Aim This study aimed to describe the levels of exposure to different forms of prednisolone in children with asthma or asthma-like symptoms, ages six months to seventeen years old, treated with crushed tablets, oro-dispersible tablets, a liquid suspension, or whole tablets. Methods Participants were randomised to received two out of four different formulations on successive days using a single-center, open-label, two-period, cross-over design. Saliva samples were collected to measure prednisolone concentrations, and a population pharmacokinetic model was used to analyse the data. The bioequivalence of the test drug to the whole tablet was determined using the 90% confidence interval of the ratios of area under the curve (AUC) and maximum concentration (Cmax). Results This study enrolled 41 children, with a mean age of 4.9 years ± 3.7 and a mean weight of 21.8 kg ± 10.9; 61% were boys. The pharmacokinetic data were best described by a two-compartment model using plasma concentrations calculated from saliva. The population mean clearance was 317±156 ml/min/70kg, with a mean half-life of 5.3 ± 3.2 hours and a volume of distribution of 141 L/70kg. The liquid suspension demonstrated bioequivalence to the control (whole tablets) in terms of AUC. However, none of the tested formulations were bioequivalent regarding to Cmax. Conclusion The tested formulations did not exhibit bioequivalence (AUC and Cmax) when compared to the whole tablet. Using different prednisolone formulations interchangeably may be challenging, especially in a paediatric population where inter-individual and residual variability in the pharmacokinetics seemed to have significant impact on exposure.