Parental supervision: a popular myth.

There must be few categories of childhood injury for which parental supervision, or the lack of it, has not been invoked as a contributory factor. Since 1990 alone, lack ofparental supervision has been proposed as a cause of road traffic injury,' electrical burn injury,2 and infant walker related injury.3 In most cases the identification of the aetiological role of parental supervision is intuitive rather than empirical. Very few ofthe reports that recommend greater parental supervision are controlled studies and it is not that unusual for injury researchers to make inferences about the value of parental supervision without even measuring it. The question of measurement is an important one and the lack of validated methods for measuring parental supervision should raise concern about the utility of the concept. As the renowned physicist Lord Kelvin observed 'until you have measured it, you don't know what you are talking about'. The aim of this paper is to examine just what is being talked about when parental supervision is invoked as a cause ofchildhood injury. Is there any evidence that supervision by parents is more effective than supervision by other responsible persons? If not, are calls for greater parental supervision just an example of semantic sloppiness, or are they covert statements about values? Data are presented to support the hypothesis that greater reductions in child injury rates will be achieved by strategies that result in less (rather than more) parental supervision.