Relevance. Symptom of bleeding in elderly age has a number of features of its manifestation, not fully associated with the severity of gingivitis and periodontitis. It is associated withpolymorbidity, polypragmasy, age-dependent involutional processes, as well as with immune system aging.Purpose. To study the influence of confounders on the interpretation of the gingival bleeding index in elderly patients with periodontitis.Materials and methods. The clinical and laboratory examination of 64 elderly patients with chronic generalized periodontitis, divided into 2 clinical groups, was conducted. The conventional treatment was administered in both patient groups. The patients in the second cohort, along with other medications, received Polyoxidonium.Each clinical group was divided into two subgroups: “a” and “b” (depending on the NSIADintake). The comparison group was also formed.It consisted of 25 patients of middle age with a similar degree of periodontitis. To assess the bleeding symptom, the Mühlemann-Cowell (1975) bleeding index was determined, as well as its reduction (%). The study of local immunity included gingival fluid sampling andestimation of IL-1β and IL-10 concentration (pg / ml). Clinical and laboratory examination of patients was carried out within the time constraints: before the start of treatment, after 7, 14, 21, and 30 days.Results. During the initial examination and statistical analysis of the bleeding index mean values in elderly patients, higher values of the coefficient of variation were found (25.07%) than in the comparison group (16.24%).It confirmed the influence of confounders over this indicator.Moreover, before treatment, in both groups, significant differences between subgroups remained (which verified the effect of non-steroidal anti-inflammatory drugs), which disappeared in the subsequent stages of the examination. In further periods of observation, a decrease in the bleeding index was found, as well as the concentration of cytokines, and in the 2nd group, these indicators were statistically significantly lower than in the 1st (p < 0.05).Conclusion. The study showed a more significanthaemostatic effect of immunomodulation compared with conventional treatment. When analyzing bleeding index in elderly patients, confounders must be considered, otherwise, the findings, which do not fully correspond to the level of inflammation in the marginal periodontium, will be incorrectly interpreted.