Correlation does not equal causation

Fri, Jul 9th 2021, 07:33 AM

In an earlier article I made the statement, “Lies, damn lies and statistics”. This was in response to persons giving numbers without any sensible statistical analysis. This habit of throwing numbers around and drawing non-sensical conclusions seems to be endemic in our world today.

The gist of the July 7, 2021 headlines of both the Tribune and The Nassau Guardian stated that the increase in both hospital admissions and death was higher in unvaccinated COVID patients.

Taking the raw number of COVID hospital admissions and deaths in the vaccinated versus the unvaccinated, then making a statement as to morbidity and mortality is laughable at best and dangerous at worse.

For example, suppose 10 persons (eight vaccinated and two unvaccinated) go to a bar and have a few drinks, then the eight vaccinated persons become involved in car accidents and the two unvaccinated persons do not.

One could make the erroneous causation statement that vaccinated persons have a higher rate of vehicular accidents than unvaccinated persons.

On the surface this looks fine. However, vaccination status probably had nothing to do with the accidents. That is, vaccination could be seen as correlated to the accidents, but actually had absolutely nothing to do with causing the accidents.

The same statistical mistakes may be – note I say may be – being made with regard to COVID hospital admissions and deaths.

What we do know, with regard to COVID infections (and all respiratory infections) is that coexisting diseases such as obesity, hypertension, diabetes mellitus and others definitely cause an increase in morbidity and mortality in COVID hospital admissions and death.

Click here to read more at The Nassau Guardian

 Sponsored Ads