Based on the results of previous work, we have proposed two mathematical models to predict the low effectiveness of glucocorticoids in patients hospitalized for treatment of exacerbation of chronic obstructive pulmonary disease (COPD). The aim of this study was to compare their characteristics. In total, 28 COPD patients who were admitted to hospital due to exacerbations were included in this study. Determination of the regression equation, including neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR) and eosinophil count in the blood of examination group of patients showed high diagnostic efficiency (78.6%), sensitivity (81.8%) and specificity (76.5%) of the mathematical model. Another model, which includes blood eosinophil count, PLR and MIF (macrophage migration inhibitory factor) level, had diagnostic efficiency of 85.7%, sensitivity of 81.8%, and specificity of 88.2%. The first model, which includes only parameters of complete blood count, can be calculated after the initial examination by the attending physician. The second model, which requires the determination of blood plasma MIF level by enzyme-linked immunosorbent assay and, accordingly, more time, but has higher efficiency and specificity, should be used after a second examination by the attending physician.