states.

From the NIH: People <65 years old had 30- to 100-fold lower risk of COVID-19 death than those ≥65 years old in 11 European countries and Canada, 16- to 52-fold lower risk in US locations, and less than 10-fold in India and Mexico.

The fraud of science literature. Finally, don't trust even 'peer reviewed by experts' articles that lack additional independent confirmation. Journals are notorious for bias, incestuous professional relationships between the authors and editors, an absence of truly blind reviews, and an unwillingness to publish disconfirmed hypotheses. Here is what a leading editor has this to say about his own editorial profession when it comes to practicing good science:

Politics play a larger part in Covid reporting, especially in reporting bad news, than do the data themselves. There are scant comparisons to death rates in nursing homes without Covid, or of deaths from influenza. And national opinion sours as a result: A recent national survey found that Americans estimated that the deaths from Covid totaled nine percent of the population, or 30 million Americans dead from Covid, or 200 times the actual deaths at that time. When all the data are in, I'd like to some 'expert' test this hypothesis: The 'Covid' death rate is positively correlated with the level of 'progressively' (i.e., socialist or liberal politics) of the state and city. This means we should expect to see death rates higher in 'progressive' cities such as New York, Seattle, Minneapolis, Madison, San Francisco, and Detroit than in more conservative, red states and cities.

To be clear, the situation remains unclear, uncertain. Even claims about Covid myths will be controversial, and some of those myths may later be confirmed as true. Chloroquine is just such a 'myth,' as the claims and evidence go back and forth. The political rhetoric of both parties' conventions will do little to clear the picture. As noted by the Lancet editor, even the supposedly 'scientific' journals are not free from bias or unreliability.