Background
The increasing prevalence of anxiety disorders underscores the critical importance of effective assessment and management strategies. While established questionnaires like the Hamilton Anxiety Rating Scale (HARS) and the Beck Anxiety Inventory (BAI) are widely used, there remains a need for instruments that explore the nuanced, qualitative features of anxiety, which are essential for personalized treatment approaches.

Methods
This study presents findings based on the Brief Anxiety Structure Questionnaire (BASQ), which is designed to evaluate behavioral manifestations, cognitive aspects, and personality traits associated with generalized anxiety disorder (GAD). Data from a Phase III clinical trial of the anxiolytic Aviandr (maritupirdine) were analyzed using machine learning techniques to develop predictive models and construct an “ideal patient profile”.

Results
Among the tested algorithms of machine learning, the decision tree model demonstrated the highest accuracy in identifying the most influential BASQ questions for therapy selection. The BASQ questionnaire revealed qualitative aspects of anxiety and personality traits, providing a deeper understanding of the structure of anxiety and supporting more personalized treatment strategies. Specific questions most strongly correlated with the effectiveness of Aviandr treatment were also identified.

Conclusion
The findings from this study suggest that integrating qualitative parameters into clinical assessment may optimize therapy for anxiety disorders. Future research will focus on further elucidating the relationship between patient anxiety characteristics and treatment effectiveness.