Economist Steven Levitt and author Stephen Dubner teamed up to analyze and challenge a variety of economic beliefs and practices, including a cost-to-value comparison of add-on CRs versus seat belts. Their 2005 book, Freakonomics, raised many questions, and the pair even wrote a paper on their analysis (though it was never published in a peer-reviewed journal). Their new sequel, SuperFreakonomics, expands on the earlier theme with additional assertions that further challenge the value of CRs for children over age two.
In a nutshell, the authors state that their analysis of data from NHTSA’s Fatality Analysis Reporting System (FARS) and General Estimates System (GES) found that the value of CRs over lap-shoulder belts is marginal for children ages 2 through 6. They continue that this claim is further supported by more complete state data from New Jersey and Wisconsin. Their conclusion is that, though CRs must be purchased to follow the law in all states, they do not believe that the protective benefits of CRs relative to seat belts for children ages 2 through 6 truly justifies the amount spent annually on CRs.
These claims have grabbed mainstream media attention, but several respected organizations have dismissed them as inaccurate. Data analyses that have been published in peer-reviewed journals have shown there is indeed a significant protective benefit of CRs for children ages 2 through 6. One example is a study published after the release of the first Freakonomics, “Effectiveness of Child Safety Seats vs Seat Belts in Reducing Risk for Death in Children in Passenger Vehicle Crashes” (Archives of Pediatrics & Adolescent Medicine. 2006. 160: 617-621). That scientific research showed that CRs are associated with a 28 percent reduction in risk of death in children ages 2 through 6. Even when cases of serious misuse were included, CRs reduced deaths 21 percent more than seat belts alone.
The Freakonomics claims about injuries are also questionable, simply due to the sources for that data. At this point, much of the injury data that is available, such as the national and state data used by the authors is based on police reporting. It is the investigating officer that determines whether an injury exists and its severity. (In Wisconsin, that data is also linked to discharge information from participating healthcare facilities.) While these data sources are used for many of our government statistics, NHTSA analyses are validated with other sources and rarely rely on this officer-coded data as the sole measure for CR use.
The authors found, using the mainly police-generated injury data, that there is little difference between the performance of CRs and lap-shoulder belts for the most serious injury categories, but that CRs do protect against less severe injuries. Conversely, Partners for Child Passenger Safety research shows that age- and size-appropriate CR use reduces serious injury risk by 71 percent (as compared to seat belts alone) in children ages 1 to 3. It further found that booster seats reduce the risk of injury by 45 percent for children ages 4 to 8
(see related article on this page).
The claims in SuperFreakonomics are also based on the results of comparative sled tests commissioned by the authors in an effort to provide further validation of their findings. Spokespeople from The Children’s Hospital of Philadelphia, however, point out some serious shortfalls of using the FMVSS 213 bench and crash dummies to simulate children wearing seat belts in real crashes. For instance, on the sled the dummy and seat belt are in optimally protective crash positions, which is frequently not the case when a small child uses a seat belt in real life. Also, current dummies cannot measure abdominal injury, and the pelvic construction does not allow the dummy to “submarine” (slide out from under the belt) like a 6-year-old human might.
The authors stress that they want children to be protected, but that “cheaper” ways of doing this would include making improvements to vehicle environments, thus also reducing add-on CR misuse. However, vehicle manufacturers claim that fewer than 5 percent of the vehicles they sell are used to transport children. Money spent to add built-in features for children would raise the cost of vehicles for many people who do not need those features.
One thing advocates and the authors do agree on: ultimately, the goal is to have equipment that keeps children safe while minimizing cost as well as misuse.
— Joe Colella
©Safe Ride News Nov/Dec 2009