What if we routinely make health recommendations which do not affect outcomes?

Interesting, if true and I have no reason to doubt its truthfulness.  From (Influenza) virus gonna virus by Alex Berenson.  The subtitle is "We increased flu shots over 15-fold between 1980 and 2020. You’ll never BELIEVE what happened next!"

The basic message is that we had, as a nation, a huge increase in the number of people being vaccinated for flu each year but no change in flu cases.  

Here are the annual deaths from flu.

Click to enlarge.  

Of course we need to look at the rate, not the raw number of deaths.  The US population was 180 million in 1960, 230 million in 1980 and it is 330 million in 2020.  

That still leaves us with a 15 fold increase in flu shots since 1980 and only an 83% increase in population.  83% increase in population versus a 1,500% increase in flu shots with no discernible impact in the absolute number of deaths.  

If the percent of the population getting flu shots increases dramatically but the death rate declines only a small amount, that's a pretty weak correlation.  Perhaps, ball-parking it, possibly about a 6% reduction in the death rate from flu given a massive increase in flu shots.

We don't have the data in this article to do a real estimation of efficacy but it has widely been known for a long time that it is very difficult to anticipate flu variants season-to-season and that the vaccines are low efficacy.  The risk return ratio for the elderly and otherwise morbidly afflicted is probably reasonably positive even with a low efficacy rate because the risks are so low.  For everyone else?  Not necessarily a solid equation.

Which reflects a longstanding weakness of the CDC, a refusal to measure costs and benefits in a rigorous fashion which would allow people to make informed choices based on age and health.  There's just the blanket admonition to get vaccinated.  

Given the comprehensive public health fiasco around Covid-19, it makes sense to revisit flu vaccines.  They are more traditional vaccines, they are more familiar.  But we have been as cavalier about cost risk ratios there as we were during Covid-19.

Hopefully the CDC learns but there are reasons to believe that institutional ass-covering might prevent that from happening.  

UPDATE:

And lickety-split, Berenson has an update addressing the issue of rates of infection.  The initial interpretation of the implications of the original data remain true when more complete data is provided.