Tip - If you are using a phone, set the "Desktop Site" option in your browser   

2022-12-20

Simple messages that can be understood in a few moments are it seems to me the way forward.

Life isn't always complicated, and those who wish to obscure the simple truth are prone to embellishing it with unwanted complexity in the hope that the simple truth will get lost in the verbiage.

Here is some very simple truth (3 minutes) about our national energy policy, brought to us by GB News.


Here is another very simple truth about the Covid pandemic.

Imagine that you have been tasked to discover the reliability of a test for a disease - how would you go about it?

One way to calibrate the test would be to assemble a group of healthy people known to be free of the disease and to test these people.

Say you have 100 people and the test correctly identifies 98 of them as being negative, and two positive for the disease. Now, in principle, you know that the test is 98% accurate. You also know that 100% of the positive results are, in this exercise, wrong. Yes the real world is a bit more complex and you might want to run more investigations on the two positive test subjects, but let's assume you did that too and the test results were corroborated.

Now, if you apply this test at random in the general asymptomatic population, what result would you expect? Obviously at least 2% of the tests should, on average, be positive - but we would assume that most or all of these were the expected false positive results.

What does that tell us about the drive for "Zero Covid"?

What if you don't actually know the accuracy of the test in the first place because you didn't bother to calibrate it, and you got 2% positive results in the general asymptomatic population - say 2000 positives in 100000 tests - would you be justified in calling a pandemic?

What if you got 5000 positives in 100000 tests? Or 15000 positives?

You may simply have calibrated the test rather than tested the population.

In reality, to establish the validity of the test, you would also need to test the other contingent - those known to have the disease - to see how the test performs for them.

That's more tricky if the symptoms are not specific to the disease, but in the scale of things, we don't need to do that in order to establish that applying an uncalibrated test in the general asymptomatic population is a fools errand.

Still doubtful? Now factor in the difficulty of ensuring that (a) the test is stringently specified to ensure that all the testing labs are working to the same exact specification (b) all the staff at the labs and in the chain of custody of the test specimens are fully trained in both the test and procedures to avoid cross-contamination, errors in equipment calibration, etc etc (c) that all lab staff and procedures are rigorously monitored to ensure compliance with the standard procedures and specifications.

Would scaling up the operation on a population-wide basis likely introduce many more errors that could significantly invalidate the results?

In the early stages of the pandemic, all the news media were full of reports about the numbers of deaths and hospitalisations "due to Covid".

As soon as these dropped off in May / June 2020, all the reporting switched seamlessly to numbers of "cases" as tests were rolled out into the general population.

Make of that what you will.