You are right that correlation is a good starting point for investigation, but it is a terrible point to stop and publish a finding publicly.
100% of people who ate broccoli between 1500 and 1850 died, therefore broccoli kills you?
There are better forms of correlation that are worth following up on (people with this…