Thursday, July 30, 2009

Distorting science for traffic

Photo by kikisdad (flickr)

Like everyone, I have my pet peeves - those small things that will annoy me to no end even though most people would think the thing it is no big deal. One of my largest pet peeves is when media outlets or blogs do not accurately report on scientific research or they skew the results to make for a more sensational story.

The world of autism has more than its fair share of misinformation and really doesn't need more.

Take for example a recent post by Lisa Jo Rudy at entitled Autism Caused by Breast Milk?!. From the opening line of -
Could breastfeeding cause autism?!
to the closing line of -
Would it have an impact on your decision to breast feed your child?
the post talks about the possible link between breast feeding and autism based on the result of a "new" study (the study is over 2 years only).

Yet, if you go read the actual study (open access) what you find is a study about exposure to PCBs and developmental issues. Specifically, it looks at how the rat brains develop differently when they are exposed to a high level of PCBs very early in life. This study is an excellent example of how it is possible for chemicals in the environment to cause developmental problems.

But it certainly isn't about breast feeding causing autism as the title of post implies. If you read the study you see that the researchers fed the rats the PCBs almost from the start of pregnancy to when the infant rats were weaned. So the link to breast feeding comes into play because PCBs can be passed from the mother to the infant in breast milk. But then again they can also be passed during the prenatal period across the placenta - the study doesn't differentiate between the two methods.

So the real point of the study is that PCBs can be passed from the mother to her offspring and, if the exposure is high enough, can cause developmental problems. When it comes to the "dangers" of breast feeding, the authors have this to say -
Although numerous studies have shown that breastfeeding of infants is superior to existing alternatives, these data suggest the possibility that, in extreme cases, for mothers with high levels of exposure to PCBs and/or other closely related toxicants that bioaccumulate in breast milk, lactational exposure of genetically at-risk infants may in fact not be in the infant’s best interest.
Notice the use of the words "possible", "extreme cases", "high levels of exposure", "genetically at-risk" followed by a "may not be in the infant's best interest". Slam dunk against breast feeding, right?

It would have been much more accurate to report that exposure to high levels of PCBs early in life may cause developmental problems but that wouldn't be a shocking as breast feeding causing autism, now would it?

No comments:

Post a Comment