The graph shows the change in standardised mortality rates (little adjustments for the structure of the population) for England & Wales, which have some of the most closely analysed mortality in the world.
> the later death spike may have been infant mortality.
Infant mortality occurs mostly within the first few years of life, and you reach the point of lowest mortality at about age 10. You then start to accelerate up quickly as testosterone kicks in!
> you wouldn't have thought it would happen in both places at the same time?
Remember the major population centres in both countries were huge maritime cities with a roaring intercontinental passenger traffic that covered the Atlantic in a week, just enough time for the whole boat to become infected.
> They based their graphs on a previous five year average, I can guarantee that showed what they wanted to show.
These graphs are all derived from analyses done by the Continuous Mortality Investigation of the IFOA (Institute & Faculty of Actuaries), and have been released weekly since March. This one is newsworthy as it covers 2020 to year end. I imagine five years is enough to show the natural variation, but not so many that it makes it impossible to read.
Mortality improvements (changes in mortality rates, as opposed to the rates themselves) are slippery things to estimate. They require an order of magnitude more data, and as you can see from the graph even at the best of times are extremely volatile. They are usually the sum of all the different things that are operating on causes of death in the population: on the one hand, better cancer treatments, medicines, and so on; on the other, continually worsening lifestyle factors like obesity and alcohol.
