Oxycodone hydrochloride (HCl) extended release (ER) tablet is an abuse-deterrent formulation that uses a physical barrier to make it more difficult to crush tablets prior to abuse via various routes. A previously conducted in vivo pharmacokinetics (PK) study showed that particle size exhibited significant effects on PK. Here, a computational modeling study using a novel combined computational fluid dynamics and physiologically based PK model was applied to better understand the mechanisms that produce differences in PK according to particle size and formulation type for nasally insufflated oxycodone HCl immediate release (IR) and ER tablets. Dissolution data were collected using a United States Pharmacopeia (USP) Apparatus 4 to support model parameterization. The in vitro dissolution data showed that the number of powder layers in the bead-based system impacted the observed dissolution pattern for the finely milled (106-500 μm) ER formulations, but not the finely milled IR (106-500 μm) or coarsely milled ER (500-1000 μm) formulations. The model was validated via comparison of PK predictions with available in vivo PK data for finely milled (106-500 μm) IR and ER formulations in the 30 mg strength, a coarsely milled (500-1000 μm) ER formulation in the 30 mg strength, and a finely milled ER formulation in the 80 mg strength. Model predictions showed relative differences no greater than 3.3 % for maximum plasma concentration (C