Even reputable journals carry grossly exaggerated results

Early evidence is a particular problem
the scream

Grossly exaggerated results have been found in more than one in three early clinical trials, according to an analysis of almost 1000 randomised controlled trials (RCTs) from reputable journals.

The findings, published in Mayo Clinic Proceedings, suggest decision makers should treat early evidence with caution.