For drug companies this is actually extremely bad. Amgen and Bayer have spent billions trying to replicate drugs that just don't work.
For a long time I just blamed the pharmaceutical companies for why so many "cures" reported in research never made it to production. I thought they were just out for a profit and had no interest in cures. However based on some recent experience this is definitely not true for all of these companies. Some of them have tried very hard to bring these cures to market and spent billions of dollars trying to do it. The problem is that it turns out nearly all of these "cures" are just false. They can not be replicated to any degree.
This kind of behavior by scientists is also driving up the costs of healthcare. These companies continue to try to bring promising research to market but the success is so low that it drives up the costs of the drugs they do manage to make.
I have even worked with some researchers that just had no idea about statistical methods. They had never had any education on the subject. As a result their conclusions were worth as much as the paper they printed it on. The problem is that at least some of them really thought their data was good and showed what they wanted to show. Far too many scientists do NOT have the statistical training required of engineers and while it would help the field a lot I doubt they will do it because that level of math would cut out far too many people for the schools to allow that.
I don't really see a way to fix this unfortunately. If you change the incentive system to reward checking other peoples results you will get a bunch of simple and obvious papers that check something else. If there is a solution to this I don't really know what it looks like but we need to find it because this is seriously hurting the rest of the society and it makes it very hard for the actual legitimate scientists to spread real research that helps us all.