Background. Currently, there is a growing number of diagnosed mental disorders worldwide. The specialist conducting the diagnosis works under time constraints as determined by regulations per patient. Given the variability of symptoms presented in various mental disorders, as well as their potential comorbidity, establishing a diagnosis on the ground of patient-reported complaints is problematic.

Objective. To analyze the challenges of differential diagnosis of mental disorders (on examples of affective disorders, eating disorders, and schizophrenia) based on the primarily available affective symptoms.

Methods. Searching for publications in databases (PubMed, eLIBRARY, Google Scholar) was conducted using the keywords "differential diagnosis", "comorbidity", "affective symptoms", and "symptom specificity". A theoretical review and analysis of the literature on the role and significance of affective symptoms in the differential diagnosis of mental disorders, their prevalence in the population, and the approaches and diagnostic tools used.

Results. The analysis of sources allowed to summarize data on the diagnosis of various mental disorders based on patient complaints and affective symptoms. Existing studies show that a significant part of complaints and affective symptoms are not specific to a particular mental disorder and cannot be classified without additional diagnostic information. The analysis highlights the value of differential diagnostics and the limitations associated with its use. Promising approaches to the diagnosis of mental disorders are discussed.

Conclusions. The complex nature of the diagnostic process requires reliance on both anamnestic data and the analysis of psychometric indicators as well as the causal relationships between them. The use of network analysis of the relationships between symptoms may have the potential to improve current diagnostic models and categories. Transdiagnostic factors associated with the manifestation of mental disorder symptoms require further research. The potential for integrating transdiagnostic approaches into modern diagnostic models appears promising; however, finding common ground between them requires further effort from the professional community.