Maybe you've read the alarming medical headline recently, stating that children born by In Vitro fertilization (IVF) are at an increased risk of developing cancer.
Maybe you've read that these kids have a 42% higher chance of developing cancer than their non-IVF peers.
Maybe you've read that kids born by IVF have a shocking 87% chance of receiving a cancer diagnosis before the age of 3 than their non-IVF peers.
If you haven't read it yet, you will.
Statistically, or mathematically, there's nothing refutable about these figures. They're just enormously misleading.
The truth is this: A team of Swedish researchers did a retrospective analysis of the medical records of 26,692 kids born by IVF between 1982 and 2005. Using pre-established epidemiological figures, they expected to find 38 cancer diagnoses.
Instead, they found 53. 15 more than they expected.
Unless you read the actual paper published in the journal Pediatrics, you won't read that three previous, peer-reviewed studies from Sweden failed to find any such increased risk.
You won't read that the authors themselves said they failed to determine whether IVF had any actual role in causing cancer.
Finally, you won't read that members of the same research group behind this study were responsible for one of those previous studies.
This last point made me wonder if the authors had some agenda; maybe they're anti-IVF, some people are because of the procedure's requisite destruction of embryos. But such an agenda would be near impossible to prove. The study's funding, always a great place to look, was provided by a grant from the Evy and Gunnar Sandberg Foundation, which appears to fund many pediatric studies in Sweden, but which is otherwise rather opaque.
Bottom line, what is the point of this research?
I have no clue. But here's what I do know about it:
In medical journalism, headlines-- however accurate-- bear the real currency of the content, and will therefore always trump science. All you can do to fight this is to read beyond the headlines, or don't read them at all.