Frequency of childhood injuries: first results of the Boulogne-Billancourtregistry (France)

Background: Boulogne-Billancourt database is in France the only permanent a nd specific childhood injury surveillance system. Integrated in an safe com munity program designed by the World Health Organization (WHO), the aims of a focal database are to provide precise information intended to implement prevention actions. Method: All children less than sixteen years old living in Boulogne-Billanc ourt, victims of injuries occurring in that town, and requiring hospital ca re, are targeted. Results: Two thousand five hundred and eighty accidents have been computed between 1(st) January 1998 and 31 December 1999 i.e. an average annual inci dence of 79.7/1000 children. As in other similar databases, domestic accide nts are prevalent in the average (40%), even though after five years old sc hool injuries become the most frequent ones (36%). Falls are the main mecha nism involved in injury at any age (44 to 54%). Distribution of other mecha nisms varies with respect to age: foreign body prevalence is maximum betwee n one and four (2.5%), struck-collisions (38%) between ten and fifteen year s old. A significant increase of falls involving roller-skates or skateboar ds has been observed between 1998 and 1999 (3.8% versus 1.1%, p<0.001). Pre dominanting injury causes are contusions (38%), mainly head trauma (10%), o pen wounds (16%), fractures (10%), and sprains (6%), Two hundred and twenty four injuried children have been admitted in 1998 205 in 1999, i.e. respec tively 13.8/1000 and 12.7/1000 children dwelling in Boulogne. Overall recur rence rare is 32%, reaching 47% after eleven years old. A significant relat ionship has been observed between injury recurrence and problems related to education (odds ratio: 2.95; 95% CI: 1.75-5.00), atypical parenting situat ion (odds ratio: 1.70; 95% CI: 1.20-2.30), and family discord (odds ratio: 1.62; 95% CI: 1.05-2.50). Household accidents are the most avoidable ones i n parents' opinion. This is not the case for road-traffic accidents, given evidence of the lack of means of protection observed in bicycle injuries. Conclusion: In spite of methodological issues such as control of exhaustivi ty of data, or difficulties to maintain a long-term surveillance sytem, thi s project worth carrying on with regard to provided information and usefuln ess for prevention. Other similar community experiences should be implement ed with the aim of setting up a national-wide surveillance system, based on an homogenous data collection.