Full access to raw data becomes more and more necessary.
The number of scientists which manipulate data increases tremendously, one solution could be the obligation to make raw data accessible on the internet. A second solution is to make mandatory the publication of all the randomized controlled trials that have been registered even if they conclude to less benefit or more harm for the studied innovative procedure. A third solution is to be very circumspect if the authors have already patented their own research topics. For example prostate cancer screening and lung cancer screening are two field of highly interesting the biomedical industry and associated with large amount of potential revenue.
The latest example is a Prostate cancer screening marker research which appeared to be entirely fake. Moreover the scientist involved had already sold the discovery to a bio engineering start-up. All that stuff ended up before a court, you can see a report of the lawsuit here. The lung cancer screening has become also a competition for X-ray industrials and searchers as you can see here.
When are at stake the academic glory, the economic greed and some time no more than just finding a job for a post doc searcher the pressure to publish always surpasses the sense of honour and the Hippocratic oath?
The BMJ journal has just been recently whistle-blowing the problem in a in-depth review of the question. Bogus evidences are not only dangerous for patients but also lead to unjustified health care spending deepening the nation debt crisis.
More information on the topic is findable below: