[Creation of a statistical model for prediction of Hodgkin's lymphoma in young adults]

The aim of this study was to create a multi-dimensional correlation matrix of determinants giving information on the degree of influence on survival of young adults with Hodgkin’s lymphoma each variable as well as the effect of the interaction of these variables with each other. 87 patients with Hodgkin’s lymphoma at the age of 19 to 29 years (mean age 24 ± 4 years) were included in the study. Multiple matrix containing the coefficients of correlation of survival and correlation coefficients of 35 analyzed factors was of 5 significant determinants (volume of tumor lesion, stage IV of disease, E-damage, accelerated erythrocyte sedimentation rate, leukocytosis). Construction of the correlation matrix in order to select factors to be included in the equation, which contained the pair correlation coefficients of overall survival and each of factors, showed that only the volume of tumor lesion (correlation coefficient 0.2570, p = 0.026) influences on sign-result. Multiple regression equation is represented as follows ȳ=0,166667-0,227273x[volume]; R2=0,0643674507. This equation allows for given values of the factor «volume of tumor lesion” to have theoretical values of resultant sign (survival), substituting the actual values of the factor in it.