Monday, 8 February 2016

The walls are closing in on incompetent and irresponsible researchers

No longer will irresponsible researchers who selectively present data be protected by their friends in high places under new proposals by the International Committee of Medical Journal Editors. (ICMJE)
The BMJ has deplored the failure of regulators to take any action on selective data presentation, and is appalled by the use of defamation laws to inhibit valid scientific discussion of research findings.
I have previously referred to the comments of Sessler and Retractionwatch that data "massaging" is a widespread and serious problem in medical research, much more serious than actual data fabrication.

In brief, the ICMJE is proposing to not only make de-identified raw data available within 6 months of publication, but also require that authors include a plan for data sharing as a component of clinical trial registration.

Journal editors plan to withhold publication of those researchers who might otherwise resort to bureaucratic obfuscation and legal thuggery to avoid public scrutiny of their work.

It can't happen soon enough.

Don't believe that data massaging is a problem in health research relevant to T1 diabetes? Look at how the abysmal research of Keys has pervaded decades of medical "thinking" in Australia. How many articles have quoted him, or quoted articles that have quoted him (and so on) to support the low cholesterol diet? More on that later.

Additional thought. Question really. Are these researchers deliberately setting out to mislead, or are they just too under-educated to realise the scientific implications of selective data presentation?
Answer: It matters not. The research needs to be withdrawn and the researcher re-educated either way. Ask any lawyer. The requirement to prove intent is exceedingly difficult to meet, and will result in no finding in many cases.

(ICMJE pic to go here.)