• child restraint
• seating position
• risk homeostasis

The “classic” in this issue is important for several reasons. First, it exemplifies the value of a deceptively simple and relatively inexpensive form of research: the observational study. Second, it illustrates how important such data can be in moving legislators to act. Third, it reminds us that insurance companies have an interest in the work of those involved in injury prevention, and should, therefore, be one of our strong supporters.

Finally, it should prompt us to take stock of our successes and failures in protecting children in cars. Once we do, it will be evident that, as the papers by Segui-Gomez et al and von Kries et al in this issue show, the battle is far from over. In the case of the former study, the method used is basically the same as that used by Williams nearly 30 years earlier.

The approach adopted by Williams seem simple, but as many have learned since this pioneering study of seat restraint use, they can be more complex than they appear. Observers need to be well trained so that two independent observers of the same event arrive at the same conclusions, not only about whether children are restrained, but also their approximate age, who else is in the car, etc. All this must be done quickly and accurately. The sites where observers are posted need to be carefully selected to avoid bias. And the conclusions drawn must be cautious, and, to be effective, well publicized. These difficulties should not intimidate other researchers. Even a relatively inexperienced community program, should be able to conduct a successful observational study with a little expert help.

As was the case with the Williams study, the data can and, where appropriate, should be used to prompt action. In this instance, it was by demonstrating the low rate of restraint use among children that politicians were eventually convinced that legislation was needed. The results persuaded them that common sense or education alone had not succeeded. For obvious reasons, most politicians would much prefer that a social good be accomplished by voluntary rather than regulatory means, and, we agree. But when data such as these are presented, there seems little alternative. Parenthetically, it is interesting to recall that the first child restraint law in the US was prompted by the work of Sanders, a pediatrician. He, no doubt, used data of this kind to persuade the legislature to pass this law. Sanders never published an account of how he went about his task, but his example reminds us of what can be accomplished by determination and making good use of available data.

The third thought prompted by the juxtaposition of the classic and the paper by Segui-Gomez et al is a mixed one: on the one hand we should be reminded of how far we have advanced since the Williams study was done—from rates of restraint use in the 1970s of about 20% to well over 80% in many areas today. Nevertheless, the battle is far from won. Not only are there countries where restraint use rates remain much, much lower—in part, no doubt, because they have failed to pass legislation—but even in areas with high use rates, studies continue to show that many preschoolers are improperly restrained. Furthermore, rates among older children are, generally, alarmingly low. And, far too many children are placed in much more dangerous, front seat. Finally, precisely how, if at all, the airbag controversy has affected all this remains to be seen.

The message is simple: much has improved since the elegant study by Williams. But much more remains to be done. To succeed, safety advocates need each other and they need allies. The Insurance Institute for Highway Safety (IIHS) is an exemplary organization that has done much to promote the interests of the industry by sponsoring research. Readers who do not yet receive IIHS reports are urged to request them. Not only do they sponsor high caliber research of the kind conducted by Williams, but they also present the results in a convincing manner. This helps ensure that the key messages are easily understood by those with the power to improve matters.