The Case for Drinking as Much Coffee as You Like
theatlantic.com
theatlantic.com
What I tell patients is, if you like coffee, go ahead and drink as much as you want and can," says Dr. Peter Martin, director of the Institute for Coffee Studies at Vanderbilt University. He's even developed a metric for monitoring your dosage: If you are having trouble sleeping, cut back on your last cup of the day. From there, he says, "If you drink that much, it's not going to do you any harm, and it might actually help you. A lot.
What else can you expect from an "institute" with a name like this?
And, surely enough, a little more digging brings this:
Nashville's Vanderbilt University is to establish an Institute for Coffee Studies this fall, funded by $6 million from the Brazilian Coffee Association and other coffee-growing organizations.
http://www.nature.com/nm/journal/v5/n3/full/nm0399_252b.html
Institute director, Peter Martin, stresses that his research group wants to maintain "arms length" from the coffee industry for ethical reasons. "We are setting up a foundation to oversee the funds which will be administered by independent scientific directors to ensure that there isn't even an appearance of conflict of interest," he insists.
In addition, the original article described a wide variety of studies done by many different institutions. I suppose it's possible that they're all just "studies" conducted by people in the pay of the nefarious coffee lobby, but we probably shouldn't start resorting to conspiracy theories without a little evidence.
Like in government, sometimes there is quid pro quo and sometimes there isn't, but I've never heard anyone come out and say, "Whoops, you caught us! We're totally taking money to sway our opinion!"
The breadth of studies makes this look pretty legit. I just don't put much faith in obligatory denials.
Ben Goldacre's book (pro science, about flaws in current system of research and science reporting) "Bad Science" is w useful read.
Too late for that.
Upton Sinclair = Chicago stockyards, abuse of animals and immigrants
Anyway, the article doesn't rest on that quote by a long shot, it goes on in enormously satisfying detail. It's a humanizing lead-in, at best. I think this is a great summary of recent research on the effects of coffee and caffeine, and it links to supporting studies on nearly every substantial point.
When you research "the effects of drinking coffee" it seems as though you are more neutral, but in fact there is still room for bias. Will you find every single effect of drinking coffee? How will you decide wether they are positive or negative? how will you decide the importance of these effects? Which effects will you research in more detail?
This weighing and ordering depends in part on the researcher, who will have a bias.
Ofcourse this is not a problem when there are lots of researchers all with their own biases, together they will paint a full picture. But if you only fund those researchers who have the kind of bias you like, then that full picture might be skewed (this is called publication bias: http://en.wikipedia.org/wiki/Publication_bias).
In this case people worry if the coffee research institute has hired biased researchers, rewards only the researchers who publish good things about coffee or even publishes selectively.
So it is a good thing to point out. And the argument is an attack on the reliability of the fundament for the argument, not necessarily on the researcher.
'Ad hominem' is a fallacy of relevance, meaning it involves dragging irrelevant information into the discussion to sway emotions. How is the source of the study's funding irrelevant?
tldr: Ad hominem != per-se fallacy.
Why do we assume that the Coffee Association is promoting bad science for their interests, rather than finding scientists who are working on things that relate to their business, and supporting those works?
Who else is going to fund coffee studies besides people with an interest (and money) to fund it? This is how the free market works; as long as the science is sound, who cares who is behind it? I'd rather the coffee industry pay for the studies from excess profits than it coming from the public coffers. It's not perfect, but it isn't full on corruption, either.
This is the exact equivalent of the coal industry finding the minority of scientists who think climate change is not man-made and funding their studies. The problem with this is obvious, and it is not good science.
>>This is how the free market works; as long as the science is sound, who cares who is behind it?
Often times, those who are funding the science have an impact on whether that science is sound. It's called a conflict of interest because the source of the funding can corrupt the motivation of the scientist to practice good science.
No it isn't. This isn't about coal or climate change. It's about whether or not it's ok for people to drink coffee.
If you think they've published bad science you should prove it. That's the awesome thing about science, it's refutable.
This is not research "related" to coffee, so your hypothesis is not valid in this instance.
My brother and sis, who both went there, love to point out that it is called "The Harvard of the South"
This next bit may be a bit cynical:
Even if they weren't funded by the coffee industry, however, they would have existential issues. If their whole purpose is to study coffee then I think that there will be a tendency to either promote coffee or to try to strike a balance. The latter is almost as bad as the former.
Maybe I'm wrong. Maybe they would have no problem publishing nothing about negative results if that was what they got, but I would bet against it. If they get a string of negative results (and don't lose their funding), they are going to start looking for studies that have a more potential upside to balance things out.
Why? Two reasons.
The first is tolerance. Caffeine tolerance reduces the sense of freshness, alertness, energy etc that coffee gives for any given dose. But tolerance isn't a linear function for all the effects of caffeeine.
The practical upshot is that, on competition day, I wind up needing more caffeine than usual to give me the extra pep to really perform well. But if I'm already tolerant to caffeine, my "pep" dose will be high enough to cause sweaty hands and to degrade my motor control -- in my case my grip becomes weaker and my pulls are not as crisp. Olympic-style weightlifting is a precision sport; I can't afford that.
Whereas, if I've weaned myself off caffeine over the preceding weeks, there's been adenosine re-regulation and I've become more sensitised to it again. So I can get all the pep and energy to push through a competition (it can be very draining with all the warm up lifts) without the sweaty hands or weaker grip or less accurate pulling technique.
The second reason to wean myself is that going cold turkey is absolute hell.
One rule I observe these days is that I don't consume caffeine after 1pm. I can't get to sleep otherwise. In practice this means two cups a day: one when I wake up, one before or during lunch. Even after the noticeable effects of caffeine have faded, it has enough of a half life to make it difficult to fall asleep in a timely fashion. Then you get stuck in the caffeine-sleeplessness-caffeine feedback loop. Worth avoiding.
Second biggest issue is that I tend to get hypoglycemic an hour or so later if I don't have a carby snack at the same time.
A cup a day (with a cup of tea in the morning) is the best balance of enjoyment versus drawbacks that I've found. If I want more I brew up a cup of decaf.
Also I find an Aeropress is easy enough to be convenient but not so easy as to trigger overconsumption (Cf. K-Cup). It doesn't hurt that it makes a great cup of coffee.
When you don't drink it all the time even a small cup goes a really long way, at least for me.
Today I'm still at about four cups a day. Similar to you, I use a Chemex instead of an automatic machine to slow my consumption.
It's worth noting, though, that what you put into coffee may not be so good for you. I sat down and did the math and realized that I was taking in almost 1500 calories/day in milk and sugar that I put in my coffee. I've since switched to drinking it black or with just a bit of skim milk (to cool it - the first cup out of a french press is scalding!), and have lost a significant amount of weight just from that change alone.
Double espresso with a dash of Baileys.. mmm..
I just came across a recipe for freezing coffee in ice cube trays, and then putting it in with Baileys and vanilla vodka.
This sounds like a winner for the annual "It's a Wonderful Life" every time they say George Bailey drinking game.
I finally found some excellent decaf (Ethiopian Sidamo at Zombie Runner [1] in Palo Alto), so I can continue indulging the habit.
[1] http://www.zombierunner.com/store/categories/cafe_zombie/cof...
It seemed to take a few months for my body to transition off caffeine. I recommend switching to tea for the first couple of weeks, since that'll at least prevent the terrible headaches.
I definitely feel healthier for quitting the habit, though I still miss "real" coffee sometimes. :(
Frightening to think of those withdrawl symptoms for such a seemingly innocuous drug.
First three days I was sleepy, then I recovered to exactly the same productivity level.
Then I decided to make me more productive and drunk a 0.25 Adrenaline Rush. My symptoms were nothing less than caffeine overdose.
I wanted to work but was unable to do anything because of total lack of focus. So I wanted to sleep and unable to sleep because fo hyperalertness. Then I wanted to work because I cannot sleep, etc. The cycle lasted for six hours at least and it was very unpleasant. I was even unable to read!
Right now I take a long walk after work if I want to be really productive next day.
[from the article]
>That there were no major differences in risk reduction between regular and decaf coffee suggests there's something in it, aside from its caffeine content, that could be contributing to these observed benefits
It could merely be that people who drink coffee eat less food, and that food is a less healthy drug
Don't get me wrong, it's clearly not crystal meth, but Caffeine overdoses are real and can be dangerous, especially in a situation where you are drinking a boatload of coffee. I'm sure I'm not the only person on this board that has had some minor heart palpitations during an all-nighter after one too many coffees or Coke Zeros.
[1] http://en.wikipedia.org/wiki/Caffeine#Caffeine_intoxication
I guess I was a little sloppy in labeling what I'm talking about a "caffeine overdose", although I think what I described happening to me (heart palpatations, dizziness) probably happens much more often than physicians are aware of, and could be classified as a minor caffeine overdose. Most people logically conclude they've had too much coffee and too little sleep, and just drink some water and go to bed like I did. They don't go to the hospital over it.
What I'm really saying is that people who consume massive amounts of coffee are usually trying to compensate for exhaustion or sleep deprivation. In fact, there's sort of a feedback loop, where massive amounts of caffeine make restful sleep much more difficult, which then leads to more caffeine consumption and possibly to a downward spiral in health. Studies have shown that a perfectly healthy person lacking in sleep is in sort of a pre-diabetic state already, so when you pile on massive amounts of stimulants at the same time, it's clearly bad for the heart. It won't lead to you being declared DOA of a caffeine overdose, but it could exacerbate heart disease, or help its onset, or even cause a heart attack.
Don't get me wrong, as I mentioned I'm a huge caffeine addict. I drink Coke Zero by the gallon. However, the picture the author is painting that massive ingestion of caffeine is consequence free is simply not accurate, and in fact dangerous.
But you can bet your bottom dollar that drinking a dozen cups of coffee by lunchtime will have some pretty interesting results. You'll feel your heart racing and probably experience something like extreme ADD. Probably headaches, nausea, crazy jitters. The severity depends on your tolerance to caffeine, of course. I get nasty headaches 6-8 hours after waking if I don't get my morning coffee, so it takes me quite a lot to get into overdose land.
[1] https://en.wikipedia.org/wiki/Caffeine#Caffeine_intoxication [2] ~150mg/kg for LD50; I weigh about 50kg
That's the gist of the article.
1. Ask HN: How much coffee do you drink?: http://news.ycombinator.com/item?id=4606007
2. Ask HN: Coffee Drinkers, How Do You Brew? http://news.ycombinator.com/item?id=4408221
> That same high dosage is also effective in fighting against colorectal cancer
I do have to wonder if it's not a primary effect of anything in the coffee, but rather a byproduct of the secondary effects.
I've actually found it to be extremely useful part of my daily routine and when I need to break the routine for whatever reason (I live nomadically so my routines are always changing), my whole day is usually a lot more uncomfortable (i.e., my digestive system and bowls feel slower).
I still haven't identified how all this works, but it's an interesting anecdote.
Coffee acts as a laxative. Interestingly, this effect is observed in decaf as well.
Source: Boekema PJ, Samsom M, et al. Coffee and gastrointestinal function: facts and fiction. A review. Scand J Gastroenterol Suppl. 1999;230:35-9. PMID 10499460
1. You still get some caffeine; decaf is about 5% of the caffeine of normal coffee.
I understand that there may in fact be organs in the human body that are not the heart, but when I hear people talking about caffeine risk, they generally seem to be referring to heart disease, so that's the kind of article I pubmed'ed.
The director who is paid to study coffee says that coffee is good for you. I am shocked, shocked, I tell you!
If you are adventurous, you should try his "bulletproof coffee" (high quality coffee with butter, but there's more to it - you can find details on his website). Butter is also good for you. Really.
This guy's combination of scientific literacy and alarmingly poor reasoning really makes me nauseous, because I can't simply write it off as mere incompetence. For instance, look at this infographic: http://www.bulletproofexec.com/wp-content/uploads/2012/04/Bu... . Specifically, look at the part comparing 1910 and 2000. That's not just a case of spurious correlation. It's like something I would make up as a wacky example of spurious correlation.
> you'll notice that he does not bridge that crucial gap between "mycotoxins are bad for you" and "my coffee has lower mycotoxin content"
I'm not sure which gap you think needs to be bridged here?
> Specifically, look at the part comparing 1910 and 2000. That's not just a case of spurious correlation. It's like something I would make up as a wacky example of spurious correlation.
I'm not sure what you're reading into this. When I look at it, I don't think "butter protects against heart problems" (which seems to be your interpretation, and I agree cannot be concluded from this data, not even remotely), but rather "butter can't be the evil it is made to be, as this statistic shows". Which is perfectly reasonable.
The thing about the QS movement, and people like Asprey, Roberts and Ferris, is that they offer very cheap experiments with easily measurable outcome, so you don't have to wait for a 30-year double blinded test that is blessed by anyone.
I've tried several things they recommend, and the vast majority of the practice checks out (regardless of the theory behind it).
I haven't tried his coffee, but I did start putting butter in my Lavazza coffee, and as a result lost weight and am feeling much better. You don't have to be a "believer" to try and see if things work for you. I don't think he's right about everything, but a lot of the counterintuitive things him (and Ferris, and Roberts) suggest actually work.
I said it explicitly: "there is simply no good reason to believe that mycotoxin levels in ordinary coffee are high enough to have any measurable effect on humans who drink it."
>When I look at it, I don't think "butter protects against heart problems" (which seems to be your interpretation, and I agree cannot be concluded from this data, not even remotely), but rather "butter can't be the evil it is made to be, as this statistic shows". Which is perfectly reasonable.
It isn't perfectly reasonable. Life expectancy in 1910 was 48 for men, and 51 for women. The average age for heart attacks today is 66 for and 70 for women. Some, but not all of that is infant mortality, yes, but the fact is that if people in 1910 had less heart disease than people today, it is likely largely because people in 1910 often didn't long enough for heart disease to become a problem.
And that's just one factor that the comparison fails to control for. The way people lived and ate in 1910 was completely different from the way people lived and ate in 2000, and it is quite likely that those changes simply overwhelm the change with respect to one particular kind of saturated fat. That doesn't mean that butter fat is harmless. It just means that you can't see the harm if you also change everything else.
As a side note, where are these statistics about heart disease in 1910 coming from? We get the top-level URL for dietheartpublishing.com (which is low on data, high on sensationalist paleo-quackery), but no link to specific data. Are we relying on what doctors were able to diagnose in 1910? Because that's a whole other level of flawed.
>I've tried several things they recommend, and the vast majority of the practice checks out (regardless of the theory behind it).
What you think you've experienced through personal trials is not significant evidence.
I don't keep the links handy, but a one second google brings up this: http://www.ncbi.nlm.nih.gov/pubmed/16284846 and a tens of other references. Coffee was not studied as extensively as far as I know, but if you have reason to believe that it is better protected than wheat and maize, please let me know why. The only non-asprey link about coffee toxin studies I could find is a single article from the 1980 that says "I looked, I didn't find, I stopped looking".
> Some, but not all of that is infant mortality, yes, but the fact is that if people in 1910 had less heart disease than people today, it is likely largely because people in 1910 often didn't long enough for heart disease to become a problem.
Well, in 1860-1910, US white infant mortality rate was coming down from 21% to 10% (with black mortality rate, when documented, being 50% higher than the white one). I don't know if that data is included in the "1910 life expectancy", but it is likely that it is. In Italy, it was going down from 25% to 13% over the same period. That's enough to make the "48/51" lifespan you quote meaningless (sources I found say 54, but whatever). http://eh.net/encyclopedia/article/haines.demography
If you consider "life expectancy of anyone aged > 4", then all of a sudden 50 years jumps to 62 years; and if you consider "life expectancy of anyone aged > 20", then it will surely increase the 1910 number to 70 or more, while hardly nudging today's numbers. I agree that his statistic is not conclusive in any way, but your interpretation is not valid either.
> That doesn't mean that butter fat is harmless. It just means that you can't see the harm if you also change everything else.
That is true. And yet, there is virtually no direct evidence of harm from butter than I can find. (Neither is there from cholesterol intake, for that matter - or salt intake, with the exception of a well defined 20% of the population - that is, if you're not in those 20%, there is no evidence of salt harm).
> Are we relying on what doctors were able to diagnose in 1910? Because that's a whole other level of flawed.
Yes, that's true. But that throws away any study done before the early 1990s - which is also flawed. We're damned if we do, damned if we don't, and everyone gets to pick their own favourite data and interpretation.
> What you think you've experienced through personal trials is not significant evidence.
What I discovered is significant for my daily life, is significant for my daily life (duh, it's a tautology). I'm not saying you should believe it because it worked for me. I'm saying you should try and figure what works for you.
FWIW, my ex was a medical doctor and I regularly read BMJ and NEMJ for a while. That was before Ionnadis's paper, but I remember thinking back then that 70% of what was published was statistically invalid.
Much of what is considered significant evidence in medicine and health today, is not significant evidence. If the experiment is simple and cheap enough, you should do it yourself.
And I'm saying that that's the same mantra that homeopaths, chiropractors and other peddlers of pseudoscience repeat. "Figure out what works for you" has implications dangerously close to relativism. To find out what works, you need science. "Try it yourself" isn't science.
I figured out that wheat is bad for me. As in, I've stopped taking in wheat (much harder than it sounds, because it's used as a filler in just about everything), upped my overall caloric intake (mostly coconut fat and butter), and lost 30 pounds in a month feeling much much better. As in, better than in years. That happened 24 months ago, and it stayed off.
My doctor, seeing it, said it was impossible - because every test HE had at his disposal said I did not have a wheat allergy (or intolerance, or whatever other variations of "bad for you" medicine has).
If you claim that losing 30 pounds in one month and it staying off for two years since is placebo, I'm going to laugh at you, and so will everyone else who actually has a brain and is not religiously following some mantras.
A friend of mine is allergic to peppers of any kind. Not deathly allergic, but excruciating-pain allergic. 20 years ago or so, it was not a standard diagnosis, and some doctors therefore suggested it was psychosomatic (because apparently, that was the only "scientific" explanation). Whatever it is she has, has since become a standard diagnosis (she's missing an enzyme, and it is a recessive genetic trait). Are you seriously claiming that she should have eaten peppers all these years because her self diagnosis was "not scientific"? I would give you the benefit of doubt that you do not, although your response indicates otherwise.
The "science" you talk about is a religion, and if you refrain from self experimenting because "it is not science", then you are a religious zealot. Physics is science; Chemistry is science; Nutrition and Medicine are religions that are controlled by politics and money, that are informed by science, but are definitely not following it.
Yes, science is a great thing. It's hardly related to modern nutrition, and missing from some areas of modern medicine.
Doctors and nutritionists have been claiming for years that salt is bad for you, and dietary cholesterol is too (to the point that you should limit e.g. eggs to no more than two a week). If you can find scientific evidence anywhere that, say, 25 eggs/day, is not healthy, then please show it. Because no one I know is aware of any. Or any evidence for calorie balance (carb=4kcal/g, fiber=0kcal/g, ...) theory for that matter. Add nutrition experts and "MDs" to your list of homeopaths and other pseudoscience peddlers - they're only better on the average because their pool of (mandated!) beliefs is culled more often -- but definitely not often enough to give it the halo of science that you assume it has.
After reading your comment, I assure you that the feeling is mutual.