The accurate determination of drug solubility in supercritical fluids and its correlation are crucially important for development of supercritical technology in pharmaceutical industries. In this research, a new empirical equation was proposed to correlate solute solubility with temperature, pressure and density of pure supercritical CO2. The accuracy of the proposed model in comparison with six other empirical equations was evaluated using 25 published solubility data sets by calculating the average absolute relative deviation (AARD). Due to these data sets, the mean AARD for the proposed model was 10.64 (+/- 4.18); where, the values for other equations were 13.97 (+/- 6.79), 13.16 (+/- 6.31), 13.33 (+/- 5.36), 14.64 (+/- 13.94), 29.62 (+/- 20.28) and 12.64 (+/- 6.04). To validate proposed model, the solubility of Clobetasol Propionate (CP) and Desoxycorticosterone acetate (DC) as two corticosteroids was measured and modeled at temperatures ranging from 315 to 345 K and pressures from 152 to 305 bar by a static method. The acquired mole fraction results varied in the range of 1 x 10(-6) to 2.1 x 10(-5) for CP and 5 x 10(-5) to 1.03 x 10(-4) for DC. The AARD of proposed model for CP solubilities was 4.95% and for DC was 5.29% that was in reasonable range according to other published solubility data. (C) 2013 Elsevier B.V. All rights reserved.