What happened with the "undereat/calorie restriction for longevity" that has played again and again in the last 2 decades?

With the rather important caveat that the data came exclusively from people actively receiving scans due to presenting with already ongoing health concerns and also that the subjects were already primarily elderly.

This seems like a case study in how to demonstrate that correlation has no discernible relationship whatsoever to causation.