If you don't look at the study, how will you know if they accounted for that or not? There are techniques available to control for correlations like that.

> There are techniques available to control for correlations like that.

If we knew all the correlations, we might stand a pretty good chance of controlling for them, but we don't know them. We might hypothesize that drinking 1% milk is associated with higher rates of exercise, and we might go out and measure that against a control group. But what if it's also associated with drinking less alcohol? Or not smoking? Or anything else.

We don't know all of the correlations, because human beings are complex, and we can't hold all other variables static, which is why nutrition science has been so abysmal. Only if we first hypothesize the correlation, or collect the data necessary to make the connection, can we even attempt to control for it.

> Only if we first hypothesize the correlation, or collect the data necessary to make the connection, can we even attempt to control for it.

If only we had some way to know which confounding factors the original authors accounted for. Oh wait, that's what RTFAing achieves.

That's not really the point. They attempted to control for demographics, and for dietary covariates like protein, total dietary fat, fiber, and saturated fat. But that is a tiny sliver of the possibly correlated factors.

Did they control for the possibility that 1% milk drinkers learned to drink 1% milk at a young age, and that population happens to live in areas further away from sources of pollution, even in comparison with other households in the same income quintile? Did they control for blood lead levels in childhood?

The point is that we don't know what the covariates are. We have no way to even approach the problem.

I'm with GP. I'd read the study if they did an experiment - randomly assigned milk to people and watched the impact - but there's too little value in these purely observational studies from nutritionists. The field hasn't grown up yet.

> There were a dozen covariates controlled in the present study (age, gender, race, household size, smoking, body mass index, MET-minutes of total physical activity, alcohol use, grams of protein consumed per kilogram of body weight, percentage of energy derived from dietary fat, grams of dietary fiber consumed per 1000 kilocalories, and percentage of total energy derived from saturated fat.

How would one handle that many factors? I am very curious what the techniques for doing so would be.

I'm no expert but my initial reaction would be that you need more than 5834 adults in a study that controls for all of those factors. We don't really know how household size, for example, would affect the results, so wouldn't you need a bunch of test data to analyze all of the combinations? Or did they only study people who are the same in most of them? (i.e. 45, male, with a household size of 4, nonsmoking, etc...)

Would love to hear from someone more informed on this than myself.

It depends on the effect sizes, but in most contexts 5k participants is considered a very sufficient sample. You don’t need millions of participants to study things like this.

The actual paper seems inconsistent and contradictory. For example, from Section 4:

> Fourth, adjusting for differences in demographic, lifestyle, and dietary covariates has little influence on the relationship between milk fat intake and biological aging. Fifth, there is no relationship between milk fat intake and telomere length among adults who consumed low levels of total saturated fat (tertile 1), but it is strong among adults who consumed moderate or high levels of total saturated fat.

WTF? Am I misunderstanding something?

Source: https://new.hindawi.com/journals/omcl/2019/1574021/

I can’t speak for this specific study, but in general that sounds like a statistical interaction, which is very common in statistical modeling.

There are of course other confounding variables that this study likely does not take into account (whether because they can’t measure it, or they chose not to include it)

True, the article at least doesn't mention if milk fat consumption correlated with other health factors. It mentions that the study considered frequency of consumption, but doesn't say if that affected results. Like much science reporting, it makes more questions than it answers.

In my opinion, drinking nonfat milk is a strong indicator that you're willing to follow dietary advice even if it's unpleasant. I can't stand the stuff.

Same here. Low-fat/nonfat milk tastes of nothing. I'd rather drink water. I found it works best to hydrate with actual water (ie. drink enough water to never feel thirst) and then if I want the taste, I'll just have a glass of full fat milk.

> I found it works best to hydrate with actual water

Water hydrates the best.

Milk (even skim) has proteins. So its not just hydration.

Or it's a symptom of what you were given as a child. I find 1% bearable, but don't like it. Cannot stand 2% or whole milk.

Is it a meta-analysis? The spin put on meta-analysis "studies" is one of the worst new trends.

As soon as any behavior is widely reported as healthy health conscious people will start doing it and if there wasn't a correlation before there will be now.

There are a lot of people who exercise so they can do things like get dessert instead of skipping it, or drink whole milk instead of low fat.

If you're exercising so you can make big concessions in your daily diet, you're delusional. Take a look at the calories you burn doing high intensity exercise vs. those in your preferred dessert.

Not saying it's a good or bad decision, just that it's what motivates a lot of people. Who otherwise probably wouldn't exercise as much or at all. And that to presume a (near) universal correlation between eating healthy and proper exercise is wrong.