For the past few weeks, news sources have reported on a study that suggests people may live longer lives on less sleep than has previously been thought most healthy. I first read about it in my local paper on Wednesday, but a Google News search turned up stories that were nearly two weeks old. However, as many news reports or their headlines about scientific studies can so often be, some of them may be misleading at best, and at worst utterly wrong.
“Sleeping as little as 5 hours a night might be good for you,” reads a representative Oct. 2 headline on the Web site for a Fox News station in New York City, myfoxny. According to the story, an article in the journal Sleep Medicine by Dr. Daniel F. Kripke and a team at the University of California at San Diego did a follow-up on a study of 459 women who participated in a sleep and mortality study 14 years ago. At the time, they were between the ages of 50 and 81. This year, researchers found that survivors were much more likely to have been getting less than the recommended eight hours of nightly sleep -- considerably less. Women who got less than five hours of sleep a night and women who got more than six-and-a-half had two to four times the mortality rate of women in between … the ones who slept from five to six-and-a-half hours per night.
Those are the bare, factual findings. In some places, though, they were sloppily reported. Few newspaper reporters or editors have been trained in the sciences, or in proper interpretation of statistics (even a decent course of logic would help), so what they write about scientific studies often goes overboard. I recall one twenty years ago, that found a correlation between coffee drinking and frequency of sex, which gave rise to lots of jokes and commentary that we should all drink more coffee.
That this study may have found a correlation between less than eight or seven hours of sleep a night and the longevity of some people does not mean everyone should automatically switch to sleeping fewer hours. One of the mantras of science students is (or should be) CORRELATION IS NOT CAUSATION. In other words, just because two phenomena happen to coincide or coexist, it does not follow that one causes the other, or that there’s any meaningful relationship between them at all. By that logic, if I notice that over the past few weeks the sun has risen shortly after my alarm clock has rung, that does not mean I can reset my alarm an hour ahead and expect the sun to rise an hour earlier.
In this case, the women who sleep only five to six-and-a-half hours may also be inherently more sturdy (genetics) or lead generally more healthy lifestyles (no smoking, lots of exercise, etc.) than the women who get what has traditionally been regarded as the healthiest eight hours. Perhaps the women who have been getting eight hours of sleep a night find it hard to get out of bed because they’re exhausted from stress.
The report in my local paper was more careful than Fox News NY: “Longer lives linked with getting 5-6½ hours of sleep” the headline reads, and the text included the critical caveat “Kripke says more study is needed, especially to find whether deliberately sleeping less might help you live longer.” The phrase “more study is needed” is always a good thing for a researcher to say, and in this case it’s especially true, since nobody has any idea why the mid-level sleepers showed a surprising longevity. The best (and most) that many scientific studies find, when they establish a possible correlation, is where further study should be focused.
Unfortunately, the Fox News NY story not only omits that caveat but ends with a more dicey proposition, supposedly from the scientific report itself: “The study concluded, ‘People who sleep five or six hours may be reassured.’ ” And in fact, I saw at least one of my Facebook friends say as much yesterday after reading the story in the local paper. But the crucial word there is may, because there’s no reassurance to be had if you’re getting less sleep due to an overstressed lifestyle and you fail to achieve balance with good nutrition and exercise.
Misinterpretation of scientific studies can occur at any level: a simple-minded reporter may overinterpret the results; or the reporter may get it right, but an editor may slap on a simplified headline that is misleading; or readers may misinterpret accurately reported stories and heads. The last is probably the most common, since it is not difficult to find misreadings and overblown interpretations in the electronic comments that follow news stories online.
It’s important to read everything in the news with a critical eye, not with your prejudices and hopes intact.
Posts
No comments:
Post a Comment