Don't believe the hype: When that DATA seems just too good

Life lessons from medical science

Publications and careers

The initial victims are often unwitting researchers who submit their work to such serious sounding journals with the expectation they are publishing in the peer-reviewed literature. Many of them are from low and middle income countries where research funds are extremely tight – the lower charges levied by these journals are attractive in comparison to the large fees charged by the real thing.

And, in an environment when career advance is often linked to publication record, having a bunch of nice sounding publications on a CV does little harm so long as nobody digs beneath the surface.

Thanks to people like Jeffrey Beall, scientists are beginning to wise up. But there’s something of an arms race going on here. Some of the larger publishers are becoming more sophisticated. A standard quality metric for a journal is the impact factor — think of this as the origin of Google’s page rank and you’ll not be far off the truth.

The publishers have responded by creating their own rating agency. Or else they have bought up struggling “real” journals as cover.

So much for the publishing side of life, but what about when the data and claims that are published by journals are open to question? Let’s look at GcMAF.

GcMAF (Globulin component Macrophage Activating Factor) is a blood product that has been promoted by various companies and websites as, among other things, a cure for cancer, HIV and autism.

It is possible for people to make amazing claims about medicines based on what looks like solid, clinical data.

For example, in the case of treatments for cancer, somebody might claim reductions in tumour growth based on unvalidated proxy measures (for example reductions in a protein called nagalese) rather than direct confirmation using CT or MRI imaging; publishing individual patient cases where there is no biopsy-confirmed disease, or not giving precise information on the disease (e.g. disease sub-type and staging before and after treatment etc); others might claim there has been a clinical trial when there has been no formal trial and no official records to confirm that patients have actually been treated.

Image manipulation software could also be put to good use to generate convincing looking Western blots and other laboratory data. But the difference is that the readers of legitimate journals subject the published papers to a much higher degree of scrutiny.