Biostatistics Research Today is a free monthly online journal that collates and summarizes the latest research about Biostatistics, including details on statistics, uncertainty, probability, modeling.

Accuracy of weight estimation methods for children.

DuBois D, Baldwin S, King WD

Division of Pediatric Emergency Medicine, Department of Pediatrics, University of Alabama at Birmingham, Birmingham, AL 35233, USA. ddubois1@creighton.edu

OBJECTIVES: To evaluate differences in accuracy of 2 weight estimation methods for children when compared with measured weights: the Broselow-Luten tape (patient's height as the predictor) and the devised weight estimation method (DWEM) (patient's height and body habitus as predictors). METHODS: Information was obtained prospectively on a convenience sample of patients presenting through triage on nonconsecutive days at the Children's Hospital Emergency Department. Weight was measured in kilograms, and a measured length or height in centimeters was obtained, as well as 2 independent assessments of body habitus. Weights were then estimated using the Broselow-Luten tape and the DWEM. This study evaluated 4 separate weight classes: less than or equal to 10 kg, 10.1 to 20 kg, 20.1 to 36 kg, and 36.1 kg or more. One hundred children were recruited into each weight class, for a total of 400 children. Comparisons of estimations with measured weights were made using the Pearson correlation coefficient method. Mean percentage errors were calculated for weight estimations by both methods. RESULTS: Both the Broselow-Luten and DWEM weight estimations when compared with measured weights showed statistical correlation (using the Pearson correlation coefficient). However, the Broselow-Luten method had a negative mean percentage error in all weight classes, and the DWEM had a negative mean percentage error in classes greater than 20 kg, indicating an underestimation of weight in those classes. CONCLUSIONS: Although both the Broselow-Luten and DWEM weight estimations show statistical correlation with measured weights, the Broselow-Luten method underestimates weights in all weight classes, and the DWEM underestimates weights in the weight classes greater than 20 kg.

Published 17 April 2007 in Pediatr Emerg Care, 23(4): 227-30.
Full-text of this article is available online (may require subscription).

© 2005-2008 Biostatistics Research Today. All Rights Reserved.