As a multidisciplinary journal we believe it important to document the dimensions of a problem from the perspective of colleagues in other disciplines. In the Special Feature (p 272) in this issue, Chamberlain and Solomon, both lawyers, set to rest the myth that the hardcore drinking driver is the underlying problem in the drunk driving epidemic. As they note, focusing only or mainly on the hardcore, repeat offender tends to marginalise the problem. Yet this is often done and is used by some as an excuse to inhibit the sort of legislation so badly needed,

The authors' findings make it clear that a large percentage of so-called social drinkers have blood alcohol concentrations in the hardcore range, and that infrequent drinkers have a much higher relative risk of a crash than do heavy drinkers with the same concentrations. Their recommendation that we discard value labels such as “social drinkers” and “hardcore drinkers” is an excellent one. If adopted by legislators this would go a long way to improving the grave situation in Canada and elsewhere.

Understandably, most injury research focuses on causes and how to prevent their consequences. The third part of the classic prevention trilogy, tertiary prevention, is too often overlooked. Thus, we welcome papers that remind readers of the consequences of injuries, physical, psychological, and economic, and how these too might be prevented.

One such paper is that of Ward et al (p 297). It examines an old, but still unresolved question—what happens when children or adolescents are exposed to violence? I was surprised at how prevalent such exposure is in the study setting and it was also unexpected to learn that the psychological consequences of this experience differ depending on the type of violence to which the adolescent is exposed. Regardless of type, however, the mental health problems that so often follow are sustained and have wide effects. In light of these findings, there is an urgent need to know more about the effectiveness of violence prevention programs. So far, too few such programs have been evaluated.

Another side of the coin (can there be more than two?) is the report by Hammig et al (p 312) regarding predictors of injury from fighting. Among adolescent males involved in fights in the preceding year, 47% injured others and 18% were themselves injured. The greatest dangers were associated with fights that took place in groups or with strangers or those where a weapon had been used. Here again, solutions are not evident. They may, however, prove to be remarkably similar to those required to reduce bullying or exposure to violence.

Consistent with the theme of this issue's editorial is the contribution by Hemenway et al (p 282) describing US national attitudes concerning gun carrying. This exemplifies the importance of knowing all or many of the ingredients in a large puzzle before one can hope to solve it. The puzzle in this instance is death and injury from handguns, especially in the US. In this study, two surveys three years apart showed that Americans feel less safe as gun carrying increases. The paper reports growing opposition to permissive practices with respect to possession of guns and growing sentiment in support of prohibitions against carrying weapons in public places. Not wishing to oversimplify a complex issue, given the public support this study indicates, it should now be somewhat easier to persuade policy makers, the judiciary, and police to adopt stronger measures.

The report by Steenbergen et al (p 286) is yet another timely reminder that having an efficacious preventive strategy is one thing; ensuring that it is fully implemented, quite another. In this instance, the focus is on a graduated driver licence program to which the barriers reported appear widespread. They range from lack of awareness of some restrictions by parents and drivers, to difficulties with enforcement of the law's provisions because of the views of police officers and judges. Assuming these barriers are found elsewhere, as noted in my editorial, it is evident that to succeed in this arena as in so many others, we need to enlist the help of others. How to do so remains poorly understood. Even heavy handed legislative measures will be less effective than we expect if barriers to their implementation are not removed.

A companion piece to the previous paper (Begg et al; p 292) provides further evidence that graduated driver licensing systems, when they include components such as night time curfews, restriction on the carrying of young passengers, and a low blood alcohol limit, are highly effective. This study uses a pre-post design to conclude that one element in particular—the night time curfew—contributes heavily to reductions in serious crashes in young drivers. Teasing out the effective components of a preventive program is a worthy exercise and should be done more often, if not routinely when such programs are being assessed.

Another facet of the need to better understand each of the steps needed to successfully implement an intervention is exemplified by the paper by Jaye et al (p 302). This report is based on a qualitative study of the views and behaviours of plumbers in New Zealand towards achieving safe hot tap water. Scalds are a growing concern and affect the elderly as well as the very young. Although many scalds are from boiled liquids, a significant proportion come from tap water that is heated more than is needed. The solution appears simple—lower the temperature of heating tanks. But it is evident that unless the views of key links in the chain are taken into account, in this case those of plumbers, trying achieve the goal will have limited success.

We have published many papers addressing economic issues related to injury prevention in part because I still harbour the possibly foolish notion that policy makers are influenced by cost savings, especially when these savings are large. In addition to illustrating the magnitude of savings from a smoke alarm distribution program, the report by Haddix et al (p 276) reminds us that most effective prevention programs combine education with a passive measure, preferably one that is free. In this case that measure is alarms given to families in high risk areas. The total net savings from this approach are such that it is difficult to imagine legislators concerned about reducing costs to their community, quite apart from reducing death and suffering, ignoring this potential bonanza.

Falls are one of the most common injuries involving all age groups, and although many have relatively minor consequences, some can be extremely serious. Apart from falls in the elderly this is a neglected research topic and thus the report by Ellis and Trent is welcome but perplexing (p 316). Do fall injuries differ among races and if so why? The answer to the first question is, yes, at least in California they do, but perhaps not in the direction one might predict. Same level falls resulting in hospitalizations are almost four times higher among whites than among other racial groups studied. They are also more likely to be serious and, not surprisingly, involve a large consumption of hospital resources. What is not at all clear is why this difference exists. The usual culprit, differences in exposure, may well be implicated but the reasons are not obvious. Nor am I convinced that better prevention will be achieved by targeting the high risk group. The jury is still out.

Hendricks and Adekoya's paper (p 307) highlights some unexpected dangers of farm animals. Using a large survey in the US, the results showed that a majority of those injured were farm residents (no surprise). But as a city dweller, I had not anticipated the number that were work related or that nearly half involved children under age 10. I was also surprised to learn that 37% involved horses and 31% cattle—beasts that I thought were relatively safe. Here again, the problem is clear but the solution is not. Readers' suggestions would be most welcome.

One important function of this journal is to help advance the field in every respect, including refining research tools. Thus, this issue includes two important methodological contributions. Based on other publications, it seemed reasonable to assume that a useful supplementary source of data, especially for unusual injuries such as drowning, is newspaper reports. However, the findings of Baullinger et al (p 339) suggests otherwise. After comparing newspaper reports with death certificates and hospital discharge data, they conclude that at least in the case of drowning, reliance on newspaper reports alone would greatly underestimate the size of the problem. Not only would the resulting morbidity and mortality be minimized, but so would several important risk factors. The moral of this story is not to make ready assumptions about data sources no matter what their reputation may be.

The other methodological contribution has a more encouraging bottom line. The report by LeMier, Cummings, and West (p 334) shows that E coded hospital discharge records can be remarkably accurate. In particular, two components of the E code, the mechanism and intent of injury, agree well with other data sources. This is good news indeed. All that remains is to persuade hospitals everywhere to proceed quickly to require E codes for every injury discharge. Here too, the problem is not knowing what to do, but how to get it accomplished. How reluctant hospitals can be persuaded is unclear, but for those hospitals who use the argument that such coding is inaccurate as an excuse for hanging back, this paper should be of great help.

Although case reports are of limited value, there are instances when they serve a vital purpose. In a previous issue we published one such series drawing attention to a new hazard: self strangulation from cloth towel dispensers. This attracted much press attention. In this issue we publish a case series to bring to our readers attention a problem that has been recognised previously but which remains underestimated—the danger to young children from reversing vehicles. These tragic events occur most often in the child's own driveway, but the report by Mayr et al (p 327) suggests that farmyards are also involved. Most of these injuries are serious and likely to have immense psychological effects on the driver.

David and Goel's report (p 329) about hand injuries to sugar cane crushers in India is also illuminating because it reminds us that in many societies—perhaps all—most injuries are still viewed as bad luck or God's will. This may be less surprising than the reluctance of so many of these workers to take protective measures, such as wearing gloves or perceiving the need for safer machines. This complacency echoes the attitude of many workers in western society as well.