Annals of Bad Health Reporting: Diet Soda Edition
Josh Bloom in Science 2.0 sets things straight:
The Iowa group studied 60,000 middle-aged women over a ten-year period. Data were accumulated from questionnaires—a notoriously unreliable method of data gathering. But this isn’t a tiny fraction of the problem.
At the end of the study period the group took a look at the health of women who did, and did not drink diet soda. Lo and behold! Of women who drank two or more diet drinks per day, 8.5 percent had some sort of heart disease. But, for women who either drank fewer or no diet drinks that number was only 7 percent. Uh-oh. Smoking gun?
Not even close. Because buried at the bottom of the article is what is really going on: The women who drank more diet soda were less healthy to begin with. They were more likely to be overweight, to smoke, and to have high blood pressure than the other group.
So, let’s correct the headline a bit: “Sick People are More Likely to Die.” Accurate headline, but it won’t sell many papers.
I pray to the gods that there is a special circle in Hell for health reporters.