I think this chart's main value is in showing the change. Reporting methods on covid deaths may be flawed, but if they are it's not unlikely those for flu and pneumonia are similarly flawed.Would be interesting if we could split out the different diseases and understand if the reporting methodology has changed in any way.
I am reminded of a chart I saw recently that someone was trying to show supporting the fact that shoplifting in San Fran is on the downturn because the number of thefts is declining - until people dug in to find that it is declining because the number of thefts being reported was changed and cops aren't accepting the reported thefts below a certain threshold.