Sifting through the risks and benefits of coffee.

Health and medicine explained.
July 27 2011 7:24 AM

Café or Nay?

Some studies say coffee is good for you; others say it's bad. The scientists are just as confused as we are.

(Continued from Page 1)

And then there are the studies suggesting that coffee is good for you. Coffee drinkers are less likely to get liver cancer, endometrial cancer, and prostate cancer than abstainers, and coffee has also been associated with a reduced risk of Alzheimer's and Parkinson's diseases.

Trying to assess coffee through the lens of these studies is an exercise in frustration. Is a potential drop in Alzheimer's risk worth a possible increase in urinary tract cancer risk? Framing the question this way quickly leads to an Andy Rooney-style mind overload: Is coffee good for us or bad for us? Who knows?

It's easy to wonder whether the scientific ambivalence about coffee simply mirrors our own tortured attitudes about the drink. Drinking coffee is pleasurable and mind-altering. (If it weren't a potent drug, we wouldn't be drinking it in the first place.) Anything that gives us a buzz must be a little naughty, and yet that shame at needing our daily fix only adds to coffee's allure. It may seem odd to suggest that peer-reviewed science would be in thrall to our moral qualms, but consider the similar tangles of data over the health risks and benefits of drinking wine, eating chocolate, or even smoking marijuana. The scientists are telling us what we want to hear: Pleasure and pain go hand in hand.

Advertisement

In fact, there's every reason to view the research on coffee with skepticism. Nearly all the studies on coffee and human health suffer from a common problem, says Melissa Wellons, a University of Alabama physician who co-authored a recent review article on the benefits and risks of caffeinated beverages: They're observational. Such studies take groups of people with and without a condition and look for differences between them. That's one way to find associations, but it can't establish mechanisms or causality. For instance, the prostate cancer study that made headlines in May enrolled 47,911 men and surveyed them about their lifestyles and dietary habits over a 20-year stretch. During that time, 642 of the participants developed advanced prostate cancer. When researchers sorted the surveys, they found there were fewer heavy coffee drinkers in the group with advanced prostate cancers than in the group without. Indeed, men who drank six or more cups of coffee per day were 60 percent less likely to be in the group with the deadly cancer. That could mean that drinking huge amounts of coffee will cut your risk of having advanced prostate cancer by 60 percent. Or it could mean that having advanced prostate cancer decreased your likelihood of being a heavy coffee drinker by 60 percent. It's hard to know, a priori, which explanation is correct.

If you divide the observational studies into two piles—"coffee is good for you" and "coffee is bad for you"—the bulk of the flimsy evidence falls on the good-for-you side. More to the point, there just isn't any persuasive evidence that coffee is harmful, says Wellons, even though people have been drinking it for centuries. That's not to say coffee can't produce a minor array of inconveniences. Caffeine at doses of 65 mg or more can help relieve headaches, but habitual caffeine use can spur chronic headaches, too. Coffee makes you more alert, but it can also provoke insomnia, anxiety, and even tremors at high doses. Regular drinkers seem to have a decreased risk of developing diabetes, but for anyone who already has the disease, a hefty cup of coffee can make things worse. Up to three cups of coffee per day may slightly lower the risk of heart attack, but drinking a large dose might also trigger a heart attack if you're already at risk.

Rest assured, we'll see more conflicting research in years to come. There's nothing especially remarkable about coffee that makes it difficult to study. Scientists could get more convincing answers if they ran intervention trials—by assigning a bunch of people to either drink coffee or abstain from it for years or decades, and then checking which group got more cancer and other diseases. But such studies take lots of money and time, and they're unlikely to happen because they're so expensive. (Industry has little incentive to bankroll this research, and the government seems to have more pressing priorities.) Instead, we're bound to get more headline-grabbing observational data that hint at what we already think we know: Coffee is scary and wonderful.

Christie Aschwanden is an award-winning writer and contributing editor for Runner's World. She blogs about science at Last Word on Nothing.