Both new guidelines concerning elemental impurity limit concentrations and novel analytical procedures for elemental analysis of pharmaceutical substances are being adopted by the United States Pharmacopeia (USP). This paper outlines the optimization of two sample preparation procedures for organic pharmaceutical excipient products, one relying on direct dissolution in an aqueous medium and another on microwave-assisted acid digestion. Optimization particularly involved the stabilization and reliable determination of Hg, Pd and Os. Analyte concentrations and spike recoveries were determined in two organic pharmaceutical excipients using a quadrupole ICP-MS instrument equipped with a collision–reaction cell. Aqueous dissolution was achieved with 0.009mMKBrO3 in 1% (v/v) HNO3 and 1% (v/v) HCl and microwave-assisted acid digestion was performed with aqua regia. Polyatomic interferences originally hampering the accurate determination of 51V, 52Cr, 53Cr, 55Mn, 56Fe and 75As were eliminated through introduction of He/H2 collision–reaction gas at a flow rate of 4.5–5.1 ml min-1 and applying a kinetic energy barrier of 3 V between the hexapole collision–reaction cell and the quadrupole analyzer. Limits of detection were more than a factor of ten lower than the proposed limit concentrations for parenteral administration and the impurity concentrations in the products tested never exceeded the limit concentrations defined by USP. Elemental spike recoveries ranged between 93.8 and 109.9% when standard addition was used for calibration, with relative standard deviations (RSDs) < 12.7%. It is expected that the proposed procedures can be directly implemented for routine impurity analysis of a broad spectrum of pharmaceutical substances.