Background According to the Advanced Paediatric Life Support (APLS) guidelines, the weight of children between the ages of 1 and 10 years can be safely estimated using the formula: weight=2 × (age +4). Although several alternative methods have been proposed to be more accurate, they are not widely used. In this study we compare the accuracy of the standard APLS guideline to alternative age based methods of weight estimation.

Methods Data was obtained retrospectively from the recorded weights of 140 314 children aged between 0 and 16 years attending a large UK Paediatric Emergency Department. The accuracy of the APLS guideline and alternative formulae were audited by comparing predictions with measured weight. For each method the percentage of study population predicted to measured weight +/−10% was derived along with mean, absolute, root mean squared and percentage error of weight estimation.

Results Between the ages of 1 and 10 years, the APLS method correctly predicted 35% of patients within +/−10% of measured weight. The Best Guess (BG) method formula (2×age +10) estimated 44% of weights to +/−10%. Under 1 year the BG infant formula ((age in months+9)/2) correctly predicted 45% of patients +/−10% (table 1).

Conclusion Although all age based methods of weight estimation are inaccurate, this study suggests the most widely used, APLS, is the least accurate. For optimal resuscitation of children, clearly, the most accurate formula is desirable. The results of our study support findings of previous studies suggesting a change in current practice is required. We suggest adoption of a modified form of the Australian BG method using the formula: weight=2 × age+10 between the ages of 1 to 10 years.