Abstract
The pediatrician routinely collects data on the occurrence of accidents. Can these data be used to predict later accidents? Are there other characteristics of children and families that improve prediction? We used cohort data on 10,394 British children collected at ages five and ten to assess the stability of accidents requiring medical attention between birth and five and accidents between five and ten reported by the mother. Pearson product-moment correlations were used to assess stability. Other predictors at age five of later accidents were identified through a stepwise multiple regression performed on half the sample and replicated on the second half. The correlation between accidents in the two time periods was .18. Of the children with three or more accidents reported at five, 15% had three or more accidents reported at ten, compared with 2.4% of those with no early accidents. Accidents before five were the best predictors of accidents to age ten followed by male sex, aggressive behavior, young maternal age, many older and few younger children. Family characteristics not associated once the other variables were included were social class, crowding, maternal distress, marital status, and overactive behavior at age five. While only 7% of the variance was explained by early accidents and the other characteristics, the relative risk of three or more accidents from five to ten was 6.3 for those with three or more preschool accidents compared with those with no early accidents.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Bijur, P., Stewart-Brown, S., Goldlng, J. et al. 513 The stability of childhood accidents: Can accidents in preschoolers predict accidents in school-age children?. Pediatr Res 19, 196 (1985). https://doi.org/10.1203/00006450-198504000-00543
Issue Date:
DOI: https://doi.org/10.1203/00006450-198504000-00543