Kevin Lomangino is the managing editor of HealthNewsReview.org. He tweets as @KLomangino.
I can’t remember a week that has featured so much useless reporting about studies that are meaningless to the average reader.
All the stories featured clickbait-y diet topics like alcohol, chocolate, coffee, and fiber.
All were based on observational studies that can only show associations, not cause-and-effect, and which are prone to drawing conclusions that are later shown to be unreliable.
Most lacked discussion of important study limitations, contained misleading statistics, or otherwise gave readers an incomplete assessment of the research.
I’d call it nothing short of a crisis of crap health news.
Not all observational studies are meaningless (e.g., smoking and cancer), and when the associations are big enough and seen often enough, they can wisely be used to guide health recommendations and decision-making.
But this was not the case with any of the studies that were reported on this week.
I can only hope that by calling attention to the problem, more readers will be warned off of these stories and journalists will think twice about covering them — or cover them more skeptically than was the case this week. We offer a primer on writing about observational studies and one that explains the importance of absolute risk figures.
On Tuesday the big headline was “Just one alcoholic drink a day increases breast cancer risk” — but nobody explained what a “5% increase” might look like using numbers that actually mean something.
On Wednesday the headlines asked, “Could chocolate guard against an irregular heartbeat?” But nobody mentioned the significant problems with diet studies based on questionnaires.
On Thursday we saw all-caps claims that “COFFEE REDUCES LIVER CANCER RISK: FIVE CUPS A DAY LINKED TO 50 PERCENT DROP IN HEPATOCELLULAR CARCINOMA.” What’s a “50 percent drop” mean? Newsweek claims the rate was reduced from “50 out of every 1,000 people” overall to only “33 people in every 1,000” who drank coffee. But how can that possibly be correct? The National Cancer Institute pegs the rate of liver cancer in the U.S. at 8.6 per 100,000 people per year, making the disease exponentially less common — and the supposed benefits of this “natural medicine” orders of magnitude smaller — than Newsweek’s claims suggest.
And lastly, somewhere along the long line the New York Times informed us, “High-Fiber Diet Tied to Less Knee Arthritis.” I can’t explain the problem with this story any better than the Times commenter identified as Stuart from Dallas, Texas.
“Increasing dietary fiber is one of the most economical ways to reduce the pain of knee osteoarthritis,…”
This conclusion from a researcher surprised me. Surely he knows enough not to fall into the causation vs. correlation trap? .. Is it possible that people who are conscientious about getting enough fiber are also conscientious about other elements of self care that impact their knees, e.g. their diets in general and maintaining a healthy weight? Perhaps the root cause is conscientiousness which results in eating enough fiber as a part of a balanced diet, a healthy weight and healthy knees.
I hope by now it’s apparent how this particular media con works and why it keeps getting repeated ad nauseum. It’s a simple four-step process:
Lather, rinse, repeat.