Falls in hospitalized children.

PURPOSE

To identify risk factors in hospitalized children using two validated adult fall risk scales.

METHODS

For each fall patient identified, a non-fall patient was matched for chronological age and year of hospitalization. The retrospective chart review coded the presence or absence of risk factors and completed both the Morse Fall Scale and Hendrich II Fall Risk Model.

FINDINGS

Overall, a significant difference was detected between pediatric fallers and non-fallers on the Morse Fall Scale. The Morse Fall Scale's sensitivity, specificity and false positive rate suggested that the scale does little better than chance at detecting true fallers and may inaccurately identify non-fallers as high risk.

CONCLUSIONS

Episodes of disorientation and fall history were the best predictors of pediatric falls for this sample.