It Doesn't Work Like That?

The UK PM was probably influenced by an 'anti-lockdown paper' from Oxford's Centre for Evidence-Based Medicine, when making his now notorious WhatsApp quip. He said "I must say I have been slightly rocked by some of the data on Covid fatalities. The median age is 82-81 for men and 85 for women. That is above life expectancy. So get Covid and live longer" (https://www.theguardian.com/politics/2021/jul/20/boris-johnsons-quip-get-covid-live-longer-an-echo-of-oxford-study). David Spiegelhalter and Anthony Masters (respectively from the Winton Centre for Risk, Cambridge and the Royal Statistical Society) lead the Classics graduate gently through the genesis of the basic error (https://www.theguardian.com/theobserver/commentisfree/2021/jul/24/get-covid-and-live-longer-no-it-doesnt-work-like-that). Spigelhalter and Masters reveal that 'Boris' confuses 2 different kinds of average (a median and a mean). The median age at death from Covid measure, is the middle-ranked value for the range. The life expectancy value is a mean value, obtained by adding together all the ages at death for a large population and dividing that total by the number of people in the sample. Life expectancy actually improves as an individual ages. They have, in effect, avoided earlier causes of death. For example, once a man reaches 79, his life expectancy, in the UK, currently rises to about 88 (something to look forward to?). What the figures actually show is that the risk factors for dying from Covid are remarkably similar to those of dying from something else. The virus consequently acts as a multiplier for other lethal risks. Every Covid death actually results in a mean average loss of 10 years of life. As Spigelhalter and Masters note, the people who died of Covid were certainly not at 'death's door' when they died. Presumably, they don't teach the difference between medians and means to Classics students? The difference should, however, be obvious to anyone purporting to speak for a 'Centre for Evidence-Based Medicine'. It seems to me that people have been playing fast and loose with the numbers.