Hacker News Reader

It could be scientifically dubious for reasons that are not related to the scientists' methods or integrity directly. For example if the funding source makes it less likely that they would have published a negative result.

Such factors are not evident by simply examining the study, but still allows exploiting science to promote a message.

That’s a valid and fair point, unless the study was registered ahead of time to protect against that.

This!

Funding can introduce biases independently from the intent of the scientist.

Let’s say scientist A assuming something that positively biases their research and scientist B assumes something that negatively biases their results. By choosing which person gets vastly more funding you can leverage those differences.

Of course that’s hardly the only influence funding sources can introduce, but I don’t want to suggest anyone specific did something wrong without evidence.

What rubs me wrong about the study is that they're inconsistent in their approach to dairy versus plant-based drinks.

Their sections on the health effects of dairy are very high level, focusing on meta-analyses and controlled studies. They go out of their way to emphasize how rare lactose intolerance and milk allergies are. However, when they get to talking about plant-based drinks, they're happy to reference individual cases of malnutrition as somehow meaningful, and make up for the absence of research by attacking the ingredients lists of some specific drinks.

In the end they acknowledge that the studies on plant-based drinks are lacking, but they give the two a far from even-handed treatment.

So, yes, the funding doesn't necessarily bias the results, but it seems to me that it has here.

Plant based “milks” are seriously lacking nutritionally compared to dairy. You don’t need a study for that, reading the back of the carton will do. Almond milk, especially, is flavored water. Still, I agree about the inconsistency you point out.

Almond milk is idiotic in many ways, especially considering how much water it takes to grow them.

Oat milk is a much better option.

> insulting to the scientist

Sounds very hermetic as a rejoinder. This way the cigarette industry could promote healthy smoking, because you know, scientists.

But you are right, lets have a closer look:

"Ian Givens is recipient of research grants from UK Biotechnology and Biological Sciences Research Council (BBSRC), UK Medical Research Council (MRC), Arla Foods UK, AAK-UK (both in kind), The Barham Benevolent Foundation, Volac UK, DSM Switzerland and Global Dairy Platform. He is a consultant for The Bio-competence Centre of Healthy Dairy Products, Tartu, Estonia, and in the recent past for The Dairy Council (London). Arne Astrup is recipient of research grants from Arla Foods, DK; Danish Dairy Research Foundation; Global Dairy Platform; Danish Agriculture & Food Council; GEIE European Milk Forum, France. He is member of advisory boards for Dutch Beer Knowledge Institute, NL; IKEA, SV; Lucozade Ribena Suntory Ltd, UK; McCain Foods Limited, USA; McDonald's, USA; Weight Watchers, USA. He is a consultant for Nestlé Research Center, Switzerland; Nongfu Spring Water, China. Astrup receives honoraria as Associate Editor of American Journal of Clinical Nutrition, and for membership of the Editorial Boards of Annals of Nutrition and Metabolism and Annual Review of Nutrition. He is recipient of travel expenses and/or modest honoraria (<$2,000) for lectures given at meetings supported by corporate sponsors. He received financial support from dairy organisations for attendance at the Eurofed Lipids Congress (2014) in France and the meeting of The Federation of European Nutrition Societies (2015) in Germany."

So basically they covered Givens and Astrup.

The very fact that scientific research has its ethics and is bound to be open to conflicts of interest should make everyone skeptical of the outcome: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2700754/

Regarding other studies: https://nutritionfacts.org/topics/milk/

Can you point to an actual error in the design of the study (please reference away, I want to know study flaws) and not just attack the scientist?

Study says it assesses the "totality of scientific evidence" but conveniently ignores all of the scientific evidence for the negative side effects of dairy. It is a very stupid study.

>conveniently ignores all of the scientific evidence

You should justify this assertion by highlighting the ignoring you claim occurs in the study. Here is a study with similar conclusions whose authors do not appear to have conflicts:

https://link.springer.com/article/10.1007/s00394-012-0418-1

The studies aren't doing the same thing. Yours looks at only the effects on obesity, cardiovascular, and metabolic disease. The OP's study purports to look at all health effects.

That study has a far narrower scope, and neither looked at all health impacts from milk consumption because the literature isn’t there.

“Milk and dairy products: good or bad for human health? An assessment of the totality of scientific evidence” did look at many health impacts and all cause mortality which is a proxy for heath. But it didn’t particularly get into why it was dismissing studies showing decreased lifespan associated with milk consumption beyond listing studies showing the reverse. By comparison it spent a while focusing on why the association with colorectal cancer may not be an issue.

That said, there isn’t a good way to measure all health impacts from specific foods because total health is so complicated. On top of this there are many confounding factors, presumably Truffles could be linked to health because consumption is linked to wealth. But nobody wants to publish “Milk: not extremely poisonous, more research needed.”

The meta analysis of dairy that I have seen suggest there are health benefits, especially for cultivated dairy (cheese, yogurt, etc). Do you know otherwise?

There is nothing wrong with consuming a moderate amount of dairy products.

However the justification for this must be that you like their taste, not for supposed health benefits.

Dairy contains a lot of nutrients and it is certainly preferable to many kinds of industrially-made food that are filled with dubious ingredients, colorants, flavors, sweeteners, oils and so on.

On the other hand, in order to claim "health benefits", it would have been necessary for dairy to be a superior source for the included nutrients, which it is not.

For proteins, a better source is lean meat, e.g. turkey breast or chicken breast, for fat much better sources are the vegetable fat sources, which allow the selection of an optimal fatty acid profile, for calcium a much better source is a supplement like calcium citrate powder, which allows the choice of any desired daily intake without disturbing the daily intakes for the other nutrients.

It is good to enjoy dairy, but one should not try to rationalize this preference as being motivated by health concerns.

It depends what you’d substitute into your diet instead. Meta studies I’ve seen show lower mortality for people consuming cheese, butter, yogurt, sour cream, etc. Which suggests to me that the average person chooses worse substitutions.

I don’t see why you’d include it in your diet if you don’t like it. But assuming you do, the evidence seems to suggest it won’t hurt you. It may even be better than common substitutes (like butter over margarine.)

I agree that I have also been quite surprised by their bold claim that their meta-analysis has included "the totality of available scientific evidence".

An example that this claim is false is that there is no mention about the studies that have identified a correlation between the galactose intake (which comes from the lactose contained in dairy) and eye cataracts.

From my anecdotal experience, because I have seen at least 4 cases of eye cataracts at people for whom dairy was a major part of their food, I believe that this association of galactose consumption with eye cataracts is true.

Fortunately, nowadays for those with enough money and with access to a modern ophthalmology clinic eye cataracts are no longer debilitating due to the surgical replacement of the lenses. Nevertheless, it cannot be considered as a minor health issue, because not everybody is so lucky.

I guess it depends on whether you believe in the funding effect.

If you do think funding biases the work, then insulting scientists abusing science to push products isn't too bad, right?

>Yes there’s a conflict of interest. No, it does not necessarily invalidate the result.

Yes, same way a mail from an Nigerian ex-prince wanting to get money out of the country and will compensate you from helping them, might still be legitimate. You never know.

But if we can't go and rerun the research ourselves, it's best to see the red flag and assume it's rubbish than accept it in the off chance "the dozens of conflicts of interests might not have influenced it".

Conflict of interest in studies is not invoked to mean the study IS invalid, it's invoked to say the mean the study is very likely invalid.

Probability matters here. How many Nigerian princes that reach out to random strangers are for real? 0 to a large number of significant digits. How many studies with conflicts of interest are invalid? I suspect less than half until you show me numbers to the contrary.

>How many studies with conflicts of interest are invalid? I suspect less than half

More studies are invalid than valid, so studies with conflicts of invalid would put it much further below 20% chance...