The objectives of this paper were to choose a psychometric scale that meets requirements for "comparison standard" in elaboration of new methods for psychometric assessment of late depression and to evaluate distortion of results for standard scale used in elderly patients. Based on the literature analysis, authors chosed the Montgomery-Asberg Depression Rating Scale (MADRS) as an etalon. The comparative analysis of MADRS scores and results of psychopathological evaluation revealed distortions of MADRS scores related to elder age. Qualitative and quantitative distortions were assessed for every MADRS item and for the total score. The analysis of separate components revealed that 7 out of 10 MADRS items were significantly changed and correlated with the age of patients while the total score did not change. This allows to use MADRS total score in further studies without any corrections. Application of the MADRS as a comparative scale in the symptomatic analysis of late depression needs the use of correction coefficients.