Yesterday, there was a lively debate in my comment section about the AAP's new car seat recommendations. Both sides made excellent points and I urge all to read the wisdom of my commenters. I really need to pull out one comment made by bj, because she actually read the studies that the AAP used to make their recommendations. With her science background, she shredded those studies nicely. 

bj writes:

I plan on looking at the data about rear-facing car seats, but the Durban group paper rebutting the Leavitt data is very bad science. The paper has misleading cites, purposeful obfuscation, a non-significant result that is treated positively because they "know" that boosters must be good regardless of the data, and a self-serving plea for more research after negative findings.

The Durban cites about 9-12 year olds in booster seats do not show, or in one case even address, that boosters are safer for that age group. instead they confound age and car location and seat belt use (9-12 year olds are safer belted in the back 2-12 year olds are safer in boosters.)

Durban lead the team wrote AAP regulations. I am deeply troubled by the quality of the science in the papers I've read so far. Now on the other hand, I think it unlikely that the having your kid in a booster will be more dangerous for your child physically, so I'm comfortable making my own decision based on the what I read of the science.

(PS: Many of these papers are publicly accessible, and cited in the guidelines — though misleadingly in some instances.)