(This article was originally published at Statistical Modeling, Causal Inference, and Social Science, and syndicated at StatsBlogs.)
Paul Alper writes:
You recall the University of Maryland chocolate milk cure for concussion [Bigmilk Strikes Again].
A new version of the same sloppiness is discussed here.
Alper is linking to a news article, “University of Iowa ignores questions about its oregano ‘cure’ for cancer-wasting syndrome,” by Eric Holland, who writes:
At the beginning of 2016, HealthNewsReview.org contributor Andrew Holtz and I spent months chasing down the facts behind a too-good-to-be-true news release about the benefits of chocolate milk for recovery from concussions.
Our reporting led to scrutiny from national news outlets and sparked a months-long internal investigation by the University of Maryland. That investigation uncovered dozens of problems with how the university conducts its research and how it communicates study findings to the public. . . .
But now it seems those lessons haven’t been universally heeded. Last week we came across a news release from the University of Iowa that made an unusually spicy claim: researchers had discovered compounds derived from oregano and thyme that are a possible “cure” for cachexia, the wasting disease affecting patients with cancer and other illnesses.
Our review of that news release shows researchers have done nothing of the sort. . . .
Holland continues:
In this case we can’t even tell what kind of study was performed. The lead researcher, quoted in the release, states that the “discovery was a serendipitous finding,” which suggests it was an observation rather than a true research finding based on an experimental design.
Another similarity with Maryland: Iowa’s release was linked to the licensing of the institution’s intellectual property to a commercial firm, in this case a pharmaceutical company planning on making an over-the-counter drug to combat cachexia. This raises concerns over conflicts of interest on the part of the researchers that weren’t addressed in the news release.
There are differences, of course, between the Maryland and Iowa cases. Maryland’s research involved children and had problems involving adequate informed consent. Iowa’s research involved mice, rather than humans, but failed to show any effects that applied to patient treatments. No true clinical trials were done in either case.
But the similarities are striking – a final one being initial hesitance to engage the news media in any meaningful way.
For patients with cachexia, and their loved ones, this is serious business. Hopes may be lifted and then dashed when what they believe to be a solution proves to be wishful thinking.
Public research universities should not be involved in this kind of exploitation. And they should have policies that ensure promotional messages are accurate, balanced, and complete.
To its credit, the University of Maryland took a hard look in the mirror when confronted with evidence that it was not not meeting this standard [policies that ensure promotional messages are accurate, balanced, and complete].
The University of Iowa still has time to learn from this example. Its reputation as a leading research institution may well depend on it.
Is that really true, that the university’s reputation will depend on its good behavior?
I fear not. After all, Cornell University went all-in on that discredited food researcher in their business school, and they’re also famous for that ESP paper published by one of their psychology professors. And Princeton has the editor of the notorious papers on himmicanes, air rage, and ages ending in 9, but that university’s reputation seems to be just fine.
Long term, sure, enough bad press and your reputation will tank. This seems to be happening to some extent with the Lancet, after publishing a series of notorious papers over the period of more than a decade, and also to PNAS, which seemingly can’t resist publishing crappy clickbait.
But universities seem to be able to withstand bad news. Actually, what seems to be happening is that universities and higher education and science more generally are getting discredited by the Marc Hausers and Anil Pottis of the world, even while the particular institutions do just fine. Seems kind of unfair, also frustrating. The antics of a Brian Wansink may well harm the reputation of science as a whole more than it will damage Cornell University—which leaves all of us on the outside screaming in frustration, while within Cornell, they’re like, Yeah, that Wansink guy gets on TV a lot. Cool, let’s give him another raise!
Not that I can talk, being at Columbia, home of Dr. Oz and that plagiarizing history professor.
The post Big Oregano strikes again appeared first on Statistical Modeling, Causal Inference, and Social Science.
Please comment on the article here: Statistical Modeling, Causal Inference, and Social Science
