I may be completely off-base here, but I remember reading the Injury Prevention article and one of the follow-up articles written about it, so I thought I'd toss in my :twocents:. I still think that the 5 times number does refer to all accidents, just not side impacts; however, it is specifically referring to children aged 12-23 months.
Marylin Bull and Dennis Durbin recommend that "Parents may be helped to understand the importance of using the convertible car safety seat in the rear-facing position longer than 1 year if they are counseled that children are 5 times safer than when riding in a forward-facing seat into the second year of life." They make no mention in their recommendation of the "5 times safer" referring only to side impacts. They noted earlier in the same article, "The odds of severe injury for forward-facing infants under 12 months of age were 1.79 times higher than for rear-facing infants; for children 12 to 23 months old, the odds were 5.32 times higher." (from
http://pediatrics.aappublications.org/cgi/content/full/121/3/619) They cite the Injury Prevention article for the 5.32 number, and I think it is this number that has prompted the more recent recommendation of rear-facing until 2 instead of 1, like in the AAP newsletter article in April.
So, unless I am misinterpreting the data and statements made by Bull and Durbin (entirely possible - I haven't run a statistical analysis in years, so I'm rusty
), I think it is accurate to use the 5 times number when talking about kids ages 12-23 months. Bull and Durbin want parents to be told that their 1 year olds are 5 times safer, but I don't see anything where they limited that to side impact crashes. The 5.53 number in the original article (553% mentioned above) does refer to side impacts only for all kids under 2, but based on the way Bull and Durbin discuss the 5.32 number in this article and the abstract of the study itself, I think the 5.32 number is a separate statistic referring to 12-23 month children in all accident types considered in the study.