Many drugs are unlicensed in children and consequently their doses have been scaled down from those used in adults. The objective of this study was to compare the performance of three scaling models in predicting maintenance doses for children from those used in adults. Three scaling models based on body weight (BW), body surface area (BSA) and BW(0.75) were used to predict maintenance doses across the paediatric age band from the equivalent adult doses for 30 different drugs. The predicted doses were compared with those in the British National Formulary for children 2006 (BNFc). Percentage error and mean squared prediction error were used as a measure of precision, and mean prediction error was used as a measure of bias. In the 1-month and 12-month age groups, the different approaches ranked on their bias (least bias first) were BW<BW(0.75)<BSA and on their precision (most precise first) were BW>BW(0.75)>BSA. The BSA and BW(0.75) methods predicted doses up to 2.86-fold higher than the BNFc in the 1-month and 1-year age group. In the 7-year and 12-year age groups, BW(0.75) and BSA performed better than BW for precision and bias, and no predictions were more than 1.8-fold higher than the BNFc. The BW method tended to also under-predict dose across the paediatric age range. Dose scaling should only be used as a last resort for determining a suitable dose in children. No single method was suitable across the entire paediatric age range.
Author(s): Trevor Johnson