In our survey, the most common reasons for not supported claims were false statements and exaggeration of efficacy.
Biased statements included disease mongering, speculations regarding surrogate markers stated as facts, lack of acknowledgment of RIO design flaws, and exaggerated statements of rimonabant's efficacy and safety.
Press release statements matched 93 of the 687 statements (14%), including 16% (30/191) of efficacy and 15% (22/150) of safety statements.
Likewise, 9 summaries also mentioned the following type of erroneous statement: "The efficacy of stimulant agents confirms that the neurotransmitter abnormalities seen in ADHD are primarily catecholaminergic in origin".
When I asked for scientific evidence to back up Gyngell's claims, she made the remarkable statement that her "personal" evidence of efficacy for recovery programs was as valid as the scientific studies I was quoting.
A claim was rated as not being supported by the cited study, if one of the following criteria applied: false statement, absence of relation, exaggeration of efficacy, unjustified generalization or unjustified transfer to humans.
To make further statements regarding efficacy, the analysis of a larger number of patients is required.
This, of course, is not a statement about efficacy of those programs – based on the 2013 data submitted to the rankings team, Y Combinator would still have topped the list if they had not transitioned models.
Given the known limitations of standards of care and guideline statements, their efficacy in preventing harm should also be empirically tested.
Reasons for not supported claims were false statements in 4 claims (24%), exaggerations of efficacy in 4 claims (24%), unjustified generalizations in 3 claims (19%), absence of relation in 3 claims (19%), and unjustified transfer to humans in 2 (13%) claims.
