Background: According to a concept of continuum of psychosis, it is postulated that schizophrenia, depressive disorder and bipolar disorder represent different clinical presentations of a common pathophysiological mechanism. Recent studies have proven increased serum concentrations of inflammatory markers in patients with schizophrenia, depressive disorder and bipolar disorder. Clinical presentation of given disorders links them to an activation of immunological processes as well. Up to our best knowledge, comprehensive study of all mentioned psychiatric conditions in light of cytokine hypothesis has not been published yet in order to assess whether inflammation could have been a common pathophysiological mechanism in terms of continuum of psychosis. The aim of this research was to investigate serum concentrations of IL-6, TNF-α and CRP in patients with depressive disorder, schizophrenia and bipolar disorder, and to compare given concentrations with serum concentrations of IL-6, TNF-α and CRP in a control group. The next step was to investigate the relationship of serum concentrations of inflammatory markers with severity of symptoms in patients with depression, schizophrenia and bipolar disorder. Methods: Study included a total of 245 subjects, aged 18 to 60, divided into four study groups: 51 healthy subjects in control group, 76 subjects diagnosed with schizophrenia, 84 subjects diagnosed with depressive disorder, and 34 subjects diagnosed with bipolar disorder. Diagnoses and severity of symptoms were assessed with psychiatric diagnostic rating scales MINI, PANSS, HAMD and YMRS. IL-6 was determined by standardized ELISA procedure, TNF-α was determined by immunoenzyme method, and CRP was determined by immunoturbidimetric method. Results: There is a statistically proven difference of serum concentrations of IL-6 between patients with depressive disorder and patients with bipolar disorder. Difference of serum concentrations of TNF-α between all three investigated groups of patients and control group is statistically proven. Difference of serum concentration of CRP between patients with depressive disorder and control group is statistically proven as well. Serum concentration of TNF-α is in correlation with severity of symptoms of bipolar disorder, while in case of severity of symptoms of schizophrenia and depressive disorder we found no correlation. Moreover, we observed no correlation between serum concentrations of IL-6 and CRP with severity of symptoms of schizophrenia, depressive disorder or bipolar disorder. Conclusion: According to our results, the relationship between serum concentrations of IL-6, TNF-α and CRP and clinical presentation of schizophrenia, depressive disorder and bipolar disorder did not prove continuum hypothesis. TNF-α could be a potential marker for investigated psychiatric disorders which is a subject for further investigation.