STATS ARTICLES 2006
The Year in Health Fads: Why Graphics are Bad for You
Trevor Butterworth, December 28, 2006
Op-Art in The New York Times confuses rather than clarifies the year in health news
Yesterday’s New York Times featured an op-ed-op-art, which held out a grail-like promise to readers:
“It is easy to get lost in medical news. Research often contradicts itself, and can leave us wondering what path to follow based on the latest findings. (Drink less, or have more red wine? Take a baby aspirin every day to prevent heart disease, or skip it for fear of stomach trouble?) The past year has provided its share of surprising and sometimes confusing health headlines. The following is a month-by-month guide to what we learned in 2006, and what we might (or might not) want to do about it.”
Unfortunately, it would appear that, if not lost themselves, the contributors decision to illustrate our common quest for clarity has the effect of exposing the perils of using pretty graphics and limited text to summarize complex health stories.
The most egregious was December’s entry, “Don’t take antidepressants if you’re so depressed you’re suicidal.” It noted that “antidepressants may increase the risk of suicide in people up to age 25” and summed up: “So if you’re really depressed, you can no longer take your medicine… More things to be down about.”
While there are some individuals who may have an increase in suicide risk from taking certain medications, there is a growing literature that suggests when the prescription of the newer antidepressants goes up, suicide rates go down.
And the op-ed-op-art didn’t even mention that this increased risk isn’t related to all antidepressants but only to those in the class known as serotonin-re-uptake inhibitors (SRI’s). However, because these medications are very difficult to overdose on compared to earlier antidepressants, they may, in fact, be safer for most people.
While data from clinical trials does suggest increased risk of suicidal behavior with SRI’s, it is unclear who is most likely to be affected, how this risk relates to completed suicide, and what it actually means. Most suicides occur amongst people who are not taking antidepressants [reg. required].
Finally, the suicide risk from taking antidepressants is believed to be most acute during medication changes, so the idea that if you start becoming suicidal you should stop taking your medications is not only inaccurate but potentially dangerous.
July’s entry, “Smoke marijuana… ok, not quite,” reports on a rat study which found that while rats given marijuana were not more likely to try heroin if given the option, they were more likely to become addicted if they did try it.
No mention that this study is problematic, as rats, unlike humans, don’t like marijuana and have to be forced to take it. Stress, however, is linked with a greater risk of addiction to opioids – so this may merely represent a stress response and not a “gateway effect.”
When it comes to marijuana, it is far better to rely on data from humans, and embarrassingly for the Times, a more recent study in humans found that those who take marijuana earlier in their drug-using careers are no more likely to become addicts later than those who do not. This confirmed the Institute of Medicine’s earlier conclusion that the science does not support the idea that marijuana is a gateway to more harmful addictions.
The March entry – “Sleeping pills can make you fat. So can sleep loss” – reported that the drug Ambien “caused some users to go on nocturnal food binges,” but that getting less sleep is linked with weight gain in women. It didn’t mention that the food binges associated with Ambien appear to be rare, while sleep loss in general is common. There were 26 million prescriptions written for Ambien in 2005, but only 1,000 reports of associated binge-eating. There are also many other sleep medications that are not associated with this problem.
A similar failure to examine relative risk can be found in the August entry on soy, which said that both the FDA and the American Heart Association have stopped recommending “a soy-rich diet as good for the heart.” But the article didn’t mention that if the alternative is a meat-heavy diet that includes a lot of animal-based fat, as is the case for many Americans, soy is still likely to be far healthier in terms of preventing cardiovascular disease (See STATS Maia Szalavitz’s “How Healthy is Soy” in the June 2006 issue of Psychology Today for more details).
It’s extraordinarily hard to sum up complex data in sound bites; but even though the Times could have expanded on these segments on the web, it just went for more graphics, linking each item to a PDF file.