When you become a world expert in a very narrow area of a field - like when you do a PhD - I found you discover that around 50% of papers are either pointless, misleading or wrong.
Let me be clear - the largest proportion of those are the pointless ones ( the reader already knew what's in the paper ).
Some of this is because research is hard, a lot of it is because of the immense pressure to publish and the strong bias for positive messages ( this approach is better, we discovered X ) to be published, rather than stuff that says - we tried this but it didn't work etc.
> I hope this is not as bad in medicine and health related research.
Same pressure, same human factors.
There is of course an additional factor in cases like this - if your research hints at a link between talc and cancer - is it ethical not to publish while you wait another 5 years for a longer study?