2023-01-21
"2022 Was a Normal Year for Mortality in England and Wales"
or
"Lies, Damned Lies, and Statistics"?
Whoa!
With "excess deaths" allegedly running at circa 1000 per week above the expected rate, how can this be?
And what is "Age-standardised mortality" anyway?
"Definition:
The age-standardized mortality rate is a weighted average of the age-specific mortality rates per 100 000 persons, where the weights are the proportions of persons in the corresponding age groups of the WHO standard population"
Whilst this may be a useful concept for the WHO when comparing how one nation is faring with another nation with respect to deaths from a particular cause across the various age groups, I suggest that using this concept to compare deaths in successive years within a single population is simply inappropriate. The WHO "standard population" of itself has nothing to tell us about the population of England and Wales in successive years.
"Source: Office for National Statistics - Monthly mortality analysis ...
- Age-standardised mortality rates per 100,000 people, standardised to the 2013 European Standard Population"
My comment would be that whilst the "2013 European Standard Population" might be better (or worse) than the "WHO standard Population" it still doesn't tell us anything about England and Wales, and is equally inappropriate.
To say that "older people tend to die more than younger people" is a truism and we correct for that by working in deaths per 100,000 within age group, and comparing age group by age group separately across the years.
But we still need a way to work out the total summation across age groups - if 100 per 100,000 of the 40-49s die, how do we add that to 87 per 100,000 of the 30-39s dying?
If we really want to count the real life cost of deaths over different age groups over successive years then the the proper conversion would be to multiply the deaths in each age group by the number of lost years (approximated by counting the years up to our expected life-span).
So if our expected lifespan is 85 years, and 100 people per 100,000 die in age group 40 to 49 years, then we would count life lost as (approx) 100 x (85 - 45) = 4000 years per 100,000 people in that age group. If there are 1M people in the 40-49 group then that equates to 40,000 life-years lost and we can simply add all these lost life-years across all age groups together to get the total life-years lost of that population in that year.
If you want to be really persnickety you could then scale the result for each year to get a figure of lost life-years per 100,000 of total population, to account for a rising or falling population overall.
I'm not convinced that anybody has done that yet.
Until somebody can come up with these calculations, in my view we don't have a valid measure by which we can compare "excess life lost" across age groups over consecutive years within a single population, and the argument is therefore largely irrelevant misleading and futile.