Slashdot Mirror
Study Says Coffee Protects Against Cirrhosis
An anonymous reader writes "Good news for those who like both coffee and alcohol. In a recent study of more than 125,000 people an Oakland, CA medical team found that consuming coffee seems to help protect against alcoholic cirrhosis. The study was done based on people enrolled in a private northern California health care plan between 1978 and 1985." From the article: "People drinking one cup of coffee per day were, on average, 20% less likely to develop alcoholic cirrhosis. For people drinking two or three cups the reduction was 40%, and for those drinking four or more cups of coffee a day the reduction in risk was 80%."
we drink neither and break our social and behavioral substance dependencies.
What doesn't kill you today only makes you stronger - until they find out that it too can kill you!
Physics is nothing like religion. If it was, we'd have an easier time trying to raise money!
Does it bother anyone else that the data in question is 21 years old? 1985 seems like an eternity ago - this from a guy born in 1982. I'm not a statistician or a doctor, but couldn't there have been a myriad of things that happened in between 1985 and now? Furthermore, if you drink coffee, most people I know drink at least 2 cups daily so I'm not sure you can draw any meaningful distinctions between 1 and 2 cups. Also, what about other caffeine sources like soda?
Based on the way that study is described, it doesn't sound as though the data necessarily supports a clear-cut causality between coffee-drinking and cirrhosis reduction. They based the results on a questionnaire, after all, and many of those are far too broad (and too sloppily answered) to give precise data about an individual's real consumption of either alcohol or coffee.
The most that this data proves is a correlation between higher reported coffee consumption and reduced cirrhosis-- and there are a ton of other reasons why that might be the case. Maybe heavy drinkers of alcohol tend to under-report their consumption of other harmful substances (like caffeine) out of guilt. Maybe higher caffeine consumption makes heavy drinkers drink a little less. Maybe coffee-drinking indicates a more white-collar lifestyle, which in turn might indicate better education and healthier life habits, any of which might itself be responsible for the diminished cirrhosis. As usual, the pop-sci treatment jumps to an easy causal conclusion that's far from being warranted by the facts.
Selection bias. They have no way for controlling the quantity of alcohol that these people consume NOR the number of years for which they consume it. Even alcoholics all drink different amounts of alcohol. Therefore you have no way of knowing if the coffee drinkers also tend to drink less alcohol (ergo, less damage to the liver), which is a very plausible explanation. The authors even admit they don't have a biologically plausible theory for why coffee might protect the liver.
And 20% is nothing with a sample size this small. An 80% drop when they drink 4 or more cups of coffee? Who has room for alcohol when they've drunk 4 cups of coffee per day? I'm willing to bet there's a huge drop in cirrhosis rates when someone eats a lot too.
There's a huge difference between association and causality, but lack of distinction results in hasty and flawed interpretations.
"The only normal people are the ones you don't know very well."