Study: Drinking any amount of alcohol causes damage to the brain

“There is no such thing as a ‘safe’ level of drinking, with increased consumption of alcohol associated with poorer brain health… In an observational study, which has not yet been peer-reviewed, researchers from the University of Oxford studied the relationship between the self-reported alcohol intake of some 25,000 people in the UK, and their brain scans.” - CNN

Discussion

….to see what the paper looks like after peer review. It’s interesting, and ethically suspect, to release one’s results before this. It’s an interesting hypothesis—the loss of brain volume correlates to alcohol use—but thinking things through statistically, your sample size goes to infinity as the hypothetical stimulus goes to zero. So you cannot prove that there is “no such thing as a safe level”, as it’s statistically impossible. You might prove “even one drink per day is correlated with brain tissue loss and likely cognitive decline”, but not “no safe level”.

There is also a problem with saying that there is “no safe level” when they haven’t actually correlated alcohol use with any actual disease. What’s meant is “it appears to correlate with loss of brain tissue, which we believe correlates with cognitive decline and other issues”—but that’s not exactly proof of disease.

Which might be part of why they’re releasing this study before peer review. They’re likely going to need to tone down the language and claims if and when the study actually comes out.

Aspiring to be a stick in the mud.

[Bert Perry]

….to see what the paper looks like after peer review. It’s interesting, and ethically suspect, to release one’s results before this.

In physics we have what’s called the ArXiv server that holds hundreds of thousands of pre-print papers so that we can all see what each other is doing. Most never get peer-reviewed and published. Is that unethical?

Does the ArXiv server get released to CNN and Fox? No doubt that there is sharing of ideas before peer review, and that’s fine. Releasing the ostensible results to the press before peer review is what I view as borderline unethical.

Aspiring to be a stick in the mud.

The peer review process of “science” is not actually science. “Peer review” is when a journal hires a few people to read a paper they are thinking of publishing and saying what they think. If these handful of people like it, it gets published. If not, your dreams of a publication are squashed. This is not some rigorous review of thousands of scientists carefully combing through your data to ensure the sanctity of science. It is a few guys making a few extra bucks.

One example: years ago my wife was trying to get her MS degree thesis published in a journal. It was a review of a new SAS function (SAS is a high-end certified statistics package) to show that it performed accurately. Any way, she edited her dissertation to be a journal article and sent it off to the publisher. It got chosen for peer-review. Later it got rejected with horrid remarks about how bad the procedures were that she followed, etc. It was a real stinger. Ironically, that month the journal published an almost identical article written with nearly the same processes and results. It turned out the “peer reviewer” of my wife’s article was the graduate advisor of the guy who got his paper published. In other words, he tore my wife’s article down to make sure his student’s got published.

Short version of the story? “Peer-review” mean very little.

[Bert Perry]

Does the ArXiv server get released to CNN and Fox? No doubt that there is sharing of ideas before peer review, and that’s fine. Releasing the ostensible results to the press before peer review is what I view as borderline unethical.

Actually, yes. Many of the articles in the media about new space finds or new theoretical discoveries are from pre-prints…

….that peer review isn’t perfect, but I still see a big difference between what’s done in this study and what you’re describing in the physics/astronomy community. In the latter, you’re making things public so that others can see it, comment on it—it’s really a form of peer review. In the former, you’re skipping that step while creating the impression that the research is indeed authoritative, and in this particular case, it’s a conclusion that one cannot prove statistically, nor does it necessarily follow medically even if you reformatted the hypothesis.

In other words, it’s a study that will take it on the chin in any honest, serious process of peer review. Or at least should. Again, granted that peer review can sometimes be full of political games, but it’s supposed to catch obvious “oopsies” like I’ve mentioned above.

You’re also going to have some serious looks at the size of the effect found, and what confounding variables there might be there. Reading here, it would seem that you’ve got ~15% standard deviation for some of these measurements, so for an effect of < 1%, your sample size—and the scrutiny of coufounding factors—is going to be a huge deal.

Aspiring to be a stick in the mud.

[Mark_Smith]

The peer review process of “science” is not actually science. “Peer review” is when a journal hires a few people to read a paper they are thinking of publishing and saying what they think.

You are way off base here. The peer review process is critical to the scientific process, notwithstanding your wife’s experience. Non-peer reviewed science should always be suspect at a certain level. Not that it may or may not be good science. Peer review is not without flaws. But independent research that has not been reviewed by any other person, does not have the same standing as a well documented and well peer reviewed research. This is how science establishes itself throughout the scientific process. You begin to learn this even in High School where you have a lab partner who works along side of you. I never could publish a result in informally and internally when I was a scientist without a fellow scientist signing off on my lab notebook. This is R&D 101. I have been published and have had my stuff peer reviewed. And there is no doubt sloppy stuff out there. Mostly from poor journals with poor editorial boards or incompetent scientists, but most of us know which of those to stay away from.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1084042/

[dgszweda]
Mark_Smith wrote:

The peer review process of “science” is not actually science. “Peer review” is when a journal hires a few people to read a paper they are thinking of publishing and saying what they think.

You are way off base here. The peer review process is critical to the scientific process, notwithstanding your wife’s experience. Non-peer reviewed science should always be suspect at a certain level. Not that it may or may not be good science. Peer review is not without flaws. But independent research that has not been reviewed by any other person, does not have the same standing as a well documented and well peer reviewed research. This is how science establishes itself throughout the scientific process. You begin to learn this even in High School where you have a lab partner who works along side of you. I never could publish a result in informally and internally when I was a scientist without a fellow scientist signing off on my lab notebook. This is R&D 101. I have been published and have had my stuff peer reviewed. And there is no doubt sloppy stuff out there. Mostly from poor journals with poor editorial boards or incompetent scientists, but most of us know which of those to stay away from.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1084042/

You overreacted to what I wrote.

Calculated risks:

  • Climb a mountain
  • Explore a cave
  • Snow ski
  • Water ski
  • Scuba diving
  • Buy Bitcoin
  • Invest in a stock
  • Drive a motorcycle
  • See how fast car can go (100 mph plus)
  • Ice skating (I witnessed a terrible accident on ice - head injury)
  • Drive on frozen lake
  • Have a glass of wine