In the paper, the authors present a method for determining the distribution class to which a selected random vector with medical parameters as components belongs. The method is based on the statistical significance test. The optimal selection problem for the significance level where the probability of the vector identification error is minimal has been solved. In order to tackle the problem, the authors used the prior information on belonging the vector components to the definite distribution class in which the statistical relationship between the medical parameters was taken into account. The developed mathematical model of patient condition should serve as support of decision-making on further treatment tactics.