This train of thought was brought on today when I was surfing the web and came across this piece about stress during pregnancy. The piece talks about a few conflicting studies about whether or not stress is actually harmful or not during pregnancy.
The first study they mention concludes:
The weight of evidence suggests that moderate levels of stress and anxiety do none of the things we fear. They seem not to affect whether women are able to conceive, whether they carry the fetus to term, or whether their kids reach normal developmental milestones. (If anything, some maternal stress during pregnancy seems to make kids mature a little faster.)
While the second study they cite has the opposite to say:
“After controlling for other conditions that could cause stillbirths, we found that experience of a stressful event in all four factors increased the risk of stillbirths by about two fold,” explains Hogue.
Also, there was an increased risk of stillbirths associated with an increasing number of events, suggesting a ‘dose-response’ effect of stress on stillbirth risk.”
What are we supposed to believe?
Now it turns out this type of thing is very common in science. Just think about anything you ever heard about diet. A few examples: ‘Saturated fat is going to kill us all,’ to the ‘turns out saturated fat is not that bad‘. Or the classic, “Eggs are as bad as smoking‘ vs ‘Eggs are Actually good for us‘. If you are interested you could Google ‘the cholesterol myth’ to sort through tons of this stuff.
You can find this sort of conflicting studies all over the place for just about any subject. To make matters worse, the more controversial the study’s conclusions, the more journalists will tend to sensationalize it (Gasp!). This is often done without any regard to the structure or validity of the study. To a lay person, a story about a study proving cholesterol is bad for us sounds pretty convincing.
The sad truth is that somebody could run a study that proves just about anything. Remember all those studies funded by the tobacco companies that never seemed to be able to prove that tobacco was harmful? *cough, cough.. lies*
So what can we do about it?
The best thing to do is actually read the study and see if it has any merit. Now this does requires some knowledge of statistics, and the scientific method, not to mention, the time to actually read all the studies… Which, lets be honest, most of us don’t have.
This is a skill and the more you do it the better you will be at calling “BullSh$t.” One of the most common errors that is made with many studies is that they do not actually prove causality, only show a correlation between one or more variables. Basically, just becuase it one variable appears to affect the variable being studied, does not actually mean that it is. It could just be a coincidence (one form of correlation), and have no direct effect (causation). This is a vital concept to internalize when it comes to trying to figure out if a study actually holds any water or not, and my explanation is a bit lacking. Here is a good Wikipedia article on the topic for a better explanation
The easier path is to find a few authority figures, or thought leaders, in the topic at hand and let them do the hard work for you. You want to pick good ones though. The old saying, “who is the more foolish? The fool, or the person who follows him?,” comes to mind here :)
Most of all, you should really take everything you hear about a study proving this or that with a grain of salt. It may be great research, with a valid conclusion. If it is, it is very likely you will hear more about it. Science is great, and it has provided us with an enormous understanding of the world we live in. However, it was often very wrong along the way.
What do you thing about the all the headlining studies? Do you believe them?
Leave a Reply