When you complain that circumstantial situations can effect the results of a study, this is why anecdotal evidence is not useful and not why actual studies aren't useful. This is the very reason scientist try to include as many subjects as possible - to average out or otherwise eliminate other possible factors. Furthermore no scientists takes one study and says "done." Instead they expect other studies to try and reproduce the results.
Unfortunately the problem is media reporting that take a single study and make a catchy click-baity headline from it. This is why you don't hear doctors stating that any of the items you listed cause cancer, because the results have not been consistently reproduced. What you're hearing are pop-sci hacks.