I've made the mistake many times myself. I've learned that in research it's critical to structure your question to be appropriate, and also to design your study to be according to gold standard if it's a study, or to focus on the quality of your experiment because that determines the quality of the result. It's fun to think about some questions but if it's not falsifiable then we cannot take as much from the result.
A lot of studies in medical research that I read aren't well designed and seem to be commercial studies paid for by supplement or other industries. In some cases for example there will be studies on mice and then the media will take that result and act as if it applies to humans. Other times the study isn't up to the gold standard of sufficiently randomized control trial, with sufficient sample sizes. And also studies are almost never reproduced because there is no money in reproducing a study.
Do you think medicine would be more effective as a field if falsifiability were strictly adhered to in practice? I think medical studies can produce useful results at times even with the problems you mention but I wonder if by solving the problems if it would slow progress?
The dilemma really seems to be related to the increasing commercial entanglement of medicine as you said. It is a trade-off between speed of research in discovering ever so more complex, specialized and effective treatments (which is highly desirable) and scientific integrity. One saving grace though is the ability to aggregate several studies on the same topic into a meta-study which may assess and improve the significance of results. In important trials this is done very frequently.
Judging from other fields of science I would also not be surprised if some journals did not uphold their standards of due diligence as much as one might expect. There is a lot of money and influence involved in scientific publication and one might accuse some publishing companies to prioritize commercial interest over scientific.