New Study Links Caffeinated Coffee To Vision Loss
dsinc writes "A new study suggests caffeinated coffee drinkers should limit their intake to reduce their chances of developing vision loss or blindness. According to a scientific paper in Investigative Ophthalmology & Visual Science, heavy caffeinated coffee consumption is associated with an increased risk of developing exfoliation glaucoma (abstract), the leading cause of secondary glaucoma worldwide. 'Scandinavian populations have the highest frequencies of exfoliation syndrome and glaucoma,' said author Jae Hee Kang, ScD, of Channing Division of Network Medicine at Brigham and Women's Hospital in Boston, Mass. 'Because Scandinavian populations also have the highest consumption of caffeinated coffee in the world, and our research group has previously found that greater caffeinated coffee intake was associated with increased risk of primary open-angle glaucoma, we conducted this study to evaluate whether the risk of exfoliation glaucoma or glaucoma suspect may be different by coffee consumption.'"
Good for you... Coffee is bad for you... Coffee is good for you... Coffee is bad for you...
Coffee is making me sea sick.
From the abstract:
Compared with participants whose cumulatively updated total caffeine consumption was <125 mg/day, participants who consumed 500 mg/day had a trend toward increased risk of EG/EGS that was not statistically significant (RR = 1.43; 95% confidence interval [CI], 0.98–2.08); P trend = 0.06).
If it's not statistically significant, then how can we take this seriously?
No, it means that you should stop reading dumb clinical articles taken out of context on Slashdot.
This is just one of those hundreds of thousands of medical articles trawling the data for a correlation so somebody can chase after another grant. According to TFA, they reviewed records of almost 79000 people and came up with 360 cases of this particular form of glaucoma. Then they take the self reported caffeine intake, adjust for 'other confounders' (waves hands) and come up with a weak (Relative Risk 1.4) association that is barely statistically significant and likely not clinically significant at all.
Hrumphh. Not impressed
(Goes back to quaffing his Nuclear Waste level caffeinated beverage)
Faster! Faster! Faster would be better!
Sorry, their conclusions are just not statistically justified.
Let me review what they found:
Compared total caffeine consumption of less than 125 mg/day to greater than 500 mg/day: no significant result
Compared abstain from caffeinated coffee to greater than 3 cups of caffeinated coffee daily: glaucoma relative risk in the interval 1.09 to 2.54
Compared consumption of (caffeinated soda, caffeinated tea, decaffeinated coffee or chocolate) to non-consumers of same: no significant result
That relative risk that they quote as being significant has a confidence interval with a lower end of 1.09; which is only barely above 1.0 (1.0= no effect). So, they studied one particular variety of one particular minor disease (of many health effects). Finding one effect at a trivial level is meaningless.
Ob xkcd: http://xkcd.com/882/
http://www.geoffreylandis.com