Is coffee good for you?

From an article about the health benefits of coffee:

Another obstacle in pinpointing the benefits of coffee is that it's difficult to isolate the effects of coffee from other healthy habits or lifestyles associated with coffee drinking. A 1999 study of coffee and tea consumption in Scotland, for instance, found that coffee drinkers were younger, had higher incomes, and were healthier in general than tea drinkers.

Coffee drinkers in the U.S. seem to fit a similar profile. Seventy percent of Americans with an annual household income of $150,000 or more drink coffee, compared with 54 percent of Americans in a household making less than $25,000 a year, according to consumer market research firm Experian Simmons.

Research has shown for decades that poorer people are more likely to die from virtually any cause than people with a higher socioeconomic status. Wealthier people are more likely to be physically active and eat healthier, and less likely to smoke -- behaviors that could prevent some of the conditions assumed to be affected by coffee.

"That's the problem [with most of the studies done on coffee]," says Vinson. "There's no perfect study out there because they can't control all the variables. The problem with a human study is everybody's different."

A similar warning should be in every single news article about the supposed health benefits of anything, because nearly all conventionally healthy activities correlate with each other. Unfortunately, such a warning only gets stuck into an article about stuff that’s not normally considered healthy.

The only time I’m confident that a health claim is true is when the condition in question is (1) not conventionally considered healthy; and (2) is not prevalent among the upper classes. Thus the continuing research which shows that it’s healthier to be a few pounds overweight is almost certainly the truth. Most everything else is in doubt.

Comments

[AllanF]

To the health claims you can be confident of, you can add saturated fat is healthy (http://wholehealthsource.blogspot.com/2010/04/full-fat-dairy-for-cardiovascular.html) and whole grains are not (http://wholehealthsource.blogspot.com/2010/03/grains-as-food-update.html).

[Rosscros]

Have they found any negative health indications associated with coffee? If they haven't, it's an enjoyable as a stimulant, even if it's not a miracle cure.

[ironrailsironweights.wordpress.com]

Being a few pounds overweight is not the sort of condition rarely found among the upper classes, as opposed to being outright obese, so your condition (2) doesn't really apply.

Peter

[Brandon Thomson]

There's probably not much risk to a few extra pounds, but a lot of studies don't control for the fact that people who have fatal diseases tend to lose weight as a result of the disease more often than they tend to gain it as a result of the disease. So studies which show it's healthier to be overweight may just be showing that people who are overweight are less likely to already have cancer, etc...

[Gorilla]

I've heard there were studies that showed (a phrase which should warn you not to believe the rest of the sentence) coffee drinking may cause stomach cancer. It turned out though that the population studied were stomach patients at a hospital somewhere, and there were really two group: those with stomach ulcers who were ordered not to drink coffee, and those with stomach cancer who weren't. So of course there appeared to be a correlation between coffee and stomach cancer...

[Miguel Madeira]

Simple solution - instead of making these studies in Scotland or in UK, they should make them in Portugal (where everybody drinks coffee).

[Gil]

Coffee is especially healthful? Pffff, you wish! In light moderation it's probably harmless but in high quantities? Probably not. Next thing people will hoping there's going to say it's healthy to be diabetic.

[Matt_Sanchez]

People use coffee to stay up late and work. The rich are more likely to have to work long hours at the store, business, office, investment firm, medical clinic, research lab, etc. Being rich aint easy.

[rob]

If people didn't actually like it, coffee would be a weird health nut thing. Chop up seeds, soak them in hot water, and drink the bitter water.

"Have they found any negative health indications associated with coffee?"

Last I heard, it's sort of bad for people with type II diabetes.

"A similar warning should be in every single news article about the supposed health benefits of anything, because nearly all conventionally healthy activities correlate with each other."

By making this objection, you are conceding that benefits from the chunk of those activities that are healthy outweigh the harm of the negative ones.

As for the benefits of being overwieght, at what age, and by what standard? Old people who are very thin are usually sick. At younger ages, smokers, drunks, and other drug addicts are thinner. At younger ages, the few pounds overweight probably does include people carrying actual muscle, and of course they're healthier.

People also vary in how vulnerable they are to problems caused by being fat. By 70 or so, the more vulnerable tend to be already dead.

[mentalculus]

"Thus the continuing research which shows that it’s healthier to be a few pounds overweight is almost certainly the truth."

No, these are just like the studies popping up saying wine and chocolate are healthy. Newspapers run them because they tell people what they want to hear.

[Joe Morrison]

95% of good health is heredity. Beyond that, the key to feeling good and staying well is moderation in all things.

[Hendrik Verwoerd]

Wrong!

Get some people for a study who don't drink coffee. Feed them coffee. See what happens.

This has been done for fish oil.

And there is no "it's good for you" or "it's bad for you". There is only "in the study population there was an improvement in the mean of ..." or "in the study population there was a worsening in the mean of ...".

I KNOW that in MY case coffee is BAD. Even one cup raises my blood pressure for a dozen hours.

[bbartlog]

To inject some HBD into it - you should really just start ignoring health claims like this that aren't carefully qualified by describing the population (including racial makeup, but preferably also any relevant genetic markers) of the study group. In the case of coffee, some of this has already been done: there is an extant study that was done in Brazil to look at differences in heart health impact for coffee consumption in slow and fast metabolizers. These were identified by some specific SNPs. As I recall the takeaway was that for fast metabolizers, moderate consumption was protective against heart disease. But more generally, the historical differences between peoples in food consumption and adaptation thereto are so large than any study that just ignores them is basically retarded (and don't forget the gut flora...).

[bbartlog]

'By making this objection, you are conceding that benefits from the chunk of those activities that are healthy outweigh the harm of the negative ones.'

Maybe, but he's also pointing out that identifying the healthy activities might be quite difficult. Do you remember the Scottish study where they showed that high IQ people tend to be healthier than low IQ ones? They tried really hard to account for actual differences in behavior (less smoking among the intelligent, etc.), but the residual effect that they couldn't impute to specific behavioral differences was enormous!

'As for the benefits of being overweight, at what age, and by what standard?'

The most striking difference that I recall is the difference in survival for older people when they suffer some mishap that requires hospitalization; especially for the very elderly (>80 yrs) being overweight, as defined by standard BMI measures (>30 is overweight IIRC), was associated with better outcomes than being regular weight or underweight (underweight being really, really bad for the elderly).
There are selection effect problems written all over that, of course. Maybe the skinny old people are so healthy on balance that they hardly ever get sick in the first place, and maybe by the time they're 80 all the more vulnerable fat people are already dead. And of course 'underweight' includes 'grossly cachectic' alongside the slim and spry.
But I still suspect that optimal (healthy) weight should be age-adjusted. If you look at the CDC charts for age versus weight, there's quite an obvious increase as people get older (not that anyone with eyes in their head should need the CDC to tell them this). And generally I would refuse to believe that evolution is so brain damaged that this is a purely accidental and deleterious effect. People are probably 'supposed to' gain some weight into middle age and older, from an evolutionary standpoint.

[Hendrik Verwoerd]

"95% of good health is heredity."

WRONG!

The correlation of age at death in identical twins is trivial at best.

[sfer]

I've heard there were studies that showed (a phrase which should warn you not to believe the rest of the sentence) coffee drinking may cause stomach cancer. It turned out though that the population studied were stomach patients at a hospital somewhere, and there were really two group: those with stomach ulcers who were ordered not to drink coffee, and those with stomach cancer who weren't. So of course there appeared to be a correlation between coffee and stomach cancer...

A discussion of this is in pages 183-186 of this book:

http://books.google.com/books?id=GseHgIbJo4gC&pg=PA184&lpg=PA184&dq=epidemiology+gordis+coffee&source=bl&ots=fFzvDK1rMh&sig=mWG4hp4AwMEgJokeBM9dOFQV8zk&hl=en&ei=rTvaS-7fHcOC8gatpaBQ&sa=X&oi=book_result&ct=result&resnum=1&ved=0CAsQ6AEwAA#v=onepage&q&f=false

[rob]

"Maybe, but he's also pointing out that identifying the healthy activities might be quite difficult. Do you remember the Scottish study where they showed that high IQ people tend to be healthier than low IQ ones?"

Identifying the healthy activities is hard for fluid behaviors and things with relatively low marginal effects. Coffee might be sort of good for you, or sort of bad. Smoking crystal meth is unhealthy.

"Get some people for a study who don't drink coffee. Feed them coffee. See what happens."

Hendrik, yes, that's very true. Similar warnings should accompany alcohol studies: don't try to drink a glass of wine a day if you've had an alcohol problem.

Bbartlog, are people who put on the middle age spread healthier than people who don't?

[Joe Morrison]

""95% of good health is heredity."

WRONG!"

Death and bad health aren't as strongly connected as you might think. My 77 year old mother-in-law has been in and out of hospitals with kidney and heart problems for the past 30 years. If she dies tomorrow she will have lived a more or less average lifespan. Her mother was sickly, too. Twin lifespans in Amish communities might be a more informative statistic. Indeed, the average Amish person without modern health care is nearly identical to an American with health insurance living within 5 miles of a hospital so go figure. Maybe a lot of geriatric medicine is just make work for doctors.